Effect of Harmonic Current

i

FOREWORDS Dean of Faculty of Engineering, University of Indonesia

The Quality in Research (QIR) Conference is the annual event organized by the Faculty of Engineering, University of Indonesia. Since started in 1998, it has become an excellent forum of discussion for all researchers from research institutions and universities all over the nation of Indonesia. The 1st and 6th QIR Conferences had been successfully organized as the high quality national conferences, and starting from the 7th QIR conference, has been organized to invite international research papers.

The 10th Quality in Research International Conference having a theme of “Research for future better life” is to provide an international forum for exchange of the knowledge, information, experience and result as well as the review of progress and discussion on the state of the art and the future trend various issues and the developments in the multi-fields of scientific and technology. The main purposes of this conference are to provide a forum for free discussion of new ideas, development and applications, including techniques and methods to stimulate and inspire pioneering work, to provide opportunities for students and young engineers to meet their experienced peer and to provide a meeting that will enforce progress, stimulate growth and advance the state of knowledge in the multi-fields of science and technology.

We would like to express our heartiest to thank to all authors and participants for their active participations in the 10th on Quality in Research (QIR) International Conference 2007, and also to all the paper reviewers, member of the technical committees, and member of the organizing committees, for their support to the success of this conference. Last but not the least; we would also like to invite all participants to the next on Quality in Research (QIR) Conference.

Faculty of Engineering, University of Indonesia Dean, Prof. Ir. Rinaldy Dalimi, Ph.D

ii

FOREWORDS

Chairman of 10th International Conference on QIR 2007

The 10th Quality in Research International Conference will provide an international forum for exchange the knowledge, information, experience and recent researches of various fields. With a strong support and presentations from academic, industry and entrepreneurs, the conference will provide an ideal platform to learn various fields and understand technological trends in the region.

The 10th Quality in Research (QIR) International Conference has a theme of “Research for Better Future Life” being the third time to go internationally, has invited limited papers from other nations such as Korea and Malaysia. The conference is organized in parallel sessions focusing on the 8 (eight) research areas such that many researchers and peer groups may focus their discussion on the relevant topics. All submitted papers had been reviewed by the technical committees and had been arranged into 8 (eight) sub-themes according to the following fields:

‐ Energy, Process and Environmental Engineering and Management: Energy and environmental issues, combustion technology, fluids mechanics and thermal fluid machinery, thermodynamics and heat transfer, geotechnical and environmental engineering, etc.

‐ Industrial, Manufacturing, Material Engineering, and Management: Production Engineering, Supply Chain Management, Innovation System, Maintenance System, Quality Management System, Human Factors Engineering, Organizational System, Fabrication and Industrial Automation, Manufacturing System: Control Management and Information Technology, etc

‐ Biomaterial, Biomedical Engineering and Biotechnology: Biomedical numerical modeling, Biomaterial, Biosensor, Biocompatibility, Biomechanics, Biotechnology, Biomedical Instrumentation, Biomedical Imaging

‐ Design and Infrastructure Engineering and Management: Product design and development, composite: Materials and applications, structural dynamics, mechanics of materials, Construction Management, Public Infrastructures and Services, Structural Engineering, etc

‐ Special Session on Electronics Engineering ‐ Information and Computation Technology ‐ Sustainable Architecture ‐ Nanomaterials and Nanotechnology: Nano structured material, Nanotechnology,

Nanocomposite, MEMS, Self Assembled Monolayer, Thin Film, etc The main purposes of this conference are to provide a forum for free discussion of new ideas, development and applications, including techniques and methods to stimulate and inspire pioneering work, to provide Opportunities for students and young engineers to meet their experienced peer and to provide opportunities for students and young engineers to meet their experienced peer and to provide a meeting that will enforce progress, stimulate growth and advance the state of knowledge in the multi-fields of science and technology. Depok, 4 December 2007 The Organizing Committee, Chairman,

iii

Gunawan Wibisono, Ph.D

Steering Committee

1. Prof. Dr-Ing. Axel Hunger, Universitaet, Duisburg-Essen, Germany 2. Prof. Dr. Carlo Morandi, Universida Degli Studi de Parma, Italy 3. Prof. Dr. Iwao Sasase, KEIO University, Japan 4. Prof. Kim Kyoo-ho, Yeungnam University, Korea 5. Prof. Dr. Ir. Irwan Katili, University of Indonesia 6. Prof. Dr. Ir. Bambang Suryawan, MT, University of Indonesia 7. Prof. Dr. Ir. Dadang Gunawan, M.Eng, University of Indonesia 8. Prof. Dr. Ir. Johny W Soedarsono, DEA, University of Indonesia 9. Prof. Ir. Gunawan Tjahjono, M.Arch, Ph.D, University of Indonesia 10. Prof. Dr. Ir. M. Nasikin, University of Indonesia 11. Isti Surjandari, Ph.D, University of Indonesia 12. Prof. Dr. Ir. Budi Susilo Soepandji, University of Indonesia 13. Prof. Dr. Ir. Sutanto Soehodho, University of Indonesia 14. Prof. Dr. Ir. Sulistyoweny Widanarko, Dilp. SE. MPH, University of Indonesia 15. Prof. Dr. Ir. I Made Kartika Dipl. Ing, University of Indonesia 16. Prof. Dr. Ir. Tresna P. Soemardi, University of Indonesia 17. Prof. Dr. Ir. Sardy, M.Eng, M.Sc, University of Indonesia 18. Prof. Dr. Ir. Bagio Budiardjo M.Sc, University of Indonesia 19. Prof. Dr. Ir. Djoko Hartanto, M.Sc, University of Indonesia 20. Prof. Dr. Ir. Eddy Siradj, M.Eng, University of Indonesia 21. Dr. Ir. Kemas Ridwan K, University of Indonesia 22. Prof. Dr. Widodo Wahyu P, DEA, University of Indonesia 23. Ir. Boy Nurtjahyo M.,MSIE, University of Indonesia 24. Dr. Ir. Dedi Prihadi DEA, University of Indonesia 25. Ir. Hendri D.S. Budiono, M.Eng, University of Indonesia 26. Dr. Ir. Sigit Pranowo Hadiwardoyo, DEA, University of Indonesia 27. Dr. Ir. Herr Soeryantono, University of Indonesia 2288.. Prof. Rinaldy Dalimi, Ph.D, University of Indonesia

Chairman of the Conference Ir. Gunawan Wibisono, M.Sc, Ph.D Technical Committee

1. Ir. Gunawan Wibisono, M.Sc, Ph.D 2. Dr. Yosia Irwan 3. Dr. Engkos Kosasih 4. Purnomo Sidi P, Ph.D 5. Abdul Muis, Ph.D 6. Badrul Munir, Ph.D 7. Tania Surya Utami, MT 8. Ir. Beatrianis, M.Si

Foreword From The Dean of Faculty Engineering, UIForeword From Chairman of 10th International Conference on QIR 2007The Committee of 10th International Conference on QIR 2007

Paper No. Title and Name of Author(s)

EPE-01 The Effect of Harmonics Current to Performance of The Over Current Relay

by: Syafrudin and Dahaman

EPE-02 Effect of The Harmonic Current Component to Active Power Losses on Power Transformer

by: Syafrudin and Dahaman I

EPE-03 Optimize placing of passive filter and reduce harmonic in power system, Case study at PT ISPAT INDOby: Iwa Garniwa, Budi S, Adrianto

EPE-04Convergence Approximation Of ABR Formulation For Josephson’s Tunneling in UO2 Chain Reaction at 45.7 MW Candu Nuclear Reactorby: Moh. Hardiyanto

EPE-05 Energy Conservation on Steel Industri By Increasing Efficiency and Electric Power Quality

by: Iwa Garniwa, Budi S, Fauzan H JEPE-06 Energy Conservation on Hotel Building (Case Study in Bandung)

by: Rudy S and Aji Nur W

EPE-07 Indonesia Energy Revolution Scenario and Projection using MESAP/Planet Simulation Model

by: Rinaldy D, Bayu I, and Sven TeskeEPE-08 Ocean wave power plant with air pressure

by: Massus Subekti and Vivian K LEPE-09 The Time for Indonesia to Optimize Geothermal Energy

by: Massus Subekti

EPE-10 Incorporating energy Commodity Price Volatility in economic Cost of Supply Analysis for Electricity Expansion Planby: Emil E. Dardak

EPE-11 Effects of Chimney Depth and Downjet Height in A Coal briquette Stove on the Co Emission and Ignition Timeby: Dijan S., Yulianto S.N., and Dian N. Kusuma

EPE-12 Scale up Reactor for Bio-gasoline Production from Crude Palm Oilby: Bambang HS., Anondho W., and M. Nasikin

EPE-13 Flame Lift-up on A Bunsen Burner; A Preliminary Studyby: C Prapti Mahandari and I Made Kartika D

EPE-14 Study of Bio-Diesel of Coconut and Corn Ethyl Ester Use By processing With The Processor Series Type in Diesel Engine Performanceby: B. Sugiarto, Sanggul H. Siregar, and Yanuar C.

EPE-15 On The Measurement of Smoke Production rate of Tropical woodby: Yulianto S. N.

EPE-16 Catalytic Ozonation of Endosulfan in Water With Activated carbonby: Enjarlis, Setijo Bismo, Slamet, and Roekmijati WS

TABLE OF CONTENTS

EPE-17 Hydrocarbon Selectivity of Fisher Trosch Synthesis from syngs with ratio of H2 to Co=1.0over Co-Fe/AI2O3 bimetalic catalysts: effect of Fe:Co ratioby: Dewi Tristantini and Borje Gevert

EPE-18 The deactivation behaviour of HZSM-5 catalys at various temperature of acetone conversion to mono-aromatic hydrocarbonby: Setiadi, Prilly F J, and Toshinori K

EPE-19 CO2 Absorpation Through Hollow Fiber Membrane Gas-Liquid Contactorsby: Sutrasno K

EPE-20 Enzymmatic and acid hydrolysis of bagasse for ethanol production by simultanious sacharificationand fermentationby: Samsuri, Gozan, Nasikin, and Prasetya

EPE-22 Biogrease using modified palm oil as base oil and thickener lithium soapby: Sukirno, M. Nasikin, B. Heru, Rizqon F, Marius, Dizzi

EPE-21 A general Application of A Direct Method For Multivariable MPC Control Strategy in Chemical Processby: A. Ahmad, A. Wahid

EPE-23 Model for Oxidation and Combustion of Ethanol in Wide range of Pressures, Temperatures and Equivalence Ratiosby: Yuswan Muharam and Soraya S

EPE-24 Water and air Circulation System in Trimaran Fishing Veseel Live fish holdsby: Sunaryo, Anita Muslih

EPE-25 Local Protein Structure Comparisonby: Ford Lumban Gaol and Belawati W

EPE-26 Antibacterial activity Analysis of Dellenia indiaca's Kernel Extract for Escherichia coli: A Preliminary Studyby: Rita Arbianti, Tania S U, and Ifa P

EPE-27 Antioxidant Activity Analysis for Polar Extract of Dillena Indica'S Kernelby: Tania Surya Utami, rit Arbianti, Lina FR

EPE-28 Understanding The Metaspaces of Street Vendors in The Cities: Temporality, Strategies and Tacticsby: Yandi Andri, Paramita A.

EPE-29 Bringing Sustainability Into The Curriculum of Architectural design Studioby: Paramita A and Yandi Andri.

EPE-30 LED Technology in Architectural Ligthing of Building'S Façadeby: Siti Hanjarinto, Gregorius A.G. A.

EPE-31 Form in The new Architecture Perspectiveby: Yandi Andri Yatmo and Kristanti Dewi Paramita

EPE-32 Utilization Effectiveness of Jakarta Municipal Asset (Case Study: East Jakarta Municipal Youth Center)by: Azrar Hadi and Ii Karunia

EPE-33 Water Quality Index Indicator of Water Pullutin. A Case Study in west Tarum Channel as Sources of Water Supplyby: Djoko M. Hartono, Sulistyoweni W., Dwita Sutjiningsih M.

EPE-34 Non Linear Dynamic Identification of Waste Heat Boiler Unit Based on Adaptive Neuro Fuzzy Inference System (ANFIS)by: Yuliati

EPE-35 Load Pattern and Realibility Assessment of Power Generation Systemby: Suparman, Zuhal, and Rinaldy D.

ISSN: 1411-1284

Proceeding 10th Int’l QIR 4-6 Dec 2007 EPE-01 1/4

The Effect of Harmonics Current to Performance of The Over Current Relay

Syafrudin(1)(2) and Dahaman Ishak(1)

(1) School of Electric and Electronic Engineering – Universiti Sains Malaysia

(2) Fac. of Engineering – Electric Department, University of Tanjungpura, Jl. A. Yani, Pontianak email : [email protected] , [email protected] and [email protected]

Abstract- The over current relay protection on the power system is a device that can measure and responds to abnormal condition and control circuit breaker to isolate the faulty section of the system. The over current relays are designed to operate with sinusoidal current waveform. Now, the current waveform on the power system have been distorted non-sinusoidal and lot of containing harmonic current component, caused by non-liner loads. For harmonic current condition of system the performances of that over current relay to be abnormal. In this paper, the effect of current harmonic distortion percentage to over current relay protection performances is discussed. In this experiment used an inverse time over current relay type of induction disc relay. The non-sinusoidal load currents that consist of different harmonic spectra were applied to inverse time over current relay and these nonlinear-load currents were recorded by a power analyzer instrument “FLUKE 42”. According to the experiment results, the pick up current and the operating time of the inverse time over current relay increase proportionally to the total harmonic distortion (THD) value of the non-sinusoidal current. It is concluded that, this type of relay cannot protect the system reliably due to the harmonic components of current. Keywords- Non-linear load, Harmonic current, inverse time over current relay.

I. INTRODUCTION istorted current waveform content harmonics current

are introduced into power distribution system due to non-linear loads or system components presenting non-linear behavior. The increasing use highly non-linear electric equipment (non-linear loads), the advancement of semiconductor technology and the use of more power electronic equipment such as converters for adjustable speed drives (ASD), switch mode power supplies, power semiconductor controller, computers, photocopy machines, TV set and etc have caused an increasing concern over the generation of harmonic current and its effects on the

power system. Harmful effect of the harmonic current on power distribution system like increasing losses, interferences to communication system, induce abnormal noise, overheat and burning in various devices[1].

Harmonic current on power distribution system also can cause some problems in over current protection system that provided complete protection and reliability for the power system. Most of the relay manufacturers design their relays for pure sinusoidal current waveform and voltages[2]. The well known characteristics of relays are not valid under distorted waveform. Each harmonic frequency component could produce an independent and cumulative effect, causing the pick up value of the relay to change depending on the magnitudes of harmonic components. Therefore, the relays can not protect the line or transformer reliably when harmonics are involved. In this paper is purposed to determining of the inverse time over current relay performances for non-linear load current that containing of different total harmonic current distortion. In this experiment, non-sinusoidal wave current that containing harmonic were applied to the inverse time over current relay and these currents, pick up current and the operating time were recorded by a power analyzer instrument.

II. PERFORMANCE OF THE INVERSE TIME OVER CURRENT RELAY

Inverse time over current relay is used for over current and short circuit protection of line or transformer in power distribution systems. In the Fig.1 shown the construction an electromagnetic inverse time over current relay type induction disc unit. All of the operating energy for the relay is applied to the primary pole coil. The current flowing in the primary coil will produce a primary flux ФU. The primary flux will induce an emf in the secondary coil. The emf in the secondary coil will cause a current to flow through the winding of the lower magnet. The secondary current lags behind the secondary emf. This current creates a magnetic field in the lower magnet ФL. Both ФL anf ФU will act on the induction disc and cause it to rotate.

D

ISSN: 1411-1284


Fig 1. (a) construction of induction over current relay, (b) induction disc and contacts (c) moving contact position.

The torque created by the magnetic field is counteracted by the tension of a spiral spring. When the turning torque overcomes the force of the spring, the relay will operate. It is determines the minimum operating current of relay. The standard current-time characteristic curve of inverse time over current relay is shown in Fig.2. Where the extremely inverse relay saturates at four multiples of pickup, very inverse at two multiples and the moderatetly inverse at pickup (IEEE standard inverse time, 1996:10)[3].

Fig.2 Time-Current curve of inverse time over current relay The non-sinusoidal current or distorted current consist of any numbers harmonic current component. Each frequency is order multiple of the fundamental frequency. If this current flow into the relays, than each current harmonic component contributing increasing frequency. Increasing the frequency of the input current results in little change in the current that is produce in the lag coil circuit. However, the flux in this coil will decrease in inverse proportion to the frequency increase. Similarly, the flux in the other coil decreases because of the lowered magnetomotive force across it. BY decreasing the magnetizing current for frequency increment and constant lag coil circuit,

the effect is the main coil and the secondary coil fluxes draw closer in phase. This slow down the disc rotation, cause the pick up to increase and ultimately causes the efficiency of the electromagnet to deteriorate to the point of non-operation. Harmonics current have serious effect on inverse time over current relay pic up current value and operation time.

III. EXPERIMENTAL RESULTS

The experiment system configuration as illustrated in Fig.3.

Fig.3 Experiment system configuration Current of non-linear loads was isolated by a current transformer (CT) that has a transformation rate 5A to 5A. All the harmonic amplitude levels and the total harmonic distortion (THD) recorded by Fluke power analyzer 42. The operating time of the inverse time over current relay is inversely proportional to the current. Technical information of this relay is given in Table I below:

Table I. Technical Information of Relay Frequency 50 Hz Accuracy of pick up current ± 5% Resetting current as % of setting 92% Contact stop time (after de-energizing)

0.08s

Consumption at rated current 5VA Over current adjustment ranges 1.00A-1.25A-1.50A-

2.00A-2.50A-3.00A-4.00A

As non-linear load for generate harmonic currents is used a single phase full wave rectifier included with large variable capacitor C, see Fig 3. For generate various values of total harmonic current distortion can do with various values of capacitor C, and with varies resistive load of rectifiers can be controlled magnitude of rectifier nput current distorted or current flow through relay.

The waveform of the non-linear load that contain harmonic current is shown in Fig 4. This waveform can be expressed in mathematic form as :

ISSN: 1411-1284


∑=

−+−=,....7,5,3

11 )(sin2)(sin2)(h

hhS thItIti φωφω

(1) where: iS(t) is distorted current non-sinusoidal I1 is current fundamental component Ih is current harmonic component orde h h is harmonic orde ( h = 3, 5, 7, …..) ω is angular frequency ( fπω 2= )

f is line system frequency Ф1 is phase angle of current fundamental component Фh is phase angle of current harmonic component orde h In the rms value form, the current in (1) can be expressed as:

∑=

+=,....7,5,3

1h

hS III

(2) And total harmonic distortion of current is :

1

~

2

2

I

I

THD hh

i

∑==

(3) In this experiment, the rms value and THD of the current (non-linear loads) are adjusted by changing the value of capacitor C and resistor of rectifier. The experiment result as effect of harmonic current to inverse time over current relay is given in Table II. In each mode is set for varies of the THD current values for a constant value pick up current setting of relay. The true rms current value Is, total harmonic distortion value (THDi), the rms value fundamental current I1 and operating time of relay were measured for each mode. For all modes, the pick up current of the relay was adjustable to 1 Ampere. Table II. Pick up current values and THDi values of relay

Mode THDi (%) I 1 (A) I S (A) 0 6.00 1.07 1.12 1 32.32 1.14 1.26 2 48.21 1.22 1.34 3 68.99 1.32 1.69 4 71.05 1.38 1.70 5 86.21 1.41 1.93

Table II. Operating time of relay for current is same 2 A

Mode THDi (%) I S (A) I S (A) 0 6.00 2.00 4.71 1 32.32 2.00 6.32 2 48.21 2.00 6.99 3 68.99 2.00 8.82 4 71.05 2.00 9.76 5 86.21 2.00 14.96

According to the experiment results, the pick up current of inverse time over current relay was affected by the presence of harmonic current (THD) and the pick up value was increase by non-sinusoidal currents. The increment depends on the total harmonic distortion (THD) value of the non-linear load current. With respect to the experimental results in Table III, the pick up current increased as long as THD value of current was increase. When THDi was above 86%, the relay which was set to 1 A, operated at 1.9 A. This shows that the relay can not perform a proper protection function and causes damage or heating-up depending on the rms value of the current and process time in power system components such as transmission lines, motors and transformers. The has different operating time value for the same relay current value in different modes as shown in Table III. Since the relay induction disc rotates rather slowly because of non-linear load current harmonic components, rel;ay operating time will increase as long as THD value of current increase. Power system components that are protected by this relay will be damaged because of the increase in the operating time.

IV. CONCLUSIONS Base on experiment results, the inverse time over current relay induction disc type can not operate effectively with non-sinusoidal currents that consist of several harmonic components (THD). Relay pick up current will increase as long as THD value of the current increase as shown in Table II. As given in Table III, the THD of current changes in spite of rms value of relay. Non-sinusoidal current remains constant and relay operating time increase as long as THD value of current increase. Distortion increment leads to increase in operating time of relay as shown in Table III. Thus, power system components can exposed to heat and finally can damaged. Acknowledgment The authors wish to thank University Sains Malaysia – School of Electric and Electronic Engineering – MOSTI that have providing the research grant (FRGS research Grant No 203/PELEC/6071144) and others facilities which have utilized in this research. The authors would like to thanks all power group members-LPKEE-USM.

ISSN: 1411-1284


REFERENCES [1] Bishop M.T and Baranowski, J.F, “Evaluating

harmonic-Induced Transformer Heating ”, IEEE Transactions on Power Delivery, Vol. 10, No.1 305-310. April 1996.

[2] Elham B.Makram and Regan B Hainnes, “Effect of Harmonic Distortion in Reactive Power Measurement”, IEEE Transaction Ind. Appl, Vol.28, No.4, July/August 1992.

[3] Power System Relaying Committee of IEEE, PES, IEEE Std.C37112,“IEEE Standard Inverse-time Characteristic equation for over current relays”, 1996

[4] Elmore. W.A. Kramer, CA and Zohholl. S.E, “Effect of waveform Distortion on Protective Relays”, IEEE Trans. On Inds. App, Vol 29, No.2, March/April, pp 404-411.

ISSN: 1411-1284


Effect of Harmonic Current Component to Active Power Losses on Power Transformer

Syafrudin1,2 and Dahaman1

1.School of Electric and Electronic Engineering – Universiti Sains Malaysia 2 Electric Department of Engineering faculty – University of Tanjungpura

email : [email protected]/my

Abstract– Power Transformer on the electric power distribution system used as changer of voltage levels and also as connector between of the high voltage lines with the low voltage lines or the consumer voltage 220V/380V. in the recent year, most of loads that connected on the power distribution system have non-linear characteristic (non-linear loads) and containing high current harmonic component, In this paper, the effect of current harmonic to active power losses on power transformer is presented. In this case, assumed that the transformer voltage is sinusoidal without containing harmonic component and wave form current transformer is distorted non-sinusoidal and containing harmonic that generated by non-linear loads. In the experimental, the transformer active power losses determined from different value of input and output power of transformer for linear load with sinusoidal current waveform without harmonic and also for non-linear loads with current waveform distorted with harmonic content THD. This experiment results show that the transformer active power losses will be increase proportionally to percentage of the total harmonic current distortion THD. Keywords– Transformer power losses, non-linear loads, total harmonic current distortion THD,

I. INTRODUCTION

n the electric power distribution system, power transformer functioned as a voltage changer from the

high voltage lines or middle voltage lines 20KV-11KV to low voltage lines or consumer voltage 380V/220V, and also functioned as loads connected. Generally, loads on the modern electric power distribution system have the non-linear current characteristic. It is caused by non-linear loads such as electrics equipments in the static converter to electric motor speed drive, computer, system illumination and lighting use electronic ballast, AC and DC controlled power supplies, battery charger, etc. Those non-linier loads cause line current waveform distorted non-sinusoidal and containing current harmonics

component. Several of the research result expressed that more than 60% of the loads are connected on the power distribution system is non-linear loads with total harmonic current distortion THD content more than 76%, estimated that this presented will be increase year to years [ 1]. Thereby the loads current that flow trough on power transformer also distorted non-sinusoidal and containing current harmonic component. The distortion current containing harmonic component have high effective value compared to the current without harmonic sinusoidal [ 2]. Considering the active power losses of the power transformer consist of the copper power losses in primary and secondary winding are function of the loads current and core losses that function of frequency consist eddy current lose and hysterisis lose, they losses is represented in thermal form [ 3]. Therefore, the transformer active power losse depend on total harmonic distortion of loads current that flow on this transformer. In this paper, A conducted an experiment to determining of the effect current harmonic component to active power losses at one power transformer. Transformer active power losses are very importance to be known because it is used for determining of a transformer operation capacities. Implementation of the transformer active power losses is in heat form, so that the high power lose can be causing the transformer over heat and dammed. Transformer active power losses determined with get different values between active power input and output that transformer. This experiment do for transformer with linear load and also non-linear load. To show the percentage of total current harmonic effect to active power losses of this power transformer, so experiment do for any non-linear load with same active power and varies current harmonic content.

II. MODELING AND ANALYSIS A. Harmonic Current Distortion Generally, ac current waveform which distorted due to non-linier load is in non-sinusoidal form and periodic wave[ 4]. Base on Fourier series analysis, the non-sinusoidal current waveform will be consisted of one current fundamental component and any harmonic current

O

ISSN: 1411-1284


component that have multiple fundamental frequency. In mathematic form can be expressed as :

∑=

−+−=,....7,5,3

11 )(sin2)(sin2)(h

hhS thItIti φωφω (1)

where: iS(t) is distorted current non-sinusoidal I1 is current fundamental component Ih is current harmonic component orde h h is harmonic orde ( h = 3, 5, 7, …..) ω is angular frequency ( fπω 2= )

f is line system frequency Ф1 is phase angle of current fundamental component Фh is phase angle of current harmonic component orde h In the rms value form, the current in (1) can be expressed as:

∑=

+=,....7,5,3

1h

hS III (2)

Percentage of current harmonic content or total harmonic current distortion THD is defined as bellow :

1

...9,7,5,3

2

I

I

THD

h

hh

i

∑== (3)

B. Modeling of Transformer Equivalent Circuit A power Transformer consist of primary winding, secondary winding and steel core. Power transformer equivalent circuit can be draw in figure Fig.1 bellow:

Fig 1. Circuit equivalent of a power transformer where: Vp is primary voltage or input voltage Vs is secondary voltage or output voltage Ip is primary current or input current Is is secondary current or output current (load current) Rp is primary winding resistance Rs is secondary winding resistance Xp is primary winding reactance Xs is secondary winding reactance Xm is magnetizing reactance RC adalah resistansi rugi-rugi inti If the load of transformer is non-linear loads, than the load current distorted non-sinusoidal and content harmonic current component expressed as:

∑=

+=,....7,5,3

1h

hsss III (4)

And active power losses of transformer on secondary winding is :

s

h

hhsssss RIIRIP

2

...9,7,5,31

2

+==∆ ∑

= (5)

Primary current Ip consist of secondary current and no-load current of transformer Io, is :

oh

hssP IIII +

+= ∑

= ,....7,5,31

(6)

While the no-load active power losses consist of or active power core lose, active power by eddy current and hysterisis and magnetizing lose. All that losses depend on material of core and frequency of system. Where the active power lose of hysterisis is:

∑=

=∆h

hhhy fBKP

,...7,5,3,1

6.1max1 (7)

and active power lose of the eddy current is :

∑∞

=

=∆,...7,5,3,1

22 max

hhed fBKP (8)

where : K1 and K2 is material constants of transformer Hence, the active power losses on primary side can be expressed as :

p

h

hhppppp RIIRIP

2

...9,7,5,31

2

+==∆ ∑

= (9)

Thereby, from equation (5),(7),(8) and (9) hence the total active power losses a power transformer can be expressed as bellow:

edhySp PPPPP ∆+∆+∆+∆=∆ (10)

This equation showing that the current harmonic component is very strong influence to active power losses value of a power transformer. In this paper, that active power losses due to harmonic current component of the power transformer is determined.

III. EXPERIMENTAL RESULTS AND DISCUSSION A. Experiment Experiment circuit for determining of active power losses due to harmonic current on a power transformer is shown in figure Fig 2. In this experiment do tested to single-phase power transformer 2000VA, 220V/220V, 50 Hz with it parameters are bellow : Rp = 6.8 Ω, Rs = 6.8 Ω, Lp = 0.56 mH Ω, Ls = 0.56 Ω, Xm = 1.87 mH. In this experiment, single phase-transformer is supplied by single-phase voltage source 120 V, 50 Hz. The fist step, transformer is given pure resistive or zero THD current with

Vp

Rp Ip Xp

RC Xm

N1 N2

Xs Rs Is

Vs

Io

ISSN: 1411-1284


a certain power load Ps. In order the load current flow trough transformer is sinusoidal without harmonic content. Then, the power input Pp and output Ps of that transformer are recorded. Active power losses can get from different values of that power input and power output. The next step, load changed by non-liner load, it is a single phase full wave rectifier with smoothing capacitors such as in figure Fig.3. With the same active power of load, recorded of input and output THD current and power Pp and Ps. Active power losses of the transformer get from different value of Pp and Ps. This experiment repeated again for several THD current values. Overall of the experiment results in measurement data are given in Table I, II, III and IV. For example, in Figure Fig.4 and 5 also given output currents waveform or load current with their current harmonic spectrum of this power transformer. In the Fig.6 show a graph of THD current content vs transformer active power losses.

Fig 2. Experiment circuit

L N

C

(a)

00.005 0.01 0.015 0.025 0.030 0.035 0.0400.020

V

I

(b)

Gambar 3. (a) Penyearah satu fasa gelombang penuh sebagai beban non-linier (b) bentuk gelombang arus dan tegangannya input

Table I. Measurement Data for Linear loads (Current THD=0%)

Input Output Vp Ip Pp THDip Vs Is Ps THDis

120 V

15.09 A

1.805 KW

0.40 % 116V 10.99 A

1.671 KW

0.40 %

Table II. Measurement Data for Linear loads (Current THD=32.4%)

Input Output Vp Ip Pp THDip Vs Is Ps THDis

120 V

17.65 A

1.838 KW

31.2% 109V 16.87 A

1.671 KW

32.4 %

Table III. Measurement Data for Linear loads

(Current THD=42.1%) Input Output

Vp Ip Pp THDip Vs Is Ps THDis 120 V

20.31 A

1.853 KW

40.0 % 92V 13.25 A

167 KW

42.1 %

Table IV. Measurement Data for Linear loads



24.56 A

1.886 KW

64.30% 86V 15.82 A

167 KW

68.0%

Table V. Measurement Data for Linear loads



28.39 A

1.907 KW

76.4% 79V 16.71 A

1.67 KW

81.0%

(a)

Power Analyzer 1 Power Analyzer 2

Trafo

Beban Non-Linier

Sumber tegangan

Is

Vs

ISSN: 1411-1284


(b)

Fig 4. Output current waveform without harmonic THD=0% and their spectrum harmonic.

(a)

(b)

Fig 5. Output current waveform without harmonic THD=0% and their spectrum harmonic.

THD vs Active Power Losses

0

0.05

0.1

0.15

0.2

0.25

0 20 40 60 80 100

% THD Current

Act

ive

Po

wer

Lo

sses

[kW

]

Fig 6. Graphic of THD load current vs active power losses transformer

B. Discussion Total active power losses of power transformer can be found from different values of active power input Pp and output Ps of that transformer. From the Table I, measurement data to linear load where current flow through transformer without harmonic THD=0%), power input of transformer is Pp = 1.320 kW and power output is Ps = 1,099 kW. Hence, active power losses of transformer can calculated as ∆P = Pp - Ps = 0.221 kW. In the same way, base on data in Table II the active power losses for THD current of 32.4% is ∆P = 0.134 kW and then from data in Table III, IV and V for THD current 42.1%, 68.0% and 81.0% respectively, the active power losses of transformer is with 0.167 kW, 0.182 kW, 0.215 kW and 0.236 kW respectively. Normally, power losses a transformer not more than 10% of total transformer rating. In this case, transformer active power losses have achieving more than that. Therefore, increasing THD current of load (non-linear load) can cause the transformer over heating in the power distribution system even though the transformer still under rating operation.

IV. CONCLUSIONS In this paper have been conducted investigate to a power transformer to determining of harmonic current effect to increase the active power losses. The transformer active power losses can be determine with different values of input and output active power. In this experiment, the active power losses increase of the power transformer is proportional to total harmonic current distortion THD of transformer loads. Finally, in order to operate the power transformer for several operating under THD percentage of non-linear loads, it is has proper design in terms of harmonic standard, proper design in terms of loading capacity heating and insulation lifetime.

ISSN: 1411-1284


Acknowledgment The authors wish to thank University Sains Malaysia – School of Electric and Electronic Engineering – MOSTI that have providing the research grant (FRGS research Grant No 203/PELEC/6071144) and others facilities which have utilized in this research. The authors would like to thank all power group members-LPKEE-USM.

REFERENCES [1] M.Salih TACI and M.Hadi Sarul, “The effect of The Harmonic

Components Upon Transformer Active Losses In case of Non-Sinusoidal Source”, IEEE, New York, 2000.

[2] Taci,M.S , “Determining The harmonic Effects of Non-Linear Loads on Parallel Connected Transfoemer in Terms Of Power Factor”, IEEE Power Quality’98, Hydrabad-India.

[3] Bishop M.T and Baranowski, J.F, “Evaluating harmonic-Induced Transformer Heating ”, IEEE Transactions on Power Delivery, Vol. 10, No.1 305-310. April 1996.

[4] Elham B.Makram and Regan B Hainnes, “Effect of Harmonic Distortion in Reactive Power Measurement ”, IEEE Transaction Ind. Appl, Vol.28, No.4, July/August 1992.

ISSN: 1411-1284


OPTIMIZE PLACING OF PASSIVE FILTER AND REDUCE HARMONIC IN POWER SYSTEM

Case Study at PT. ISPAT INDO

Iwa Garniwa MK, Budi Sudiarto; Adrianto

Electrical Engineering Department, University of Indonesia Depok 16424, Indonesia

Abstract-One of the solutions to reduce harmonic distortion is using passive filter. The most common type of passive filter is the single-tuned “notch” filter. This is the most economical type and is frequently sufficient for the application. This paper contents optimizing of placing a passive filter in a power system so it can be obtain the maximum and most efficient filterazation to reduce losses in medium voltage (MV). How to reduce harmonic and improve power factor using single tuned notch filter is also studied in this paper. The object of this research is an power system which has 3 panel of low voltage (LV) and the each of them has a different non linear loads. First panel, TR 20, has a maximum THDi (Total Harmonic Distortion) equal to 7.28 % and maximum THDv equal to 2.46 % (both in phase 3). Second panel, TR 21, has a maximum THDi equal to 62.41 % (phase 1) and has maximum THDv equal to 9.5 % (phase 3). Third panel, TR 23, has a maximum THDi equal to 58.37 % (phase 1) and maximum THDv equal to 8.26 % (phase 3). The MV from the simulation has THDi equal to 34.14 % and THDv equal to 1.57 %. I. Introduction Current from the harmonic load causes extra heat, isolation failure, operation failure, etc. One of the solutions to solve this problem is using passive filter at the source of the harmonic load. Using this passive filter, harmonic distortion expected to decrease until the limit of tolerance so that the power system can work properly. Research methodology that is been used in this paper starts with literature study of design passive filter so it can be implemented to non linear load. The result of the design filter will be simulated using ETAP Power Station 4.0.0. Output of the simulation will be analyzed and then we can make the best advice to the company which the data is being used. The company as the object research is PT.Ispat Indo. PT. Ispat Indo is the biggest steel company in East Java with the output production

billet and wire rod as the final product. The data were taken using power and harmonic analyzer equipment. II. Harmonic and Filter Harmonic distortion is caused by nonlinear devices in the power system. A nonlinear devices is one in which the current is not proportional to the applied voltage.

Fig 1. Distortion current cause by non linear loads

The sum of sinusoidal waves which create non sinusoidal wave can be solving using Fourier analyses.

)2sin(2)(1

0 n

n

nn ftnYYtY ϕπ −+= ∑

∞=

= (1)

Where Yo = Amplitude of DC current component Yn = RMS value from harmonic component F = fundamental frequency Φn= phase angle from harmonic component

ISSN: 1411-1284


Fig 2. Fourier series representation of a distorted

waveform

Total Harmonic Distortion (THD) expressing distortion level generated by all harmonic components and defined as follow:

1

1

2max

M

M

THD

h

nn∑

>= (2)

Where: THD = Total Harmonic Distortion Mn = RMS value of current or voltage –n orde M1 = RMS value of current or voltage at fundamental frequency There are two kinds of filter, active filter and passive filter. Passive filter is common use in industry. The weakness using passive filter are depends on impedance, system frequency, and component tolerance. Two purposes using filter are:

1. To reduce harmonic in AC network. 2. To provide reactive power for the harmonic

load or other loads. There are two kind of passive filter, series filter and shunt filter. Shunt filters that is commonly use are single tuned filter and order two damped filter. These two filters are the simplest way of design and the cheapest to implementation. Passive filter contents inductance, capacitance which combines to control harmonic. Filter passive are commonly use and not expensive. Using passive filter is possible to interact with the power system which can disturb it. It is important to check all of the possibility interaction when filter is being designed. The shunt filter works by short-circuiting harmonic currents as close to the source of the distortion. This keeps the currents out of the power system. This filters connected to the bus bar where reduce of harmonic voltage determined and create a filter bank.

Fig 3. Passive filter in a power system

There are two parameters need to be considerate when determine value of R, L, and C: a. Quality factor (Q) b. Deviation relative frequency (δ)

III. Measurement and Data Program we use to simulate the power system is ETAP 4.0.0.

Fig 4. Power system design using

ETAP 4.0.0

The data has been recorded using 9625 Power Measurement and transferred to the personal computer using 9625 Power Measurement Support Software. The data of MV we have it from the simulation software (ETAP 4.0.0).

Table 1. Data from the panels when the THDi maximum

TR 20 TR 21 TR 23 TM

THDi 7.28% 62.41% 58.37% 34.14%

Current 1344 313.3 A 372.5 A 92 A

THDv 2.37% 3.70% 3.42% 1.57%

ISSN: 1411-1284


Power Factor

0.68 0.68 0.7 0.66

(a)

(b)

Fig 4. (a) Current distortion wave in TR 20, 21, and 23

(b) Current distortion spectrum in TR 20, 21, and 23 Using IEC 61000-3-6 and 61000-3-4 standard, the limit of THDv is 8 %. From the data, we can determine the panels which need a passive filter are TR 21, TR 23, and MV. These panels have THDv and distortion current over the limit. IV. Analyze and Solution By using the equation below, we can determine the losses in power system:

RIPlosses2= (3)

Where I = total current R = Impedance P = power loss If the impedance of the system constant, we can decrease the losses by decreasing the current.

Single tuned “notch” filter is use to reduce harmonic. The basis parameters of design filter as follow:

Table 3. TR 21 Filter harmonic data

THDi Maximum (%) 76 Phasa 1

THDv with highest THDi

(%) 4.46

THDv with highest

harmonic order (%) 3.3 Order 5

PFCC kVAR ON 166

Table 4. TR 23 Filter harmonic data


THDv with highest THDi

(%) 4.75

THDv with highest

harmonic order(%) 3.56 Order 5

PFCC kVAR ON 189

Table 5. TM Filter harmonic data


THDv with highestTHDi

(%) 1.47

THDv with highest

harmonic order (%) 1.23 Order 5

PFCC kVAR ON 878

The steps to design a filter based on the table above as follow: 1. Election of frequency filter.

From the measurement, the highest harmonic level happened at order 5. Tuning of order degraded by a little below order in fact so that n = 4.8. This degradation is conducted as tolerance to filter component and also variation of system impedance.

2. Enumeration of capacitor size measure and resonance frequency. The formula of current at the fundamental frequency for capacitor bank as follow:

actual

actualFLcap

kV

kVARI

3= (4)

ISSN: 1411-1284


Formula to look for capacitive impedance as follow:

rated

rated

C M

kVX

var

2

= (5)

For TR 20, nominal voltage phase to phase is 430 V. For TR 21 and 23 is 730 V and for MV is 11 kV. Inductor impedance determined with formula following:

2n

XX C

R = (6)

Because there is reactor, the fundamental current increase which can expressed by:

( )RC

busFLfilter

XX

VI

−=

3 (7)

Because the current increased, the reactive power compensation increase and we can determine using:

FLfilterbusplied IVkVAR **3sup = (8)

Table 6. Filter specification

THDi Maximum Filter

Spesification TR 21 TR 23 TM

System

frequency 50 50 50

Rating of

Capasitor Bank 166 189 878

Rating of Bank

Capacitor

Current 131.25 150.45 46.11

Tuning

harmonic filter 4.8 4.8 4.8

Tuning system

frequency 240 240 240

Capacitor

Impedance 3.211 2.801 137.742

Reactor

Impedansi 0.139 0.122 5.978

Filter Full Load

Current 137.19 157.3 48.2

There are 4 variations when placing the filter. The results of simulation using ETAP 4.0.0 in TM as follow:

Table 7. Filter at TR 21

Before After

Difference

THDi (%) 34.14 28 6.14

THDv (%) 1.57 1.26 0.31

Current

(A) 92 84 8

Power

factor 0.66 0.71 -0.05

Table 8. Filter at TR 23

Before After

Differense

THDi (%) 34.14 27 7.14

THDv (%) 1.57 1.21 0.36

Current

(A) 92 83 9

Power

factor 0.66 0.72 -0.06

Table 9. Filter at TR 21 and TR 23

Before After

Difference

THDi (%) 34.14 22 12.14 THDv (%) 1.57 0.98 0.59

Current (A) 92 76 16

Faktor Daya 0.66 0.77 -0.11

Table 10. Filter at TM

Before After

Difference THDi (%) 34.14 27 7.14

THDv (%) 1.57 0.98 0.59 Current

(A) 92 62 30 Power factor 0.66 0.97 -0.31

ISSN: 1411-1284


From the result of simulation above, we can determine that the biggest difference in decreasing current is when the filter located at TM. The current was decreased 30 A to the final value 62 A. This happen because harmonic distortion was reduced by the passive filter which the filter designed based on the entire load (TR 20, TR 21, and TR 23). This makes the filter design more precision. Reactive power that is supplied by the capacitor bank also based on the entire load so it’s enough to increase the power factor value to 0.97 which can avoid the penalty. This also helps make the losses in TM decrease. The THDv was decreased to 0.98 % from the original value 1.57 %. The THDi also decreased to 27 % from the original value 31.14 %. These data shows us that passive filter can reduce harmonic, increase power factor, and reduce the power loss.

V. Conclusion 1. Single tuned filter can be use to reduce

harmonic distortion, increase power factor, and reduce power loss.

2. Optimize placing of passive filter can make passive filter more efficient and effective.

3. Placing the passive filter at the MV in the object of this research is the best place to minimize the power loss.

4. The results using filter at MV are: THDi = 27 % THDv = 0.98 % Current = 62 A Power factor = 0.97

VI. Reference 1. Roger C. Dugan..[et al]. 2002. Electric Power

System Quality. New York : McGraw-Hill 2. Cahyadi, Rifky. 2003. Upaya Menghemat

Energi dengan Mereduksi Harmonisa. Depok : Skripsi Departemen Teknik Elektro Universitas Indonesia

3. Setiabudy, Rudy. 2006. The Design Of Passive Filter To Overcome Harmonic Distortion. Depok : Proceedings. IJJSS 2006

4. Zunaedi, Totok. 1999. Eliminasi Harmonik Dan Kompensasi Daya Reaktif di PT. Engenys Steel. Depok : Skripsi Departemen Teknik Elektro Universitas Indonesia.

ISSN: 1411-1284


CONVERGENCE APPROXIMATION OF ABR FORMULATION FOR JOSEPHSON’S TUNNELING IN UO2 CHAIN REACTION

AT 45.7 MW CANDU NUCLEAR REACTOR

Moh. Hardiyanto Department of Educational Physics – FTMIPA - UNINDRA Large Hadron Collider Laboratory, CERN – Lyon, France

E-mail : [email protected]

Abstract - The convergence of free covariant

equation in Einstein’s space with quantum condition is studied using the ABR (Abrikosov-Balseiro-Russell) formulation in convergence approximation for Josephson tunneling is important role for determine of neutrino particle existing, especially after Cerenkov’s effect for 45.7 megawatts nuclear reactor based on UO2 chain reaction. This approaching will be solved the problem for determine the value of interstellar Electrical Conductivity (EC) on UO2 chain reaction, then the post condition of muon has been known exactly. In this research shown the value of EC is 4.32 µeV at 378 tesla magnetic field for 2.1 x 104 currie/mm fast thermal neutron floating in 45.7 megawatts adjusted power of Canadian Deuterium Uranium (CANDU) nuclear reactor. The pictures had resulted by special Electron-Scanning-Nuclear-Absorbtion (ESNA) shown any possibilities of Josephson’s tunneling must be boundary by muon particles without neutrino particle existing for 350 – 456 tesla magnetic field on UO2 more enrichment nuclear fuel at CANDU, whereas this research has purpose for provide the mathematical formulation to boundary of muon’s moving at nuclear research reactor to a high degree of accuracy and with Catch-Nuc, one of nuclear beam equipment has a few important value of experimental effort.

Keywords - ABR formulation, EC value, UO2 chain reaction, magnetic field value

I. INTRODUCTION Its is known that for muon-hadron

scattering, the close-coupling equations [1] have been used extensively in the sophisticated computation in Abrikosov-Balseiro-Russell (ABR) formulation [2]. In the close-coupling equations, the complete wave functions are expanded in target states on Einstein’s space. In this paper the target states are constructured in the finite L2 basis space following the results of Abellian system and next formulated by ABR without Dirac’s condition. The previous studied the convergence of the approximation target states for discrete and continuum cases in quantum condition for UO2 chain reaction at 45.7 megawatts (MW) adjusted power in Canadian Deuterium Uranium (CANDU) nuclear research reactor. The advanced studied its application to the ABR equations which results in pseudo state close-coupling approximations for bound-free transition in convergence structure for Josephson’s tunneling. For first step using by

Einstein’s space for eliminary construction of ABR formulation in UO2 chain reaction.

In this paper, the Abellian system was conducted for covariant equation to Einstein’s space and ignores the Dirac’s condition for 300 – 410 tesla magnetic fields on fast thermal neutron, to study the convergence behavior of the free covariant equation part. Its possible analytical computation is also addressed after finding the asymptotic behavior.

II. Pseudostate Close-Coupling Approximations

The close-coupling equations in ABR formulation require to study is given by

[ ] ∑∞

=

±±± =⟩∈−+−1

0k

kjkjj fVfEH (1)

where

20 2

1 ∆−=H (2)

the total energy of the system connecting the initial and final state fi → satisfies

221 2

1

2

1ffi kkE +=∈+=∈ (3)

±jf are channel functions related to the

complete wave function as

)()(),( 2

1121 rfrrr j

jj

±∞

=

± ∑= φψ (4)

in which the + (-) superscript refers to singlet

(triplet) scattering and )( 1rjφ are muon eigen

functions.

The channel potential ±jkV is given by

±±± += jkjkjk WUV (5)

where

ISSN: 1411-1284


−±

−+−

>= ±± ∫ ∫

2121

1

2

211

2121

)(1

)(11

)(

rrrr

r

rrrr

rrdrdrU

k

k

jjk

φ

φ

φ

(6)

and

iijji

ijkjk EW φγφδ )( −∈+∈= ±± ∑ (7)

with

1=±iγ 14 =iγ (8)

If replace the target states by a set of pseudo states (as consequences of finite basis set) in order to solve the equations approximately. We has to modify the close-coupling equations for Einstein’s space through for normal condition after Cerenkov’s effect at 315 tesla magnetic field since the exchange term in the form (7) is valid only for exact target states. The form appropriate to pseudo states is given by

NjNkNjNkjk rHrHW φφφφ ><+><=± )()( 21

∑=

± <−>+N

iNiiNijk E

1

)( φγφδ (9)

where iNφ label the pseudo states and H(r1) is the

target non-linier for Hamiltonian operator in Abellian.

In calculating the pseudo state close-coupling approximation one must consider the error present in the quadrature rule approximation and the close-coupling potentials when pseudo states are employed. The first error will not be discussed here. The second error will be discussed in association with the direct-potential component of the channel potentials since they are regarded is being responsible for the dominant scattering process especially at higher energy.

III. Convergence Approximation of ABR Formulation

In the momentum re-presentation, the derivative of ABR formulation for muon state in UO2 chain reaction is given by

VGVVT B 02 += (10)

Here G0 is the diagonal matrix of free channel Green’s functions. This second ABR formulation approximation has been used by Einstein’s space in Abellian system, especially the fast thermal neutron floating at 350 tesla magnetic field to test the suitability of pseudo states expansions with initial and first states chosen either the ground or 2s, 2p excited states.

The actual Josephson tunneling element from (10) require to study is

[ ]∫ ∆•

−

+−

∆−=

drrer

kkV

jri

iij

jiij

)()(

2),( 2

φφδ (11)

where ji kk −=∆ . The indices i and j can be

either discrete or continuous after Cerenkov’s effect on 2.1 x 104 currie/mm and the range of 315 – 390 tesla magnetic field. For free covariant equation in quantum condition for UO2 enrichment reaction potentials one uses

∫ ∆•−∆−=∆ drrerI pri

qpq )()(2)( .2 φφ (12)

where pφ and qφ are initial and final continuum

states having the momenta p and q respectively, given by

∑∞

=

+

=

0

.)()12(

2

1)(

q

Nq

iq

q

qr

rqPrUei

qrr

qq

qq

l

ll

ll

lδ

φ (13)

and normalized to a δ function in 3)2/( πq . The

approximate target wave function NqpU is given by

qq

qq

nnq

N

n

Nq

xpn

n

qBU

ll

ll

lφ)(

22(

)1(

)(

11

0

+−

= ++Γ+Γ

>=

∑ (14)

where

γπθ

γ)2/(2/)1(2 )1(

)1(2)(

−+−

−+Γ=

ex

iqB

q

q

q

p

qp

ll

(15)

and

.....0.

)()()( 122/1

==− +−+

q

nr

n

n

rLerr qqq

l

lll λλϕ λ

(16)

solving the integral (15) after quite lengthy derivation one finally obtains

)(

..,.

12

3

)1(

)(32)(

qpqpq

qp

qp

iLm

mmML

pq

ci

pqV

llll

ll

δδ

π

−++

−

−∆

−=∆

∑

)(

000)12(4

)12)(12(2/1

∆−

+++

Lpqqpqp

qp

qp

ILMmm

LL

x

ll

llll

π (17)

where

ISSN: 1411-1284


)()()(

)()22()22(

)(

11,1

0.

2

qmpn

MN

mnq

qp

Lpq

xPxPqB

pBI

pp

p

qp

ll

l

ll

l

ll

+−−

=

++

∑

+Γ+Γ=∆ λ

∆+∆∆ −−+ ∑ )()2(Re 11 iiix vL λ

).:22.22:

2()2( 2

∆+∆+++

−−−++−++Γ

ii

mnvFvx

qp

qpqp

λλ

λλ

ll

llll

(18)

IV. METHODOLOGY

This research using by high degree accuracy mathematical approaching and a few experimental efforts with special Electron-Scanning-Nuclear-Absorbtion (ESNA) and Catch-Nuc, for see the results of the mathematical approximation, in case is ABR formulation in Josephson’s tunneling for UO2 chain reaction at 350 – 456 tesla magnetic field and determine of Electrical Conductivity (EC) value in 2.1 x 104 fast thermal neutron floating.

A few mathematical equations in quantum condition will be explain in this section. ABR Formulation non Abellian

The metric tensor is defined through the line element ds as follows:

βααβ dxdxgds =2 (19)

Hence, different metric shall lead to different properties of space-time. For a space time where the metric is defined through

( )[ ]

++

−

−=

)sin(1

22222

2

222

φθθ ddrkr

dr

tRdtds

(20)

the space satisfies the properties that it is homogeneous and isotropic. Such properties agree with the nuclear structure in microscopic principles and are also supported by experimental data. Accordingly the above metric, called the ABR Formulation without Abellian system, becomes a standard model of UO2 structure. In the above expression R(t) denote the scale factor and k is a constant. The universe is closed if k > 0, open if k < 0 and flat if k = 0 and R(t) sometimes are rescaled in such a way that the value of k assume one of the three values -1, 0, or +1. The Cartesian form of the above line element.

−++

−+

+

−+

−=

2

22

2

2

22

22

2

22

11

11

11

)(

kr

kzdz

kr

ky

dykr

kxdx

tR

dtds

(21)

∂∂

−∂∂

+∂∂

=Γ µβγ

βµγ

γµβαµα

βγx

g

x

g

x

gg

2

1(22)

Combining (19) and (21), one finds the explicit forms of non-zero components of the Riemannian curvature tensors as:

.00 )(

)(ab

ab tR

tRR δ=

−+=

200 1

)()(kr

xkxtRtRR ba

abba δ (23)

Λ+Λ= ∫ AAdAATrk

SM 3

2

4π (24)

)2()1( 00'),2,1('

ρρδρρ

EEK −−∈∈∈∈

= (25)

( ) Ζ∈= ∫ − 31224

1MwZ dggTrS

π (26)

∫−=1

0

)()( daaxFaxxA µνµ

µ (27)

∫=Γ1

0

)()( xRdxx λλλ αβµν

µαβν (28)

01'

)(2

1)(

'

)2(^

Eqgn

s

QQ

s

s

ll

∑=

−−

=

ρρ

ρρ β (29)

V. RESULT AND DISCUSSION It is shown in the final equation that beside the double series n and m which are finite, the ABR function F2 is formally defined by a double series expansion with radius of convergence severely restricted. Because the occurrence of –n and –m which are finite in the arguments, there is no problems arise about its convergence. The remaining is to compute (22) numerically in order to see how rapid the convergence is since the expression of Riemannian curvature tensor is

similar with the expression of µνF in the non

Abellian gauge. Such a formula (18) may also be obtained in the general theory of relativity, the fast

ISSN: 1411-1284


thermal neutron break into UO2 matrix is show in Fig. 1.

Fig.1. The floating of muon moving in UO2 matrix at 378 tesla

magnetic field (Courtesy of CANDU nuclear reactor, Canada)

To obtain this formula, let us impose the ABR gauge condition on the Christoffel symbol as follows:

0)( =Γ xx aβµ

µ (30)

If we make the transformation of µµ λxx →

where λ : [ ]1,0 is a parameter could obtain:

)(

)()(

x

xxxRx a

λ

λλλαβν

αβνµ

µβµν

µ

Γ+

Γ∂= (31)

and if use the relation:

).()(

)(

)()()(

xFxx

xFx

x

xF

d

xdxF

d

d

λλ

λλ

λλ

λλλλ

λλ

µµ

µµ

µ

µ

∂=∂

∂∂∂=

(32)

))(()( xd

dxRx λλ

λλλ α

βναβµν

µ Γ= (33)

Using the above formula, we shall derive

the Christoffel symbol and contrast them with the Christoffel symbol previously derive from the ABR formulation on Abellian system and the re-polarization of UO2 matrix will give in Fig. 2.

Fig.2. The floating of muon moving

in UO2 matrix at 378 tesla magnetic field and fast thermal neutron at 2.1 x 104 currie/mm floating (Courtesy of CANDU nuclear reactor, Canada)

We have seen that the ABR gauge which

satisfy the Josephson’s tunneling effect can be deduced from (23). We shall refer the symbol of

FSαβνΓ for the ABR gauge in Christoffel symbol

to distinguish with that derived from the Robertson-Walker Riemannian tensor, the fast thermal neutron at 2.1 x 104 currie/mm floating will shown up at Fig. 3, especially without the neutrino particle.

++−

++

+−++

−

=+=

+

+

∞

∑

)!4(

)2()1(

)!3(

)1(2)1(

)!2(

)1(

)(0),(),(

2

1

)(

n

n

n

n

n

rFkntrBtrE

n

nn

n

(34)

Fig.3. The floating of muon moving in UO2 matrix at 378 tesla magnetic field and fast

thermal neutron at 2.1 x 104 currie/mm floating without neutrino particle

(Courtesy of CANDU nuclear reactor, Canada)

Based on Fig. 3, if integrate the equation by parts, we will obtain the series form of the equation as

∫ ∑∞

= +−=

1

0 0 )!2(

)1()(

nn

nn

xd

d

nxRxd ρα

βρνρ

λλλλ (35)

Let us now choose a scale factor as: R(t) = emt. This is an asymptotic value of the Friedmann model of UO2 nuclear structure with

3/Λ=m , where Λ is the UO2 constant. Using by special ESNA, we get the Josephson’s tunneling with Abellian operator such as the Fig. 4.

ISSN: 1411-1284


Fig.4. The Josephson’s tunneling

in UO2 matrix at 378 tesla magnetic field (Courtesy of CANDU nuclear reactor, Canada)

The singularity for the first turns out to be worse compared to the later since the standard one has only a singularity factor from EC value at interstellar of muon’s moving for 2.1 x 104 fast thermal neutron floating at range 350 – 456 tesla magnetic field without neutrino particle existing

after Cerenkov’s of ( ) Vekr µ32.41/1 2 =− .

At least on the pedagogical point of view one then becomes aware that the ABR formulation in UO2 chain reaction for a given Riemannian curvature tensor are not unique.

VI. CONCLUSIONS

Investigations and research using by ABR formulation in convergence approximation and ESNA also Catch-Nuc equipments based on UO2 chain reaction by Josephson tunneling at 45.7 MW CANDU nuclear research reactor has a few result, expressed:

a. The strength of fast thermal neutron floating is 2.1 x 104 currie/mm.

b. The values of Electrical Conductivity (EC) for interstellar muon’s moving is 4.32 µeV at 378 tesla magnetic field.

c. These equations have a high degree accuracy, so they could be determine of muon’s moving after Cerenkov’s effect for 45.7 MW adjusted power.

ACKNOWLEDGMENT The author wish to thank Canadian Deuterium Uranium (CANDU) Nuclear Reactor team for contributions to this work.

REFERENCES [1] Bransden,B.H.Scott,T.,Shingal,R.

and Roychoudhury,R.K., Journal Phys.B 15, 2003

[2] Bray I. and Stelbovics.A.T., Phys.Rev. A. 46, 2002

[3] Burke. P.G and Seaton M.J., Math. Journal of Phys.10 , 2002

[4] CANDU nuclear reactor, UO2 Interference Structure, Canada, 2003

[5] Flinder, and Stelbovics.A.T.,Test of Close-Coupling in the poet Model, Procedings of the Sixth International Symposium on Correlations and polarization in Electronic and Atomic Collisions and (e,2e) Reactions, Univ. of California, USA (2001).

[6] Helèna, Duprix, Structure of ABR formulation in Dirac’s condition, Universitè du Quèbec, Canada, 2003

[7] Madison,D.H.and Callaway.J.J., Journal Phys B 20, 2002

[8] Stelbovics,A.T., and Abrikosov G., American J.Phys.43, 2003

About Author:

Lecturer of Department of Educational Physics and Researcher at Large Hadron Collider (LHC) Laboratory – Lyon - France, and now take a doctoral program at Universite du Quebec, Canada.

ISSN: 1411-1284


Energy Conservation on Steel Industry by Increasing Efficiency and Electric Power Quality

Iwa Garniwa; Budi Sudiarto; Fauzan Hanif Jufri

Faculty of Engineering, University of Indonesia

e-mail : [email protected] ; [email protected] ; [email protected]

Abstract – Electrical energy conservation is

electric energy usage with high efficiency by minimize losses at all production process, start from generation, transmission, distribution, and consumption.

Industry sectors included steel industries are large energy consumers because industry sectors use about 30% primary energy. This mean energy conservation on industry sector gives a significant role for thrift the energy.

This paper did some research for energy conservation opportunity on demand side focused in electrical energy conservation, especially in compact energy industries, that is steel industries. This research takes the some opportunity by increasing efficiency on processes, power systems, and electric power quality on electrical systems. Energy consumption efficiency for every process can be explained from Specific Energy Consumption (SEC), that is a value meant ratio between total energy consumption to total product yield.

The recommendations yield from this paper will be followed by feasibility study from technical sides and economic sides, using Life-Cycle Costing Analysis. The opportunities can be high cost or investment, or low cost or no cost.

The data on this paper get from survey to one of steel industries, that is PT SERMANI STEEL, which produce galvanized iron sheet, and PT SURABAYA WIRE which produce nails and wire rod.

Keywords– Energy Conservation, steel industry, electric power quality, energy efficiency

I. INTRODUCTION

ndustry sectors are large energy consumers because they use about 30% primary energy. This mean

energy conservation on industry sectors give a significant role for thrift the energy. Besides that, industry sector consist of large consumers which a small number, so energy conservation can be easier than do at other consumers. Energy sectors can be classified based on its energy consumption, those are

compact energy, for example steel industries, cement industries, and melting factory. Processes on steel industries use much energy. Those are used for electric motor, convert electrical energy to mechanical energy, and for arc furnace, convert electrical energy to heat energy. Electrical energy usage can be reduced by increase or modify another factors used in processes. As an example, increasing power factor, efficient loading on motors and transformers, or increasing the fluid pressure on furnace, etc.

II. BASIC THEORY [1] Energy intensity in producing products defined as Specific Energy Consumption (SEC), it is ratio between energy consumed to yield products to amount of products yield. The unit is kWh/Ton. By conserve energy, we hope that the SEC will decrease, which mean decrease the energy consumption without decrease the product yield. 2.1 Improving power factor

The reactive power has to be supplied by the utility even though it is not actually registered by the power meter (as real power used). The magnitude of this reactive power increases as the power factor decreases. To account for the loss of energy due to the reactive power, most utilities have established rate structures that penalize any user that has low power factor. Therefore, significant savings in the utility costs can be achieved by improving the power factor. This improvement can be obtained by adding a set of capacitors to the entire electrical system. See figure 1,

Figure 1. Improving power factor by using capacitor The size of these capacitors (Q compensator) is typically measured in KVAR (the same unit as the reactive power) and can be determined, as shown in figure 1, using the power triangle analysis :

I

ISSN: 1411-1284


tan tan

1 1tan cos tan cos

P Q Q PC ex comp ex comp

P P PF PFC ex comp

ϕ ϕ

= − = −

− − = −

(1)

Where P stands for the real power (measured in kW), while Qe and Qc stand for the reactive power before and after the retrofit, respectively. The thrift yield from improving power factor can be defined as [2] :

2

% 1 100%exreduction

comp

PFLoss

PF

= − ×

(2)

2.2 Reducing harmonic distortion

Harmonic distortion (based on IEC 702-07-43) is waveform distortion which frequency multiplied from its fundamental frequency in integer number.

Figure 2. Harmonic distortion waveform

To quantify the level of distortion for voltage and

current a dimensionless number, referred to as the total harmonic distortion (THD), is determined through a Fourier series analysis of the voltage and current waveforms and is respectively defined as:

2 2111

hR M S M M T H Dh

h

→ ∞= = −∑

> (3)

Harmonic distortion will increase the RMS value both current and voltage, so that will increase power loss in the systems. Besides that, harmonic distortion will reduce equipment lifetime, three phase neutral current overload, etc. 2.3 Replacement motor with energy-efficient motor

Induction motor – commonly used in industrial sectors – has an inductive characteristic, produced by its magnetic field to rotate the rotor. One parameter is typically important to identify an electric motor during full-load operation, the conversion efficiency of the motor (η). This efficiency expresses the mechanical power as a fraction of the real electric power consumed by the motor.

mechanical

electrical

P

Pη = (4)

Due to various losses (friction, core losses due to the alternating of the magnetic field and resistive losses through the windings), the motor efficiency has

typical values ranging from 75% to 95%, depending on the size of the motor. In the above definition, PM stands for the mechanical power output of the motor, expressed in kW or horsepower (HP), which is the most important factor in selecting a motor.

Table 1. Typical motor efficiencies (Hoshide, 1994)

The improved efficiency of the high/premium-

motors is mainly due to better design with use of better materials to reduce losses, which however comes with a higher price (about 10 to 30% more than standard-efficiency motors). This fact partially explains why only one-fifth of the motors sold in the US are energy-efficient. 2.4 Use Adjustable-Speed Driver (ASD)

The use of adjustable-speed drivers (ASDs) offers numerous advantages to drive systems' energy efficiency. They can offer the potential for higher efficiencies, lower operating costs, easier control and minimal maintenance. The most frequently used type of adjustable-speed drivers is the frequency controlled ASDs offering the greatest advantages. They consist of two types: direct and indirect inverters. The first ones convert frequency and voltage in one step, whereas the latter use an intermediary DC link. ASDs can take the saving about 50% in reheating furnace 2.5 Reducing peak of power demand

This method used by controlling the demand in consumer side. For reducing peak of demand power, it can be determined by demand factor and load factor. If demand factor is low, it means that the connected load is not efficient. Demand factor describe as

maximum demanddemand factor =

connected load (5)

Life-Cycle Cost Analysis [3] LCC = I + R - R + E + OM&Rcpl es (6)

[ ]NS = ∆E + ∆W + ∆OM&R - ∆ I + ∆R - ∆RA:BC 0 cpl es

(7)

∆E + ∆OM&RSIR = A:BC ∆I + ∆R - ∆R0 cpl es

(8)

( ) ( )1

AIRR = 1 + r SIR -1N (9)

ISSN: 1411-1284


( )

∆E + ∆OM&R - ∆R + ∆Rt t cpl est t 011

Itdt

≥ ∆∑

+=

y (10)

Where, LCC = present value of LCC I = present value of investment Rcpl = present value of capital replacement cost

Res =present value of residual cost

E = present value of energy cost OM&R = present value of operation, maintanance, repairement cost

:NSA BC = Net Saving

∆E = ( )E - EBC A = energy cost saving


3.1 PT SERMANI STEEL The SEC of PT SERMANI STEEL before energy conservation done is :

KES Bulan kWh Lembar kG kWh/Lbr kWh/kG

September 2006 58.600 333.768 1094.322 0,18 0,05 Oktober 2006 63.400 206.265 667.272 0,31 0,10 November 2006 50.900 321.493 1.027.163 0,16 0,05 Desember 2006 61.400 314.480 998.718 0,20 0,06 Januari 2007 60.300 371.483 1.135.617 0,16 0,05 Februari 2007 69.200 315.667 998.460 0,22 0,07 Maret 2007 59.000 310.368 1.053.246 0,19 0,06 April 2007 67.100 188.735 647.882 0,36 0,10 Mei 2007 47.400 354.857 1.135.782 0,13 0,04 Juni 2007 69.900 226.372 722.879 0,31 0,10 Juli 2007 53.600 342.927 766.234 0,16 0,07 Agustus 2007 47.400 119.191 417.582 0,40 0,11

Rata-rata 59.016.67 283.801 888.763,1 0,23 0,066

Table 2. SEC of PT SERMANI STEEL before energy conservation

The energy conservations which passed by LCC analysis are :

energy saving

investmentaverage saving

pay back periode

(kWh/month) (Rp) (Rp/month) (month)Reducing connected load

0 13,935,000 2,550,428 3

Improving power factor

625.68 20,000,000 486,084 20

energy conservation

Table 3. Electrical energy conservation of PT SERMANI

STEEL Using ASDs and replacing the motor did not pass the economic requirement, so they are not feasible for current time. However, if there will be a plan for reconstruction and has big enough money for investment, it will be best to use ASDs and premium-efficiency motor. While harmonic distortion is just about 2 % – 3 %. The new SEC after energy conservation are :

before energy conservation

after energy conservation

saving

Product yield (Ton) 888.76 888.76 Product Cost (Rp) 54,458,719.00 51,422,207.00 3,036,512.00 Energy Conumption (kWh) 59,016.00 58,390.12 625.88 Rp/Ton 61,274.94 57,858.37 3,416.57 SEC (kWh/Ton) 66.40 65.70 0.70 SEC (kWh/kG) 0.0664 0.0657 0.000704

Table 4. SEC of PT SERMANI STEEL after energy

conservation From table 4, SEC of PT SERMANI STEEL decrease 0,7 kWh/ton a month. 3.2 PT SURABAYA WIRE The SEC of PT SURABAYA WIRE before energy conservation done is :

Bulan kWh kG KES (kWh/kG)

Agustus 2006 160.500 2.187.689 0,07 September 2006 206.100 3.286.985 0,06 Oktober 2006 244.500 2.128.311 0,11 November 2006 173.790 3.703.988 0,05 Desember 2006 284.940 2.908.885 0,10 Januari 2007 167.040 2.417.828 0,07 Februari 2007 190.950 2.158.409 0,09 Maret 2007 173.220 2.941.461 0,06 April 2007 159.690 1.714.668 0,09

Mei 2007 88.620 1.790.362 0,05 Juni2007 139.110 2.593.985 0,05 Juli 2007 181.560 2.216.294 0,08

Rata-rata 180.835 2.504.072 0,07

Table 5. SEC of PT SURABAYA WIRE before energy conservation

The energy conservations which passed by LCC analysis are :

energy saving

investmentaverage saving

pay back periode

(kWh/month) (Rp) (Rp/month) (month)reducing harmonic distortion

14,291,280 25,000,000 3,488,358 7

improving power factor

4,087,459 40,000,000 654,139 24

energy conservation

Table 6. Electrical energy conservation of PT SURABAYA

WIRE Demand factor of PT SURABAYA WIRE is 90,1% so that the connected load could not be reduced. Usage of ASDs and replacement of premium-efficiency motor do not pass the LCC Analysis for current time. The new SEC after energy conservation are :

before energy conservation

after energy conservation

saving

Product yield (Ton) 2,504 2,504 - Product Cost (Rp) 136,364,807 126,256,500.55 10,108,306.45 Energy Conumption (kWh) 180,835 162,456.26 18,378.74 Rp/Ton 54,457.22 50,420.48 4,036.75 SEC (kWh/Ton) 72.22 64.88 7.34 SEC (kWh/kG) 0.0722 0.0649 0.0073

Table 7. SEC of PT SURABAYA WIRE after energy

conservation

ISSN: 1411-1284


From table 7, SEC of PT SURABAYA WIRE decrease 7,34 kWh/ton a month.

IV. CONCLUSIONS

TThheerree aarree ff iivvee mmeetthhooddss ttoo ccoonnsseerrvvee eelleeccttrriiccaall eenneerrggyy iinn iinndduussttrriiaall ppoowweerr ssyysstteemm,, tthheeyy aarree iimmpprroovviinngg ppoowweerr ffaaccttoorr,, rreedduucciinngg hhaarrmmoonniicc ddiissttoorrttiioonn,, uussiinngg AAddjjuussttaabbllee--SSppeeeedd DDrriivveerr,, rreeppllaacciinngg mmoottoorr wwiitthh hhiigghh--eeff ff iicciieennccyy oonnee,, aanndd ccoonnttrrooll ll iinngg ppeeaakk ooff ppoowweerr ddeemmaanndd.. TThheeyy ccoonnssiisstt ooff hhiigghh ccoosstt mmeetthhoodd aanndd llooww ccoosstt mmeetthhoodd.. HHiigghh ccoosstt mmeetthhoodd nneeeedd aann aaddvvaannccee ssttuuddyy ttoo ttaakkee tthhee ddeecciissiioonn,, wwhhii llee llooww ccoosstt nnoott.. SSppeeccii ff iicc EEnneerrggyy CCoonnssuummppttiioonn ((SSEECC)) ooff PPTT SSEERRMMAANNII SSTTEEEELL ddeeccrreeaassee aabboouutt 00..77 kkWWhh//ttoonn aa mmoonntthh aaff tteerr ddooiinngg eenneerrggyy ccoonnsseerrvvaattiioonn.. TThhee mmeetthhooddss uusseedd aarree rreedduucciinngg ddeemmaanndd ccaappaaccii ttyy ((ccoonnnneecctteedd llooaadd)) aanndd iimmpprroovviinngg ppoowweerr ffaaccttoorr uussiinngg ccaappaaccii ttoorr bbaannkk.. TToottaall eenneerrggyy ssaavviinngg iiss 662255 kkWWhh aa mmoonntthh,, aanndd aavveerraaggee ssaavviinngg iinn 1100 yyeeaarrss oobbttaaiinneedd iiss RRpp.. 33..003366..551122,,-- aa mmoonntthh.. PPTT SSUURRAABBAAYYAA WWIIRREE ddeeccrreeaassee SSEECC aabboouutt 77,,3344 kkWWhh//ttoonn iinn aa mmoonntthh aaff tteerr iimmpprroovviinngg ppoowweerr ffaaccttoorr aanndd rreedduucciinngg hhaarrmmoonniicc ddiissttoorrttiioonn aass sstteeppss iinn ccoonnsseerrvveerr eelleeccttrriiccaall eenneerrggyy.. TToottaall eenneerrggyy ssaavviinngg iiss 1188..337788..773399 kkWWhh aa mmoonntthh wwii tthh tthhee aavveerraaggee ssaavviinngg ggeett iinn 1100 yyeeaarrss iiss RRpp.. 44..114422..449977,,-- aammoonntthh

REFERENCES [1] Centre for Renewable Energy Sources, “Energy Audit

Guide, Part B : Systems Retrofit for Energy Efficiency”, European Commission, Directorate General for Employment and Social Affairs, European Social Fund, Athens, 2000

[2] Dugan, Roger C, et al, “Electrical Power Systems Quality”, McGraw Hill, New York, 2003

[3] Fuller, Sieglinde K, and Petersen, Stephen R, “Life-Cycle Costing Manual, for The Federal Energy Management Program”, NIST Handbook 135, US Departement of Commerce, Washington, 1995

ISSN: 1411-1284


Energy Conservation On Hotel Building (Case Study in Bandung)

Rudy Setiabudy, Aji Nur Widyanto

Electrical Engineering Department, University of Indonesia

Depok 16424, Indonesia Email : [email protected] ; [email protected]

ABSTRACT-Energy conservation interpreted as exploiting energy the efficient, rational and effective without lessening energy needed, productivity and comfortable. Energy conservation can be done by energy audit. The scope of an energy audit, the complexity of calculations, and the level of economic evaluation are all issues that may be handled differently by each individual auditor and should be defined prior to beginning any audit activities. An energy audit can be simply defined as a process to evaluate where a building uses energy, and identify opportunities to reduce consumption. energy consumption in hotels is among the highest in the non-residential building sector in terms of absolute values (for example, 280 kWh/m2 in Greece, 420 kWh/m2 in France). In this case study, there are two opportunity of thrift able to be done by this Hotel, installation filter harmonic and degradation of energy with pay back period equal to 6 months. The total opportunity can save the cost per month equal to Rp 18.221.200,- key words : Energy, Audit, Conservation I. INTRODUCTION

ourism’s relationship with the environment is complex and may involve many activities that can have adverse environmental effects, especially

related to greenhouse gas emissions and climate change. However, tourism also has the potential to raise awareness about environmental values and can serve as a tool to finance the protection of natural areas and increase their economic importance. The two sides of this relationship can be seen in the many municipalities of the Mediterranean basin that are, to a large extent, dependent on tourism and therefore also dependent on the quality of their natural resources to attract visitors. In these communities, energy consumption in hotels is among the highest in the non-residential building sector in terms of absolute values (for example, 280 kWh/m2 in Greece, 420 kWh/m2 in France). In line with the increasing of the development followed with growth of economics of Indonesia, hence directly will bring the big challenge for supplying energy, more than else development

followed with improvement of usage energy with infinitude. Especially in town - metropolis, at this time a lot of woke up by the office block, the shopping centre and also hotel require energy big enough. Governmental therefore have more than two decade ago, blazed the way Konservasi Energi Nasional Program’s as indivisible shares from Kebijakan Energi Nasional. Energy conservation interpreted as exploiting energy the efficient, rational and effective without lessening energy needed, productivity and comfortable. II. BASIC THEORY

Energy audits can mean different things to different individuals. The scope of an energy audit, the complexity of calculations, and the level of economic evaluation are all issues that may be handled differently by each individual auditor and should be defined prior to beginning any audit activities. An energy audit can be simply defined as a process to evaluate where a building uses energy, and identify opportunities to reduce consumption. There is a direct relationship to the cost of the audit, how much data will be collected and analyzed, and the number of conservation opportunities identified. Thus, a first distinction is made between cost of the audit which determines the type of audit to be performed. The second distinction is made between the type of facility. For example, a building audit may emphasize the building envelope, lighting, heating, and ventilation requirements. An organized approach to auditing will help us collect useful information and reduce the amount of time spent evaluating your facility. By splitting the audit process into three distinct components, pre-site work, the site visit, and post-site work, it becomes easier to allocate our time for each step and leads to a more comprehensive and useful audit report. The following sections describe the tasks associated with each step of the audit process. Pre-Site Work Pre-site work is important in getting to know basic aspects of the building. This preparation will help ensure the most effective use of our on-site time and minimize disruptions to building personnel. A thorough pre-site review will also reduce the time

T

ISSN: 1411-1284


required to complete the on-site portion of the audit. The pre-site review of building systems and operation should generate a list of specific questions and issues to be discussed during the actual visit to the facility. Pre-site Tasks

1. Collect and review two years of utility energy data.

2. Obtain mechanical, architectural, and electrical drawings and specifications for the original building as well as for any additions or remodeling work that may have been done.

3. Draw a simple floor plan of the building on 8-1/2 x 11 or 11 x 17 inch paper.

4. Calculate the gross square footage using outside building dimensions multiplied by the number of stories.

5. Use audit data forms to collect, organize and document all pertinent building and equipment data

6. Develop a building profile narrative that includes age, occupancy, description, and existing conditions of architectural, mechanical, and electrical systems.

7. Calculate the Energy Use Index (EUI) in Btu/sqft/year and compare it with EUIs of similar building.

The Site Visit With pre-site work completed, we should have a basic understanding of the building and its systems. The site visit will be spent inspecting actual systems and answering specific questions from your pre-site review. Plan to spend at least a full day on-site for each building. The amount of time required will vary depending on the completeness of the pre-site information collected, the complexity of the building and systems, and the need for testing of equipment. Small buildings may take less time. Larger buildings can take two days or more. Here are some steps to help you conduct an effective audit:

1. Have all necessary tools available on site. 2. Prior to touring the facility, sit down with the

building manager to review energy consumption profiles and discuss aspects of the facility we aren't able to see such as occupancy schedules, operation and maintenance practices, and future plans that may have an impact on energy consumption.

3. Confirm the floor plan on our drawing to the actual building and note major changes.

4. Fill out the audit data sheets. 5. Look at the systems relating to the ECMs

and O&Ms on your preliminary list. 6. Take pictures as you walk through the

building. Post-Site Work

Post-site work is a necessary and important step to ensure the audit will be a useful planning tool. The auditor needs to evaluate the information gathered during the site visit, research possible conservation opportunities, organize the audit into a comprehensive report, and make recommendations on mechanical, structural, operational and maintenance improvements. Post-site work includes the following steps:

1. Immediately after the audit, review and clarify your notes.

2. Review and revise your proposed ECM and O&M lists.

3. Process your photos and paste or import pictures on 8-1 /2 x 11 inch pages.

4. Organize all charts, graphs, building descriptions, audit data sheets, notes and photos into a 3 ring binder.

Electrical System Distribution Audit The inefficient operation of electrical distribution systems stems mainly from a low power factor. Power factor correction is cost-effective when utility penalties are imposed. Low power factors can be improved with power factor correction devices and high-efficiency motors. Additional energy can be saved by installing energy-efficient transformers and replacing existing motors with small and/or higher efficiency motors, or by installing variable-speed motor drives.

a. Power Factor The total power requirement of a load is

made up of two components, namely, the resistive part and the reactive part. The resistive portion of a load can not be added directly to the reactive component since it is essentially ninety degrees out of phase with the other. The pure resistive power is known as the watt, while the reactive power is referred to as the reactive volt amperes. To compute the total volt ampere load it is necessary to analyze the power triangle indicated below:

P(W att)

Q L (VAr)

Q c(VAr)

S = P + Q L

°ϕ

Fig 1. Power Triangle

P Pcos S 2 2Q P

ϕ = =+

(1)

ISSN: 1411-1284


Q = Volt Amperes Reactive (VAR) P = Watts S = Volt Amperes (VA) Cosφ = Power Factor Amperes Factor

b. Harmonic Distortion wave (IEC 702-07-43) is

transformation a signal which not be intended and generally not realize with fundamental signal reference 50 / 60 Hz. The prime of Wave distortion for example harmonic, DC component and inter harmonic. Harmonic component or habit referred as harmonic is wave having original number fold frequency to fundamental frequency. Equation of Frequency harmonic is:

h = n x F Hz (2) with : h: -n order harmonic frequency. F: fundamental frequency of system (50Hz/6 Hz) n: harmonic order.

Fig. 2 Forming process of Harmonic Distortion

wave

Wave which is distortion formed by result of merger fundamental wave with a few harmonic wave which also in form of sinusoidal. Value of harmonic amplitude of wave is some percentage of amplitude fundamental waves.

Harmonic spectrum distribution all harmonic component amplitude as function from its harmonic order and illustrated to use histogram. At Fig. 2, can be seen example of harmonic spectrum. It can be said that spectrum represent current comparison or harmonic frequency voltage to fundamental frequency voltage or current.

Fig. 3. Harmonic Spectrum

Besides can be explained graph, wave

which distortion by harmonic earn also explained mathematically use Fourier analysis. Periodic wave which not in form of sinusoidal can be expressed as amount of expressed fundamental frequency harmonic series with the following Fourier analysis [4]:

0 0 01

( ) ( cos sin )n nn

f t a a n t b n t∞

== + ω + ω∑

(3) with :

00

1( )

T

a f t dtT

= ∫

(4) = fundamental value of f(t) to one

period that is from 0 till T

00

2( )cos

T

na f t n tdtT

= ω∫

(5) = 2 x average value of f(t) cos n to one period wave

00

2( )sin

T

nb f t n tdtT

= ω∫

(6) = 2 x average value of f (t) sin n to one period wave n = harmonic index

The Fourier equation can be used to

break distortion wave which have become waving harmonic wave and basis. This matter becomes basis in analyzing harmonic at electric power system. On the contrary if we know voltage amplitude value or current every order, hence can be obtained by total current or voltage as follows:

112rmsV V= and

112

rmsI I=

(7) max

max

22 2 2 2

1 2 31

1 1...

2 2

h

rms h hh

V V V V V V=

= = + + + +∑

(8)

ISSN: 1411-1284


max

max

22 2 2 2

1 2 31

1 1...

2 2

h

rms h hh

II I I I I=

= = + + + +∑

(9)

Total Harmonic Distortion or THD expressing distortion level generated by all harmonic components and defined as follows:

2

2

1

n

nn

M

THDM

=∞

==∑

(10) with : THD = Total Harmonic Distortion Mn = RMS value of current or voltage –n

orde M1 = RMS value of current or voltage at

fundamental frequency III. Measurement and Data

This hotel subscribe to electrics from PLN equal to 1385 kVA (tariff faction B3). From tables and curve consume the electrics during one year, average of usage kWh is around 429.180 kWh per month. At figure.1 seen the usage energy of electricity in April is highest usage around 383.046 kWh and the lowest usage in March around 265.038 kWh. For lowest billing happened in March (210 million rupiah) with usage energy of electrics equal to 265.038 kWh.

Table 1. electric bill

Pemakaian PemakaianAUGUST 2,005 309,400 73,360 382,760 252,722,665 SEPTEMBER 2,005 305,320 72,460 377,780 259,218,405 OCTOBER 2,005 278,530 64,616 343,146 231,867,505 NOVEMBER 2,005 233,770 54,718 288,488 211,957,895 DECEMBER 2,005 298,000 68,748 366,748 291,198,465 JANUARY 2,006 291,822 65,250 357,072 284,333,365 FEBRUARY 2,006 254,590 59,334 313,924 255,776,595 MARCH 2,006 214,024 51,014 265,038 210,742,695 APRIL 2,006 311,342 71,704 383,046 258,970,710 MAY 2,006 283,096 64,280 347,376 266,925,980 JUNI 2,006 297,348 69,440 366,788 283,009,865 JULI 2,006 262,924 62,176 325,100 271,952,190

BULANPEMBAYARAN

TAHUN BIAYATOTALLWBP WBP

Konsumsi Energi

-

50,000

100,000

150,000

200,000

250,000

300,000

350,000

400,000

450,000

Agst. 05 Sept. 05 Okt. 05 Nov. 05 Des. 05 Jan. 06 Febr. 06 Mar. 06 Aprl. 06 Mei. 06 Jun. 06 Jul. 06

Bulan

En

erg

i (k

Wh

)

-

50,000,000

100,000,000

150,000,000

200,000,000

250,000,000

300,000,000

350,000,000

Bia

ya (

Rp

)

Energi Biaya Fig. 3 Graphic of electric energy and bill

Measurement consumes electrics energy and

harmonic levels have been done at electrics panel in company making of bearing. Then, the data has been recorded by ”9625 Power Measurement” copied to computer personal and accessed to use ”9625 Power Measurement Support Software”.

Obtained by Measurement data for as follows:

a. Harmonic of Voltage Value of Total Harmonic Distortion (THD) of

Voltage at the panel as follows:

Table 2. Value of Total Harmonic Distortion (THD) of Voltage

VOLTAGE HARMONIC

U1 (%) U2 (%) U3 (%) Average 1.82 1.95 1.95 Maximum 2.35 2.59 2.36 Minimum 1.51 1.61 1.63 By using maximum Voltage THD equal to 3 %,

seen that voltage harmonic level at electrics panel have in tolerance range.

Hereunder represent each phase of distortion voltage wave at the electrics panel which have harmonic. Fig. 4 hereunder represents waveform noted by measuring instrument.

Fig. 4. voltage wave

2. Current and Level of Harmonic Current Value of Total Harmonic Distortion (THD) of

Current at the panel as follows:

Table 3. Value of Total Harmonic Distortion (THD) of Current

CURRENT HARMONIC

I1 (%) I2 (%) I3 (%) Average 14.04 14.04 15.15

Maximum 20.51 20.51 24.40

Minimum 8.33 8.33 8.26 By using maximum Current THD equal to 10

%, seen that current harmonic level have passed tolerance range. Fig. 5 represents each phase of harmonic distortion current wave which have and noted by measuring instrument.

ISSN: 1411-1284


Fig. 5. Current wave

Compared to voltage wave, influence harmonic

distortion current wave stronger destroying. This matter can be seen from level of harmonic current (% THD) larger than level of harmonic voltage (% THD).

3. Power Factor Value of Power Factor at the panel as follows:

Table 4. Value of Power Factor

PF1 PF2 PF3

Average 0.99 0.99 – 0.98

Maximum 1.00 0.99 – 0.99

Minimum – 1.00 0.98 – 0.97

By using minimum power factor equal to 0,85,

seen that ower level have in tolerance range. 4. Total Power

Fig. 7 represents of total power which have and noted by measuring instrument.

Konsumsi Energi

-

100.000

200.000

300.000

400.000

500.000

600.000

700.000

0:00

1:00

2:00

3:00

4:00

5:00

6:00

7:00

8:00

9:00

10:0

011:

0012

:00

13:00

14:0

015:

0016

:00

17:00

18:0

019:

0020

:00

21:00

22:0

023:

00

Waktu

En

erg

i (W

)

Harian Hari Kerja Hari Libur Fig. 6. Total power curve

From graphic, seen the result measurement

of electrics load PLN, the maximal load almost come near 770,5 kW with mean power factor equal to 0,98 hence used energy at this Hotel almost come near 786,22 kVA or to load contract PLN (1385 kVA) have weared the load equal to 56,77 %.

IV. Analyze and Solution

Identify energy conservation based on specification data of electrics installation and result data of measurement electric energy characteristic, hence some opportunities of thrift able to be done as follows :

a. Installation Filter Harmonik From data and measurement, maximum

level of current harmonic have passed the tolerance (< 10 %). To overcome the mentioned needed by usage filter harmonic. This Harmonic current height will cause some losses at equipments operation, overheating, neutral overloading, degradation lifetime equipments and improvement of consumption kwh ( current) so that to solve can be done with installation filter harmonic, thereby deductible current harmonic till under 5 %.

Based on calculation, Cost effective for external time of peak load ( LWBP)kWh/month = Rp.8.136.000,- and Cost effective for time of peak load ( WBP)kWh/month = Rp. 2.275.200,-. So the total Cost effective kWh/month = Rp. 8.136.000 + Rp. 2.275.200 = Rp. 10.411.200,- or equal Rp. 124.934.400,- per annum.

b. Degradation subscribe from PLN

Demand factor is the number show comparison between maximum electricities (peak load) with electricity attached (Subscribed kVA). Pursuant to record measurement which have been done in September 2006, we get calculation of peak load/burden operate for is 786,22 kVA. With subscribe capacity from PLN equal to 1385 kVA hence level of Demand Factor to month of September 2006 is

Demand Factor = 786 kVA

1385 kVA = 56,77 %

From data energy consumption (kWh a yearlong, seen the highest consumption kwh happened in April, is so that needed to do approach of peak load during the month of September to month April as follows :

383046786,72 797,69

377780x= = kVA

Relate at calculation of peak load (kVA), hence to determine customer capacity (KVA) can be countable by adding reserves KVA equal to 20 % from highest load (kVA), so that = 120 % x 797,69 kVA = 957,22 kVA

ISSN: 1411-1284


Degradation subscribe from 1385 kVA become 1110 kVA, hence can be done by the cost-saving kVA equal to : Cost effective kVA/month

=( 1385 – 1110) x cost/kVA = 275 x Rp. 28.400 = Rp. 7.810.000

Cost effective kVA/annum = 12 x Rp. 7.810.000 = Rp. 93.720.000

Degradation subscribe from 1385 kVA

become 1110 kVA need an investment around Rp 53.695.000,- so the simple pay back period : 53.695.000

7.810.000= = 6,875 month ≈ 7 month

V. Conclusion

1. Demand factor of electrics in this Hotel only 56,77 %

2. Peak load oh this hotel equal to 786,22 kVA

3. There are two opportunity of thrift able to be done by this Hotel, installation filter harmonic and degradation of energy with pay back period equal to 6 months.

4. The total opportunity can save the cost per month equal to Rp 18.221.200,-

References

1. Fan Wang. 2001. on Power Quality and

Protection. Göteborg : Chalmers University of Technology

2. Roger C. Dugan..[et al]. 2002. Electrical Power Systems Quality. New York : McGraw-Hill

3. Theraja B.L et.al. 1997. A Text Book of Electrical Technology. New Delhi : S. Chand & Company Ltd

4. Thumann, Albert. William J. Handbook of energy audits, Younger—6th ed.

ISSN: 1411-1284


Indonesia Energy Revolution Scenario and Projection using MESAP/Planet Simulation Model

Rinaldy Dalimi1, Bayu Indrawan2, Sven Teske3

1 Electrical Engineering Department , University of Indonesia 2 Engineering Center, University of Indonesia

Tel. 08194831323, fax. 021-7863506, email : [email protected] 3 Climate & Energy Unit Greenpeace International

Abstract – Two scenarios up to the year 2050 are outlined in this report. The Reference Scenario is based on the reference scenario published by the International Energy Agency in World Energy Outlook 2006, extrapolated forward from 2030 have been calculated using the MESAP/PlaNet simulation model. The Energy Revolution Scenario has a target for the reduction of worldwide per

capita carbon dioxide emissions to less than 1.3 tonnes per year by 2050.

Indonesia Energy Revolution Scenario describes a development pathway which turns the present situation into a sustainable energy supply. Exploitation of the existing large energy efficiency potential will reduce primary energy demand from the current 6,900 PJ/a (2004) to 9,500 PJ/a in 2050. This compares with a demand of 16,000 PJ/a in the Reference Scenario.

The electricity sector will have the strongest growth in renewable energy utilisation. By 2050, more than 60% of electricity will be produced from renewable energy sources. A capacity of 112 GW will produce 367 TWh/a of electricity. In the heat supply sector, the contribution of renewables will continue to grow, reaching more than 65% by 2050. By 2050 over 40% of primary energy demand will be covered by renewable energy sources.

Keywords - Indonesia Energy Revolution Scenario, Reference Scenario, Alternative Scenario, MESAP/PlaNet simulation model, Renewable Energy.

I. INTRODUCTION

he climate change imperative demands nothing short of an Energy Revolution. At the core of this

revolution will be a change in the way that energy is both produced and distributed. The five key principles behind this shift will be to implement clean, renewable solutions, especially through decentralised energy systems, to respect the natural limits of the environment, to phase out dirty, unsustainable energy

sources, to create equity in the use of resources and to decouple growth from the consumption of fossil fuels.

Decentralised energy systems, where power and heat are produced close to the point of final use - avoiding the current waste of energy during conversion and distribution - will be central to the Energy Revolution, as will the need to provide electricity to the two billion people around the world to whom access is presently denied.

II. BASIC THEORY

The development of future global energy demand is determined by three key factors:

• Population development: the number of people consuming energy or using energy services.

• Economic development, for which Gross Domestic Product (GDP) is the most commonly used indicator. In general, an increase in GDP triggers an increase in energy demand.

• Energy intensity: how much energy is required to produce a unit of GDP.

Both the Reference and Energy Revolution scenarios are based on the same projections of population and economic development. The future development of energy intensity, however, differs between the two, taking into account the measures to increase energy efficiency under the Energy Revolution Scenario.

III. EXPERIMENTAL RESULTS Combining the projections on population development, GDP growth and energy intensity results in future development pathways for final energy demand in Indonesia. These are shown in Figure below for both the Reference and Energy Revolution Scenarios. Under the Reference Scenario, total energy demand will more than double from the current 4,500 PJ/a to 11,300 PJ/a in 2050. In the Energy Revolution Scenario, we expect a much

T

ISSN: 1411-1284


slower increase to 6,500 PJ/a in 2050, which is about 45% more than today, but reducing by 40% the projected consumption under the Reference Scenario. The accelerated increase of energy efficiency, which is a crucial prerequisite for achieving a sufficiently large share of renewable energy sources in energy supply, is beneficial not only for the environment but also from an economic point of view. Taking into account the full service life, in most cases the implementation of energy efficiency measures saves costs compared to additional energy supply. The mobilisation of energy saving potential leads directly to a reduction in costs. A dedicated energy efficiency strategy therefore helps to compensate in part for the additional costs required during the market introduction phase of renewable energy sources.

IV. CONCLUSIONS The Energy Revolution scenario shows a pathway to a more sustainable future for Indonesia’s energy sector. Its main features are improvements in energy efficiency and a higher share of renewable energy. This report demonstrates that renewable energy is not a dream for the future – it is real, mature and can be deployed on a large scale. Decades of technological progress have seen renewable energy technologies such as wind turbines, solar photovoltaic panels, biomass power plants and solar thermal collectors move steadily into the mainstream.

REFERENCES [1] Baur, Jorg, “MESAP - Methodological Concept of

PlaNet“, Institut für Energiewirtschaft und Rationelle Energieanwendung, Stuttgart, 1998

[2] Sven Teske, “Energy Revolution – A Sustainable

World Energy Outlook”, Greenpeace International, 2007.

3a4737f339

Reference Scenario

0

2.000

4.000

6.000

8.000

10.000

12.000

2004 2010 2020 2030 2040 2050

Transport Other Sectors Industry

PJ/

a

0

2.000

4.000

6.000

8.000

10.000

12.000

2004 2010 2020 2030 2040 2050


PJ/

a

Energy Revolution Scenario

PJ/

a

0

2.000

4.000

6.000

8.000

10.000

12.000

2004 2010 2020 2030 2040 2050


PJ/

a

0

2.000

4.000

6.000

8.000

10.000

12.000

2004 2010 2020 2030 2040 2050


ISSN: 1411-1284


Ocean Wave Power Plant With Air Pressure

Massus Subekti*, Vivian Karim Ladesiψ

Fac. of Engineering, University of Indonesia, Telp. 08881547426, email: [email protected]; [email protected]

Fac. of Oceaninc, Sepuluh November Institut of Technology Telp. 08563100083, email: [email protected]

Abstrak- This Paper study about device a power plant of sea wave with pressure air, this generating advantage among other things source energi abundance especially the coastal area and archipelago area, high closeness energy and the potential area which unlimited. done wave observation of area to be used for reference, then design the effective model to convert sea wave energy become observation data input so that will be got turbine energy which couple with generator

Result of research indicate that by using calculation Boyansi, level of force float (F1) at secondary ponton have diameter 30 cm is 0.14 kN or 140 N. With l1 = 4 m and l2 = 2m, hence got force at cylinder equal to 280 N. with Wide of Output Surface 7,85.10-5 m2, Force of cylinder output is 0.27 N. Correct territorial water for the use model this conversion is coast of between 10 m until 100 m from coast lip which adapted by contour sea floor of wave and coast break condition which possible happened, for this case is coastal area Lampung. Level of Energy which convertion very hinge from high of waves that happened, so that in the end will have an in with the level of yielded electricity.

Keywords– Ocean Wave, Power Plant

I. INTRODUCTION

n this time, most of energy used come from fossil fuel, that is oil fuel, gas, and coal, used fossil fuel can

destroying the environment wich induce for green house beside that, fossil fuel is nonrenewable and unsustainable source energy category.

This problems increase complex when a lot of society not benefit from electric current, especially on purilies, isle, seaboard and not yet been reached by netork of electric are, because the location is far from electric power station. This condition of couse can pursue economic growth in this are. For this problem, alternative souce energy which reach to this area.

One of alternative souce energy the society neds is exploiting sea wave energy. The profit from this anergy is enough for supplying high density of energy, unlimited

of potential area, needn’t fuel, no pollution, renewable enrgy, more produces of energy, and cheap cost.

II. BASIC THEORY

Ocean Wave

Wave is one form of transition or switch simple energy wich that energy received from original source kinds of sea wave based on generator:

1. Bodies moving on surface make a wave at sea surface with low periode from low energy

2. Swells on the sea surface from blown wind in the sea surface

3. Tsunami from disruption of earthguake 4. The tide causing flooding from month and sun

gravitation field One of its, the important and use in this research is

swells Many types of wave in the world and difficult to

describe by matematis because unlinear, tree dimension and its for wich random. But, there are some approach can be used, among the others is Airy theory, stokes, mich, Gersner, konidal, and Single, with each other having different definition.

SWELL (linear wave) is a certain of wave as relativie have great length and small height. SWELL is represent wave was energy. Although high wave (H) more small than long wave (λ), energy from high SWELL relative. For the comparison between height wave to length wave(H/λ) ≤ 1/50.[1]

.

I

ISSN: 1411-1284


Figure 1. Sket of wave

Float Building Theory The observed rom based low of static of float building theory, because of that the se water fluid not get out from Archimedes principle where afloat buiding gotten immersed to water will experience force comparable with water volume removed, this mean weight of object float at this fluid same as water volume remove. level of force that happened at object following equqtion:

F = g V ρ [2] (1) Where : F = Force float (N)

g = Gravitation V = Volume

ρ = Specific gravity fluid While, g ρ = 10,06 kN/m3 atau 1,026 t/m3 (2) So that the level of force will influence by volume of

object float. At This research, the secunder pontoon ude is circle

form, if the assumed is pontoon have diameter (d) so pontoon volume is

Vponton = 4/3 π (d/2)3 (3)

Wind Energy Convert Large of energy was transferred by wind energy to turbine rotor shaft depend on air density, wide turbine rotor are and air speed, and can be formulated:

E = ½ m v2 [2] (4) dimana : E = energy (joule)

m = mass of air (kg) v = wind speed (m/detik)

Sum up mass which through of place by wide A m2 and move with speed m/detik, can be formulated:

m = A v q [2] (5) where : A =Wide (m2)

q = air density (kg/m3) large of power which tranfered by turbine heve

efficiency η can be formulated: [3] P = E (per et of time)

= η A ½ q v3 = η r2 ½ q v3 (kW) [3] (6) dimana : P = Forve (kW)

η = efficiency of turbine rotor r2 = wide area of turbien rotor (m2) r = radius of turbien rotor (m) q = air density (kg/m3) v = wind speed (m/det)

III. EXPERIMENTAL RESULTS Diagram Alur

Research step path in the form of diagram can be described by following :

Figure 2. Groove diagram of the research

Detailedly step of research: 1. Literature Study

Collecting circumstantial and more complete literature-literatur about theory and method which can be utilized in research

2. Models Scheme of conversion Energy Determining of model convert energy which is assuming effective and efficient

3. Determining size measure model Determining wich size measure model have making to be conducted a calculation

4. Determination result of and the compatible location Simulation/ Case study And place which is suited for application

S

H/

C λ

H

h

x

z

SWL : Still water Level

Literature Study

START

Scheme Model.

Size measure Model

Result of.

FINISH

Yes

No

?

ISSN: 1411-1284


L = 3m

P = 5 m

T = 2m

Ponton Sek

Model and the Size Measure Convert 1. Convert Energy sketch

Figure 3 Convert energy sketch ( seen from front)

2. Diagram Block

Figure 4. Block of diagram of generating device

3. Principle Energy yielded by secondary ponton from sea wave channelled to pistone through lever. Piston will depress air come into air tube 1& 2. the high pressured air used to turn around turbine which have been tied on a generator. then the Air thrown

4. Primary Ponton

Figure 5. Primary ponton dimension

Figure 6. High of Primary Ponton from sea floor

Figure 7. Place Arrangement of primary Ponton ( seen from the)

5. Secondary ponton & Activator Lever

Figure 8. Dimension of activator lever.

By using calculation Boyansi (Eq.1), level of force float (F1) at secondary ponton have diameter 30 cm (V =1.4137x 10-5 is 0.14 kN or 140 N. With l1 = 4 m and l2 = 2m, hence got force at cylinder equal to 280 N

Primary Ponton

7 m

Φ

F1

F2

4

2

piston

G

Generator

Air turbin

Air save 2

Air Save 1

ISSN: 1411-1284


F2

6. Cylinder

Figure 9. Dimension of cylinder

Diameter of piston (φ piston) = 32 cm Wide of piston surface = 0,08 m2 Diameter of cylinder output = 1 cm Wide of Output Surface = 7,85.10-5 m2 Force of cylinder output (F3) = 0.27 N Long of Cylinder = 0,5 m Air volume @ cylinder = 0,04 m3 Air volume 20 cylinder = 0,8 m3 6. Air Save Diameter = 0,25 m Long = 1 m Volume Air Save 1 &2 = 0,098 m3 Diameter output = 5 cm Surface output = 19.63 cm2

= 1.96.10-3 m2

IV. CONCLUSIONS 1. Ocean Wave Power Plant enough to be used

especially in archipelago region which not yet been touched by an electrics network

2. Correct territorial water for the use model this conversion is coastal area with distance 10 m until 100 m from coast lip which is adapted for contour of sea wave floor condition and coast break which possible.

3. Level of energy converted very depended from high of coastal wave that happened, so that in the end will have an in with the level of yielded electricity

REFERENCES

[1]. Triatmodjo, Bambang, ”Teknik Pantai”, Beta Offset, Yogyakarata, 1999

[2]. Giancoli, Daoglas, “Fisika 1”, 5th edition, Erlanggga, Jakarta, 1998

[3]. Abdul Kadir, “Energi, Sumber daya, Inovasi, Tenaga listrik dan Potensi ekonomi”, Edisi Kedua, UI Press, Jakarta, 1995

[4]. Hans Jk, “Ocean Energy Reecovery”, Honolulu. Hawai 1989

[5]. Lewis, Edwar V.. “Principles Of Navar Architecture”, Printed in The USA, Jersey City,1988

[6]. www.oceanpowertechnologies.com/pdf/senate_hearing_paper.pdf

[7]. www.Wavegen.co.uk

φ: 32cm

30cm φ:1cm

F3

ISSN: 1411-1284


T

The Time for Indonesia to Optimize Geothermal Energy

Oleh: Massus Subekti*

* Fac. of Engineering University of Indonesia Telp. 021-4712137, email: [email protected]; [email protected]

Abstract- world oil price almost 100 US dollar per barel progressively push world society to call away the attention new source energy utilize to anticipate period rare of petroleum. One of source energy which nowadays start to become world attention is geothermal, besides big enough potential and the environmental friendliness, convert goethermal become electrics own efesiency which high enough.

exploiting of Goethermal directly have reached 27.825 MWt in 71 state in the year 2004. While for the power station have reached 8.200 MWe in 21 state in the world. among other things the United States of America, French, Germany, and Indonesian.

Indonesia represent state with biggest potency geothemal in the world that is ± 40 % or equal to 27.000 MWe, but only 807 MWe or equal ± 3% have been exploited for power station. Some constraint like weakness still the infrastructure, low price sell, weakness clarity punish, governance bureaucracy which the condition still KKN, subsidize fuel oil and also weakness awareness socialize, making growth PLTP Indonesia walk very tardy.

Some step which ought to be done among other things gift of incentive to development PLTP, covering: reduction PBB, area retribution, PPH, PPN, Iease import and the licensing amenity. Revise legislation about geothermal in order to give clarity punish to all investor, reduction step by step subsidize to BBM in order to the price sell of geothemal more kompetitif, facility of research activity, seminar and discussion in around technological development of PLTP in Indonesia, and which do not less important planning inwroughtly and continual in development of national energy in order to there continuity in sill of energy of primary in Indonesia Keywords– Geothermal, Indonesia

I. INTRODUCTION The price of petroleum which keeps increasing has forced the world citizen to start

the change their attention towards the latest energy resources in order to anticipate the extinction of petroleum era. One of the energy resources which

become the attention of the world is geothermal since not only the energy that is produced can be conversed into electric energy, but its potential is also sufficient and friendly for the environment.

Several countries who have already applied geothermal energy such as the United States, England, France, Italy, Sweden, Swiss, Germany, New Zealand, Australia, and Japan

II. BASIC THEORY

According to etymological term, Geothermal is

originated from two words, Geo which means “earth” and thermal which means “heat”. Therefore geothermal means the heat that comes from the earth. According to the regulation in UU No. 27/ 2003 about geothermal, it is stated that geothermal is the resource of heat energy contained in hot water, steam and rocks along with the mineral substances and other gases which genetically cannot be separated in geothermal system and its function needs a mining process.

The movements of the earth layers which collide with each other causes the radioactive process in earth’s layers depth and make a resulting magma formation with reservoir more than 2000o C. Every year rain and snow trickles absorb into the earth layers, and are held in rocks layers which are already exposed to heat waves and magma. These rocks layers are called geothermal reservoir which range between 200o – 300o C. The water cycle which happen every year cause the reservoir rocks layers as the place to create geothermal energy which can be produced repeatedly in a very long time.

In the earth’s skin, sometimes the water stream can be very close with hot rocks where the temperature can reach until 148o C. The water does not turn into steam because there is no contact with air. When the hot water comes up into the earth surface because of a space or a crack on the earth’s skin, the hot water will come up, which is usually called hot spring. This hot spring is commonly used for hot spring pools, and many of these become a tourism object. [1]


A. The History of Geothermal Energy in Indonesia

ISSN: 1411-1284


The exploration of geothermal in Indonesia was started in 1920, Netherland built 5 shallow wells with depth of +66 meters in West Java. Out of 5 wells which were built, there is only 1 well left which still produce superheated steam, that is the locomotive well in Lumajang. In 1930, the inventory of geothermal was once stopped because of economic depression. Then, in 1974 Indonesia started its geothermal exploration again.[2]

B. The Potency of Indonesian Geothermal

Indonesia is a country with the largest geothermal potential in the world as much as 40% from the world reserve which equals with 27.000 MWe. This potential is spread especially in Sumatra, Java, Bali, Sulawesi, Nusa Tenggara Barat, and Nusa Tenggara Timur. From that potential, only 807 MWe or + 3% which is already exploited for geothermal power plant. (See Table 1)

Table 1. The Geothermal Potential in Several

Countries[3]

Exploitation No. Country Potential

(MWe) (%) 1 Iceland 5800 202 3.5 2 USA 22990 2534 11 3 Indonesia 27000 807 3 4 Philippines 4335 1931 44.5 5 Japan 20000 535.25 2.7 6 Mexico 6000 953 15 7 New Zealand 3650 435 11.99

Based on the location of Indonesia geothermal potential, West Java is at the highest rank with 40 locations and followed by NTT with 18 locations, Aceh with 17 locations, and North Sumatra, West Sumatra, and North Sulawesi with 16 locations in each. (Table 2)

Table 2. The Locations of Geothermal Potential in

Indonesia [4] No Province sum No Province sum 1 Aceh 17 14 East Java 11 2 North Sumatara 16 15 Bali 5 3 West Sumatra 16 16 NTB 3 4 Riau 1 17 NTT 18 5 Jambi 8 18 Nort Sulawesi 5 6 Bengkulu 4 19 Gorontalo 2 7 Bangka Belitung 3 20 Central Sulws 14 8 South Sumatra 6 21 South Sulws 16 9 Lampung 13 22 SouthEast Sul 13 10 Banten 5 23 Nort Maluku 9 11 West Jawa 40 24 Maluku 6 12 Central Jawa 14 25 Papua 2 13 Yogyakarta 1 26 West Kalmt 3

Total Lokasi 251 Source: The Directorate General of Minerals, Coal, and

Geothermal, ESDM

Based on the category of geothermal potential, the total amount of Indonesia geothermal resources is 14.080,5 MWe, with 9.467,5 MWe of which is speculative and the rest of 4.613 MWe is hypothetic resource. Whereas the total reserved potential is 13.060 MWe. Java is the island with the largest geothermal potential with estimated reserve reaching for 1.837 MWe with installed capacities of 785 MWe. (Table 3)

Table 3. Indonesia Geothermal Energy Potential [4]

Resource(MWe)

Reserve (MWe) Location

Speculative

Hypothetic

Estimated

Feasible

Verified

Capacities

Sumatera 5,63 2.353 5.433 15 389 2 Jawa 2.362,5 1.591 2,86 603 1.837 785 Bali-Nusa Tenggara

175 427 871 - 14 -

Sulawesi 925 125 721 110 65 20 Maluku 275 117 142 - - - Kaliman 50 - - - - - Papua 50 - - - - -

9.467,5 4.613 10.027 728 2.305 Sum 251 lokasi 14.080,5 13.060

807 MWe

27.140,5 Source: The Directorate General of Minerals, Coal, and

Geothermal, ESDM From the existed potential, there is only 2.7%

which has been exploited while 3.1% is being in the exploration phase, 30% is still in the geological survey phase, and the rest of 63,7% is still being in the preliminary survey phase. (Chart 1) [5]

Chart 1. Percentage of Indonesia Geothermal Energy Potential

C. Indonesia Geothermal Power Plant Indonesia Geothermal Power Plant Capacity

The exploitation of Indonesia geothermal energy is enhancing together with the increase of electric energy necessities. This improvement of usage is shown from the increasing electric energy capacity usage from geothermal energy as much as 32,25 MW in 1982; it increased into 142,42 MW in1990; increased into 587,5 MW in 1998 and 807 MW in 2004. (Chart 2) [6]

Percentage of Indonesia Geothermal Energy Potential

Eksploit asi

2.7%

Eksplorasi

3.1%

Survei Geologi

30%

Survei Awal

63.7%

ISSN: 1411-1284


Source : various source

Indonesia Geothermal Power Plant Capacity

32.25142.42

807

587.5

0200400

600800

1000

1982 1990 1998 2004

Year

MW

Chart 2. Indonesia Geothermal Power Plant Capacity

Current Indonesia Geothermal Power Plant

At present, there are 7 Geothermal Power Plants (GPP) which have been operating; 140 MW in Kamojang, 330 MW at Gunung Salak, 145 MW in Darajat, 110 MW in Wayang Sindu, 60 MW in Dieng (Java) and 2 MW in Sibayak (North Sumatra) and 20 MW in Lahendong (North Sulawesi) (Table 4)

Table 4. Current Indonesia Geothermal Power Plant

No Region Company Capacity (MW)

1 Kamojang Pertamina 140 2 Gunung Salak Chevron G.Salak 330 3 Darajat Chevron Darajad 145 4 Wayang Windu Star Energy 110 5 Dieng Geodipa 60 6 Sibayak Pertamina 2 7 Lahendong Pertamina 20

T O T A L 807 Source: Department of Energy and Mineral Resources

Exploitation Plan of Geothermal Power Plant (GPP)

At this moment there are 3 places which are still in the phase of exploitation plan, they are; Lahendong II GPP in North Sulawesi with 20 MW capacities which is planned to activate in the end of 2007, and Wayang Windu II GPP in West Java with 110 MW capacities which is planned to operate in 2008, and Bedugul II GPP with 175 MW capacities which cannot be confirmed when to operate yet, due to the contradiction from the residents of Bali itself. (table 5). [7], [8]

Table 5. Geothermal Power Plants in Exploitation Plan

No Name Capacities

(MW) Year

1 Lahendong II [7] (Nort Sulawesi)

20 2007

2 Wayang Windu II [8] (West Java)

110 2008

3 Bedugul (Bali) 175 ? Source: various sources

Geothermal Power Plants in Exploration Phase At present, there are 14 GPP which are still in the

exploration phase, such as; Seulawah GPP with 160 MWe and Jaboi GPP with 50 MWe in Nanggroe Aceh Darussalam, Sekincau GPP with 238 MWe in Lampung, Cisolok GPP with 45 MWe, Tangkuban Perahu GPP with 220 MWe, Gunung Tampomas GPP with 20 – 50 MWe in West Java, Ungaran GPP with 50 MWe in Central Java, Ngebel Gunung Wilis GPP with 120 MWe in East Java, Mataloko GPP with 65 MWe in NTT [9]

Table 6. List of Geothermal Power Plants in

Exploration Phase [10]

No Region Company Capa city

(MWe) 1 Seulawah NAD 160 2 Jaboi NAD 50 3 Sekincau Lampung 238 4 Cisolok West Java 45 5 Tangkuban Perahu West Java 220 6 G. Tampomas, sumedang West Java 20-50 7 Ungaran Central Java 50 8 Ngebel, gunung Wilis East Java 120 9 Mataloko NTT 65 10 Atadei NTT 40 11 Marana south-east

Selawesi 40

12 Suwana Gorontalo 65 13 Jailolo North Maluku 75 14 Songa-Wayaua North Maluku 140

T O T A L 1358 Source: www.esdm.go.id Indonesia Geothermal Energy Development Plan

Referring to the Blue Print of National Energy Management 2005 – 2025, the government has planned mixed energy which in 2003 the geothermal energy will be raised from 1.4% until 3.8% from all national energy consumption. In order to achieve that target, the government has made the Road Map of Indonesia Geothermal Energy Development 2004 – 2025 by enhancing the capacity of GPP from year to year. In 2008, it is aimed that the capacities will be installed for 2000 MW whereas in 2012, 2016, and 2020 it is aimed to reach 3442 MW, 4600 MW and 6000 MW. Therefore, in 2025, the capacity is intended to reach for 9500 MW.

Constraints in Indonesia Geothermal Development

A number of steps that should be taken for Indonesia geothermal development, such as; 1. The distance between the location of geothermal

resources and the capacity load makes the need of massive infrastructural supports is important.

2. The flaws in infrastructure, especially the transmission system that connects Geothermal Power Plant with capacity load.

ISSN: 1411-1284


3. A significant opening investment. The geothermal development needs enormous investment especially in exploration phase which affects the aspects of expense and value from the whole project, and the decide of the outcome steam price. Geothermal Energy Association (GEA), the Energy Department of the United States of America noted that making an exploration of geothermal well with the capacity of 55MWe will need investment which costs US$ 5.8 million. This includes the cost as much as US$ 432 thousand for preliminary study, US $ 1.2 million for exploration, and US$ 4.2 million for properness study.[11]

4. The fuel subsidy that makes the selling price becomes less competitive.

5. Very low selling price which is not comparable with the expended investment.

6. The requirement of advanced technology, which results in the use of geothermal exploration technology from out of the country.

7. The weakness in legal transparency that can create the possibility of contract violation.

8. Bureaucracy constraints. It is difficult to pass government bureaucracy in who is not corruption-free yet.

9. The weakness in community awareness and government role in socializing the importance of geothermal power plant development.

The Strategies of Geothermal Power Plant Development in Indonesia

A number of steps that should be taken for Indonesia geothermal development, such as; 1. The distribution of incentives toward GPP

development that include the decrease in PBB, region retribution, PPH, PPN, Import Tax and permission ease.

2. The revision of legalizations about geothermal usage in order to give legal transparency to the investors.

3. The distribution of buying guarantee after the steam is discovered. Based on the consideration that geothermal is not an exportable commodity, but it can be useful merely for domestic needs.

4. The fiscal policy of free-tax policy in order to import the operational goods (PDRI) for the necessity of geothermal effort is necessary to be given in order to support axisting contract project so the production capacity can be added in accordance with the contract commitment.

5. The periodical decrease of fuel subsidy so that the selling price of geothermal will be more competitive.

6. Facilitating researches, seminars and discussions about the development of geothermal power plant technology in Indonesia.

7. Supporting the improvement of geothermal usage for direct significance.

8. Integrated and continuous planning in improving national energy, consequently, there will be a synergy in the development of primary energy in Indonesia

IV. CONCLUSIONS

If it is viewed from the existing potential, the

factor of technological progress, the increase of world’s oil price and the efforts to reduce the glass house effects, the future of geothermal energy in Indonesia is very promising. On the other hand, considering the high cost of investment needed, it seems that the development will not be the same as expected. It needs not only concrete steps from the government but also complete support from the community so that the prospect of geothermal energy in Indonesia is compatible with the expectation.

REFERENCES

[1]. http://www.pertamina.com/index.php?option=com_co

ntent&task= view&id=3015&Itemid=340 [2]. http://www.kompas.com/kompas-

cetak/0307/14/teropong/407920.htm [3]. www.dim.esdm.go.id/makalah/2-8%20ITB-

Nenny%20M%20S.pdf [4]. Sugiharto Harsoprayitno, Peluang Panas Bumi

Sebagai Sumber Energi Alternatif Dalam Penyediaan Tenaga Listrik Nasional, Direktorat jenderal Mineral, Batubara dan Panas Bumi, Dep ESDM

[5]. http://www.majalahtrust.com/ekonomi/sektor_riil/1166.ph

[6]. www.dim.esdm.go.id [7]. www.kompas.com [8]. rafflesia.wwf.or.id.pdf [9]. www.esdm.go.id [10]. Blue Print Pengelolaan Energi Nasional 2005-2025 [11]. http://www.republika.co.id/koran_detail.asp?id=29903

2&kat_id=&kat_id1=&kat_id2=

ISSN: 1411-1284


Incorporating Energy Commodity Price Volatility in Economic Cost of Supply Analysis for Electricity

Expansion plan

Emil Elestianto Dardak*

* Fac. Of Economics, International Programme, INDONUSA Esa Unggul University Tel. (62-21)5674090 email : [email protected]

Abstract– Rising prices of tradable energy commodity is a key risk in the provision of electricity infrastructure. Indonesia is undertaking major expansion of electricity through construction of coal-fired power plants. The current abundance of coal supply, including the technology to utilize low calorie coal, is the main rationale behind the selection of coal. At the same time, the electricity utility (PLN), is faced with the need to pursue least cost expansion, and is given no clear incentive to take into consideration energy mix and security in its expansion plan. Thus, whenever coal-fired power plant can be pursued as the financially least cost expansion, any other generation options, even those promising hedge against commodity price volatility, such as renewable energy, lose out to coal. This paper seeks to conduct a cost comparison between coal-fired power plant and geothermal power as a renewable energy, in which the aspect of coal price volatility is taken into consideration. Economic cost of supply curve will be used to conduct the analysis. Keywords– energy, electricity, renewable energy, cost of supply, price volatility

I. INTRODUCTION

nergy security is becoming increasingly pertinent along with the significant hike in fossil fuel price,

particularly oil. A Presidential Decree no.5 has mandated an energy mix with more than 10% contribution coming from new and renewable energy. However, we see coal dominating the immediate expansion plan with the 10,000MW crash program. At the same time, renewable energy has grown very little. Among the most significant barriers to the implementation of renewable energy is the cost of provision. Global Environmental Facility has in the past provided incremental cost financing to provide incentives for development of renewable energy and to cover such incremental cost vis-à-vis conventional fuel

based generation. With the ratification of Kyoto Protocol, a global mechanism named Clean Development Mechanism was established to reward reduction in carbon emission achieved by substituting less clean energy technology with cleaner technology. Nevertheless, only 4 years away from the end of the Kyoto Protocol in 2012, Indonesia has not been able to utilize this mechanism to upscale its renewable generation. Geothermal, is available in huge potential in Indonesia, with an estimated size of 27GW. However, only 4 percent of this potential has been developed and only 2 projects have or will soon benefit from carbon credits, which are Lahendong and Darajat. Therefore, it can be argued that the provision of carbon credit cannot enhance viability of renewable energy to a level that makes it competitive to substitute less clean fossil fuel technology. On the other hand, the volatility of fossil fuel price has led many to question about the reliability of such generation for the longer term. This is where renewable energy is believed to possess a comparative advantage by nature, as its price is relatively stable. Taking geothermal as an example, geothermal promises stable price over the long term. However, least cost expansion planning conducted by the state power utility in Indonesia reveals bare minimum geothermal generation, as reflected in the five-year Electricity Provision General Plan. The competitiveness of renewable energy is clearly low if reference is made to this least cost planning, however, this contradicts the notion that renewable energy can become competitive with its stability of price. It is argued that the lack of quantification of such commodity price volatility hedge benefit is the main reason for its relative absence in the terms of large scale contribution for future power generation plan. It is however understandable to argue that the site-specific characteristics of geothermal project costs have made it difficult to incorporate in the least cost

E

ISSN: 1411-1284


expansion planning model. In addition, other factors may have hindered the development of geothermal, which includes the level of risk, time taken from reconnaissance to reach commissioning, and other possible factors that may affect the comparison between geothermal and other generation options. Nevertheless, it is important that quantification of the benefit of hedge against volatility can be made to allow for such incorporation in the expansion planning for power utilities.

II. BASIC THEORY

A study by Wiser and Bolinger (2004) look into the volatility of fossil fuel prices and how this enhances the competitiveness of renewable energy with its relatively certain cost of supply. Their study focuses on the volatility of gas-fired power plants in the USA, and on how to appropriately compare the levelized cost of fixed-price RE to the levelized cost of variable-price gas-fired generation. Wiser and Bolinger also point out the common current practice in analytic studies and utility planning, as previously mentioned in the case of Indonesian power utility. The practice is to compare the cost of these resources based on an inherently uncertain – and notoriously inaccurate – fuel price forecast. Wiser and Bolinger suggest that a more appropriate approach be the comparison of the levelized cost of RE to the levelized cost of gas-fired generation based on a guaranteed price of natural gas that can be locked in with forward, futures, or swap contracts. They test the reliability of such contracts as reference for long term price forecasts based on futures, forward and swap gas prices that were priced in November 2000 – November 2003. The finding shows that price forecasts are consistently made below the forward price. Hence, to lock in a coal price, ones must pay a premium. At the same time, price forecasts have been inaccurate, and often underestimated the actual price of fuel commodity. Thus, to obtain the benefit of hedge against price volatility, utility must refer to a forward price if non renewable energy or fossil fuel based generation is selected. Figure 1 shows the comparison for contract price and spot price for the Northern Appalachian prices between January 2000 and September 2002. Contract prices are generally higher than spot prices, whereby contract prices are more stable and are priced at a premium that range between $1 and $2 per short ton. In terms of percentage the premium ranges from 4% to 9%, with the exception of minor cases where spot price is higher than contract price. If we take an average of those two numbers, the base premium for locking in coal price is 6.5%.

Figure 1 – Comparison of Contract and Spot Prices for Coal

In case of renewable energy such as geothermal, the cost of fuel is not applicable since the source of energy is contained in the reservoir where it is non-tradeable and has no fluctuations of price. There is possible fluctuation in operations and maintenance cost due to inflation, but since such consideration is not made in coal power calculation, this is not going to affect the comparison results. There are indeed make up wells requirements to maintain steam supply, but this can be accounted for in the initial capital costing.

III. EXPERIMENTAL RESULTS A comparison of cost for coal power and levelized cost for renewable energy (geothermal) was made. Coal power cannot actually be levelized as cost inevitably increases along with long term fuel price increase. Geothermal energy has inherent risk and thus a low case and high case scenario were used in this calculation. The assumptions used for calculating cost of coal power plant: i) Capacity: 300MW, ii) Heating Value: 0.4 ton/MWh, iii) operations and maintenance: $30/KW, iv)capital cost: $1,000/KW, iv) lifetime: 25 years, and v) annual operation hours: 7,000 hours. The assumptions used to calculate geothermal power (low case) is i) Capacity: 300MW, ii) operations and maintenance: $18/KW, iii) capital cost: $1,800/KW, iv) lifetime: 25 years, and v) annual operation hours: 8,760 hours. The assumptions used to calculate geothermal power (high case) is i) Capacity: 300MW, ii) operations and maintenance: $18/KW, iii) capital cost: $1,800/KW, iv) lifetime: 25 years, and v) annual operation hours: 8,760 hours.

ISSN: 1411-1284


In the calculation of coal power plant cost, the coal price used is the forward coal price that is charged at an average premium of 6.5% of the spot price. The forecasted spot price is indexed to inflation at an average of 6% per year. Based on this calculation, we obtain the results where every year, coal power price must change due to change in coal fuel price. Using the above assumptions at the discount rate of 12%, the cost for first year when spot price of coal is estimated at $45/ton is 4.7 cents. It is important to note that forward price is used as volatility of price within the year can be hedged using such contract, and this makes the comparison with renewable energy more robust. The cost towards the end of the lifetime at 25 years, reaches 10.6 cents due to the increase in spot coal price at $182/ton. If we compare this with spot price based calculation, the first year cost would be 4.6 cents, and the cost towards the end of the lifetime reaches 10.1 cents, which is around 5 percent lower than the price based on forward contracts for coal that is priced at 6.5 percent premium. In the case of geothermal, the low case gives a levelized cost of 3.71 cents per kWh. Levelized cost is possibly calculated in the case of geothermal since there is no associated fuel price increase relevant to the costing. Variations in capital cost is based on geological conditions, whereby depth of well and productivity are among key factors that affect the cost of upstream activities, in addition to the number of wells and success ratio of drilling. The high case for geothermal gives a levelized cost of 6.19 cents per kWh. If geothermal discount rate is assumed at 16% to reflect the higher risk for the project, the low case changes to 5.94 cents and the high case increases to 7.86 cents per kWh. If coal power plant is calculated conventionally, using spot price at a fixed level, the levelized cost would be 4.61, in which a low case geothermal would be competitive while a high case geothermal would not be competitive when discount rate is set at 12%, which is equal to coal. A mid-range between the two cases for geothermal would also not be competitive to this levelized cost of coal. It is true that in the first year cost for coal in the previous calculation of coal based on contract price, coal is still more competitive than geothermal with the exception of low case. But along with the increase in coal price, the cost per kWh increases and makes geothermal more competitive than coal.

When the discount rate is set at 16%, a conventional calculation of coal would make both geothermal power not competitive. If we compare both cases, in the case of 12% discount rate, a high case geothermal would only be viable when the project reaches its 11th year when the coal power price reaches 6.2 cents. In the case of 16% discount rate, a low case geothermal would only be viable when the project reaches its 10th year when coal power price reaches 6 cents, and the high case geothermal would only be viable when the project reaches 18th year, when coal power price reaches 8 cents. The challenge now is in finding the levelized cost for coal power plant. Since the price cannot be levelized, there needs to be a benchmark for cost over the lifetime of the project. To use the first year figure would underestimate the cost per kWh, while using the final year figure would overestimate since such price would only be valid for the final year of operation. As a solution, the benchmark price is calculated based on the average per kWh price over the lifetime of the project. The price is also normalized with the discount rate to reflect the present value of higher power price that reflects higher purchasing power increase along with economic growth as indexed by inflation. Inflation is assumed at an average of 6 percent per annum. Interestingly, if we use inflation to normalize future price, the cost of coal power becomes lower as the increase in price is less than the discounted rate. Thus, on the basis of future paying ability increasing proportionate to inflation, an increase in coal price may be acceptable and the ascending price structure may allow for cheaper initial power price that may fit with the existing purchasing ability. The price of 1.06 cents in the 25th year actually becomes very low when present value is calculated at 6% discount rate, whereby the normalized price is 2.5 cents. Thus, when the benchmark price is calculated at the average per kWh price over the lifetime of the project, the cost is at 3.25 cents.

IV. CONCLUSIONS Renewable energy provides the benefit of energy security and hedge against price volatility. Price volatility can be contained through locking in of price using forward contract instruments. Such contract for coal would create premium at an average of 6.5% of the spot price. Thus, comparison between renewable energy and coal price must be made using this contract price, particularly since utility seek to avoid volatility in ensuring security of energy supply.

ISSN: 1411-1284


Conventional method of assuming constant price and trying to arrive at a levelized cost for coal fired power plant is certainly underestimating the true cost of provision over the lifetime of the project. Thus, comparing renewable energy with coal becomes difficult and biased towards coal that is currently priced lower in the market. The use of inflated price to reflect long term price increase has fundamentally changed the results. Conventional levelized cost would give 4.61 cents per kWh, while a levelized cost would give 4.7 cents in the first year going upwards to more than 10 cents in the end of the 25th year. Competitiveness of geothermal increases if coal price increase is taken into account. However, if coal price increase is normalized with inflation figures to account for increasing ability to pay, the increase caused by coal price can be offset by the increase in purchasing power. This notion requires careful examination and may undermine the impact in relation to environment and potential scarcity due to variability of uses other than power generation. The key finding of this analysis would remain the fact that geothermal power promises hedge, which can significantly alter the competitiveness of coal.

REFERENCES [1] Wiser, R. and Bolinger, M. (2004), The Value of

Renewable Energy as Hedge against Fuel Price Risk, World Renewable Energy Congress Proceeding, Colorado

[2] Globalcoal.com, Futures in Coal [3] Platts Coal Database

EFFECTS OF CHIMNEY DEPTH AND DOWNJET HEIGHT IN A COAL BRIQUETTE STOVE ON

THE CO EMISSION AND IGNITION TIME

Dijan Supramono*, Yulianto Sulistyo Nugroho**, and Dian Nurlita Kusuma*

*Department of Chemical Engineering, Faculty of Engineering, University of Indonesia **Department of Mechanical Engineering, Faculty of Engineering, University of Indonesia

Depok 16424, West Java, INDONESIA Phone: 021-7863516, email : [email protected]

Abstract - The utilisation of Indonesian coal as fuel has increased steadily from year to year in response to more expensive oil. Concurrently, the utilisation of coal in briquette shapes has also increased, especially by small and medium industries and low-income households. However, their increased utilisation encounters some disadvantages such as long ignition time (ignition delay) and high pollutant emissions. In terms of the CO emission, it has been found that the current briquette stoves still have high CO emission in the order of magnitude 100 ppm, which could reach 1000 ppm. These figures are well above the threshold emission value of 25 ppm for human health. Effects of the chimney depth of the coal briquette stove and downjet height installed inside the chimney on the CO emission have been investigated. Chimney is a zone above the briquette bed in the coal briquette stove. The depth of the chimney may affect the residence time of combustion gases above the briquette bed and longer chimney zone is hypothesed to allow CO to have sufficient time to convert to CO2. The downjet height is measured from the mouth of the coal briquette stove to the surface of briquette bed surface. The downjet above the coal briquette bed creates a recirculation above the briquette bed which retains combustion gases for some time in the chimney zone and enables CO to have sufficient time to convert to CO2. At the same time, unreacted hydrocarbons have also longer residence time to convert to CO2. The conditions which accommodate the conversion of CO to CO2 are encouraged which gives two benefits, lower CO emission from the briquette stoves and higher coal heat release from the briquette. Exothermic reaction from carbon in coal to CO2 is approximately three times of that from coal to CO. In this investigation, data of combustion temperature and CO emission above the coal briquette bed versus time were collected in a fire

calorimeter during the combustion of 0.8 kg coal briquettes starting from their ignition. The depth of chimney and the height of the downjet were varied respectively from 10 cm to 30 cm and from 5 cm to 20 cm. The air for combustion flowed naturally, which is also called natural draft, from the bottom side of the stove upwards. The analysis of the data found that the deeper chimney gives shorter ignition time and higher maximum temperature. However, the deeper chimney experienced steeper temperature reduction after the maximum temperature was achieved. When the maximum temperature was achieved, the CO concentration dropped. It was followed by the increase of CO concentration after the combustion gas temperature dropped. CO concentration dropped again after it reached its maximum value. Averagely the CO concentration is higher in the deeper chimney. In terms of the effect of downjet height, it was found that the higher downjet increases ignition time and maximum temperature, but reduces the average CO concentration in the combustion gases. In conclusion, the deeper chimney and shorter downjet benefit to the shorter ignition time and adversely increase the CO concentration in the combustion gases. . Keywords: Coal briquettes , Stove, Chimney, Downjet I. INTRODUCTION Statistical data published by Ministry of Energy and Mineral Resources showed that coal briquette production was predicted to increase twice in the next four years [1]. Although the utilisation of the coal briquettes has good prospect, the high CO and hydrocarbon emissions and long ignition time are still major problems. Some preliminary experiment on the related topic in the researcher’s laboratory found that the current briquette stoves still have high CO emission which could reach 700 ppm [2]. These figures are well

beyond the threshold CO emission in working areas stipulated by Minister of Workforce in 1997, i.e. 25 ppm. Human exposed to such a high CO emission for long time will have health problems because gas CO can bond haemoglobin of the blood which hinders the transport of oxygen in the blood. The effort carried out by Tekmira Bandung to reduce CO emission was by creating proper ventilation in the room where briquette stove using exhaust fan. By doing so, the gas of the product of coal briquette combustion is expelled out of the room and fresh air is drawn towards the room so that the CO content in the air drops approaching the threshold value [3]. It is well known that the ventilation system in the kitchens in most of Indonesian households is rarely equipped with exhaust fan. Without the fan, the convection of CO to the exterior air is slow and this gas can remain inside the kitchens for some time. Therefore, the introduction of a method/methods to improve the design of coal briquette stove so that the design itself controls the emission of CO and hydrocarbons. This research is concerned with how to reduce the CO gas emission which leads to achieving or approaching the threshold value of CO emission.

Emission CO originates from oxidation reactions of volatile matter as well as carbon. In order to obtain highly complete reaction towards CO2 formation 3 requirements should be fulfilled, i.e. sufficient residence time for conversion of CO to CO2, sufficient oxygen to complete oxidation reaction, and high temperature to improve reaction kinetics. The first requirement should be considered in the briquette combustion because the highest temperature of the briquette material only reaches about 700oC as a consequence of the heat absorption of the briquette combustion by high heat-capacity briquette material [3]. This temperature is much lower than that achieved by liquid or gas fuel combustion, which could reach 1200oC. These low temperatures are unfavourable for the conversion of CO to CO2 which requires high temperatures (more than 1000oC) [4]. In order to offset this disadvantage, coal briquette combustion requires much longer residence time to obtain high completion of oxidation reaction of CO to CO2, which is the controlling reaction of coal briquette combustion. The second requirement is related to the sufficient contact between hydrocarbons of volatile matter and CO gas and oxygen to react towards the CO2 formation. This is related to the mixing pattern between briquettes and air. The excessive oxygen makes the heat hardly increase the combustion gases temperature high enough to proceed the conversion of CO to CO2 and incurs partial oxidation. Products of this partial oxidation are oxygenated compounds such as -CH2O, -CHO and CO gas [5]. The first and second requirements can be enhanced if the reactions proceed at

high temperatures. Therefore, in order to reduce the emissions of CO gas and hydrocarbons, in all zones where the contact between combustion gases and oxygen in coal briquette stove occurs, the contact time is sufficiently long, the contact between oxygen and partial oxidation products is performed well, and the gas phase temperature is kept high. II. EXPERIMENTAL In this research, the first and second requirements were to be investigated by varying of chimney depth and downjet height (see Figures 1 and 2). Combustion reactions wer carried out using natural draft for air supply. Chimney is a zone above the briquette bed in the coal briquette stove. The chimney depth is measured from the mouth of the coal briquette stove to the surface of briquette bed. The hypothesis of the research is that the depth of the chimney affects the residence time of combustion gases above the briquette bed and the deeper chimney allows CO to have sufficient time to convert to CO2. The downjet is a jet installed above the coal briquette bed which directs fresh air jet downwards. The downjet above the coal briquette bed creates a recirculation above the briquette bed which retains combustion gases for some time in the chimney zone and allows CO and unreacted hydrocarbons to have sufficient time to convert to CO2. The downjet also supplies additional oxygen for the reactions of CO and unreacted hydrocarbons. The conditions which accommodate the conversion of CO to CO2 are encouraged which gives two benefits, lower CO emission from the briquette stoves and higher coal heat release from the briquette. Exothermic reaction from carbon in coal to CO2 is approximately three times of that from coal to CO. The depth of chimney was chosen 10, 20 and 30 cm by adjusting the height of grate and downjet height 5, 10, 15 cm by adjusting the length of the downjet nozzle at chimney depth of 30 cm. For each value of chimney depth or downjet height, the CO concentration in the flue gas of the stove was measured. The velocity of the air from the nozzle downjet was measured averagely 0.017 m/s. The measurement of CO concentration was performed concurrently with the measurement of temperature of combustion gases immediately above the briquette bed to obtain information of the ignition time. Using this information, effects of the chimney depth and downjet height on the ignition time, which is a measure of how difficult the coal combustion works, were obtained. The equipment and material used in the experiment are as follows: 1. Fire calorimeter The fire calorimeter facilitates temperature and CO concentration measurements. The calorimeter was built

with reference to ASTM standard E1354-1997 and ISO 5660. Figure 3. shows schematic design of fire calorimeter. A thermocouple was used to measured flue gas temperature sucked from the briquette stove. 2. Gas analyzer Quintox A gas sampling gun was inserted into the ducting to measure CO and residual oxygen concentrations in the flue gas and connected to a Gas Analyzer, Quintox –KM9106. Using data logger of the gas analyser, CO and residual oxygen concentrations can be measured at preset sampling frequency. 3. Coal briquette There were 2 coal briquettes used in the experiment, ignition-promoting briquettes and cooking briquettes. The first briquettes were used to accelerate the ignition delay of the cooking briquette combustion. They were prepared by blending the subbituminous coal powder and 15% weight of ethyl acetate and 10% cooked starch as briquette glue. The cooking briquettes were supplied from South Sumatera. The weight ratio of the first to the second briquettes was 1 : 10. The experiment was initiated by burning ignition-promoting briquettes for 10 minutes. The burnt briquettes were then laid on the top of the cooking briquettes and this was the beginning of the cooking briquette combustion. The combustion allowed to occur while CO and oxygen concentration data were logged and flue gas temperatures immediate above the briquette bed and in the calorimeter ducting were recorded.

Figure 1. Stove with chimney zone

Figure 2. Stove with downjet installed above the briquette bed III. RESULTS AND DISCUSSION 1. Chimney Depth Profile of the flue gas temperature above the briquette bed against combustion time is shown in Figure 4. At all values of chimney depth, the profiles are similar in which the temperature increases to reach a maximum value. The times when the temperatures start to increase which represents the initial time of the ignition are different for different chimney depths. The result presented in Figure 4 shows that the deeper the chimney, the earlier the ignition starts and the deepest chimney gives more abrupt increase of the flue temperature. At other depths, the increase is more gradual. In terms of the maximum flue gas temperature, the deeper the chimney, the maximum temperature is higher. The ignition is triggered by the release of volatile matters from the briquette surface. Because the ignition starts at different times among briquettes in the stove in which the briquettes close to the ignition-promoting briquettes (upper layers cooking briquettes) ignite earlier, the ignition times described in Figure 4 combine different ignition times of the briquettes in the stove. Briquette bed with the deepest chimney gives better heat propagation from the ignition-promoting briquettes to the cooking briquettes compared to others, implied by abrupt increase of the flue gas temperature. The earlier ignition time in the stove with deeper chimney shows that the stove generates earlier heat due to more oxidation reaction between oxygen and briquettes. There is a competition between the favourable condition of more natural draft due to buoyancy driving force [6] and unfavourable condition of more natural draft due to back pressure [7] in the chimney at higher temperature. It seems that back pressure is more dominant to determine the magnitude of natural draft. This is implied by the effect of chimney depth on CO concentration as shown in Figure 5.

Figure 5 shows that. CO molecules start to evolve when the ignition starts to occur and its concentration drops when the ignition reaches its maximum temperature. It shows that CO is converted to CO2 if high temperature is available in the flue gas as shown by corresponding temperature data in Figure 4. The combustion reaction of coal briquettes is initiated by fast reaction of coal to gas CO and slow reaction of CO to CO2. The kinetics of the late reaction is enhanced if high temperature prevails. The CO curve rises again after the maximum temperature passes for some time followed by the gradual reduction of CO concentration as mass of briquettes also reduces. Visually, Figure 5 shows that deeper chimney gives more CO production. This is in line with what has been mentioned in the previous paragraph, in which deeper chimney prevents the natural draft than shallower chimney. Consequently, CO concentration in the flue gas produced by the deeper chimney is more than that by shallower chimney. The average calculation of the CO emission found that the CO concentration average of the stove at chimney depths of 30cm, 20cm and 10 cm respectively are 456, 287 and 235ppm. High CO concentration measured in this experiment suggests that further investigation is required to reduce this amount. One way to achieve this may be by preheating the line of the downjet so that the air exiting the downjet is hot enough to function as additional reactant in the briquette surface and it is expected that CO concentration can go lower.

Figure 3. Schematic design of fire calorimeter

2. Downjet Height Figure 6 shows the effect of downjet height on the flue gas temperature right above the briquette bed. The graphs show that the stove with higher position of the downjet from the surface of the briquette bed, the later the ignition starts to occur. The velocity of air exiting the downjet reduces when the air flows further down away from the downjet nozzle [8]. The location of downjet near the surface of the briquette bed benefits the convection of the heat from the ignition-promoting

briquettes to the cooking briquette beneath the earlier briquettes. Therefore, the stove with lower position of the downjet gives earlier ignition. However, once ignition occurs, the heat is absorbed by the downjet air (quenching). The lower the position of the downjet, the effect of absorption is stronger, so that the maximum ignition temperature is lower. This gives consequences on the CO emission where lower maximum ignition temperature, which corresponds to the lower average flue gas temperature, results in higher CO emission as shown in Figure 7. The averages of CO emission as the downjet height is varied 10, 15 and 20 cm respectively are 747, 433 and 301ppm. The quenching whould have been avoided if the downjet air had been preheated such as by passing the line of downjet through the combusting briquette bed. IV. CONCLUSIONS The experiment has been carried out using coal briquettes to determine effect of chimney depth and downjet height on the ignition time and CO emission. The air supplied to the briquettes using natural draft. The experiment gives conclusions that 1. The deeper chimney gives earlier ignition time and

higher maximum temperature. However, due to the backpressure resulting from the higher combustion gas temperature in the more confined high chimney, such favourable temperature even gives adverse effect on the CO emission.

2. The lower downjet gives earlier ignition to the coal briquettes. However, once the ignition has commenced, the cool air exiting the downjet at lower position gives adverse effect on the combustion oxidation.

3. Further investigation is needed to improve the effect of the downjet on the CO emissions. One possible is to preheat the air in the downjet line before ejecting the air into the chimney zone.

Figure 4. Effect of chimney depth on combustion gas

temperature

Figure 5. Effect of chimney depth on CO emission in flue gas

Figure 6. Effect of downjet height on combustion gas temperature

CO Emissions at Varying Downjet Heights

0

200

400

600

800

1000

1200

0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000

Time (s)

CO

(p

pm

)

Jet 10cm

Jet 15cm

Jet 20cm

Figure 7. Effect of downjet height on CO emission in the flue

gas V. REFERENCES [1]. DESM, Planning of production and utilisation of coal

briquettes 2006-2010, Departement of Energy and Mineral Resorces Indonesia, Desember 2005

[2]. Wibowo, T. and Supramono, D., Effects of shape, size and carbonisation of coal briquettes on the ignition time, heat release and pollutant emissions, Student Final Project, 2007.

[3]. Balia, L., Research and development of coal briquettes in Indonesia, Research Centre of Mineral and Coal Technology, Departement of Energy and Mineral Resorces Indonesia, 1996

[4]. Makino, A. Drag coefficient of a slowly moving carbon particle undergoing combustion, Combustion Science and Technology, vol. 81, pp. 169-192, 1992.

[5]. Turns, S.R., An Introduction to combustion, concepts and applications, McGraw-Hill, 2nd edition, 2000

[6]. ASHRAE Handbook of Fundamentals, Chapter 26 by American Society of Heating, Refrigeration and Air-Conditioning Engineers (ASHRAE). Atlanta, GA: 2001.

[7]. B. Lewis, and G. von Elbe, Combustion, Flames and Explosions of Gases, Academic Press, Inc, 1987. [8]. Evans, D.G., Downjet combustion of coal at low burning rates, Inst. Fuel Conf., Newcastle, N.S.W., 1966,

ISSN: 1411-1284


Scale Up Reactor for Bio-gasoline Production from Crude Palm Oil

Bambang Heru Susanto, Anondho Wijanarko and Mohammad Nasikin

Department of Chemical Engineering, the University of Indonesia

Kampus UI Depok, Indonesia 16424

Abstract-The process of synthesizing bio-gasoline from crude palm oil through cracking reaction was conducted using a liquid phase batch reactor at atmospheric pressure. The reaction temperatures were between 300oC and 320oC, while the reaction time were between 1 and 2 hours for each temperature. In laboratory scale, at 300oC, H-zeolite that was prepared from natural zeolite Klinoptilolite type, was used with the catalyst/reactant (CPO) weight ratio 1:75 to produce 10,3% bio-gasoline in yield that have bio-gasoline fraction around 93.5% and octane number of 122. Catalyst was prepared by an ion exchange method using NH4NO3 aqueous solution. Addition of 5% bio-gasoline to gasoline increases the gasoline performance similar to the Pertamax.

This bio-gasoline production method was scale up to the near commercial scale and for first step, the dimension of the reactor was increased to 10 L. The reactor dimension scale up was done base on hydrodynamic calculation with the assumption of constant ratio of mixing power per volume. For better homogenized temperature and composition in the reactor, this scale up uses a helical ribbon impeller in this mixing reactor, the value of mixing time is around 10 second. For producing biogasoline that can also be used as an additive gasoline, the optimum temperature of this reaction is similar to the result in laboratory scale. This reaction produced 12.2% bio-gasoline in yield that had octane number of 106 and gasoline fraction around 34.3% (dominated with aromatic and olefin product). The octane number of this product was lower to the laboratory scale product, and for using as an additive of gasoline to Pertamax grade gasoline, it need an increasing volume of additive up to 10%.

Key word: scale up, H-zeolite, bio-gasoline, crude palm oil (CPO)

Introduction The possibilities of producing fuels

from renewable resources such as biomass are recently one the major issues in global research activities (Idem et al., 1997; Twaig et al., 1999; Katikaneni et al., 1996). These kinds of technologies are more environmental friendly and contribute effectively toward sustainable development (Yusoff, 2004). Vegetable oils have been the main subject of these researches (Twaig et al., 1999). Vegetable oils mostly consist of triglyceride, a hydrocarbon molecules formed when a propane molecule bonded with three fatty acids at each of the carbon atoms. This molecule

has a quite similar structure with the hydrocarbons in crude oil. Therefore it is a very potential alternative for the crude oil in producing hydrocarbon fuels. Among the vegetable oils, palm oil is known to be one of the most potential alternatives because of its high amount of production, especially in tropical country like Indonesia and Malaysia (Yusoff, 2004).

Bio-gasoline was one of renewable hydrocarbon fuel product from vegetable oil such as crude palm oil (CPO), and have characteristic and composition is similar to the commercial gasoline and may be used as commercial additive to increase the octane number of premium gasoline and an attempt to produce these bio-gasoline from palm oil through cracking reaction like commonly done in crude oil refinery recently have not been done. This was one of the reasons to do this research. This research was done to study the possibility of applying the cracking reaction, in the process of liquid fuel production from palm oil, using cracking catalyst in a different reaction condition than the common condition applied in crude oil refineries. In refineries, the cracking reaction is commonly done in gas phase reactor with high pressure. In this research, the reaction was done in liquid phase reactor with the reaction temperature below the initial boiling point of palm oil and in atmospheric pressure.

Some researches have been done to produce liquid fuels from palm oil. Bhatia, et al have proved that catalytic cracking of palm oil using HZSM-5 catalyst at 350oC could produce light products such as methane, ethane, gasoline, kerosene, diesel oil, and BTX (benzene, toluene, xylene) (Twaig et al., 1999). He also proved that liquid fuels can also be produced from palm oil-based fatty acid mixture which is the waste from palm oil industry and also from used cooking oil (Ooi et al., 2004a; Ooi et al., 2004b). Other attempts with different researchers, different methods, and different catalysts have also been done to utilize the palm oil in producing liquid fuels efficiently (Idem et al., 1997; Twaig et al., 1999; Katikaneni et al., 1996; Yusoff, 2004; Ooi et al., 2004a; Ooi et al., 2004b; Twaig et al., 2003a; Twaig et al., 2003b; Twaig et al., 2004, Wijanarko and Nasikin, 2007). This research was

ISSN: 1411-1284


design to produce liquid fuels from CPO by catalytic cracking using H-Zeolite at 300oC for producing the renewable fuel in gasoline range.

Material and Method The catalyst used for this experiment

was a natural H-zeolite, that was prepared from Indonesian Lampoong natural zeolite, mostly consist of clinoptilolite type zeolite, by ion exchange method. For 25 kg H-zeolite preparation, natural zeolite was processed around 50 h at 500 rpm mixing rate using ammonium nitrate solution (1 N) as the source of proton. thereafter this produced NH4-zeolite form was dried at 110oC around 4 h and then calcined at 520oC for 5 hours to omit NH3 species for produce H-Zeolite. Then, produced H-zeolite was characterized by BET to find the catalyst pore diameter.

The laboratory scale process of synthesizing bio-gasoline from crude palm oil through cracking reaction was conducted using a liquid phase batch reactor with stirrer and heating jacket at atmospheric pressure, the reaction temperatures were between 300oC and 320oC, while the reaction time were between 1 and 2 hours for each temperature. In laboratory scale, at 300oC, it used catalyst/reactant (CPO) weight ratio 1:75. The reactor was also equipped with a reflux tunnel in order not to make it over-pressured without losing any light hydrocarbon products formed in the reaction. The support of the catalyst was prepared by an ion exchange method using NH4NO3 aqueous solution. The density and boiling point of the reaction products were measured to evaluate the reaction. The

products were also analyzed using FTIR spectrometry and GCMS. To obtain the bio-gasoline, sample of the reactor product was distilled in an atmospheric distillation column. The distillations were done twice to get light hydrocarbons with as high concentration as possible. An addition of 5% bio-gasoline that was obtained from previous research result was capable to increase gasoline performance similar to commercial Pertamax. The schematic illustration of this laboratory scale process was shown in Fig. 1.

The pilot scale process of synthesizing bio-gasoline from crude palm oil through cracking reaction was conducted using a liquid phase batch reactor with double helical ribbon stirrer and heating jacket at the similar condition to laboratory scale condition. To obtain the products, sample of the reactor product was also purified with the similar process of laboratory scale process. The products were also analyzed using FTIR and GCMS. The reactor design illustration of this pilot scale process including its detail sizing was shown in Fig. 2.

Result and Discussion To evaluate the cracking of triglyceride

and to ensure whether cracking really occurred, the FTIR spectra of one of the reactor products is compared to the FTIR spectra of palm oil before the reaction. To do the comparison, the absorbance peak of C=O bond in the range of 1740-1745 cm-1 is used as reference because the amount of this bond did not significantly changed during reaction (Silverstein and Webster,

Fig. 1. Schematic illustration of the experimental steps.

1997). Fig. 3 shows that relative to the reference peak, the peak at 1161 cm-1 in the spectra of reactor product is dissapeared which was identified in the spectra of palm oil (not shown). It means there was C–O group during the reaction was cracking. The peak at 3000 cm-1 that was dominant in the spectra of palm oil (not shown), have a significant decreasing after

catalitic cracking, and this indicated that the decreasing of carbon atom in CPO molecular structrure was occured. Although at interval 800

ISSN: 1411-1284


Fig. 2. Batch Reactor Design using Helical Ribbon

Impeller Type (measurement unit in cm)

– 1000 cm-1 and also at 700 cm-1 was seem as a little peak in FTIR result of bio-gasoline product, it was indicated that formation of double bond fungsional group and aromatic group during cracking reaction was occured. The other results in pilot scale and also result in smaler case were same to the above explanation. To identify the hydrocarbon content of the bio-gasoline, GC-MS analysis was conducted. GC-MS analysis was also used to identify the hydrocarbon content of palm oil before the reaction for comparison. The

results of the analysis are shown in Table 1. Bio-gasoline was one of renewable hydrocarbon fuel product from vegetable oil such as crude palm oil (CPO), and have characteristic and composition is similar to the commercial gasoline and may be used as commercial additive to increase the octane number of premium gasoline. The yield of cracking reaction in producing gasoline-grade hydrocarbon is defined as follow:

%100%)( ×=P

YVolYield (1)

where P is the palm oil feed volume and Y is the gasoline-grade hydrocarbons volume in the reactor products. The process of synthesizing bio-gasoline from crude palm oil through cracking reaction was conducted using a liquid phase batch reactor at atmospheric pressure, the reaction temperatures were between 300oC and 320oC, while the reaction time were between 1 and 2 hours for each temperature. In laboratory scale, at 300oC, a H-zeolite was used with the catalyst/reactant (CPO) weight ratio 1:75 to produce 10,3% bio-gasoline in yield that have gasoline fraction around 93.5% and octane number of 122. An addition of 5% bio-gasoline that was obtained from laboratory scale research result was capable to increase premium gasoline performance similar to commercial Pertamax.

Fig. 3. FTIR result of Bio-gasoline Product at 300oC.

This bio-gasoline production method was developed from laboratory scale up to the near commercial scale and for first step, the dimension of the reactor was increased to 10 L and it would be increase to higher volume. The reactor dimension scale up was done base on hydrodynamic calculation with the assumption

of ratio of mixing power per volume is constant. For better homogenized temperature and composition in the reactor, this scale up found that using helical ribbon impeller in this mixing reactor, the value of mixing time is around 10 second.

ISSN: 1411-1284


Tabel 1. GC-MS Analysis Results Comparison of the CPO and bio-gasoline.

% Berat Scale Up 10 L

Komponen Skala Labarotarium 300 oC 310 oC 320 oC

Parafin/olefin C1 – C4 6,13 9,44 5,9 4,32

C5 6,10 0,38 0,22 0,23 C6 1,76 0,41 1,29 1,13 C7 0,84 4,30 5,27 6,78 C8 52,63 1,12 11,62 7,05 C9 3,09 5,40 9,42 4.57 C10 7,25 16,87 12,33 14,77 C11 15,68 1,53 16,98 6,41 C12 - 2,21 3,73 2,15 C13 - 3,28 4,06 2,41 C14 - 2.84 4,05 4,88 C15 - 14.10 6,71 7,5 C16 1.55 23,2 5,39 21,65 C17 - 5,45 1,81 1,81 C18 4,97 4,65 2,01 4,32 C19 - 0,15 0,09 3,69 C20 - 2,61 - -

Aromatik C6 - - 0,13 0,46 C7 - - 0,88 0,62 C8 - - 0,43 0,67 C9 - - 2,21 0,14 C10 - 0,15 1,35 0,45 C11 - 0,39 2.39 1,59 C12 - 0,44 1,21 0,55 C13 - 0,76 0,64 0,66 C14 - 0,32 - 1,19 C15 - - - - C16 - - 0,18 -

Physical Properties

Octane Number 122,4 106 98 98

Density (g/cm3) 0,77 0.84 0.81 0.84

Viscosity (poise) 0,054 0.0048 0.0167 0.0359

RVP (kPa) 48,32 - - -

Distillation Temp. (oC) 255 280 280 280 Yields (%) 10,3 12,2 14,2 8,4

Note : Grey boxes are bio-gasoline fractions ( around 93,48; 34,28; 70,28; dan 49,42% were respectively bio-gasoline fraction in laboratory and pilot scale that was done at 300; 310; dan 320 oC) .

For producing an additive to premium gasoline, the optimum temperature of this reaction is similar to the result in laboratory scale, it was found at 300oC (catalyst/reactant weight ratio 1:75 and reaction time 1 hour). This reaction produced 12.2% bio-gasoline in yield that had octane number of 106 and gasoline fraction around 34.3% (dominated with aromatic and olefin product). The octane number of this product was lower to the laboratory scale product, and for using as an additive of premium gasoline to Pertamax grade gasoline, it need an increasing volume of up to 10%.

Conclusion Using H-zeolite catalyst, this process

had succeeded in synthesizing gasoline-grade hydrocarbons (bio-gasoline) through cracking reaction even in atmospheric pressure. The bio-gasoline produced contains C8 to C12 and also aromatics, which also composing the gasoline from crude oil, with the yield and octane number of 11.93% and 122.4 in laboratory scale and 12.2% and 106 in pilot scale at the same process condition. Addition of 5% bio-gasoline to gasoline was capable to increase premium gasoline performance similar to commercial

ISSN: 1411-1284


Pertamax. However, for using as an additive gasoline to Pertamax grade gasoline, it needs an increasing volume of additive up to 10%.

Reference Idem, R.O., Katikaneni, S.P.R., Bakhsi, N.N.,

1997. Catalytic Conversion of Canola oil to Fuels and Chemicals: Role of Catalyst Acidity, Basicity, and Shape Selectivity on Product Distribution. Fuel Process. Tech. 51, 101 - 125

Katikaneni, S.P.R., Adjaye, J.D., Idem, R.O., Bakshi, N.N., 1996. Catalytic Conversion of Canola Oil over Potassium-Impregnated HZSM-5 Catalysts: C2-C4 Olefin Production and Model Reaction Studies. Ind. Eng. Chem. Res. 35, 3332 - 3338

Ooi, Y.S., Zakaria, R., Mohamed, A.R., Bhatia, S., 2004a. Catalytic Conversion of Palm Oil-Based Fatty Acid Mixture to Liquid Fuel. Biomass Bioenerg. 27, 477 – 478

Ooi, Y.S., Zakaria, R., Mohamed, A.R., Bhatia, S., 2004b. Catalytic Cracking of Used Palm Oil and Palm Oil Fatty Acids Mixture for The Production of Liquid Fuel: Kinetic Modelling. Energ. Fuels 18, 1555

Silverstein, RM., Webster FX., 1997. Spectrometric Identification of Organic Compounds, 6th Ed. John Wiley & Sons, Inc., New York

Twaiq, F.A., Zabidi, N.A.M., Bhatia, S., 1999. Catalytic Conversion of Palm Oil to

Hydrocarbons: Performance of Various Zeolite Catalysts. Ind. Eng. Chem. Res. 38, 3230 - 3232

Twaiq, F.A., Zabidi, N.A.M., Mohamed, A.R., Bhatia, S., 2003a. Catalytic Conversion of Palm Oil Over Mesoporous Aluminosilicate MCM-41 for The Production of Liquid Hydrocarbon Fuels. Fuel Process. Tech. 84, 105 - 109

Twaiq, F.A., Mohamed, A.R., Bhatia, S., 2003b. Liquid Hydrocarbon Fuels from Palm Oil by Catalytic Cracking over Aluminosilicate Mesoporous Catalysts with Various Si/Al Ratios. Micropor. Mesopor. Matl., 64, 95 - 96

Twaiq, F.A., Mohamed, A.R., Bhatia, S., 2004. Performance of Composite Catalysts in Palm Oil Cracking for the Production of Liquid Fuels and Chemicals. Fuel Process. Tech. 85, 1283 – 1287

Wijanarko A. and M. Nasikin, Biogasoline production from palm oil by cracking using NiMo/Zeolite Catalyst, Japan-Indonesia Bilateral Symposium on Sustainable Engineering, Yogyakarta, Indonesia, May 14th, 2007

Yusoff, S., 2004. Renewable Energy from Palm Oil – Innovation on Effective Utilization of Waste. J. Cleaner Production. 14, 87 – 93

ISSN 1411-1284


Flame Lift-up on A Bunsen Burner; A Preliminary Study

Cokorda Prapti Mahandari* and I Made Kartika Dhiputra†

†Engineering Faculty University of Indonesia Kampus UI Baru 16424

Tel. 330188, fax. 3918115 email:[email protected]*

email:[email protected]†

Abstract-Some research on ring stabilizer for lean premixed turbulent flame has been done. In those researches the ring mounted flush with the exit burner port. This basic research also employed a ring which was incorporated above the tube by a ring adjuster. Propane and air flow rate were metered, mixed and introduced to the bottom of Bunsen burner tube. Propane flow rate kept unvarying and air flow rate was increased gradually. On some value of air flow rate, flame would jump to the ring and it did not attach to the exit tube any longer. This phenomenon was called flame lift-up.

This experiment was done on several variation of ring position. Air Fuel Ratios (AFR) for which the flame lift-up phenomena took place was calculated. Furthermore experiments without a ring have also conducted to compare the flame characteristic mainly the AFR for blow-off condition.

It was found that flame lift-up correlate interchange to the AFR. For the impact of ring position, it was found that there was a particular position that made flame lift-up stable very best. AFR for flame lift-up was in reversed to the position of the ring. Experiments without ring obtain that value of AFR for flame lift-up was above the blow off limit of flame without ring. In addition, from the Fuidge diagram it was found that ring did operated as a flame stabilizer. Keywords: Flame, lift-up, Bunsen, ring

I. INTRODUCTION

nderstanding of the propagation and stability of premixed flames not only is important for

fundamental combustion research but is also relevant to the development of new combustion technologies such as those associated with low-NOx emission, lean burn, micro-scale combustion, and material synthesis. In practical burners there are two important design criteria that are avoided, lift-off and flashback. Lift-off is the condition where the flame is not attached to the burner tube or port but it is stabilized at some distance from the port. Flash back is the occurrence

when the flame enters and propagates through the burner tube without quenching. Flame lifting, as the result of lift-off condition, is generally unfavorable for a number of reasons. First it increases the escape of unburned gas or incomplete combustion. Secondly, above the lifting limit, ignition becomes intricate. Furthermore accurate control of the position of a lifted flame is difficult to manage so that it resulted in a poor heat transfer characteristic. Lifted flames can also be noisy [1].

However in some situations lifted flames are encouraged as it has been shown that lifted flames can reduce NOx levels in furnaces [2]. Damage on the nozzle tip of burner that had occurred very often also attracted the application of flame lifting. Concerning on flame stability of lean premix combustion, lift-off phenomena was took place at the most lean mixture. There are two mainstream researches on lift-off conditions: keeping the flame stable at lift-off condition and reattach the flame to the burner. Maintaining flame on lift-off condition was achieved by employ bluff-body, ring stabilizer etc [3-5]. Reattach the flame to the burner was achieved using a pulsed high voltage discharge [6]. This research was in the effort of maintaining the lift-off condition stable by introducing ring stabilizer. Research on ring stabilizer so far was putting a ring exactly on the tube of the burner [5]. This research was also using a ring that introduce above the Bunsen burner tube. This work however has to be considered as a first initiative.

A quite similar phenomenon with lift-off is

introduced. As lift-off condition keeping the flame above the tube burner or port freely, this phenomenon was retained the lifted flame using ring. The occurrence when the flame jump to the ring was named lift-up. As a preliminary study, flame lift-up would be investigated on a Bunsen burner. The tube was 14 mm diameter and 38 cm height. A ring with outside diameter 30 mm, inside diameter 10 mm and thickness 6 mm was placed above the tube burner. Flow rate of fuel, in this case was propane and air were metered using rotameters whenever lift-up happened. The rotameter had been calibrated using a wet gas meter. 8 position of ring from the exit tube

U

ISSN 1411-1284


were investigated. Air Fuel Ratio and burning load was calculated than they were plotted on graphics to get the Fuidge diagram. Understanding this phenomenon would inspire the application of flame lift-up in the future.

II. BASIC THEORY

There are many definitions of combustion. It involves fuel and oxidizer. Fuel is any substance that releases energy when oxidized. Oxidizer is any oxygen-containing substance that reacts with fuel for example air. Complete combustion can be achieved by mixing the fuel and the oxidizer. To characterize the fuel-air mixture some relevant relations are needed. Air Fuel Ratio or AFR is defined as [7]:

fuelfuel

airair

fuel

air

nM

nM

m

mAFR == (1)

It is expressed in terms of mass (m), molecular weight (M) and number of mol (n). The equivalence ratio of a air-fuel mixture is defined as

actualAFR

AFR tricstoichiome=Φ (2)

One method to get cleaner combustion is lean combustion or Φ<1. Working with lean mixtures is advantageous from a pollution point of view as low flame temperature implies low potential for NO generation. However the flame may become too slow and hence prone to extinction or blow-off. Blow-off is the condition when the burning velocity is less than the mixture velocity. Simple design principle of the Bunsen burner has been incorporated in many gas appliances such as cooking stoves and gas burner. Premixed flame from Bunsen burners are relatively clean and give more intense combustion with higher effective temperature.

Burning load (BL) is defined as the power of combustion per area of burner. The equation for burning load is

A

x HVVBL fuel=

(3)

BL is burning load, W/m2 Vf is fuel flowrate, m3/s HV is heating value of fuel, J/m3 A is area of burner, m2

Flame will be unstable if the reactant flow rate is

too high. This then results in the flame’s inability to maintain its position, and blow-off occurs. The entrainment of air from the surroundings will then weaken the mixture. Flame stability can be best

described by a Fuidge diagram. Such a diagram is reproduced in Figure 1 [8].

Figure 1. Fuidge Diagram

III. EXPERIMENTAL RESULT

Flame lift-up phenomenon was initially

discovered on ring stabilizer experiment. Ring was placed above the Bunsen burner tube. Fuel flow rate was fixed and air flow rate was varied until flame jumped from the exit tube to the ring and ‘sit’ on the ring as is shown on Figure 2.

Figure 2. Flame lift-up on Bunsen Burner

This flame was completely different from the

flame on Johnson, et al research. They put a ring exactly on the exit tube [5].

ISSN 1411-1284


AFR vs ring position on flame Lift-up

25.00

30.00

35.00

40.00

45.00

50.00

55.00

0 5 10 15 20 25 30 35 40 45

Ring position (mm)

AF

R b

y vo

lum

e5 mm

10 mm

15 mm

20 mm

25 mm

30 mm

35 mm

40 mm

Figure 3. Variation of AFR flame lift-up on ring position

AFR vs Burning Load flame lift-up

0

10

20

30

40

50

60

5725 6327 6930 7532 8135 8738

Burning Load, kJ/m2.s

AF

R b

y vo

lum

e

'30 mm'

'40 mm'

'20 mm'

'blow-off'

'blow-off without ring'

Figure 4. AFR vs Burning Load flame lift-up

AFR and burning load when propane lift-up

happened was calculated based on Equation 1 and Equation 3. The experiment result is presented on

Figure 3 and Figure 4. AFRst by volume of propane or C3H8 is 23,81 [8]. It means that the combustion was

ISSN 1411-1284


on lean combustion as the minimum AFR was around 40.

Figure 3 presents AFR of flame lift-up for each

position of ring from the exit tube. 8 position of ring which were 5mm, 10 mm, 15mm, 20 mm, 25 mm, 30 mm, 35 mm and 40 mm from the exit tube were done. It is shown that the AFR for flame lift-up was depend on the position of the ring. As ring position is higher, the AFR decrease slightly. It means that flow rate of air must be higher to bring the flame ‘sit’ on the ring. Therefore flame lift-up is considered to be an equilibrium condition of burning and mixture velocity. Within the same position there was a certain range of AFR. Among the eight positions it was on 5 mm and 40 mm that lift-up found to be quite difficult. The wider AFR range implies that flame stability area was also wider. Ring position of 20 mm and 25 mm was the most stable among others. It indicates there are other reasons influence the lift-up phenomenon. It could be the turbulence and the separation flow of propane-air.

Figure 4 described the Fuidge diagram for lift-up

flame. Compare to Fuidge diagram for methane [7], the AFR for propane was very much higher. Note that AFRst methane itself was lower than AFRst of propane. In case of propane, it was 23.18 while methane was only 9.524. The lift-up curve was above blow-off curve without ring. In this condition ring was indeed considered as a flame stabilizer.

IV. CONCLUSION As a preliminary study, the stability of flame lift-

up phenomenon was investigated. It was found that lift-up will occur on lean combustion of premixed flame. From Fuidge Diagram flame, it was revealed that flame lift-up area is above and blow-off area without ring. Position of ring has been discovered influenced very significant.

Further study on geometric of ring as a blockage area should be deliberated. Temperature between the

exit tube and the ring becomes very striking to explore if this phenomenon will be implemented practically.

REFERENCES

[1] Turns, Sthepen R “ An Introduction to Combustion Concepts and Applications” Mc GrawHill International, Singapore, 1996

[2] Chao, Y.C., Yuan, T., and Tseng, C.S., Effects of flame lifting and acoustic excitation on the reduction of NOx emissions, Combust Sci. and Tech., vol. 114, pp. 49-65, 1996

[3] Andres A. Chaparro and Baki M. Cetegen, Blowoff “Characteristics of Bluff-body stabilized conical premixed flames under upstream velocity modulation,” Combustion and Flame Vol 144, Issues 1-2, pp. 318-335, 2006

[4] A. Kempf, R.P Linddtedt and Janicka, “ Large Eddy Simulation of bluffbody stabilized nonpremixed flame”, Combustion and Flame Vol 144 Issued 1-2, pp. 170-189, 2006

[5] M.R. Johnson, L.W Kostiuk, R.K. Cheng ‘ A Ring Stabilizer for Lean Premixed Turbulent Flames’ Combustion Group, Energy & Environment Division, Lawrence Berkeley Laboratory, Berkeley, California, 94720

[6] K. Criner, A. Cesson, J. Louiche, P. Vervisch, ‘ Stabilization of Turbulent Lifted Jet Flames Assited by the Pulsed High Voltage Discharge ‘ Combustion and Flame 144, pp. 422-425, 2006

[7] A C McIntosh, “Laboratory Manual, MSc Course in Combustion and Energy”, University of Leeds, 1997

[8] http://www.geoff-jones.co.uk/downloads/chemistryassignmentreport laminarpremixedflamestabilityandflammabilityandburnerdesign_msc.pdf, 11 November, 2007

[9] Drysdale, Dougal, An Introduction to Fire Dynamics, John Wiley & Sons, England, 1998

ISSN: 1411-1284


Study of Biodiesel of Coconut and Corn Ethyl Ester Use by Processing with the Processor Series Type

In Diesel Engine Performance

Bambang Sugiarto, Sanggul H. Siregar, Yanuar Chandra Fac. of Engineering, University of Indonesia, Jl. Salemba Raya 4, Jakarta 10430

Fac. of Engineering, Tel. 330188, fax. 3918115 email : [email protected]

Abstract– Biodiesel is the one of the alternative energy which can be renewed and environmental friendly. Indonesia has a big potency to develop and use biodiesel as a Diesel fuel because there is much kind of plantation resources in it. The process of biodiesel can be conducted with process of transesterification. In this experiment Biodiesel was processed from Coconut Oil and Corn oil by using processor series type. The performance test was conducted on Diesel Engine Research and Test Bed with Nissan type SD 22 engine without any modification.

The fuel are mixing between Diesel fuel and biodiesel was variated at biodiesel contain 5 %, 10 % and 20 %. The engine speed changing are 1300, 1500, 1700 and 1900 rpm while the throttle valve open in 30 %, 40 %, 50 % and 60 %. The testing result showed that the opacity value decrease when using these biodiesels.

The result also showed that biodiesel from coconout oil and corn oil can increase the thermal efficiency, brake horse power and decrease specific fuel consumption of Diesel engine test especially in variation of throttle valve open and variation of speed engine charge. Generally, from two kinds of biodiesel (coconut and corn ethyl ester), coconut ethyl ester with contain 20 % mix with Diesel fuel has the best resulted.

Keywords : Biodiesel, transesterification, emission.

I. INTRODUCTION

ncreasing of resident growth accompanied with the expanding of industries and transportation

which is the biggest user of fuel fossil. Other side, the an effect of this situation will make sources of fuel fossil sharply decrease and in a moment will finished [3]. Ironically, the situation is not followed with the degradation of fuel usage in many countries.

One way to solve this problem is improvement and exploiting of biodiesel. To obtain biodiesel, the vegetable oil or animal fat is subjected to a chemical reaction termed transesterification [6]. In that

reaction, the vegetable oil or animal fat is reacted in the presence of a catalyst (usually a base) with an alcohol (ethanol or methanol) to give the corresponding alkyl esters (or for methanol, the methyl esters). Figure 1 depicts the transesterification reaction [2].

Figure 1 : Transesterification Reaction

Biodiesel can be produced from a great variety of

feedstocks. These feedstocks include most common vegetable oils (e.g., soybean, cottonseed, palm, peanut, rapeseed/canola, sunflower, safflower, coconut, corn) and animal fats (usually tallow) as well as waste oils (e.g., used frying oils). The choice of feedstock depends largely on geography. Depending on the origin and quality of the feedstock, changes to the production process may be necessary.

Biodiesel is miscible with diesel oil in all ratios. In many countries, this has led to the use of blends of biodiesel with petrodiesel instead of neat biodiesel. Blending with diesel fuel can be denoted by acronyms such as B5, B10 and B20, which indicates a blend of 5%, 10 % and 20% biodiesel with diesel oil [5]. In this experimental, ethanol (pyroxylic spirit) is used for producing biodiesel of Coconut Oil and Corn Oil. Biodiesel has many advantages as follows : • It is renewable. • It is energy efficient. • It displaces petroleum derived diesel fuel. • It can be used in most diesel equipment with no or

only minor modifications. • It can reduce global warming gas emissions. • It can reduce tailpipe emissions, including air

toxics.

I

ISSN: 1411-1284


• It is nontoxic, biodegradable, and suitable for sensitive environments.

Blending of B5, B10 and B20 used in this

experimental without any engine modification. Properties of fuel used :

Standard Corn Ethyl Ester

Coconut Ethyl Ester

Viscosity at 40oC 2.3-6 mm2/s 4.869 2.656 Density at 15oC 0.85-0.90 0.8886 0.868 Flash Point Min 100oC 180 104 Cloud Point Max 18oC 11 -5 Pour Point - oC - 3 10 Water content Max 0.05 % 0.02 0.02 LHV kJ/kg 37500 39500 Cetane Number Min 48 50.27 63 Free glycerol Max 0.02 % 0.0135 0.002 Total glycerol Max 0.24 % 0.3235 0.043 TAN, mg KOH/gr Max 0.8 0.2168 0.17 Ester content Min 96.5 % 99,1569 99.1676

Source : BPPT, Measured by ASTM D Experimental Procedure

In the first case, tests of engine performance and exhaust gas opacity on diesel oil alone were conducted as a basis for comparison. In the second phase, testing with B5, B10 and B20 of Coconut ethyl ester and Corn ethyl ester. The experiment were conducted at variation of speed 1300, 1500, 1700 and 1900 at uniform throttle valve open 40 % and variation of throttle valve open 30%, 40%, 50% and 60% at the rated speed 1500 rpm.

The engine specification used in this experimental are given in the table below :

Model SD-22 (Nissan Motor Co, Ltd) Type Water cooled 4 cycle, diesel

engine Number of cylinder 4 Diameter of cylinder

83

Stroke 100 mm Capacity 2163 cc Compression ratio 22 : 1 Output power 47 PS Speed 3200 rpm (max)

II. BASIC THEORY

The Process of Biodiesel

In this experimental, biodiesel of Coconut Oil and Corn Oil was produced by transesterification process with ethanol (spritus) by using processor series type as shown in the figure 2 :

Figure 2. Processor series type BDP-10FG-BV

Processor specifications : MODEL BDP-10FG-BV TUBE Material Thick

Fiber glass 1,5 mm

Production Metode Batch Production Capacity 10 liter/batch ElLECTRICAL Voltage Frequency Consumption

220V AC 50 Hz 350 Watt

DIMENSION High Width Length

140 cm 25 cm 52 cm

Duration of process 21 hours

Engine Performance Some parameters noted during experiment and

used as a raw data. Result of the testing datas shown with the parameters [8] [1] of fuel consumption (FC), specific fuel consumption (SFC), brake horse power (BHP), thermal efficiency (ηth) and exhaut gas opacity opacity of B5, B10, B20 of Coconut ethyl ester and Corn ethyl ester. Furthermore, it will be compared with diesel fuel as a comparator. Following will be elaborated its calculation method [8] : • Fuel Consumption Fuel Consumption per set of time as shown with

equation below :

t

VFC g.3600

= [L/hr] (1)

• Brake Horse Power (BHP)

ISSN: 1411-1284


Measurable motor energy output that is transmissed through the shaft.

60.1000...2 Tn

BHPπ= [kW]

(2)

• Specific Fuel Consumption (SFC) Fuel Consumption per Brake Horse Power

BHP

FCSFC= [L/kW.hr]

(3)

• Thermal Efficiency (ηth ) Thermal efficiency express the fuel energy effectivity to product some work.

100.632

xQ

BHP

fth =η [%]

(4)

• Opacity of exhaust gas (k %) As known, the combustion process of raw materials in diesel engines gives out few components of exhaust gas and smoke. The measurement of the emission of exhaust gas includes measurement of the opasity of exhaust gas. Rate of smoke concentration with the percentage of light being accepted by concentration sensor. Measured in percentage scale.


Specific Fuel Consumption a. Specific fuel Consumption at the rated speed (1500

rpm) and variation throttle open (30%, 40%, 50% and 60%)

Specific Fuel ConsumptionRated speed 1500 rpm

0.1

0.2

0.2

0.3

0.3

0.4

0.4

0.5

0.5

0.6

0.6

20 30 40 50 60 70% Throttle Valve Open

SFC

[L/H

P.h

r]

Diesel Oil B5CCNT

B10CCNT B20CCNT

B5CORN B10CORN

B20CORN

THROTTLE DIESEL

OPEN OIL RESULT % RESULT % RESULT %

30 % 0.398 0.373 -6.4 0.379 -4.8 0.397 -0.3

40 % 0.257 0.259 0.6 0.256 -0.4 0.258 0.4

50 % 0.202 0.243 20.6 0.243 20.8 0.229 13.6

60 % 0.241 0.252 4.4 0.253 4.9 0.249 3.0

RESULT % RESULT % RESULT %

Increase 0.503 26.2 0.367 -7.8 0.369 -7.2

Decrease 0.243 -5.4 0.245 -4.5 0.254 -1.1

0.237 17.6 0.229 13.6 0.230 13.9

0.230 -4.7 0.254 5.2 0.237 -1.8

B5 CORN E. E B10 CORN E. E B20 CORN E. E

B5 COCONOUT E. E B10 COCONOUT E. E B20 COCONOUT E. E

The Specific Fuel Consumption of blends is almost same at throttle valve open 30 % and then it decreased at 40% of throttle valve open by average 23,5 %. The highest decrease was happened on B5 Corn ethyl ester amount 6,4 % at 30% of throttle valve open. The sfc decreased at 50% and 60% of throttle valve open. The sfc decreased with increase in bhp and increased at

50% and 60% of throttle valve open due to effect of heating value of the blends.

b. Specific fuel Consumption at the rated throttle

valve open (40 %) and variation speed (1300, 1500, 1700 and 1900 rpm)

Specific Fuel ConsumptionThrottle Valve Open (40%)

0.1

0.2

0.3

0.4

0.5

0.6

1100 1300 1500 1700 1900 2100Rotational Speed [rpm]

SFC

[L/H

P.h

r]

Diesel Oil B5CCNT

B10CCNT B20CCNT

B5CORN B10CORN

B20CORN

RPM DIESEL

OIL RESULT % RESULT % RESULT %

1300 0.225 0.232 3.1 0.237 5.0 0.234 4.0

1500 0.250 0.262 4.6 0.257 2.6 0.262 4.5

1700 0.266 0.251 -5.8 0.268 0.8 0.292 9.9

1900 0.472 0.379 -19.6 0.489 3.6 0.488 3.6

RESULT % RESULT % RESULT %

Increase 0.230 2.0 0.231 2.3 0.235 4.4

Decrease 0.230 -8.3 0.250 -0.1 0.256 2.4

0.289 8.7 0.268 0.7 0.285 7.1

0.502 6.3 0.423 -10.2 0.490 3.9

B20 CORN E. EB10 CORN E. E

B10 COCONOUT E. EB5 COCONOUT E. E B20 COCONOUT E. E

B5 CORN E. E

B5, B10 and B20 of the both biodiesels have Specific Fuel Consumption which generally higher than Diesel Oil SFC at up to 1700 rpm but it generally decreased at 1900 rpm. B5 Coconut ethyl ester was higher decrease (19.6%) at 1900 rpm. Further, all of blend is generally show similar behavior at rated of speed.

Brake Horse Power (BHP) a. Brake Horse Power at the rated speed (1500 rpm)

and variation of throttle open (30%, 40%, 50% and 60%)

Brake Horse PowerRated speed 1500 rpm

2

6

10

14

18

22

26


BH

P [H

P]

Diesel Oil B5CCNT

B10CCNT B20CCNT

B5CORN B10CORN

B20CORN

THROTTLE DIESEL


30 % 5.351 5.296 -1.0 5.425 1.4 4.835 -9.7

40 % 18.637 16.423 -11.9 20.298 8.9 19.227 3.2

50 % 21.183 21.405 1.0 21.589 1.9 20.187 -4.7

60 % 22.143 22.143 0.0 - 21.774 -1.7 21.036 -5.0

Increase RESULT % RESULT % RESULT %

Decrease 3.580 -33.1 5.351 0.0 - 8.933 66.9

20.002 7.3 17.880 -4.1 31.562 69.4

20.888 -1.4 20.789 -1.9 33.766 59.4

21.940 -0.9 21.774 -1.7 34.659 56.5


B20 COCONOUT E. EB10 COCONOUT E. EB5 COCONOUT E. E

The highest increase of bhp occurred at B20 of

Corn ethyl ester amount 69.9% at 40% of throttle valve open and and B20 of Corn ethyl ester amount 66.9 % at 40% of throttle valve open. Decrease of bhp was dominant at 60% of throttle valve open by

ISSN: 1411-1284


average 2.3%. Bhp increased with increase increase in engine load. b. Brake Horse Power (BHP) at the rated throttle valve

open (40 %) and variation speed (1300, 1500, 1700 and 1900 rpm)

Brake Horse PowerThrottle Valve Open (40%)

2468

1012141618202224

1100 1300 1500 1700 1900 2100Rotational Speed [rpm]

BH

P [

HP

]

Diesel Oil B5CCNT

B10CCNT B20CCNT

B5CORN B10CORN

B20CORN

RPM DIESEL


1300 17.911 18.391 2.7 17.591 -1.8 16.552 -7.6

1500 20.851 21.036 0.9 20.667 -0.9 19.836 -4.9

1700 11.920 13.384 12.3 12.422 4.2 10.958 -8.1

1900 4.791 6.544 36.6 4.862 1.5 4.675 -2.4


Decrease 17.591 -1.8 17.591 -1.8 17.495 -2.3

20.335 -2.5 20.408 -2.1 20.482 -1.8

11.084 -7.0 12.548 5.3 11.293 -5.3

4.534 -5.4 5.703 19.0 4.558 -4.9



The highest increases of bhp were occurred of B5 of Coconut ethyl ester (36.6%) at speed 1900, B5 Coconut oil (12.3%) at speed 1700 rpm, B5 of coconut ethyl ester (0.9%) at 1500 rpm and B5 of Coconut ethyl ester (2.7%) at speed 1300 rpm. But, for other blends decrease was occurred at speed 1300 rpm by average 2.2%. Bhp increased to a maximum (up to 1500 rpm) and decreased due to decreased engine load by effect of friction increase with engine speed and becomes dominant at higher speeds. Thermal Efficiency (ηth) a. Thermal efficiency at the rated speed (1500 rpm)

and variation of throttle open (30%, 40%, 50% and 60%)

Thermal EfficiencyRated speed 1500 rpm

5

10

15

20

25

30

35

40

20 30 40 50 60 70

% Throttle Valve Open

ηη ηηth

[%]

Diesel Oil B5CCNTB10CCNT B20CCNTB5CORN B10CORNB20CORN

THROTTLE DIESEL


30 % 17.670 19.007 7.6 18.807 6.4 18.184 2.9

40 % 27.386 27.395 0.0 - 27.849 1.7 27.981 2.2

50 % 34.903 29.123 -16.6 29.280 -16.1 31.523 -9.7

60 % 29.148 28.102 -3.6 - 28.150 -3.4 29.051 -0.3


Decrease 5.777 -67.3 8.635 -51.1 8.933 -49.4

32.277 17.9 28.853 5.4 31.562 15.2

33.706 -3.4 33.545 -3.9 33.766 -3.3

35.403 21.5 35.136 20.5 34.659 18.9



The increase of thermal efficiency is most dominant at 40 % of throttle valve open by average 8.48% and decrease is most dominant at 50% of throttle valve open by average 7.28%. The drop in thermal efficiency with in portion of blends must be attributed to the combustion process.

b. Thermal efficiency at the rated throttle valve open

40 % and variation of speed (1300 rpm, 1500 rpm, 1700 rpm and 1900 rpm)

Thermal Efficiency

Throttle Valve Open (40%)

0

10

20

30

40

1100 1300 1500 1700 1900 2100

Rotational Speed [rpm]

ηη ηηth

[%

[


RPM DIESEL


1300 31.215 30.473 -2.4 30.115 -3.5 30.794 -1.3

1500 28.100 27.028 -3.8 27.742 -1.3 27.600 -1.8

1700 26.444 28.267 6.9 26.570 0.5 24.696 -6.6

1900 14.915 18.673 25.2 14.587 -2.2 14.782 -0.9


Decrease 30.854 -1.2 31.061 -0.5 30.970 -0.8

30.922 10.0 28.621 1.9 28.428 1.2

24.539 -7.2 26.722 1.1 25.576 -3.3

14.151 -5.1 16.907 13.4 14.864 -0.3



The better thermal efficiency was most dominated by B5 Coconut ethyl ester at speed 1900 rpm by amount 25.2%. Further, for B5 Coconut ethyl ester decreased 6.9% and B10 Corn ethyl ester increased 1.1% at speed 1700 rpm. But, it generally decreased by average 1.7% for all of blends.

Exhaust Gas Opacity (%) a. Exhaust gas opacity at the rated speed (1500 rpm)

and variation throttle open (30%, 40%, 50% and 60%)

ISSN: 1411-1284


Opacity (%)Uniform Speed at 1500 rpm

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

0.40


k

Diesel Oil B5CCNT

B10CCNT B20CCNT

B5CORN B10CORNB20CORN

THROTTLE DIESEL


30 % 0.096 0.069 -28.3 0.055 -42.9 0.055 -42.9

40 % 0.180 0.096 -46.7 0.139 -22.8 0.100 -44.7

50 % 0.200 0.122 -38.8 0.122 -39.1 0.101 -49.4

60 % 0.219 0.148 -32.4 0.141 -35.6 0.110 -49.8


Decrease 0.065 -31.9 0.052 -46.1 0.044 -54.5

0.143 -20.6 0.113 -37.5 0.115 -36.4

0.142 -29.1 0.127 -36.6 0.117 -41.4

0.142 -35.2 0.159 -27.6 0.140 -36.3


B20 CORN E. EB10 CORN E. EB5 CORN E. E

All of blends have lower opacity than diesel oil which B20 Corn Oil has highest decrease (54.5%) at 30% of throttle valve open, B5 Coconut Oil opacity decreased 46.7% at 40% of throttle valve open, B20 Coconout Oil opacity decreased 49.4% at 50% of throttle valve open and B20 Coconut Oil opacity decreased 49.8% at 60% of throttle valve open. Furthermore, all of blends decreased opacity by average 46.7% as compared diesel fuel. The biodiesels/ethyl ester fuels produces less smoke, probably because the biodiesel contains oxygen, high cetane number which helps the good combustion in the cylinder. b. Exhaust gas opacity (%) at rated throttle valve open

(40 %) and variation speed (1300, 1500, 1700 and 1900 rpm)

Opacity (% )Uniform Throttle Valve Open - 40 %

0.000

0.050

0.100

0.150

0.200

0.250

0.300

1100 1300 1500 1700 1900 2100Rotational speed (rpm)

k


RPM DIESEL


1300 0.182 0.127 -30.0 0.131 -27.8 0.104 -42.7

1500 0.174 0.140 -19.8 0.132 -24.4 0.102 -41.4

1700 0.112 0.066 -41.1 0.074 -34.4 0.069 -38.8

1900 0.108 0.075 -30.2 0.079 -26.5 0.079 -26.5


Decrease 0.118 -35.0 0.127 -30.0 0.107 -41.0

0.124 -29.0 0.137 -21.3 0.139 -20.1

0.065 -42.4 0.068 -39.3 0.054 -51.8

0.073 -32.6 0.073 -32.6 0.058 -46.0

B5 CORN E. E


B20 CORN E. EB10 CORN E. E

All of blends have percentage of opacity lower than diesel oil opacity. B20 Coconout ethyl ester opacity has highest decrease (42.7%) at speed 1300 rpm, B20 Coconut ethyl ester opacity decreased 41.4% at speed 1500 rpm, B20 Corn ethyl ester opacity decreased 51.8% at speed 1700 and B20 Corn ethyl ester opacity decreased 46% at speed 1900 rpm. Furthermore, all of blends decreased opacity by average 39.7% as compared diesel fuel.

IV. CONCLUSIONS

1. Effect of biodiesel blends for the specific fuel consumption were almost similar at up to 1500 rpm but it become increased at speed 1700 to 1900 rpm which increase by average 16.7%.

2. The highest increase of bhp occurred for B20 Corn ethyl ester (66.9%) at 30% of throttle valve open and decreased at 60% of throttle open. The bhp change was resulted by engine load.

3. The thermal efficiecy increased by average 8.48% for the blends at 40% throttle valve open and most dominant decreased at 50% of throttle valve open (7.28%). While, for the rated speed, the thermal efficiency increased at 1700 rpm and 1900 rpm which B5 Coconut ethyl ester was higher (25.2%) at 1900 rpm and B10 Corn ethyl ester increased 1.1% at 1700 rpm. The drop in thermal efficiency related to combustion process.

4. Using both biodiesels (Coconut ethyl ester and Corn ethyl ester), exhaust opacity were lowest (as compared to diesel fuel), where it decreased by average 46.7% at the rated speed (1500 rpm) and 39.7% at 40 % throttle valve open. The both biodiesels contain oxygen, high cetane number which helps the good combustion in the cylinder.

5. B20 of Corn ethyl ester represent the best composition for all blends.

REFERENCES

[1] Willard W. Pulkrabek, 2004, Engineering Fundamentals

of The Internal Combustion Engine-Second Edition, Pearson-Prentice Hall, New Jersey 07458.

[2] Zulkarnain, 2007, Studi Pengaruh Karakteristik Biodiesel Minyak Goreng Curah (Sawit) dan Minyak Jagung Dengan Pereaksi Spritus dan Prosessor Jenis Sususn Terhadap Performa Mesin, Tesis, UI, Jakarta.

[3] Kepmen ESDM No. 0002, 2004, Kebijakan Energi Terbarukan dan Konservasi Energi, Jakarta.

[4] Heywood, John B, Internal Combustion Engine Fundamental, McGraw-Hill International Edition.

[5] US department of Energy, Energy and Renewable Energy, ‘Biodiesel Handling and Use Guideline’, 2004

[6] Gerhard Knothe. Et al, ’The Biodiesel Hand Book’, National

Center for Agricultural Utilization Research Agricultural

Research Service U.S. Department of Agriculture Peoria,

Illinois, U.S.A, 2005 [7] Gerhard Knothe. Et al, ’The Biodiesel, The Use of

Vegetable

ISSN: 1411-1284


Oils and Their Derivative Diesel Fuels; National Center for

Agricultural Utilization Research Agricultural Research Service U.S. Department of Agriculture Peoria, Illinois,

U.S.A. [8] Nissan D. Motor CO. Ltd, ‘Guide and Manual Book of

Diesel Engine Test Bed’, Japan.

ISSN: 1411-1284


Abstract— This paper examines the smoke production rate of tropical woods (i.e. Kayu Jati and Kayu Kamper) in the Cone Calorimeter. The samples were exposed to a constant incident heat flux of 27 kW/m2 The smoke opacity measurements obtained from the Cone Calorimeter were used to derive the rate of smoke production (RSP) of the materials. Since, the mass of the sample also simultaneously measured, the smoke extinction are per unit mass of sample consumed can also be deduced. This quantity, the so called SEA (smoke extinction area) is an important parameter in considering the fire detection and reduced visibility the occupants of a building in fire. This paper also discusses the SEA under pyrolysis and flaming conditions. The outcomes are in good agreements with those reported in the literatures. Keywords— Smoke opacity, smoke production rate, fire safety engineering, tropical wood.

I. INTRODUCTION

The early fire behavior of building products is important for many aspects of fire safety engineering. The heat release rate is a fundamental variable of fire with which almost all other emission properties are highly correlated [1]. Another important descriptor of a fire is the smoke production rate. Smoke is produced in almost all fires and presents a major hazards to life. The production of smoke and its optical properties are often measured separately based on related standard methods.

Wood and its derived products remain the most popular materials for furniture used in domestic premises. In the event of fire, wood based products contribute significantly in the initiation and growth of fire and the level of fire severity. The capability to predict the burning rate of wood in modern times has become increasingly important as fire safety engineering moves toward a performance-based approach to building design [1,2].

A fire calorimeter or also known as a cone calorimeter has been developed in Department of Mechanical Engineering University of Indonesia. The fire calorimeter is an integrated small-scale instrument to measure important fire safety parameters of wood and other building products. The instrument consists of a heater of a truncated cone shape and is capable of providing heat fluxes to the specimen in the range of 0 –

30 kW/m2, a digital electronic balance, an electric igniter, thermocouples and a temperature control system, a flow rate measurement apparatus, a sample holder and a ducting systems. Calibration of heat flux as a function of heater temperature is performed with a heat flux meter. At a sufficient distance a gas sample is taken using a portable gas analyzer to record O2, CO and CO2 levels [3].

An extensive research work has been carried out to study fire safety parameters of tropical woods and other building materials, such as the effect of sample species, sample thicknesses, grain orientation, fire spread by means of time to ignition, heat release rate, and mass loss rate [1-3]. The design of the fire calorimeter also allows other measurements to be taken as long as the analyzers are available.

This paper presents the results of smoke property measurement using the fire calorimeter. In this work, smoke opacity measurement was carried out using a recently purchased smoke meter apparatus available in Thermodynamics laboratory of Department of Mechanical Engineering University of Indonesia. The focus of this paper is to evaluate the smoke production rate of some tropical wood species. A better understanding of the pattern of smoke production and yield will be one of the key parameter towards the implementation of performance-based fire protection system in building and plant designs.

II. THEORY Smoke is generated in both smoldering and flaming combustions. Smoke from smoldering is originated from chemical degradation and volatiles evolution of carbon based material when heated to certain temperatures. It consists of high molecular weight fractions of condense aerosol / droplets of tar and high boiling liquids. Meanwhile, smoke from flaming combustion is almost entirely consists of solid particles [4].

Smoke emission is one of the basic elements for characterizing a fire environment. The combustion conditions under which smoke is produced-flaming, pyrolysis, and smoldering-affect the amount and character of the smoke. The smoke emission from a flame represents a balance between growth processes in the

On the measurement of smoke production rate of tropical wood

Yulianto Sulistyo Nugroho

Department of Mechanical Engineering, University of Indonesia Kampus UI Depok 16424, INDONESIA

Ph. +62 (021) 7270032, Fax. +62 (021) 7270033, e-mail: [email protected]

ISSN: 1411-1284


fuel-rich portion of the flame and burnout with oxygen. While it is not possible at the present time to predict the smoke emission as a function of fuel chemistry and combustion conditions, it is known that an aromatic polymer, such as polystyrene, produces more smoke than hydrocarbons with single carbon-carbon bonds, such as polypropylene. The smoke produced in flaming combustion tends to have a large content of elemental (graphitic) carbon [5].

The smoke emission, together with the flow pattern, determines the smoke concentration as smoke moves throughout a building. The effects of the smoke produced by a fire depend on the amount of smoke produced and on the properties of the smoke. It is the combination of obscuration and toxicity that present the greatest threat to the occupants of a building involved in fire. Most of the fatalities can be attributed to the inhalation of smoke and toxic gas. The effect of reduced visibility is to delay escape and increase the duration of exposure of the occupants of a building to the product of combustion. In the early stages of a fire, smoke production rate is relevant to fire detection.

The smoke properties of primary interest to the fire community are light extinction, visibility, and detection. However, the most widely measured smoke property is the light extinction coefficient. The yield of particulate smoke from a burning material may be assessed by some method: (i) filtering the smoke and determining the weight of particulate matter, (ii) collecting the smoke in a known volume and determining its optical density, and (iii) allowing the smoke to flow along a duct, measuring its optical density where plug flow has been established. Optical density may be determined by measuring the attenuation of a beam of light passing through the smoke [4].

The physical basis for light extinction measurements is the Lambert-Beer Law. In essence, the law states that there is a logarithmic dependence between the transmission of light through a substance and the concentration of the substance, and also between the transmission and the length of material that the light travels through.

Fig. 1 Absorption of a beam of light as it travels through the smoke container, adapted from [6].

According to the Lambert-Beer Law in the absence of smoke the intensity of light falling on the photocell is (Io), then in the presence of smoke, the reduced intensity (I) will be given by

( )kCLII o −= exp (1)

where k is the extinction coefficient, C is the mass concentration of smoke particles and L is the path length of the optical beam passing through the smoke. Optical density, De is then defined in natural logarithm as follows [4-7]:

kCLI

ID

oe =

−= ln (2)

Smoke extinction coefficient, k, (m-1) can be determined by a smoke meter as follows :

=I

I

Lk oln

1 (3)

Concentration of smoke can also be expressed as percentage obscuration or smoke opacity (%), given by:

%100×

−

o

o

I

II (4)

The rate of smoke production, RSP (m2/s), can be determined by

=

a

ss T

TVkRSP & (5)

where sV& is the volume flow rate of exhaust gas (m3/s), k

is smoke extinction coefficient (m-1), Ts is the gas temperature in the exhaust duct (K), and Ta is the ambient temperature (K).

In a fire calorimeter, the volume flow rate of exhaust

gas, sV& (m3/s) can be estimated using the following

expression [9]:

51,0

3105,2

∆⋅×= − p

T

TV

s

as& (6)

Since in the fire calorimeter assessment, one also measure the mass loss rate, thus a quantity called as specific extinction area, SEA (m2/kg) can also be defined as :

ISSN: 1411-1284


mRSPSEA

&

1⋅= (7)

= kg

m

kg

s

s

m 22

=

⋅

mT

TVkSEA

a

ss

&

&1⋅

= (8)

The SEA represents the smoke extinction area per mass of fuel burned.

III. EXPERIMENTAL

3.1 Experimental apparatus

This work was carried out using a fire calorimeter or also known as a cone calorimeter developed in Department of Mechanical Engineering University of Indonesia. The fire calorimeter is a small-scale instrument to measure ignition time, heat release rate and smoke production of wood and other building products. The instrument consists of a heater of a truncated cone shape and is capable of providing heat fluxes to the specimen in the range of 0 – 30 kW/m2, a digital electronic balance, an electric igniter, thermocouples and a temperature control system, a flow rate measurement apparatus, a sample holder and a ducting systems. Calibration of heat flux as a function of heater temperature is performed with a heat flux meter. At a sufficient distance a gas sample is taken using a portable gas analyzer to record O2, CO and CO2 levels. A portable smoke meter is used to measure the smoke opacity of the exhaust gas.

The basic design and working principle of the apparatus was based on the cone calorimeter standard - ASTM E 1354 [7]. The schematic diagram of the fire calorimeter was given in other publication by the author [3]. 3.2 Sample preparation

Two tropical wood samples, i.e. Kayu Jati and Kayu Kamper are used in this work. The samples are prepared with sides of 10 mm by 10 mm and 16 mm thicknesses. The wood samples were prepared with an along grain orientation (Fig. 1). All specimens are conditioned to constant mass at room temperature and relative humidity. The properties of the samples are presented in Table 1.

Fig. 2 The specimens are prepared in along grain orientation.

Table 1 Properties of the samples used

Specimen Density, ρ (kgm-3)

Kayu Kamper 621 Kayu Jati 761

3.3 Procedure

A typical run was initiated by supplying electric power to the heater which is controlled by an electronic temperature controller using a chromel-alumel thermocouple of type K. The fan speed is controlled by an inverter and set at 13,3 Hz to give a constant volume flow rate. The temperature is set to a certain value to provide a desired heat flux level. Once the temperature is stable, and other equipments, i.e. gas analyzer, smoke meter and digital electronic balance are ready, the sample is inserted to sample holder. The sample is protected from initial heat flux from the heater using a radiation shield. The experiment and data collection are started by opening the radiation shield. The transient data of temperatures, sample mass loss, exhaust gas analysis, smoke opacity measurement are recorded simultaneously through a DAQ system connected to a personal computer. In this study, the measurement are focused on the smoke produced and mass loss measurements. The samples are exposed vertically to a constant incident heat flux of 27 kW/m2.

IV. RESULTS AND DISCUSSION

Fig. 3 Smoke opacity (%) and mass loss rate (gr/s) of Jati

wood measured using fire calorimeter and smoke meter. Mass loss rate and smoke opacity curves

0 200 400 600 800 1000 12000.00

0.02

0.04

0.06

0.08

0.10

0.12

Sm

oke

opac

ity (

%)

Mass loss rate (g/s)

Mas

s lo

ss r

ate

(g/s

)

Time (s)

0

5

10

15

20Kayu Jati

Smoke opacity (%)

100 mm

100 mm 10 – 30 mm

Incident heat flux perpendicular to grain

Along grain

100 mm

100 mm 10 – 30 mm

100 mm

100 mm 10 – 30 mm


Along grain

16 mm

100 mm

100 mm 10 – 30 mm


Along grain

100 mm

100 mm 10 – 30 mm

100 mm

100 mm 10 – 30 mm


Along grain

16 mm

ISSN: 1411-1284


Fig. 3 shows a plot of sample mass loss rate (g/s) and

smoke opacity (%) against time for Kayu Jati sample. It clearly shows that the first peak of the mass loss curve occur soon after the sample was ignited. Similar pattern was found for Kayu Kamper sample (Fig. 4). It indicates that both wood species have similar physical responds to incident heating. Meanwhile, Kayu Jati has greater mass loss rate than Kayu Kamper.

0 200 400 600 800 1000 12000.00

0.02

0.04

0.06

0.08

Kayu Kamper

Sm

oke

opac

ity (

%)


Mas

s lo

ss r

ate

(g/s

)

Time (s)

0

2

4

6

8

10

12

Smoke opacity (%)

Fig. 4 Smoke opacity (%) and mass loss rate (g/s) of Kayu Kamper measured using fire calorimeter and smoke meter.

As shown on Figs 3 and 4, there was delay on the

smoke opacity measurement data compared to mass loss rate curves. This is due to differences in probes locations. The mass of the sample was measured directly below the sample, meanwhile the smoke probe was located in the exhaust duct about 1 m away from the sample. A single peak was identified for both smoke opacity curves, which is consistent with the mass loss rate patterns.

Smoke extinction area curves

According to Eq. 8, the values of smoke extinction

area, SEA (m2/g) for both samples were plotted on Figs. 5 and 6. Again, the mass loss curves were also given on the same figures to show any possible correlation between the two curves. It is observed that the Kayu Jati sample has greater values of SEA, especially in the earlier time, as clearly shown on Fig. 7.

On the basis of the mass loss rate which has proportional correlation with the heat release rate [3], the curves on Fig. 7 can be divided into two sections. The first section is representing the induction, moisture and volatile release phase. This phase is known as pyrolisis conditions where the smoke is relatively darker compared to the flaming conditions. As reported in the literature the smoke conversion factor of wood and plastic materials under pyrolysis conditions has greater values that during flaming combustion [5]. The smoke produced in flaming combustion tends to have a large content of elemental

carbon. In fact Kayu Jati has higher density value than Kayu Kamper (Table 1).

In practice, greater value of smoke extinction area, SEA means reduction in visibility. As we all concerns, visibility of exit signs, doors, and windows can be of great importance to an individual attempting to survive a fire, and plays important role on fire safety engineering consideration during building and plant design.

Fig. 5 Smoke extinction area (SEA) (m2/g) and mass loss rate (g/s) of Kayu Jati calculated according to Eq. 8.

Fig. 6 Smoke extinction area (SEA) (m2/g) and mass loss rate (g/s) of Kayu Kamper calculated according to Eq. 8.

0 200 400 600 800 1000 12000.00

0.02

0.04

0.06

0.08

Kayu Kamper


Mas

s lo

ss r

ate

(g/s

)

Time (s)

0.0

0.5

1.0

1.5

2.0

2.5

SE

A (

m2 /g

)

Smoke extinction area, SEA (m2/g)

0 200 400 600 800 1000 12000.00

0.02

0.04

0.06

0.08

0.10

0.12


Mas

s lo

ss r

ate

(g/s

)

Time (s)

0

2

4

6

8

10

12Kayu Jati

SE

A (

m2 /g

)

Smoke extinction area, SEA (m2/g)

ISSN: 1411-1284


0 200 400 600 800 1000

0

2

4

6

8

10

12

Kayu Jati Kayu Kamper

SE

A (

m2 /g

)

Time (s)

Fig. 7 Comparison of SEA values for Kayu Jati and Kayu Kamper.

V. CONCLUSION

The results of this study show that the smoke opacity

patterns correlate with the mass loss rate of the samples tested. The rate of smoke production in the earlier stages of combustion, i.e. pyrolysis condition is relatively higher than that of flaming condition. In general, the value of smoke extinction area (SEA) of Kayu Jati is greater than Kayu Kamper’s. In practice, greater value of smoke extinction area, SEA means reduction in visibility. This is of our concern on maintaining the visibility of exit signs, doors, and windows during a fire accident.

This work shows that the fire calorimeter apparatus can be used effectively in measurement of smoke properties.

ACKNOWLEDGMENT

This work has been funded by the Ministry of National Education Republic of Indonesia through Hibah Bersaing XIII (2005-2007) Research Fund. Thanks to Mr. M. Andy Fardiansyah and Mr. Candra Gupta Marihot for preparation of the samples and the smoke opacity measurements.

REFERENCES

[1] Quintiere, J. G., “A Theoretical Basis for

Flammability Properties”, Fire and Materials, 30, 175-214, 2006.

[2] Nugroho, Y.S., Soesanto, and E. Puspiartono, Effect of Sample Orientation and Thickness on Piloted Ignition of Timber, Proceedings of the Int’l. Conference on Fluid and Thermal Energy Conversion 2006, Jakarta, Dec. 10 – 14, 2006, ISSN 0854 – 9346.

[3] Nugroho, Y.S., Junaedi, A., and Kiswanto, G., “Pengembangan Kalorimeter Api untuk Karakterisasi Sifat Bakar Material”, Jurnal Teknologi, 4, 294-301, 2005.

[4] Drysdale, D., An Introduction to Fire Dynamics, 2nd Edition, John Wiley & Sons, John Wiley & Sons, 2003.

[5] Mulholland, GW, (2002), Smoke Production and Properties, Chapter 13, The SFPE Handbook of Fire Protection Engineering, 3rd Edition, pp. 2-258 – 2-268.

[6] Wikipedia, website: http://en.wikipedia.org/wiki/ Beer-Lambert_law (seen on 21st November 2007).

[7] ASTM E 1354-97, Standard Test Method for Heat and Visible Smoke Release Rates for Materials and Products Using an Oxygen Consumption Calorimeter, Annual Book of ASTM Standards, American Society for Testing Materials, Philadelphia, 1997.

[8] Myllymaki, J and Baroudi, D, Prediction of smoke production and heat release rate by convolution model, Nordtest Technical Report 1297-96, VTT Building Technology, Finland, 1999.

ISSN: 1411-1284


Catalytic Ozonation Of Endosulfan In Water With Activated Carbon

Enjarlis,a,b Setijo Bismo,a Slamet,a Roekmijati W. Soemantojoa

aDepartment of Chemical Engineering, University of Indonesia, Depok16424 Tel. 7863515

bDepartment of Chemical Engineering, Institute Technology of Indonesia, Serpong Tangerang 15311 Tel. 7561092 .fax7561092.e-mail: [email protected]

Abstract– The degradation of endosulfan by catalytic ozonation with activated carbon as catalyst was investigated at neutral pH and different temperatures. The aim of this work is to study the effect of the addition of activated carbon on the degradation rate of endosulfan, focusing on the rate constant (k) and the activation energy (Ea). Endosulfan was selected here as target organic pollutant due to its included organochlorine pesticide (OCPs) may cause serious environmental concern.The presence of the activated carbon as catalyst is effect slightly in enhancing ozonation rate of endosulfanan compared to non catalytic ozonation. The rate constant (k) of endosulfan were 11.54 x 10-2min-1at 30oC for catalytic ozonation, and 10.27 x 10-2min-1for non catalytic ozonation. The activation energy for catalytic ozonation was 6.58 kcal/mol and for non catalytic ozonation was 8,631kcal/mol. Keywords– Endosulfanan, Catalytic ozonation, Activated carbon.

I. INTRODUCTION atalytic ozonation with activated carbon as catalyst can be accelerated ozone decomposition to form OH• radical , thereby increasing even further the removal efficiency of this because OH• is powerfull, non-selective chemical oxidant, which acts very rapidly with most organic compounds [1, 2]. It has been reported that hydroxyl radicals were generated by combinations of ozone plus UV radiation, UV radiation plus H2O2, UV radiation plus Fenton’s reagent (photo-Fenton system) [3] and by mixing a few milligrams of activated carbon in ozone-containing water [4]. With such context, this study was investigated the degradation of endosulfan in water by catalytic ozonation at neutral pH and different temperatures, which is focused on (i) determining possible diffusion limitation (ii) determining the rate constants (k) and activation energy (Ea). Endosulfan was selected here as target organic pollutant due to its paradigmatic endocrine discruptors included organochlorine pesticide (OCPs) may cause serious environmental concern and health problem in animals,

including humans [5]. According to free radical reaction theory, OH• will attack chlorinated organic compounds by hydrogen abstraction or electron transfer. Then, the organic radical will decompose further to chlorinated intermediates. These intermediates are eventually oxidized by OH• to final products; organic acid and even carbon dioxide [6].

II. BASIC THEORY

The degradation of contaminant organic in presence of activated carbon is a combination of competing homogeneous and heterogeneous reaction: direct reaction of molecular ozone, and an indirect reaction involving non selective free-radicals. Both reactions take place in the solution bulk and on activated carbon surface [7]. Therefore, the total degradation rate of contaminant organic in the presence of activated carbon can be defined as the sum of the homogeneous reaction rate, (-rC)homogen,calculated in the absence of activated carbon, and the heterogeneous reaction rate, (-rC)heterogen

due to the presence of activated carbon []. The total endosulfan degradation rate can be mathematically expressed as:

rrr HeteEHomoETotalE ...+=

……(1) As discussed earlier the activated carbon enhanced the oxidation rate of contaminant by promoting the reduction of dissolved ozone into hydroxyl radicals that cause the endosulfan degradation, the rate of equation given in Tabel 1. Recently, Jans and Hoigne [1] pointed out that catalytic ozone decomposition can be cataloqued as another Advance Processes Oxidation (AOP) with a stoichiometric ratio (ozone decomposed/hyroxyl radical formed) equal to that of the other ozone involved in AOPs. Tabel 1. Rate of equation at degradation of endosulfan by ozonation in the presence activated carbon

No

With activated carbon With out activated carbon

1 (- dCE/dt)hete = f1(CE, CO3, COH•, CAc)

(- dCE/dt)homo = f4(CE, CO3, COH•)

C

ISSN: 1411-1284


2 (- dCO3/dt)hete = f2(CE, CO3, COH•, CAC)

(- dCO3/dt)homo = f5(CE, CO3, COH•)

3 (- dCOH•/dt)hete = f3(CE, CO3, COH•, CAC)

(- dCOH•/dt)homo = f6(CE, CO3, COH•)


Endosulfan (1,2,3,4,7,7-hexachlorobicyclo-2,2,1 heptene 2,3-bis-hydroxy methane-5,6 sulfite) was obtained from Chem. Service West Chester with purity 95 %. The endosulfan solution was prepared by deionozed water obtained from Aquatron Auto Still Yamato Tipe W-182. The specific surface area of activated carbon were measured using the multipoint BET of N2 adsorption in a Quantachrome Autosorb -6 with surface area 6.87.102 m2/g. Ozone was produced by a RS 09805 ozone generator with maximum ozone production capacity of 0.25 g of O3/h. The experimental instrument consists of an ozone generator, a cylindrical glass column reactor with an external jacket surrounded and a water stream which was pumped from thermostatic bath to maintain the temperature at the selected value for each experiment. The dimension is of the reactor is 450 mm high with ID 40 mm which equiped by inlet diffuser for bubbling the gas mixture, outlet gas, sampling port and magnetic stirrer. Once the experiment was started, the air-ozone mixture was fed into the flasks (KI solution) in order to determine the ozone concentration in the gas form. The reactor was filled with 300 ml demineralized-water and the pH was adjusted to 7. The temperature was set for 20, 25 and 30oC, until a predetermined of water was saturated with ozone in excess continuously by injecting ozone gas for 30 min. The process is followed by the addition 0.5 g of activated carbon (for the catalytic process) and 4.5 x 10-5 mol/L endosulfan in solution. The concentration of endosulfan and ozone presence in the system was measured at 0, 3, 6, 10, and 15 min of treatment. The dissolved ozone concentration in aqueous solution was determined by iodometric methods, which endosulfan was analyzed by Gas Chromatograph type 4C, colum silicone ov-17 3 meters, ECD (Electron Capture Detector), Shimadzu, solvent n-hexsane and mobile gas-phase N2. All experiments were carried out in duplo the presence and the absence of activated carbon. The influence of variables on the degradation of endosulfan were first assessed. Non-catalytic (blank experiment) and catalytic experiments were conducted for each experimental series. Since the catalytic ozonation process involves steps of external and internal diffusion, variables studied were: activated carbon particle size and agitation speed. From experimental result agitation speed equal or higher than 750 rpm and carbon particle equal or lower 0.2 -0.3 mm did not affect the ozonation rate. Hence, at these conditions, it can be assumed that external mass transfer rate and internal diffusion rate of ozone through pores did not control the

catalyst ozone decomposition [9]. Therefore, agitation and activated carbon particle particle were kept at 750 rpm and 0.2-0.3 mm size. At these conditions, the kinetics should only be depended on the chemical surface ozone decomposition reaction [9]. Next, a series of experiments were then carried out at different temperatures to determining Ea from k. Endosulfan degradation Fig. 1 shows the influence of the presence of activated carbon as catalyst for the degradation rate of endosulfan by catalytic and non-catalytic ozonation. From that figure, it can be seen that the degradation rate of endosulfan increased with the presence of activated carbon compared to the absence of activated carbon. This is because, the presence of activated carbon can accelerate ozone decomposition to form OH· radicals which have a high reactivity to most organic micropollutants [1, 2].

0,5

1,0

1,5

2,0

2,5

3,0

3,5

4,0

4,5

5,0

0 3 6 9 12 15

time, min

CE.1

0.-5 m

ol/

L

Fig. 1. Endosulfanan degradation rate by ozonation processes (x) non-catalytic (blank experiment)

()catalytic ozonation at condition: pH 7; temperature 20oC; agitation speed 750 min-1;activated carbon particle

0.2-0.3 mm CO3 6.372 x 10-3 O3 mol/L; CEo 5.0 x 10-5mol/L

Influence of temperature on the degradation of carbofuran Fig.2 shows the effect of temperature on degradation of endosulfan for catalytic ozonation at pH 7. From that figure, it can be stated that the higher the temperature is, the higher is the degradation rate of endosulfan. This effect is due to an increase in the rate constant of the chemical reaction. Yazgan et al [10] reported that the change of temperature gave positive effect on the removal rate of the endosulfan. Although increasing

ISSN: 1411-1284


temperature causes decreasing in dissolved ozone concentration (data are not shown), a higher oxidation rate was obtained at higher temperatures. This phenomenon can be explained by the increasing diffusion coefficient at higher temperature [10].

0

1

2

3

4

5

6

0 3 6 9 12 15

time, min

CE,x

10-5

,mo

l/L

Fig.2 The effect of temperature on degradation of endosulfan by catalytic ozonation at condition: pH 7; agitation speed 750 min-1;activated carbon particle 0.2-0.3 mm CO3 6.372 x 10-3 O3 mol/L; CEo 5.0 x 10-5mol/L; temperature 20oC, ∆ 25oC, 30oC Kinetics and Mechanism The observed endosulfan degradation in the presence of activated carbon results in a combination of competing homogeneous and heterogeneous reaction: direct reaction of molecular ozone, and an indirect reaction involving non selective free-radicals [7]. The degradation kinetics of endosulfan by homogeneous reaction in terms of molecular ozone and hydroxyl radical that is produced by the decomposition of ozone at neutral pH can be formulated as follows: Degradation rate of endosulfan homogeneous reaction:

•+=

− OHEE

OEE

o

E CCkCCkdt

dC.... 231

hom

.

(2) Degradation rate of endosulfan by heterogeneous reaction:

ACOHEE

ACOEE

ACEE

hete

E CCCkCCCkCCkdt

dC....... 5343

. ++=

−

(3) Thus, since of ozone concentration in solution in excess and the constant amount of activated carbon, the amount of OH• and O3 stayed constant during the processes (i.e., at a steady state). Therefore, Eq. (1), (2)

and (3) can be rearranged to the pseudo-first-order equation in Eq. (4) as follows:

( ) EEHete

EHomo

E

Hete

E

Homo

ECkk

dt

drr

C Total .+=+=−

(4)

with, •+=OH

EO

EEHomo CkCkk .. 231

(5)

and , •++= OHE

OE

ACEE

Hete CkCkCkk ... 5343

(6) Integrating, with initial condition CE = CEo leads to:

tktkkC

C ETotal

EHete

EHomo

E

Eo .)(ln =+=

(7) with, .EHete

EHomo

ETotal kkk +=

(8) where k E

total represents the apparent pseudo-first-order total rate constant for ozonation process in presence activated carbon. The degradation endosulfan by homogeneous reaction non catalytic as follows:

tkC

C EHomo

E

Eo .ln =

(9) Figure 3 and 4 describe the plot of ln C Eo /CE) versus time (t) with slope as k E total and k E Homo in various temperatures by catalytic and non catalytic oznation, respectively. Insert of Fig. 3 and 4 shows, typical Arrehnius plot for ln k against 1/T to endosulfan in catalytic and non catalytic ozonation processes, respectively. This linear regression indicate R2 = 0.99. From that figure, it can be stated that the higher the temperature is, the higher is the degradation rate. This effect is due to an increase in the constant rate of the chemical reaction. Similar phenomenon was reported by many researchers [10, 11]. The degradation of endosulfan in the presence of activated carbon slightly significantly enhancing the degradation rate of endosulfan with kE

total was 11.54 x 10-2min-1 and Ea 6.58 kcal/mol at 30oC compared to non-catalytic ozonation with k E

Homo was 10.27 x 10-2min-1 and Ea 8.631 kcal/mol. This is because of the using of activated carbon which small surface area was (6.87.102 m2/g) compared to which was used by the other authors. They used activated carbon with surface area was larger than 6.87.102m2/g [2, 10].

ISSN: 1411-1284


0

0,4

0,8

1,2

1,6

2

2,4

2,8

3,2

3,6

4

0 3 6 9 12 15time, min

ln (

CE

o /C

E)

-3,0

-2,5

-2,0

-1,5

-1,0

0,00325 0,0033 0,00335 0,0034 0,00345

1/T (1/.K)

ln k

tota

l, m

in-1

Fig 3 Constants rate of degradation endosulfan by catalytic ozonation at condition: pH 7; agitation speed 750 min-1;activated carbon particle size 0.2-0.3 mm; CO3 6.372.10-3 O3 mol/L; CEo 5.10-5mol/L; temperature 20oC, ∆ 25oC, 30oC. Insert: Arrhenius plot for degradation of carbofuran by catalytic. On other hand, compared to the degradation of carbofuran by catalytic ozonation using the same activated carbon as catalyst we obtained a slower degradation rate of endosulfan [12]. This is because, the reaction of endosulfan to ozone and OH• is less reactive compared to the reaction of carbofuran to ozone and OH•. In addition, the presence of activated carbon in ozonation processes does not really act as a catalyst, but rather as an initiator and/or a promotor for ozone transformation into OH• [13]. Finally, from the rate constants and activation energy given in Table 2 corresponded to the experiments completed at defferent

temperatures in the range of 20-30oC by catalytic and non-catalytic ozonation

0

0,5

1

1,5

2

2,5

3

3,5

4

4,5

0 3 6 9 12 15time, min

ln(C

Eo/

CE)

-3,3

-2,8

-2,3

-1,80,00325 0,0033 0,00335 0,0034 0,00345

1/T (1/.K)

ln k

",(m

in-1

)

Fig. 4 Constants rate of degradation of endosulfan by non catalytic ozonation at condition: pH 7; gas flow rate 0.1603 L h-1; agitation speed 750 min-1; CO3 6.372.10-3

O3 mol/L; CEo 5.10-5mol/L; temperature 20oC, ∆ 25oC, 30oC. Insert: Arrhenius plot for degradation of carbofuran by catalytic.

Table 2. Pseudo-first order constant (k), correlation coefficient (R2), activation energy of carbofuran degradation by catalytic ozonation and non-cataytic ozonation

Catalytic Ozonation Non-catalytic ozonation Temperature (oC) k.10-2

(min-1) R2 Ea (kcal/mol) R2 k. 10-2 (min-1) R2 Ea (kcal/mol) R2

30 11.54 0,99 10.27 0,99 25 9.61 0,99 6,586 0,99 7.86 0,99 8.631 0,99 20 7.97 0,99 6.32 0,99

ISSN: 1411-1284


IV. CONCLUSIONS

The presence of the activated carbon (catalytic ozonation) is sligtly significant in enhancing the degradation rate of endosulfan compared to non-catalytic ozonation processes. The rate constant (k) of endosulfan were 11.99 x 10-2min-1at 30oC for catalytic ozonation, and 10.27x10-2min-1for non catalytic ozonation. The activation energy (Ea) for catalytic ozonation were 8.631 kcal/mol and for non catalytic ozonation 6.586kcal/mol.

REFERENCES [1] Jans, U., Hoigne, J. “Activated carbon and

carbon black catalyzed transformation of aqueous ozone into OH-radicals” Ozone Sci. Eng, vol 20, pp.67-89, 1998

[2] Beltran, F. J., Rivas, F. J., Fernandez L. A., Alvarez P. M., and Montero-de-Espinosa, R., “ Kinetics of Catalytic Ozonation of Oxalic Acid in Water eith Activated Carbon”, Ind. Eng. Chem. Res. vol.4, pp. 6510 -6517, 2002.

[3] Langlais, B., David A. R., Brink, D. R., “Ozone in Water Treatment Application Engineering”, Cooperative Research Report. Florida. Lewis Publishing. 1991.

[4] Jans, U., Hoigne, J. “Activated carbon and carbon black catalyzed transformation of aqueous ozone into OH-radicals” Ozone Sci. Eng., vol 20, pp. 67-89, 1998.

[5] Extoxnet, 2000, “Pesticide Information Profiles: Endosulfan”, Extention Toxicology Network, Oregon State University. http://ace.orst.edu/info/extoxnet/pips/carbofur.htm)

[6] Benitez., F.J., Beltran-Heredia, J., Acero, J.L., and Rubio, F.J. “Oxidation of several chlorophenolic derivatives by UV irradiation and hydroxyl radicals”.J. Chem. Technol. Biotechnol., vol. 76, pp. 312-320, 2001.

[7] Valdes, H., Zaror, C.A. “Heterogeneous and homogeneous catalyti ozonation of benzothiazole promoted by activated carbon: Kinetic approach”, Chemosphere., vol. 65, pp. 1131-1136, 2006.

[8] Polo Sanchez, M., Ramos Leyva- Ramos, Rivera-Utrilla., “ Kinetics of 1,3,6-naphthalenetrisulphonic acid Ozonation in presence of activated carbon”.Carbon vol. 43, pp. 962-969, 2005.

[9] Fogler Fogler, H, S. Elements of Chemical

Reaction Engineering. 3rd ed; Prentice-Hall:

Englewood Clifft, NJ, 1999

[10] Yazgan, M. S., and Kinaci, C., “Beta-Endosulfan Removal From Water by Ozone

Oxidation”, Water Science & Tecnology. Vol. 48, no. 11, pp. 511-517, 2004.

[11] Benitez, F. J., Acero, J. l., and Real, F.J, “Degradation of carbofuran by using ozone, UV radiation and advanced oxidation processes”, Journal of Hazardous Materials vol. B89, pp. 51- 65, 2002,

[12] Enjarlis., Setijo Bismo, Slamet and Roekmijati W. Soematoyo., “Kinetics degradation of carbofuran by ozonation in the presence activated carbon”

Proceeding 14th Regional symposium Chemical engineering. UGM Yokyakarta. Indonesia, 2007.

[13] Sanchez Polo, M., Gunten, U.V., Rivera-Utrilla, J.”Efficiency of activated carbon to transform ozone into OH• radicals: Influence of operational parameters”, Water Res.vol. 39, pp. 3189-3198, 2005.

ISSN: 1411-1284


Hydrocarbon selectivity of Fischer-Tropsch synthesis from syngas with ratio of H2 to CO = 1.0 over Co-Fe/Al2O3 bimetallic catalysts : effect of Fe:Co ratio

Dewi Tristantinia and Börje Gevertb

a University of Indonesia, Department of Chemical Engineering, 16424 Depok, Indonesia. bChalmers University of Technology, Department of Chemical and Biological Engineering,

S-412 96 Goteborg, Sweden aCorresponding author: Phone: +62.217863516, Fax.:+ 62.217863515, e-mail: [email protected]

Abstract

The utilization of bio-syngas produced from low temperature gasification of biomass i.e. hydrogen to carbon monoxide ratio (H2/CO-ratio) between 1.0-1.5 directly inside a Fischer-Tropsch (FT) synthesis is an interesting option since it makes a water gas shift (WGS)-unit and CO2-removal unit prior to the FT-reactor unnecessary. Different Fe:Co ratio (12 wt% bimetals) Al2O3-supported catalysts were performed in Fischer-Tropsch synthesis with syngas H2/CO ratio = 1.0 at 20 bars and reaction temperature of 483 K

Catalyst characterizations applied in this study shows that the higher Fe content in the catalyst decreases CO chemisorbed but did not influence significantly in the decrease of the BET surface area and the degree of reduction.

The CO conversion, the hydrocarbon formation rate and the selectivity to C5+ were decreased while the selectivity to CH4 and the selectivity to CO2 were increased with increasing Fe content. The higher Fe:Co ratio in the catalyst the higher WGS activity (CO2 selectivity) though it is still far from the desired usage H2/CO ratio 1.0 (the H2/CO ratio in the feed). The result indicates that the increase of WGS activity was followed by the decrease of FT activity probably due to the competition of those reactions on the same active sites for Fe-Co bimetallic catalysts.

Key words: Bio-syngas; Hydrocarbon; Fischer-Tropsch Synthesis, Water-Gas-Shift, Cobalt, Iron, Bimetallic catalysts, Effect Fe:Co.

I. INTRODUCTION

Biomass can be converted to synthesis gas

(syngas), called bio-syngas, via a high or low temperature gasification process. This bio-syngas consists mainly of H2, CO, CO2 and CH4. It contains less hydrogen (often named H2-poor bio-syngas) in the mixture compared to syngas from fossil

sources. An option for the production of renewable automotive fuels from H2-poor bio-syngas is the Fischer-Tropsch (FT) synthesis, in which H2 and CO is converted to longer hydrocarbons (HCs) mainly over Co- or Fe-based catalysts. The purpose of the present work is to study the catalyst activity and selectivity to form hydrocarbons for a series of cobalt-iron alumina supported catalysts using syngas with 1.0 ratio of hydrogen to carbon monoxide. This process requires that the WGS occurs internally over the FT-catalyst by the water produced in the FT-reaction because that the syngas has a smaller amount of hydrogen than the reactants stoichiometry needed.

II. BASIC THEORY

When the syngas obtained upon gasification of biomass used directly inside a Fischer-Tropsch (FT) synthesis, it requires the water gas-shift occurs internally over the FT-catalyst by the water produced in the FT-reaction for that the syngas has a smaller amount of hydrogen than the reactants stoichiometry needed. The WGS reaction occurs simultaneously with the production of hydrocarbons during FT synthesis over iron-based catalysts. These two reactions are:

FTS: OHCHHCO 2222 +−−→+

(1)

WGS: 222 HCOOHCO +↔+

(2) A high WGS reaction rate could convert a larger amount of CO (source of carbon) to CO2 rather than the desired hydrocarbon products. However, from the mass balance point of view, the production of CO2 parallels the production of H2 potentially increases the rate of reaction (1) so that CO and CO2 are exchangeable through the WGS reaction [1]. Thus the relative extents of FT synthesis and WGS reactions need to be optimized for the maximum production of hydrocarbons.

In our previous studies we reported the effect of feed gas composition to FT synthesis hydrocarbon products over Co/Al2O3 catalysts. We applied three

ISSN: 1411-1284


kind of syngases with H2/CO-ratio 2.1, 1.5 and 1.0 using dry feed [2] and with the addition of water [3]. The former study indicated that lower (H2/CO-ratio) decreased the CO conversion and CH4 selectivity while the C5+ selectivity and C3-(olefin/paraffin) were slightly increased.


The FT-synthesis experiments were applied in the same reactor (pressure 20 bars, initial temperature 483 K) with the initial space velocity 200 cm3/min or gas hourly space velocity (GHSV) 12 SL/gcat.h and held for approximately 12 hours. The GHSV was lowered to 50 cm3/min (or GHSV 3 SL/gcat.h) and held for another 12 hours. The reason for this is that the flow rate is the minimum flow can be run in the reactor set up to reach the maximum conversion. The reactor temperature was increased 10 K every period after each run 12 hours until the last period at temperature 523 K. Continuing our work reported in [2,3], the catalysts we used in this study were 0%, 2.4%, 6%, 9%, and 12% Fe added to 12%, 9.6%, 6%, 3% and 0% Co give 0:20, 5:20, 10:10, 20:5 and 20:0 ratios of Fe to Co. By remembering that the total bimetal impregnated was 12% weight, we can make simplicity of percentage ratio of Fe to Co as follow: (a) 0:100 (0% Fe:12% Co), (b) 20:80 (2.4% Fe:9.6% Co), (c) 50:50 (6% Fe:6% Co), (d) 80:20 (9.6% Fe:2.4% Co), and (e) 100:0 (12% Fe:0% Co). Catalyst characterization The characterization of catalyst applied in this study is reported below including temperature program reduction (TPR), temperature program oxidation (TPO or oxygen titration), CO chemisorptions, BET surface area measurements and X-ray diffraction

IV. CONCLUSIONS

3.1. Effect of Fe:Co ratio on the hydrocarbon selectivity

CO chemisorption Table 1 shows characterization results refers to this study. The CO adsorption data show the difference between the first and second isotherm. It is obvious that the difference results (the pure chemisorptions data) show a dramatic decrease of CO adsorbed as the

Fe loading increased. It is known that when Fe is covering the surface, both the CO uptake and the reducibility decrease, even though the same adsorption stoichiometry of CO on Co and Fe could be assumed [4]. This was the case when Fe was sequentially added to a Co/TiO2 catalyst. A probable reason for the lower CO uptake in that study could be an inefficient reduction of the Fe2O3 in poor contact with Co3O4, since supported Fe2O3 requires higher reduction temperature than Co3O4. In the current study, the dramatic decrease in CO uptake with increased Fe content (catalysts a – d) is not reflected in the degree of reduction (see Table 1) or FT activity (Table 2). The reason is believed to be a failure in measuring the number of active sites in the Fe-containing catalysts, rather than indicating that bimetallic catalysts would have much higher site activities than the pure Co catalyst. A probable explanation for the low CO uptake is that the Fe is oxidized by traces of H2O or O2 from the ambient air upon evacuation at high temperature, and that the evacuation time between the two isotherms was too long. Schanke [5] performed the evacuation between a first and a second CO adsorption isotherm for Fe/Al2O3 catalysts only for two minutes, since longer evacuation times (1 – 3 h) almost completely removed the initially adsorbed CO. Anyhow, the CO adsorption results indicate that Fe is effectively covering the Co sites, reaching full coverage of the Co sites between 20 – 50 % Fe (catalysts d – e). A surface enrichment of Fe in supported Co-Fe bimetallic catalysts, and a CO chemisorptions for high Fe content (above 33 % Fe) Co-Fe bimetallic catalysts similar to that of pure Fe, is in agreement with Duvenhage and Covill e [4]. Compared to the current study, they observed relatively small effects on the CO uptake when changing Fe:Co ratio on TiO2-supported bimetallic catalysts. Furthermore, they actually observed a 60 % increase in CO uptake between a 10 wt% Co/TiO2 catalyst and a co-impregnated 0.1 wt% Fe / 10 wt% Co / TiO2 catalyst. The replacement of Co with Fe in a 10 wt% metal / TiO2 catalyst actually improved the CO uptake even when the composition was 5 wt% Fe / 5 wt% Co / TiO2, compared to the pure Co catalyst. However, the FT activity (at a H2/CO ratio of 2) was generally lower the higher the Fe content, even if the Co loading was kept constant at 10 wt%. This is explained by a lower site activity of the alloy, and/or of a surface enrichment of Fe, which is known to have a much lower activity compared to Co [4]. In another study with more Fe:Co ratios investigated, Duvenhage and Coville [6] reported that the FT activity was higher for 5 wt% Fe / 5 wt% Co / TiO2 compared to either 2.5 wt% Fe : 7.5 wt% Co or 7.5 wt% Fe : 2.5 wt% Co supported on TiO2. The 5 wt% Fe / 5 wt% Co/ TiO2 catalyst possessed the highest reducibility but the lowest CO uptake of the three, resulting in a higher turnover frequency of that alloy, however far below that of a pure Co/TiO2 catalyst.

Table 1. Characterization result of Fe-Co/Al2O3 catalysts

Catalyst Fe:Co

BET surface area, [m2/g

catalyst]

CO chemisorption (first - second

isotherm) [µmol CO/g

catalyst]

Degree of

reduction [%]

0:100 174 74 47 5:95 156 37 48 15:85 145 35 51 20:80 149 13 50 50:50 153 8 49 80:20 157 7 41 100:0 162 7 40 Al2O3 195 - -

ISSN: 1411-1284


Table 2. Activities and selectivities for Fe-Co/Al2O3 catalysts in all periods.

The degree of reduction of the catalysts is slightly

increased when incorporating small to moderate amounts of Fe (catalysts b – e) compared to the mono-metallic Co-catalyst. The same observation has been made by Duvenhage and Coville [4, 6] who reported co-impregnated bimetallic Fe-Co/TiO2 catalysts giving higher degrees of reduction than the mono-metallic Co/TiO2. Duvenhage and Coville [4] reported that upon reduction at 623 K in H2 of the calcined 5 wt% Fe / 5 wt% Co / TiO2 catalyst discussed above (prepared by incipient wetness co-impregnation with the metal nitrates), a bcc Fe-Co alloy was formed and all metal in the catalyst was fully reduced. The reducibility for the Fe-Co bimetallic catalysts in the current study are much lower, and also much closer to each other, possibly indicating stronger interaction between the metals and the support, and a poorer contact between Co and Fe.

Effect of GHSV and Fe:Co ratio on the activity and selectivity

The activity and selectivity of the catalysts in which the variation of GHSV were done at 483K (Table 2). It is well known that Fe catalyst has a lower CO conversion on FT reaction compare to Co catalyst. Table 2 shows the CO conversion, hydrocarbon formation rate and selectivity to C5+ (CO2-free), selectivity to CH4 (CO2-free), selectivity to CO2 and C3(olefin/paraffin) for different Fe-Co/Al2O3. As can be seen in Table 2, the bimetal-containing catalysts (b-d) show a decrease in hydrocarbon formation rate with an increase of Fe content. The CO conversion decreases in the same way as hydrocarbon formation rate decreases. This is most probably due to the fact

that Fe gave a negative effect to catalyst activity. Similar results were made for a series of Fe:Co bimetallic catalyst employed in FT reaction using syngas with inlet molar H2/CO-ratio 2.0-2.1 [7]. Another synchronic result was found for a Fe:Co bimetallic catalyst employed in benzene hydrogenation and thiophene desulphurization reactions reaction [4]. It seems that the addition of two FT active metals, when used together, did not simply give the additive properties expected from knowledge of the properties of the individual metals [4, 6]. Some other studies on the use of Fe/Co catalysts for FT reaction using syngas from coal or natural gas have been reported [8, 9] and biomas derived syngas [10,11].

The selectivity to C5+ also decreases following the increase of iron content in the catalysts up to 80% (a-d). This can be understood that iron tend to form shorter chain of hydrocarbons than the longer one. The selectivity to C5+ for the merely iron catalyst (e) is slightly higher than this for catalyst d. Contrary with C5+ selectivity, the selectivity to CH4 and selectivity to CO2 increase very much with the increase of Fe/Co-ratio in the catalyst. The selectivity to CH4 and the selectivity to CO2 steep up from 11.3% and 3.9% for 20F-80C (b) to 23.1% and 10.5% for 80F-20C (d), respectively. These results are opposite with Duvenhage’s [6] who got a decrease in CH4 formation when the temperature was increased. A trend of decreasing C5+- and increasing of CH4- and CO2- selectivity culminate in 80F-20C (d) catalyst and the activity was lift up back when only Fe metal in the catalyst. The Fe/Al2O3 (e) catalyst does not show a lower CO conversion either a lower hydrocarbon formation rate than those for the one with 80% Fe content (d). The

Catalyst Condition CO

conversion [%]

Hydrocarbon formation rate

[gHC/gcat.h]

Selectivity to C5+ (CO2-free) [%]

Selectivity to CH4 (CO2-free) [%]

Selectivity to CO2 [%]

(a) 0Fe-100Co Dry feed 5.7 0.20 82.2 8.8 3.4

Reduced GHSV 23.6 0.21 86.6 6.0 1.6

(b) 5Fe-95Co Dry feed 7.5 0.27 78.0 10.1 3.1

Reduced GHSV 25.4 0.23 79.7 8.4 2.3

(c) 15Fe-85Co Dry feed 7.2 0.25 78.2 10.2 3.6

Reduced GHSV 22.0 0.20 78.3 9.3 2.9

(d) 20Fe-80Co Dry feed 5.8 0.20 70.0 13.5 5.1

Reduced GHSV 20.8 0.18 72.8 11.3 3.9

(e) 50Fe-50Co Dry feed 4.4 0.15 64.9 14.8 5.5

Reduced GHSV 13.8 0.12 64.8 14.9 5.5

(f) 80Fe-20Co Dry feed 3.2 0.11 64.9 15.6 8.1

Reduced GHSV 7.7 0.06 48.1 23.1 10.5

(g) 100Fe-0Co Dry feed 4.5 0.16 59.2 16.6 6.1

Reduced GHSV 10.0 0.09 51.9 20.0 8.7

ISSN: 1411-1284


selectivity to CH4 is 19.9% and selectivity to CO2 is 8.7% for Fe-single metal catalyst (e). In short words, the performance of Fe/Al2O3 to some extent is better than 80Fe:20Co. This is possibly due to a larger amount of metallic Fe present in the former which is indicated by the TPR pattern, where the Fe-peaks are shifted to lower temperatures for catalyst (d) in the [9]. It could be also the reason due to the BET surface area and the CO chemisorptions of Fe catalyst (e) is little bit higher than those for 80Fe:20Co (d) catalyst. This result indicates that the Fe-metal is responsible to form more CO2 than longer chain hydrocarbons in the product [12].

The CO hydrogenation rate over (a) (0 % Fe : 100 % Co) with stoichiometric feed (H2/CO = 2.1) is 0.37 g hydrocarbons (HCs) /gcat,h [2, 3]. It is obvious that no catalyst in the current study, with H2/CO = 1.0, comes close to this value. The WGS equilibrium constant at 483 K is approximately 370, calculated from data in reference [13]. which means that in order to reach the WGS equilibrium either the water partial pressure or the CO partial pressure must be very low at the reactor exit. For dry experiments with no external water addition, this means that essentially all water formed in the reactor must react in the WGS reaction (1.2), giving a usage ratio of 0.5. Hence, for a feed of H2/CO = 1.0, it is not necessary to reach the WGS equilibrium in order for the usage ratio to match the feed ratio, especially not with external water addition. However, in all FT experiments presented here, the WGS reaction quote in the gas exiting the reactor has a very low value (~ 10-3 – 10-1) compared to the equilibrium constant. The un-promoted Fe catalyst is obviously not especially active for the WGS reaction. It is known that in LTFT synthesis with Fe-based catalysts, the WGS equilibrium is not reached, due to the low temperature making the WGS reaction slow. In order to improve the WGS activity, the promotion of Fe by alkali is needed [14]. Duvenhage and Coville [4, 6] reported that the relative WGS activity was tripled for a 5 wt% Fe / 5 wt% Co / TiO2 catalyst upon 0.5 wt% K promotion. However, K is a poison to Co, lowering the total activity. A complement to the current study would be to keep the Co loading constant and to add Fe in different amounts, since it is obvious that lowering the Co loading reduces the overall FT activity, even if the WGS activity is increased by the Fe content. This would be economically reasonable since iron is about thousand times less expensive than cobalt. The main reason for lower activity is of course that the more active Co is replaced by the less active Fe, and also that the FT rate over Fe catalysts is negatively affected by the water partial pressure. Furthermore, not only the FT reaction, but also the WGS reaction takes place over the Fe surface. Only for catalyst (d) (20 % Fe : 80 % Co) the effect of a relatively high WGS activity (due to a high Fe content) on the total activity (CO conversion) was positive, if compared with the other bimetallic catalysts with lower Fe content. In general though, a higher WGS activity resulted in a lower FT activity

irrespective of water partial pressure in reactor (see Figure 3), indicating that the WGS- and FT reactions take place at the same type of sites.

IV. CONCLUSIONS

Fischer-Tropsch synthesis at 20 bars and 483 K with H2-poor syngas (H2/CO ratio = 1.0) was performed over different Fe:Co ratio (12 wt% bimetal) Al2O3-supported catalysts. All catalysts had very low activity for WGS, which resulted in low productivities (per gram catalyst). Even for external water addition, the WGS activities were poor. A higher Fe:Co ratio in the catalyst resulted in a higher WGS activity, however not lowering the H2/CO usage ratio to the desired value of 1.0 (the H2/CO ratio in the feed). The results indicate that the WGS- and FT reaction were competing for the same active sites (on the Fe-containing catalysts), giving a decrease in productivity for the catalysts with highest WGS activity. The higher the Fe content, the lower was the C5+ selectivity and C3 olefin/paraffin ratio. Water addition increased C5+ selectivity and C3 olefin/paraffin ratio and reduced CH4 selectivity. Fe was enriched at the catalyst surface, probably covering the Co sites. In order to achieve a high conversion of a H2-poor syngas, a combined high WGS- and FT activity is necessary which is difficult when the two reactions are competing for the same sites. Acknowledgement This work was supported by the Swedish Energy Agency. Odd Asbjørn Lindvåg from SINTEF, Trondheim-Norway is acknowledged for his kindly help during the experimental part of the work.

REFERENCES

[1] Raje, A., Inga, J.R. Davis, B.H., Fuel 76 (1997) 273. [2] D. Tristantini, S. Lögdberg, Ø. Borg, B. Gevert, A.

Holmen, Fuel Processing Technol., 88 (2007) 7, 643-649. [3] D. Tristantini, and B. Gevert, A. proceeding 14th RSCE, UGM, Yogyakarta, 2007. (accepted). [4] D.J. Duvenhage, N.J. Coville, Appl. Catal. A 153 (1997) 43. [5] D. Schanke, Hydrogenation of CO over supported iron catalysts, PhD Thesis, The Laboratory of Industrial Chemistry, The University of Trondheim, The Norwegian Institute of Technology, Trondheim, 1986. [6] D.J. Duvenhage, N.J. Coville, Appl. Catal. A289 (2005) 231. [7] Guerrero-Ruiz, A., Sepulueda-Escribano, A., Rodriguez- Ramos, I, Appl. Catal., 81 (1992) 101. [8] Wender, I., Fuel Process. Technol. 48 (1996) 189. [9] Soled, S.L., Fiato, R.A., US Patent 4607020 (1986). [10] Mirzaei, A.A. , Habibpour, R., Kashi, E., Appl. Catal. A 296 (2005) 222. [11] Jun, K., Roh, H.,Kim, K.,Ryu, J., Lee, K., Appl. Catal. A. 259 (2004) 221. [12] Eilers, J.,Posthuma, S. A., Sie, S. T., Catal. Lett., 7

(1990) 253. [13] H. Hiller, R. Reimert, in: B. Elvers, S. Hawkins, M. Ravenscroft, J.F. Rounsaville, G. Schulz (Eds.), Ullmann’s Encyclopedia of Industrial Chemistry, vol.

ISSN: 1411-1284


A12, 5th ed., VCH Verlagsgesellschaft mbH, Germany, 1989, pp. 179-181. [14] A. Holmen, D. Schanke, G. Sundmark, Appl.

Catal. 50 (1989) 211.

ISSN: 1411-1284


The Deactivation Behaviour of HZSM-5 Catalyst at Various Temperatures of Acetone Conversion to

Mono-aromatic Hydrocarbons

Setiadi*, Prilly Fatticianita J.*, Toshinori KOJIMA(1)*

* Department of Chemical Engineering, Faculty of Engineering, University of Indoneisa, Kampus Baru UI Depok 16424, Tel. 021-7863516, email : [email protected], INDONESIA

* Department of Applied Chemistry, Faculty of Engineering, Seikei University 3-3-1, Kichijoji-Kitamachi, Musashino, Tokyo 180-8633, JAPAN

Abstract- Catalytic stability of HZSM-5 (Si/Al=25) on various temperatures (573K, 623, 673 and 723 K) have been studied to obtain the suitable temperature for the formation of monoaromatic chemicals in acetone reaction. The reaction was performed in the near atmospheric pressure with acetone velocity at 4 h-1. The results show that the catalyst stability is strongly depend on the reaction temperatures, and the catalyst generally underwent a deactivation on each temperature. However, the best result of acetone reaction is at 673 K, acetone conversion are closed to 100 % during 17 h and the yield of monoaromatic chemicals was above 70 % for 13 h of reaction. The highly decreasing of surface area (total area or micropore area) of used catalyst, the deactivation is considered caused by the pore blocking on the external surface of HZSM-5. Key words- Reaction Temperatures, Monoaromatic Chemicals, HZSM-5, Deactivation

I. INTRODUCTION For design or developing a new catalyst which can works with a long life time is attractive and challenging tasks for increasing the operational and economical benefits of an industrial process. In principle, life time of a catalyst is correlated to the extent of reaction and the selectivity to the desired product in order to meet the aspect of economically acceptable to the process in the real commercial industries. Depend on the reaction condition and the catalyst employed, life time of a catalyst is mostly ascertained by these factors.

The probability of deactivation process in the acetone conversion via aldol condensation reaction is enough high due to the formation of a condensation product will shut down the reaction over H-ZSM-5

catalyst. Aldol condensation and secondary reactions of acetone on zeolite H-ZSM-5 were extensively studied using 13C NMR spectroscopy (Xu et al., 1994) and using IR observation of acetone adsorption (Zaki et. Al, 2000). The formation of the acetone reaction condensation product is considered to the high coverage of acetone (high ratio of acetone/active site), which allows the high mobility of acetone with the possibility of the bimolecular condensation reaction product such as mesityl oxide (Panov and Fripiat, 1998). However, the condensation product has a much lower mobility than acetone. As pointed by Melo et al. (1997), the condensation product is retained in the zeolite pores because of their low volatility, of adsorption on the acid sites or because their size is close to or greater than the size of pore aperture.

Some reaction techniques were employed to increase to the catalyst’s lifetime in acetone conversion by adding the water (Podrebarac et. al, 1997; Lucas et. al, 2001). However, the systematic study to correlate the lifetime or stability of catalyst with reaction temperature is still absent. The impressive results of the effect of temperature on the acetone conversion using H-ZSM-5 have been demonstrated by Chang et al. (1981), Chang and Silvestri (1977). But, the knowledge about the activity and stability on each temperature and the typical ratio of Si/Al of HZSM-5 catalyst were nothing. For developing a catalyst for the acetone conversion to aromatic chemicals, we have reported in the previous study that the activity and stability of HZSM-5 catalyst with Si/Al ratio=25 is very high during 10 h of reaction with 100 % conversion. And 673 K is the favourable temperature for the formation of aromatic chemicals (Setiadi et al., 2003). Based on the arguments summarized above, this report presents the results of acetone conversion on the longer times of stream in more than 10 h of reaction on the various temperatures.

ISSN: 1411-1284


Fig. 1 Acetone conversion during time on stream reaction on various temperatures : 723 K (∆), 673 (O),

623 K (ÿ) and 573 K (◊).

II. EXPERIMENTAL

Fine crystal powder of HZSM-5 （Si/Al=25）was used as catalyst for the reaction of acetone aromatization. One gram of this HZSM-5 was mixed with quartz sand to avoid the excessive increase of pressure drop. This mixture was packed in a SUS 316 reactor column with 6 mm i.d and 700 mm length. Some amount of quartz sand was added at the upper part of the catalyst bed preheating and vaporising the acetone loaded into reactor. A stainless steel rod was inserted at the lower part of the reactor to strengthen the catalyst bed against the pressure due to the gas flowing through the bed. To fix completely the catalyst bed in its position, quartz wool was filled on each side of the catalyst layer to keep the possible entrainment of the powder during the experiment performed.

The reaction temperatures employed were at 723, 673, 623 and 573 K with acetone space velocity 4 h-1 at near atmospheric pressure. To examine the stability of HZSM-5 in the acetone conversion on those temperatures, HZSM-5 fresh catalysts were used when each experiment was started to run. The detail description of the experimental method have been reported in the our previous work (Setiadi et. al, 2003).

The fresh and used catalysts after reaction at each temperature were characterized by BET method for total and micropore surface area measurements. Nitrogen was used as adsorbed gas at the liquid the liquid nitrogen temperature and these measurements were carried out by the apparatus of Micromeritics Gemini model 2360. The determinations of the micropore surface area were performed by t plot method and the thickness of adsorbed layers (t) were calculated by Halsey equation.

III. RESULT AND DISCUSSION

In order to find the optimum temperature for obtaining the high activity and stability of H-ZSM-5 with Si/Al ratio of 25, the reaction of acetone conversion was done at various temperatures at 723 K, 673 K, 623 K and 573 K. The results are in Fig. 1 which are the plots of acetone conversions versus times on stream. The reason of the using of HZSM-5 with the Si/Al ratio 25 is consistent to our previous result that this ratio exhibited the effective ratio of HZSM-5 catalyst for the conversion of acetone to aromatic chemicals (Setiadi et al., 2003). It is clear shown that if the reaction was performed on the higher temperature region (more than 673 K), the activity and stability of HZSM-5 were much better than on the lower ones. Within 17 hours of reaction, the results of reactions at 673 and 723 K show that the time on stream curves of acetone conversions were

stable at 100 % on both temperatures. Unfortunately, the a- cetone conversions obtained after 17 h reaction underwent a sharp decrease to be around of 20 % especially for 673 K. It indicates that the HZSM-5 catalyst deactivated after 17 h of reaction.

In the case of the lower temperature region

(lower than 673 K), it obviously shown that the low

stabilities of H-ZSM-5 can be seen and the acetone conversions were decreased in the shorter times of stream to be less than 20 % within 12 h reaction (at 623 K) and 4 h reaction (573 K). The tendency of this deactivation at the lower temperature region, may be caused by the low ability of HZSM-5 to activate the acetone molecules than that to the high temperature region. And the acetone reaction leads to a formation of multiple condensation products with the higher rate of formation comparing to a proceed of dehydration process

Fig. 2 shows the monoaromatic yields during times on

0

20

40

60

80

100

0 5 10 15 20 25 30Time on stream [h]

Ace

tone

con

vers

ion

[%]

0

20

40

60

80

100

0 5 10 15 20 25 30

Time on stream [h]

Yie

ld [

%m

ol c

arbo

n]

Fig. 2 Yield of aromatic compound during time on stream reaction on various temperatures : 723 K (∆), 673 K (O), 623

K (ÿ) and 573 K (◊).

ISSN: 1411-1284


stream. The similar result was found is that the time on stream yield of monoaromatic compounds from the acetone conversion at 673 K were also superior than that at other temperatures during 13 h. Even though the stability of acetone conversion obtained by performing the reaction at 723 K was little bit higher than 673 K, but this time on stream yield at this temperature was lower. These results lead to a conclusion that the temperature reaction of 673 K is the suitable temperature for conversion of acetone for aromatic product formation.

In general, the deactivation can not be avoided to occur on the each temperature region and the deactivation rate is more rapid on the lower temperature region. The possibility of a formation poly condensation products with heavier molecular weight on each temperature is high and these products covered the center of active sites or to block the pore channel of HZSM-5.

To address the possible cause of the decreasing of catalyst activities, we observed the surface area of fresh catalyst and the surface of each catalyst after reaction on each reaction. Table 1 shows the result of surface area measurement for the fresh and deactivated catalysts

Table 1 Result of surface area measurement

H-ZSM-5 sample BET m2/g

Micropore m2/g

1 Fresh catalyst 321.8 209.4

2 Used Catalyst at 573 K 76.0 44.2



on each temperature. It is obviously shown that there was a dramatic change of the surface area (BET total surface area or micropore surface) of fresh HZSM-5 and deactivated catalyst. The decreasing of surface area of fresh HZSM-5 to be deactivated catalyst was about 76 % for total area and 80 % for micropore area. And the fact (not shown here) is that there was any change of physical properties from white color (fresh catalyst) toward black color (deactivated catalyst). These findings strongly suggest that the deactivation is main caused by surface blocking or pore plugging of HZSM-5. As pointed out by Lucas et al. (1997), the formation of external coke leads to block of the channel structure even at low coke contents. The relative sharply decreases of acetone conversions at all temperatures as shown in Fig. 1, it could be correlated to a evident of a high possibility of the multiple condensation product formation (coke) which may cover the external surface of HZSM-5 on the mouth of pore channels.

The deactivation could be caused not only by the evident of pore blocking by coke formatted during reaction proceed, but also the process of the dealumination process. The detection of alumina external surface could be appropriately performed by

EPMA measurements. Based on the distinction of curve lines of spatial distribution for alumina obtained for both catalyst (not shown here), it shows that the amount of external alumunium atom for used catalyst is higher than that for fresh catalyst. These result ssuggest that the dealumination process have also occurred during the acetone reaction. And also indicates that dehydration process as one step of reaction mechanism and the water produced will react with the framework aluminium atoms. Depend on the reaction type employed, the chemical species produces due to the dealumination process can be more catalytically active or inactive comparing to the tetrahedral framework aluminium atom. The excellent report described the dealumination phenomenon have been demonstrated by Masuda et al. (1998) which showed that tetrahedral aluminium atoms are extracted out of the framework and transformed into partially distorted octahedral aluminium atoms. These atoms move to the outer surface of the zeolites crystals. However, those aluminum atoms are likely not active for the acetone reaction and to allow in more completely decreasing of HZSM-5 activity as shown in Fig.1.

IV. CONCLUSIONS

In general, the deactivation process is observed

on all temperatures observed (573, 623, 673 and 723 K) over HZSM-5 (Si/Al-25) and is more rapid if the acetone reaction was performed on the lower temperature region. Based on the time on stream stability experiments, the best temperature for the formation of monoaromatic chemicals is 673 K. The stability of the catalyst achieved was close to 100 % conversion during 17 h and the yield was obtained in more than 70 % for 13 h of reaction.

The decreasing of catalyst activity should be connected to the decreasing of the surface area (BET or micropore) by pore blocking due to the formation of heavier molecular products from acetone condensation reaction.

REFFERENCES

1) Chang, Clarence D., W. H. Lang, and W.K. Bell; "Molecular Shape-Selective Catalysis in Zeolite," in Catalysis of Organic Reactions edited by William R. Moser, Marcel Dekker Inc., 73-94 (1981)

2) Clarence D. Chang and Anthony J. Silvestry, "The Conversion of Methanol and Other O-Compounds to Hydrocarbons over Zeolite Catalysts," Journal of Catalysis 47, 249 (1977)

3) Lucas, A., P. Canizares, A. Duran and A. Carrero; “ Coke Formation, Location, nature and regeneration on Dealuminated HZSM-5 type Zeolites”, Applied Catalysis A : General,156 , 299-317 (1997).

4) Lucas, A., P. Canizares, A. Duran,"Improving deactivation behaviour of HZSM-5 catalysts", App.Catal. A : General, 206, p.87-93 (2001)

ISSN: 1411-1284


5) Masuda, T., Yoshihiro Fujikata, Shin R. Mukai, Kenji Hashimoto; “ Changes in Catalytic Activity of MFI-type Zeolites Caused by Dealumination in a Steam Atmosphere”, Applied Catalysis Ａ: General, 172, 73-83 (1998).

6) Melo, L., P. Magnoux, G. Giannetto, F. Alvarez, M. Guisnet; " Transformation of acetone over a 0.4PtHMFI(60) catalyst. Reaction scheme," Journal of Molecular Catalysis A : Chemical, 124, 55-161 (1997)

7) Panov A.G.; Fripiat J.J.; "Acetone Condensation Reaction on Acid Catalysts," Journal of Catalysis, 178, no. 1, 188-197 (1998)

8) Panov, A. and J.J. Fripiat; "An Infrared Spectroscopic Study of Acetone and Mesityl Oxide Adsorption on Acid Catalyst," Langmuir, 14, 3788-3796 (1998)

9) Podrebarac, G.G., Ng F.T.T, Rempel, G.L, A kinetic Study of the Aldol Condensation of Acetone using an Anion Exchange Resin Catalyst, Chemical Engineering Science, 52, 17, 2991, (September 1997)

10) Setiadi, S., T. Tsutsui, T. Kojima," Conversion of Acetone to Aromatic Compound with HZSM-5, The journal of Japan Institute of Energy, 82 (12), (2003)

11) Xu, Teng, Eric J. Munson, and James F. Haw; "Toward a Systematic Chemistry of Organic Reactions in Zeolites: In Situ NMR Studies of Ketones," J. Am. Chem. Soc., 116, 1962-1972 (1994)

12) Zaki, M.I., M. A. Hasan, F.A. Al-Sagheer, and L. Pasupulety; "Surface Chemistry of Acetone on Metal Oxides: IR Observation of Acetone Adsorption and Consequent Surface Reactions on Silica-Alumina versus Silica and Alumina," Langmuir, 16, 430-436 (2000)

ISSN: 1411-1284


CO2 Absorption Through Hollow Fiber Membrane

Gas-Liquid Contactors

Sutrasno Kartohardjono Faculty of Engineering, University of Indonesia, Kampus Baru UI, Depok 16424

Tel. (6221) 7863516, fax. (6221) 7863515, email: [email protected]

Abstract– Hollow fiber membrane modules have been widely used as contactor and filtration devices to provide high surface area in a small volume. Membrane-based contactor devices provide an alternative technology for gas-liquid contacting operations that overcomes disadvantage of conventional contactors such as columns and towers, and in the same time offers substantially more interfacial area using hollow fiber membrane contactors. In this study, performance of hydrophobic microporous polypropylene hollow fiber membrane contactors is examined to absorb CO2 using water and dilute solution of NaOH as absorbents.

Experiment results show that CO2 flux through the membrane contactors using 0.01 M NaOH solution is 120.000 times higher than using water as solvent. The CO2 flux decreases with increasing contactor-packing fraction for the same liquid velocity using both solvents. The same effect also occurred for the mass transfer coefficients characteristics. Liquid pressure drops increase with increasing contactor-packing fraction at the same liquid flow rate due to an increase in friction between fibers and water. Based on the friction factor data the fibers surface did not behave like a smooth pipe within the range of velocity in the experiments. The friction factors obtained for the membrane contactors in this study are 4.6 to 14.6 times higher than the theoretical value based on flow of fluids in the smooth pipe. Keywords– hollow fiber, contactor, mass transfer, pressure drop

I. INTRODUCTION

ollow fiber membrane modules have been widely used as contactor and filtration devices to provide

high surface area in a small volume [1]. As a gas-liquid contactor, unlike more conventional membrane applications such as microfiltration, ultrafiltration and reverse osmosis, the driving force for separation is a concentration rather than a pressure gradient [2,3]. Therefore, only a small pressure drop across the membrane is required to ensure that gas-liquid

interface is remain immobilized at the membrane pores [2,4,5].

Membrane-based contactor devices provide an alternative technology for gas-liquid contacting operations that overcomes disadvantage of conventional contactors such as columns and towers, and in the same time offers substantially more interfacial area using a suitable membrane configuration or module such as hollow fiber membrane modules. In hollow fiber membrane contactors, gas flow rate and liquid flow rate can be varied independently of each other without flooding, loading and foaming, which are characteristics of dispersion-based gas-liquid contactors. Furthermore, all of the membrane surface area is available for contacting regardless of how low the flow rates are. Membrane-based contactors are easier to be scaled-up than other conventional mass transfer equipment.

Membrane operations usually scale linearly, so that an increasing in capacity is achieved simply by the add of membrane modules subject to the limitations of supporting equipment such as transfer pump, piping, etc. In addition, membrane-based contactors can be operated over a wide range of capacities by modular design. Small or large capacities can be obtained by using few or many modules.

However, membrane based contactors have several disadvantages such as the additional resistance of membrane it self to mass transfer performance, subject to channeling which result in a loss efficiency, and subject to fouling. Membrane Systems have a short life and a limitation in the number of equilibrium stages due to pressure drop constraints. These relatively few disadvantages are often compensated by the advantages cited above [6].

Hollow fiber membrane contactors superior to their surface area per volume compare to conventional contactors, such as absorption and extraction columns, typically 50cm2/cm3, and may be as high as 300cm2/cm3 [7]. These values are about 30 times more area than what is achievable in gas absorbers and 500 times what is obtainable in liquid/liquid extraction columns [6]. Hollow fiber membrane contactors (HFMC) have been demonstrated in a wide range of application such as in gas stripping and gas absorption. In the hydrophobic microporous

H

ISSN: 1411-1284


membrane gas/liquid contactors, the pores are filled with gas and as a result the resistance in the membrane can be neglected to the overall resistances of the contactor.

This behavior of hydrophobic membrane contactor has brought many researchers to utilize hydrophobic microporous hollow fiber membrane contactors to remove acid gases, mainly CO2 and H2S from natural gas or CO2 from flue gas stream [8]. The aims of this study are: to evaluate the effectiveness of hollow fiber membrane module as a contactor device in CO2 absorption process; and to develop a simple model, which is useful in the membrane contactor design, for characterizing the mass transfer correlation of CO2 absorption process in the membrane contactor.

II. BASIC THEORY

The mass transfer of CO2 from the lumen fiber to the solvent on the shell-side involves three steps: transfer of CO2 in the gas side, diffusion of oxygen through the membrane, and diffusion of oxygen into the liquid side. The driving force for CO2 removal from the gas side comes from the CO2 concentration gradient between the gas and the phases across the membrane.

It is more convenience to express overall mass transfer coefficient, KL, in the contactors based on the liquid phase for many gas absorption/stripping processes. For a microporous-hydrophobic membrane with gas-filled pores, the relation is,

iLim

i

iG

i

L kk

H

k

H

K

11 ++= (1)

where Hi is the Henry’s law constant for the ratio of the liquid concentration to the gas concentration in equilibrium. For sparingly soluble gases the relation is,

iLL kK

11 ≅ (2)

indicating that the liquid film controls the mass transfer.

III. EXPERIMENTAL RESULTS Hollow fibers membranes used in the experiment were made of polypropylene supplied by AKZO. The fibers are 2.7 mm in outer diameter and 0.45 mm in wall thickness. The hollow fibers modules were made by potting both ends of the fibers into 1.5 cm nipple using epoxy (Araldite). The module was connected to external shell made from PP pipe with a PP T-joint, and at the other end of the nipple was connected to vacuum pump using brass reducer 1.5 cm to 0.3 cm. The diameter of the external shell can be varied using appropriate connection to PP T-joint where in this study 1.6 cm shell diameter was used. There were 3 contactors of about 40-cm length and 10, 15 and 20

fibers used in the experiments to give membrane packing fractions in the contactors of 0.2, 0.3, and 0.4, respectively.

Schematic of experimental configuration as shown in Fig.1 was used to measure mass transfer performance and pressure drop of the membrane contactors. The liquid in the reservoir, which is water or 0.1 M NaOH solution, was initially pumped through the membrane contactor at the shell side. Meanwhile, CO2 gas was sent to the contactor through the lumen of fiber. pH and temperature of the liquid were automatically measured every 20 s from the pH-meter probe, and digital manometer measured liquid pressure differences between inlet and outlet contactor.

Figure 1. Schematic diagram of experimental set up.

The overall mass transfer coefficients from the

experiments are calculated using Equation (3),

1*

0*

lnCC

CC

A

QK

m

L

−−

= (3)

where C*, Ct and C0 are CO2 concentrations in equilibrium, liquid inlet and outlet contactor, QL is liquid flow rate to the membrane contactor, Am is outer membrane fibers surface area, and KL is the observed overall mass transfer coefficient, respectively. Friction factors of the contactors were calculated using Equation (4),

22l

vl

Pdf e

ρ∆= (4)

where f is friction factor of the module, ∆P is liquid pressure drop, de is equivalent diameter of the contactor, l is contactor length, ρ is water density, and vL is water velocity, respectively.

CO2 flux through the membrane contactors, JCO2, and the overall mass transfer coefficients, KL, from the experimental results were plotted versus water velocity vl as shown in Figure 2 and 3, respectively. As it can be seen from Figure 2, the water can absorb CO2 through the membrane contactor as high as 4.5x10-8 mole CO2 per square meter membrane area per hour. Meanwhile, 0.01 M NaOH solution can absorb CO2 through the membrane contactor as high as 5.5x10-3 mole CO2 per square meter membrane area per hour, which is 120.000 times higher than using water as solvent.

ISSN: 1411-1284


The CO2 flux decreases with increasing contactor-packing fraction for the same liquid velocity. The same effect also occurred for the mass transfer coefficient of the membrane contactors. This can be explained as in a region of low packing fraction contactor, transverse flow and surface renewal effect is seem to be more influence to the mass transfer performance rather than channeling effect, meanwhile at the higher packing density the channeling effect is more dominant [1]. From another point of view, the geometrical dependencies of boundary layer profiles might also increase the mass transfer coefficient of contactor with lower packing density. The increase is directly related to the boundary layer conditions, where the profiles of boundary layers on a curved body become thinner with an increasing degree of curvature, as for example decreasing do fiber with other conditions constant. Therefore, based on this theory, the mass transfer coefficient will increase with decreasing packing fraction of the contactor, especially at higher liquid velocities. The channeling occurred because of the existing regions of densely- and loosely-packed in the contactor which creates preferential flow around fibers in the contactor resulted mainly from uneven distribution of fibers and flow [9]. Membrane contact area might have been reduced in densely packed regions as fibers were more adhered to each other, reducing the availability of contacting surface.

1.E-08

1.E-07

1.E-06

1.E-05

1.E-04

1.E-03

1.E-02

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7

V L (m/s)

JC

O2

(mo

le/m

2 .s)

N=10N=15N=20N=10 N=15N=20

Figure 2. Variation of CO2 fluxes JCO2, in water (filled

symbols) and 0.01 M NaOH solution (open symbols) with liquid velocities vL, in the membrane contactors

To examine the dependence of CO2 transfer

coefficient on the process parameters using water as a solvent, correlation are conveniently expressed in terms of dimensionless Sherwood number, Sh, Reynolds number, Re, and Schmidt number, Sc.

CB ScASh Re= (5) In this study the Schmidt number, Sc, was not varied, so the 1/3 power-dependence in the literature was assumed [10], and Equation (5) become,

33.0Re ScASh B= (6)

1.E-07

1.E-06

1.E-05

1.E-04

1.E-03

1.E-02

0.1 1

v L (m/s)

KL

(m/s

)

N=10N=15N=20N=10N=15N=20

Figure 3. Variation of mass transfer coefficients KL, in water (filled symbols) and 0.01 M NaOH solution (open symbols)

with liquid velocities vL, in the membrane contactors

Figure 4 shows the experimental data as Sh plotted against Re at various packing fraction contactors to obtain the exponent for Reynolds number, B in Equation (5) for each contactor. Furthermore, Equation (6) can be simplified into Equation (7) to see the dependency of Reynolds number on mass transfer performance.

baSh Re= (7)

The exponent for Reynolds number, b, was obtained by the linear regression of Figure 4 and it was found that the values were ranging from 0.63 to 0.65 and have the average value of b=0.64. This value indicated that the mass transfer in the contactors is dominated by turbulent flow [10]. The values of a for each contactor were obtained from the slope of the experimental Sherwood number against Re0.64 where these values are ranging from 0.0115 to 0.0387. To obtain the correlation between a and packing fraction of the contactor,ϕ, the values of a were plotted against contactor packing fraction where the correlation between a and packing fraction is best fitted by Equation (8),

72.10026.0 −= ϕa (8)

where ϕ is packing fraction of the contactor, respectively.

Equation (8) describes the dependency of geometry of the hollow fiber membrane contactors employed in this study where the mass transfer performance is related to the packing fraction of fibers in the contactors. An empirical correlation for mass transfer performance in the hollow fiber membrane contactors used in the experiments can then be correlated in the form

ISSN: 1411-1284


64.072.1 Re0026.0 −= ϕSh (9)

To see the effects of contactor packing fraction on

pressure drop due to friction loss, the friction factors of the contactor were obtained from Equation (4) and then were plotted against Reynolds number as shown in Figure 5. The friction factor increases with decreasing contactor-packing density for the same contactor diameter due to an increase in hydraulics diameter of the contactor as expressed in Equation (4). The dependency of friction factor on Reynolds number for fluid flow in the contactor has general form,

baf Re= (10)

where f, a and b are friction factor and empirical constants from the experimental results, respectively. The exponent b has a negative value as can be seen from many literatures [11,12]. As for instance, b is –1 and -0.25 for liquid flow in the laminar region (Re<2000) and in the turbulent region (Reynolds number up to 105), respectively [11]. It can be seen here that the negative value of b becomes smaller as the flow more turbulent. In this study where the range of Reynolds number measurement in the contactor is from 800 to 3400 has b values ranging from –0.27 to –1.01 as shown in Figure 5. As a comparison, It was found that the friction factor data has value of b = -1.1 for Re<5 and b = -0.5 for 5<Re<100 based on the experiment using microporous hollow fiber membrane blood- oxygenator [12].

0.1

1.0

10.0

100 1000 10000

Re

Sh

N=10N=15N=20

Figure 4. Variation of Sherwood numbers Sh, with Reynolds

numbers Re, in the membrane contactors

The friction factor ratio of the contactor to the

smooth pipe is shown in Figure 6. These data reveal that the fibers surface did not behave like a smooth pipe within the range of velocity in the experiments. Thus viscous sub layer did not develop due to the movement of the fibers as quantitatively expressed in very low viscous drag values compare to inertial drag values. The random movement of the fibers was more pronounced at lower velocities, and as a result, the value of the friction factor is also higher at lower

velocities [4]. The friction factor ratio in Figure 6 is higher for lower packing densities due to the more movement of the fibers in low packing fraction contactor.

0.001

0.01

0.1

1

100 1000 10000

Re

f

N = 10N = 15N = 20

Figure 5. Variation of friction factor f, with Reynolds

number Re, in the membrane contactors

The friction factors obtained for the membrane

contactors in this study are 4.6 to 14.6 times higher than the theoretical value based on flow of fluids in the smooth pipe as shown in Figure 6. As comparison, It was reported that the friction factor of hollow fiber blood oxygenators was about 7 times higher than the theoretical value at Re<5 and about 31 times higher than the theoretical value at 5<Re<100 [12]. Meanwhile, other researcher reported that the friction factor of sealed end polypropylene hollow fiber contactors with packing densities 8.7 to 9.6 % were about 2 times higher than the values for pipe flow reported in the literature [4]. It can be seen here that the friction factor ratio can vary from contactor to contactor depend on the geometry of the contactors.

0

2

4

6

8

10

12

14

0 1000 2000 3000 4000

Re

f R

N = 10N = 15N = 20

Figure 6. Variation of friction factor ratio fR, with Reynolds

number Re, in the membrane contactors

IV. CONCLUSIONS

ISSN: 1411-1284


Experiments have been conducted to absorb CO2 gas using water and 0.01 M NaOH solution as solvents through hydrophobic microporous polypropylene hollow fiber membrane contactors of packing fraction 0.2, 0.3 and 0.4. The water can absorb CO2 through the membrane contactor as high as 4.5x10-8 mole CO2 per square meter membrane area per hour. Meanwhile, the 0.01 M NaOH solution can absorb CO2 through the membrane contactor as high as 5.5x10-3 mole CO2 per square meter membrane area per hour, which is 120.000 times higher than using water as solvent. The CO2 flux decreases with increasing contactor-packing fraction for the same liquid velocity using both water and 0.01 M NaOH solution as solvents. The same effect also occurred for the mass transfer coefficients of the membrane contactors.

The mass transfer performance of the membrane contactors using water as absorbent can be formulated as a function of contactor geometry and liquid flow rate. The contactor geometry is represented by packing fraction, whilst the flow rate is represented by Reynolds number. An empirical correlation for mass transfer performance in the hollow fiber membrane contactors in the range of packing fraction of 0.2 to 0.4 can be correlated in the form 64.072.1 Re0026.0 −= ϕSh .

Liquid pressure drops increase with increasing contactor-packing fraction at the same liquid flow rate due to an increase in friction between fibers and water. Based on the friction factor data the fibers surface did not behave like a smooth pipe within the range of velocity in the experiments. The friction factors obtained for the membrane contactors in this study are 4.6 to 14.6 times higher than the theoretical value based on flow of fluids in the smooth pipe.

REFERENCES [1] Lipnizki, F., and Field, R.W., Mass transfer

performance for hollow fibre modules with shell-side axial feed flow: using an engineering approach to develop a framework, Journal of Membrane Science, 193 (2001), 195-208.

[2] Dindore, V.Y., and Versteeg, G.F., Gas–liquid mass transfer in a cross-flow hollow fiber module: Analytical model and experimental validation, International Journal of Heat and Mass Transfer, 48 (16) (2005), 3352-3362.

[3] Sirkar, K.K., Other new membrane processes, in W.S.W. Ho and K.K. Sirkar (Eds.), Membrane Handbook, Chapman & Hall, New York, (1992), 885-899.

[4] Ahmed, T., Semmens, M.J., and Voss, M.A., Energy loss characteristics of parallel flow bubbleless hollow fibre membrane aerators, Journal of Membrane Science, 171 (2000), 87-96.

[5] Lim, S.P., Tan Xiaoyao and Li, K., Gas/vapor separation using membranes: Effect of pressure drop in lumen of hollow fibers, Chemical Engineering Science, 55 (4) (2000), 2641-2652.

[6] Gabelman, A., and Hwang, S.T., Hollow fiber membrane contactors, Journal of Membrane Science, 159 (1999), 61-106.

[7] Wickramasinghe, S.R., Semmens, M.J., and Cussler, E.L., Better hollow fibre contactors, Journal of Membrane Science, 62 (1991), 371-388.

[8] Mavroudi, M., Kaldis, S.P., and Sakellaropoulos, G.P., A study of mass transfer resistance in membrane gas–liquid contacting processes, Journal of Membrane Science, 272 (1-2) (2006), 103-115.

[9] Wu, J., and Chen, V., Shell-side mass transfer performance of randomly packed hollow fibre modules, Journal of Membrane Science, 172 (2000), 59-74.

[10] Costello, M.J., Fane, A.G., Hogan, P.A., and Schofield, R.W., The effect of shell side hydrodynamics on the performance of axial flow hollow fibre modules, Journal of Membrane Science, 80 (1993), 1-11.

[11] Coulson, J.M., and Richardson, J.F., Chemical Engineering, Volume 1: Fluid Flow, Heat Transfer and Mass Transfer, Pergamon Press, Oxford, New York, Toronto, Sydney, Paris, Frankfurt, 1978, 3rd ed., pp. 40-45.

[12] Wickramasinghe, S.R., Garcia, J.D., and Han, Binbing, Mass and momentum transfer in hollow fibre blood oxygenators, Journal of Membrane Science, 208 (2002), 247-256.

ISSN: 1411-1284


Enzymatic and Acid Hydrolysis of Bagasse for Ethanol Production by Simultaneous Sacharification and

Fermentation

Samsuri, M.1,2, Gozan, M.1, Nasikin, M. 1, Prasetya, B. 2

1Chemical Engineering Department, Faculty of Engineering,

University of Indonesia, Kampus UI, Depok 16424, Indonesia, 2 Centre of Biotechnology, Indonesian Institute of Science (LIPI),

Jl. Raya Cibinong km 48, Bogor, Jawab Barat, Indonesia

Email address: [email protected]

Abstract-Ethanol was produced from bagasse using enzymatic hydrolysis, acid hydrolysis and combination of Enzymatic and acid hydrolysis in Simultaneous Saccharfication and fermentation. Cellulase and xylanase were used for enzymatic hydrolysis and HCl low concentrations (1% and 0,5%) was used for acid hydrolysis.

The results showed that the highest ethanol yield with enzymatic hydrolysis with enzyme cellulase at pH 5, about 5,51g/L or about 11, 24% base on original bagasse. The highest ethanol production from bagasse by combination of enzyme cellulase and xylanase at pH 5 about 7,63 g/L or 15,25%. The highest ethanol yield after hydrolysis with HCl 0,5% is 3,24 g/L at incubation for 24 h. The highest ethanol yield after hydrolysis with HCl 1% is 3,77 g/L or 7,54% at incubation for 24 h. The highest ethanol yield after hydrolysis with combination of cellulase, xylanase and HCl 1% about 8,22 g/L or 16,45%. This yield increasing about 2% if compared with ethanol yield without acid addition. If compared with acid hydrolysis by HCl 1% increasing about 8% based on original bagasse

Key words : Bagasse, SSF, Enzymatic hydrolysis cellulose, xylanase, Acid hydrolysis HCl 1. INTRODUCTION

Most ethanol produced in the world today is derived from starch or sucrose (Gong et al., 1999). Ethanol fermentations are traditionally carried out for wine or beer production, but ethanol for transportation and industry are a large potential market and growing use. Ethanol belongs to clean combustion. Its oxygen content decreases emissions in mix combustion with gasoline. Because this ethanol is originally plant matter, its use as fuel does not contribute to the net accumulation of carbon dioxide in the atmosphere (Costello and Chum, 1998). Starch is abundant in crop materials, but expansion of ethanol production for the

purpose of automotive fuel requires feedstocks that do not compete for food or fiber (Himmel, et al., 1996).

In the other hand, lignocellulosic material is by far the most abundant raw material such as in hardwood, softwood, grasses and agricultural residues. Further potential raw materials are newsprint, office papers, municipal solid wastes, etc. Lignocellulose is a more complex substrate than starch. It consists of a mixture of carbohydrate polymers (cellulose and hemicellulose) and lignin. The carbohydrate polymers are tightly bound to lignin mainly by hydrogen bonds but also by some covalent bonds (Lee, 1997). Lignocellulose contains five major sugars, which varies with the feedstock (Hinman et al, 1989). They are the hexose; glucose, mannose, galactose and pentose; xylose and arabinose. Glucose is the most abundant carbohydrate in lignocellulosic material, xylose is the second one.

One of the sustainable alternative energy is relatively cheap production and environment friendly was development bio ethanol from waste residue agriculture. It does contain many lignocelluloses like bagasse (waste residue sugar industry). Indonesia has a lot of biomass like bagasse. Based on Indonesian Central Research Sugar Agriculture data, bagasse was produced 32% from sugarcane. 60% of it, was used by sugar industry as fuel, material for paper, fungi industry, etc. Because of that, approximately about 40% of bagasse wasn’t use. Polysaccharide of sugarcane contain more 70%, it divide to 50%-55% cellulose and 15%-20% hemicellulose. The lignin contain about 20-23%, beside that it was called ash compound (Pandey, et al, 2000).

Generaly The process for converting the lignocellulosic material to ethanol requires: (1) delignification to liberate cellulose and hemicellulose from their complex with lignin; (2) depolymerization of the carbohydrate polymer to produce free sugars; and (3) fermentation of mixed hexose and pentose sugar as monosaccharide to produce ethanol. The accurate analysis during conversion process is also important for the successful development of feasible

ISSN: 1411-1284


conversion process. Presently, it is known that among the key processes described above, the delignification of lignocellulosic raw materials is the rate-limiting and most difficult task to be solved (Lee, 1997).

Hydrolysis process is converted polysaccharide to monosaccharide. Major constituents in enzymatic hydrolysis and acid hydrolysis from lignocellulosic material like bagasse are glucose and xylose released from cellulose and hemicelluloses, respectively. Ethanol fermentation process utilizing yeast has been well developed with glucose as a carbon and energy source. Generally enzymatic hydrolysis using cellulose or xylanase, and acid hydrolysis is using HCl or H2SO4.

Simultaneous Saccharification and Fermentation (SSF) firstly described by Takagi (Takagi, et al, 1977], combine enzymatic hydrolysis or acid hydrolysis of polysaccharide with simultaneous fermentation of the sugars obtained ethanol. The advantage of this process is cellulose or hemicellulose were converted to monosaccharide, it rapidly ferments to ethanol and use one reactor for entire process so that saved the cost equipment.

This research was tried to make ethanol from bagasse by SSF process with enzymatic hydrolysis, acid hydrolysis and combination of enzymatic hydrolysis and acid hydrolysis.

2. Methods 2.1 Preparation of sample

Sugarcane bagasse (Saccharum officinarum) were milled and screened (30-60-mesh). The milled bagasse was then air-dried to a final humidity 10% and stored under dry conditions. Bagasse was taked from Indonesia (from sugar factory in Lampung, Indonesia). Preparation of material Hydrolysis HCl 0.5% and 1% was used for acid hydrolysis, Cellulase and Xylanase was were used for enzymatic hydrolysis. 2.2 Growth Stock S.cerevisiae S.cerevisiae AM12 was pre-cultured in Potato Dextrose Agar (PDA) 2%, Agar (0.25 g), Aquades (50ml) and incubation for 3-4 days on ambient temperature, and then was used as yeast for SSF process.

2.3 Preparing yeast inoculation S.cerevisiae AM12 fresh from growing stock was pre-cultured with 50 ml medium (glucose, 10 g/l; yeast extract, 1.0 g/l; KH2PO4, 0,1 g/l; MgSO4.7H2O, 0,1 g/l; and (NH4)2SO4, 0,1 g/l) in 50 ml flask. Before inoculation, medium was autoclaved for 20 minutes, and then incubation on ambient temperature for 24 hours with use by shaker.

2.4 SSF process SSF medium 5.5 ml, it containing bagasse (0.25 g), nutrient medium (2.5 ml), 0.5 Na-citrate buffer (pH 4, 4.5, 5) cellulase enzyme (10 FPU = 0.016 g) and 2.5

ml yeast inoculated. Bagasse, nutrient medium and Na-citrate buffer sterilize for 20 minutes in autoclave. However, enzyme was added without sterilization. Nutrient medium is containing 1.0 g/l (NH4)2PO4; 0.05 g/l MgSO4.7H2O and 2 g/l yeast extract. Cultivation was took and get in to micro centrifuge tube. The clean liquid from sample had taken in 24, 48, 72 and 96 hours and doing ethanol concentration test for ethanol produced. For strong acid hydrolysis using 0.5% and 1% (v/v) concentration, chloride acid was added after bagasse and then enzyme, yeast inoculated (2 ml), Na-citrate buffer pH 5 (0,5 ml) and nutrient (2 ml). 2.5 Analysis a. Analysis lignin, holocellulose and α-cellulose

Lignin analysis was done by modified klason lignin method. It is with add sulphuric acid 72% in sample and agitate until crushed, and then autoclave in 1210C for 30 minutes. After that, it filtered with paper filter, covered by aluminium foil and dry it into oven for 1 hour weighing the weight.

Holocellulose and α-cellulose analysis, it analyze by wise method. It mixed sample with sodium chlorate, acetate acid and aquades. Incubation with hot water in 800C, chilled, filtration with aquades and it wash with acetone. Then, solid residue was dried in oven on 1050C for 1-2 nights and weighing the weight. b. Measuring ethanol concentration

Ethanol concentration was measured by Gas Chromatography SUPELCOWAX-10 type (Supelco Inc., 0,53 mm ID., 15 m, 0,5 mm, FID) with temperature was maintained at 500C. Before tested, sample was took 50 µl and 200 µl distilled water added.

3. Result And Discussion

3.1 Composition Bagasse Analysis

The analysis result show that lignin contain in bagasse approximately about 24.2% from bagasse total. α-cellulose contain about 52.7% and hemicellulose contain about 17.5% in bagasse.

0 10 20 30 40 50 60

lignin

α-cellulose

Hemicellulose

Extractive chemical

composition (%)

Figure 1. Chemical composition of bagasse

ISSN: 1411-1284


Totally of hemicellulose and α-cellulose that is opportunity to convert to monosaccharide in hydrolysis process. Actually about 69% composition of bagasse is polysaccharide and able to convert to ethanol production. Theoritically Theoretically if calculated base of cellulose the maximum ethanol yield about 56,86% base α-cellulose and 28,43% base on original bagasse. If calculated base of hemicellulose maximum ethanol yield about 56,84 % base on hemicellulose and 11,37 % base on original bagasse. Totally maximum ethanol yiel base original bagasse about 40%.

3.2 Ethanol production from bagasse with

Enzymatic Hydrolysis Ethanol production with enzim cellulose

Ethanol production from bagasse using enzyme cellulase showed in figure 2. The highest ethanol yield from experiment at pH 5, about 5,51g/L or about 11, 24% base on original bagasse. Optimum pH for enzyme is 4.8 and yeast is between 4 and 5. So that SSF process pH is among 4.5-5. This condition is exactly for enzyme and yeast condition. Beside pH, temperature is also influence in SSF process. Temperature for growing yeast media is about 300-350C (Makanjuola, et al, 1992). But the cellulolytic enzyme act optimally about 400C. So, the optimum temperature range for SSF is 35-410C. At lower temperatures the rate of saccharification decreases significantly.

0.0

1.02.0

3.04.0

5.06.0

7.0

0 24 48 72 96

Incubation time (h)

Eth

anol

con

cent

ratio

ns (

g/L)

pH = 4

pH = 4,5

pH = 5

Figure 2. Ethanol production from bagasse with enzyme cellulose in variation of pH

At this research, SSF temperature was used in ambient temperature, while yield ethanol was produced is not too high because cellulase enzyme capability was decreased to transform cellulose to glucose.

The result also shows that ethanol can be obtained in 24 hours almost reach optimal concentration in SSF process. After 24 hours, ethanol concentration not too significant increase and highest ethanol concentration was obtained between 72 and 96 hours of incubation time. From this research, SSF process with pH 5 was obtained the highest ethanol concentration compare with pH 4 and pH 4,5.

Based on literature, optimum pH for SSF process is between pH 4.5 until pH 5 and the highest ethanol was obtained at pH 5. These phenomena is relevant with the literature. Based of this research for the other experiment con was continued on pH 5.

Cellulose are usually a mixture of several enzymes. At least three major groups of cellulases are involved in the hydrolysis process; (1) endogluconase which attacks regions of low crystallinity in the cellulose fiber, creating free chain-ends; (2) exogluconase or cellobiohyrolase which degrades the molecule further by removing cellobiose units from the free chain-ends; (3) ß-glicosidase which hydrolyzes cellobiose to produce glucose. In addition to the three major groups of cellulase enzymes, there are also a number of ancillary enzymes that attack hemicelluloses, such as glucuronidase, acetylesterase, xylanase V-xylosidase, galactomananase, and glucomananase (Duff and Murray, 1996). During the enzymatic hydrolysis, cellulose is degradable by the cellulases to reducing sugars that can be fermented by yeast or bacteria to ethanol Ethanol production from bagasse by combination of enzyme cellulase and xylanase. Lignocellulosic material especially bagasse consist 17,5% of hemicellulose. The highest composition of hemicelluloses is xylan, that can not convert to xylose by enzyme cellulase. Enzyme xylanase is one of alternative enzyme able to convert hemicellulose (xylan) become xylose and xylose able to convert to ethanol. Ethanol production by enzim cellulase and xylanase showed in figure 3 and 4, respectively. The highest ethanol production from bagasse by combination of enzyme cellulase and xylanase at pH 5 about 7,63 g/L or 15,25%. The yield of ethanol if compared with yield of ethanol using enzyme cellubiase without xylanase increase about 3 %. This phenomena indicated that enzyme xylanase increasing the ethanol yield. The highest ethanol yield at pH 5, this phenomena indicated pH 5 is the optimum pH for process SSF.

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

0 24 48 72 96 120

Incubation time (h)

Eth

ano

l co

nce

ntr

atio

ns

(gr/

L) pH = 4

pH = 4,5

pH = 5

Figure 3. Ethanol production from bagasse by enzyme cellulase and xylanase

ISSN: 1411-1284


0.00

2.00

4.00

6.00

8.00

10.00

12.00

14.00

16.00

0 24 48 72 96 120

Incubations time (h)

Eth

ano

l co

nce

ntr

atio

ns

(%)

pH = 4

pH = 4,5

pH = 5

Figure 4. Percentage of ethanol production from bagasse by enzyme cellulase and xylanase base on original bagasse 3.3 Ethanol production from bagasse with Acid

Hydrolysis

Concentrated acids HCl have been used to hydrolysis of lignocellulosic materials (in this research bagasse). The highest ethanol yield after hydrolysis with HCl 0,5% is 3,24 g/L at incubation for 24 h. The highest ethanol yield after hydrolysis with HCl 1% is 3,77 g/L or 7,54% at incubation for 24 h. Ethanol concentration of bagasse after SSF using hydrolysis with HCl 0,5% and 1% is showed at figure 5, respectively.

0.00

1.00

2.00

3.00

4.00

5.00

6.00

0 24 48 72 96 120Eta

no

l co

nce

ntr

atio

ns (g

r/L

)


HCl 0,5%

Figure 5. Ethanol production from bagasse by acid hydrolysis (HCl 0,5% and 1%).

Based on data in figure 5, ethanol production from bagasse by HCl hydrolysis increasing at 24 h incubation time because HCl is powerful agents for cellulose and hemicellulose hydrolysis (Sivers and Zacchi, 2002). The ethanol yield is lower than with enzyme hydrolysis because HCl was used at the lower concentrations (only 1% and 0.5%), more higher concentration HCl the yield of ethanol will be increasing that is indicated ethanol yield by HCl 1% higher than 0,5%. But unfortunately concentrated acids are toxic, corrosive and hazardous and require reactors that are resistant to corrosion (Sun and Cheng, 2002). In addition, the concentrated acid must be

recovered after hydrolysis to make the process economically feasible (Sivers and Zacchi, 1995).

Dilute acid hydrolysis has been successfully developed for pretreatment of lignocellulosic materials. The dilute sulfuric acid pretreatment can achieve high reaction rates and significantly improve cellulose hydrolysis (Esteghlalian et al., 1997). At moderate temperature, direct saccharification suffered from low yield because of sugar decomposition. High temperature in dilute acid treatment is favorable for cellulose hydrolysis (Millan, 1994). Recently developed dilute acid hydrolysis processes use less severe conditions and achieve high xylane to xylose conversion yields. Achieving high xylane to xylose conversion yields necessary to achieve favorable overall process economics because xylane accounts for up to a third of the total carbohydrate in many lignocellulosic materials (Hinman et al., 1992). There are primarily two types of dilute acid pretreatment processes: high temperature (T greater than 160oC), continuous-flow process for low solids loading (5-10% weight of substrate/weight of reaction mixture) (Brennan et al., 1986; Converese et al., 1989), and low temperature (T less than 160oC), batch process for high solids loading (10-40%) (Esteghlalian et al., 1997). Although dilute acid pretreatment can significantly improve the cellulose hydrolysis, its cost is usually higher than some physico-chemical pretreatment processes such as steam explosion. A neutralization of pH is necessary for the down stream enzymatic hydrolysis of fermentation process (Sun and Cheng., 2002). 3.4 Ethanol production from bagasse with

combination of Acid Hydrolysis

This research was done by combination of cnzyme cellulase, xylanase and additional strong acid with low concentration in SSF process. Additional strong acid was done together with enzyme. The strong acid was used is chloride acid (HCl) with concentration 0.5% and 1% (v/v). Additional strong acid will be affected in hydrolysis process of cellulose to glucose and can be to increase ethanol yield. The results show that the highest ethanol yield after hydrolysis with combination of cellulase, xylanase and HCl 1% about 8,22 g/L or 16,45%. This yield increasing about 2% if compared with ethanol yield without acid addition. If compared with acid hydrolysis by HCl 1% increasing about 8% based on original bagasse. Ethanol production from bagasse using SSF with combination of enzymatic hydrolysis (Cellulase and cellubiase) and acid hydrolysis (HCl 1% and 0,5%) showed at figure 6, respectively.

ISSN: 1411-1284


0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

0 24 48 72 96 120


Eth

ano

l co

nce

ntr

atio

ns

(g/L

)cellulase & xylanasepH=5"Cellulase, Xylanaseand HCl 0.5 %"Cellulase, xylanase& HCl 1 %"

Figure 6. Ethanol production from bagasse by enzyme cellulase, xylanase and added HCl low concentrations.

Base on this result, if the acid concentration more highly, then the hydrolysis capability will more increase. In acid hydrolysis, the degree of polymerisation cellulose will be decreased significantly. It was caused by H+ concentration (proton) in strong acid. If the concentration strong acid more high, then H+ concentration will be more high too. Hydrolysis by acid can be able to breakdown β-1, 4-glycosidic chain in cellulose. The reaction will start from proton in acid which rapidly react with glycosidic oxygen which is connect to two sugar units transform to conjugation acid. The breakdown of C-O chain and broken conjugation acid will be transform to carbonium cyclic ion. After addition water, sugar and proton will be free (Carrasco, et al, 1994).

Hemicellulose is easier to hydrolysis rather than cellulose with acid hydrolysis. Hemicellulose also easier to depolymerisation in high temperature at acid condition (Palmqvist, et al, 1996) and hemicellulose sugar such as xylose, arabinose can be free with hemicellulose hydrolysis. Hemicellulose hydrolysis in acid condition is possible to form 2-furaldehyde (furfural) form pentose sugar, it mainly from xylose and form 5-hydroxymethil-2-furaldehyde (HMF) from hexose’s sugar. Furfural compound and HMF can delay yeast work because it makes lag phase in ethanol production and decrease ethanol productivity, but there is no effect to ethanol was produced. S.cerevisae cans metabolism furfural and HMF, in order that inhibitor effect will be disappeared. It caused by membrane permeability of S.cerevisae. 4. Conclusion

Based on the result of research was obtained, and

then the conclusion: 1. Ethanol was production from bagasse using

enzyme cellulase, xylanase and HCl in low concentrations.

2. The highest ethanol yield with enzymatic hydrolysis with enzyme cellulase at pH 5, about 5,51g/L or about 11, 24% base on original bagasse.

3. The highest ethanol production from bagasse by combination of enzyme cellulase and xylanase at pH 5 about 7,63 g/L or 15,25%.

4. The highest ethanol yield after hydrolysis with HCl 0,5% is 3,24 g/L at incubation for 24 h. The highest ethanol yield after hydrolysis with HCl 1% is 3,77 g/L or 7,54% at incubation for 24 h.

5. The highest ethanol yield after hydrolysis with combination of cellulase, xylanase and HCl 1% about 8,22 g/L or 16,45%. This yield increasing about 2% if compared with ethanol yield without acid addition. If compared with acid hydrolysis by HCl 1% increasing about 8% based on original bagasse.

6. References

Carrasco, J.E.; Saiz, M.C.; Navarro, A.; Soriano, P.; Saef, F. and Martinez, M. (1994) Effect of dilute acid and steam explosion pretreatments on the cellulose structure and kinetics of cellulosic fraction hydrolysis by dilute acids in lignocellulosic materials. Applied biochemistry and biotechnology. 45/46, 23-34.

Hatakka, A. (2001) Biodegradation of lignin. In M Hofrichter and A. Steinbüchel (eds.), Biopolymers, vol. 1. Wiley-VCH, Weinheim, Germany. p. 129- 180.

Hofrichter, M. (2002) Review: Lignin conversion by manganese peroxidase (MnP). Enzyme Microb. Technol. 30:454-466.

Itoh. H., Wada. M., Honda. Y., Kuwahara. M., Watanabe. T., 2003. Bioorganosolve pretreatments for simultaneous saccharification dan fermentation of beech wood by etanolysis dan white rot fungi. Journal of Biotechnology 103, 273-280

Lee, J., 1997. Biological conversion of lignocellulosic biomass to ethanol. Biotechnology 56, 1-24.

Lynd, L.R., Bothast, R.J., Wyman, D.E. 1991. Fuel etanol from cellulosic biomass. Science 251: 1318-1323.

Martín, C., Galbe, M., Nilvebrant, N.O., Jönsson, L.J., 2002. Comparison of the fermentability of enzymatic hydrolyzed of sugarcane bagasse pretreated by steam explosion using different impregnating agents. Applied Biochemistry and Biotechnol. 99, 699-716.

Maiorella, B. L. 1985 Industrial chemical, biochemical and fuels: Ethanol. –In: Comprehensive biotechnology. The principles, applications and regulations of biotechnology in industry, agriculture and medicine. M.-M. Young (ed.), Pergamon Press Ltd. University of Waterloo, Ontario, Canada, Vol. 3. Chapter 43, 861-914.

ISSN: 1411-1284


Makanjuola, D. B., Tymon, A. and Springham, D. G. 1992. Some effects of lactic acid bacteria on laboratory-scale yeast fermentations. Enzyme Microb. Technol. 14: 350-357.

Mc Millan, J.D., 1994. Pretreatment of lignocellulosic biomass. In: Himmel, M.E., Baker, J.O., Overend, R.P (Eds), Enzymatic Conversion of Biomass for Fuels Production. American Chemical Society, Washington, DC, pp. 292-324

Palmqvist, E., Hahn-Hägerdal, B., Galbe, M., Larsson, M., Stenberg, K., Szengyel, Zs., Tengborg, C. and Zacchi, G. 1996a. Design and operation of a bench-scale process development unit for the production of ethanol from lignocellulosics. Biores. Technol. 58: 171-179.

Pandey,A. Soccol,C.R. Nigam,P. And Soccol,V.T. 2000. Biotechnological potential of agro-indistrial residues. Sugarcane bagasse. Bioresour Technol. 74: 69-80.

Sun, Y., Cheng, J. 2002. Hydrolysis of lignocellulosic materials for etanol production: review. Bioresource Technology 83, 1-11.

Takagi,M., Abe, S., Suzuki, S., Emert, G.H., Yata, N. 1977. A method for production of etanol direcly from cellulose using cellulose and yeast. Proceedings of Bioconversion symposium, Delhi, 551-571.

Wyman, C.E., 1994. Etanol from ligcellulosic biomass: Technology, economics, and opportunities. Bioresource Technology 50, 3-6.

Zadrazil, F. Brunert, H., 1982. Solid state fermentation of lignocellulose containing plant residues with Sporotrichum pulverulentum Nov. and Dichomitus squalens (Karst.) reid. European Journal of Applied Mirobiology and Biotechnology 16, 45-51

ISSN: 1411-1284


A General Application of a Direct Method for Multivariable MPC Control Strategy in Chemical

Processes

A. Ahmad1 and A. Wahid2

1. Department of Chemical Engineering, Universiti Teknologi Malaysia, Johor 2. Department of Chemical Engineering, University of Indonesia, Depok. Corresponding authors: [email protected] and [email protected]

Abstract – A general application of a direct method for model predictive control strategy is proposed for multivariable nonlinear control problem in chemical processes. The aim is to provide a solution to nonlinear control problem that is favorable in terms of industrial implementation. The scheme utilizes multiple linear models to cover wider range of operating conditions. Depending on the operating conditions, suitable models of 2x2, 3x3, and 4x4 processes are used in control computations. Servo and regulatory controls of the system are examined on constrained and unconstrained conditions. Comparisons are made to conventional controllers. The results confirmed the potentials of the proposed strategy. Keywords – Tuning, model predictive control, multivariable, nonlinear control, chemical process

I. INTRODUCTION

he cornerstone of MPC is the model [1]. It cause MPC is called MBPC (model-based predictive

control). MPC uses models in 2 ways: using a reliable model to predict effect of past control moves on P future outputs, assuming no future moves, and using the same model to compute the optimal M controller moves. Implement first move and repeat procedure.

Predictive control is now one of the most widely used advanced control methods in industry, especially in the control of processes that are constrained, multivariable and uncertain. A large number of implementation algorithms, included industrial predictive control applications [2] have a appeared in the literature.

II. TUNING THE MPC CONTROLLER

Dynamic matrix control (DMC) [3] is the most popular MPC algorithm used in the chemical process industry today. Over the past decade, DMC has been implemented on a wide range of process applications. A major part of DMC’s appeal in industry stems from

the use of a linear finite step response model of the process and a simple quadratic performance objective function. The objective function is minimized over a prediction horizon to compute the optimal controller output moves as a least-squares problem.

Tuning in conventional control strategy (P, PI, and PID) is related to obtain an optimum setting of controller parameters (Kc, Ti, and Td). Ziegler-Nichols, Lopez, Ciancone, etc. [4] are some examples of single-loop tuning in P, PI, and PID controllers. Huang, et. al. [5] has proposed a direct method for multi-loop PI/PID controller design based on FOPDT/SOPDT model of each loop.

MPC controller has certain parameters setting to achieve its optimum performance. During the time, trial-and-error efforts have been done to reach this goal until Shridhar & Cooper [6, 7] proposed a tuning strategy for unconstraint SISO and multivariable MPC. Dougherty 8and Cooper [8] proposed a non-adaptive DMC tuning strategy (see Table 1) based on all of FOPDT models in systems.

The principles of the multivariable DMC tuning strategy [8] are: 1) T – Sampling time. While a large T refers to a low

computation load, a small T refers to a properly track the evolving dynamic behavior. Too slow of a sampling rate will lead to information losses, and too fast of a sampling rate could lead to numerically sensitive procedures. Nevertheless, this method allows values of T other than the recommended value given in Table 1.

2) P (prediction horizon) and N (model horizon). Both P and N have the same setting and are related to the settling time of the slowest (the largest time constant) sub-process in the multivariable system. A large P improves the nominal stability of the closed loop. Meanwhile, a large N makes controller has long enough time to avoid the instabilities that can otherwise result since truncation of the model horizon misrepresents the effect of controller output moves in the predicted process variable profile.

3) M – Control horizon. M equals to 63.2% of the settling time of the slowest sub-process in the multivariable system. This ensures M to be long

T

ISSN: 1411-1284


enough such that the results of the control actions are clearly evidenced in the response of the measured process variable.

4) - Controlled variable weights. The setting of

those parameter are free. However, in the most case, they are set equal to one.

5) – Move suppression coefficients. Its primary

role in DMC is to suppress aggressive controller actions. When the control horizon is 1 (M = 1), no move suppression coefficient is needed (λ = 0). If the control horizon is greater than 1 (M > 1), then the analytical equation given in Table 1 is used.

Table 1 Non-adaptive DMC tuning strategy [8] Approximate the process dynamics of all controller output to measured process variable pairs with FOPDT

models:

1. Select the sample time as close as possible to:

2. Compute the prediction horizon, P; and the model horizon, N:

3. Compute a control horizon, M:

M

4. Select the controlled variable weights,, to scale process variable units to be the same. 5. Compute the move suppression coefficients, :

6. Implement DMC using the traditional step response matrix of the actual process and the initial values of the parameters computed in steps 1–6.

III. APPLICATION IN GENERAL PROCESSES Dougherty and Cooper [8] have proved the non-

adaptive DMC tuning strategy in the three 2x2 processes (general transfer function, multi-tank, and distillation column). Here, we shall illustrate comparisons between PI/PID and MPC controller strategy using Table 1 in the same and more complex processes: 2x2 process, 3x3 process and 4x4 process [5]. 3.1 Wood and Berry distillation column (2x2 process)

Wood and Berry distillation column model [5] is the most popular model of 2x2 process system. This

model consists of two matrixes: manipulated variable matrix (2x2) and disturbance matrix (2x1). Nevertheless, the first matrix (manipulated variable matrix) is commonly used by researchers [5, 9].

Table 2 shows results of PID and MPC tuning in Wood & Berry model. Responses of Wood & Berry model using these settings are shown by Figure 1. Although, MPC controller is not set in the optimum setting, MPC controller performances are better than PID controller performance. MPC controller tuning has improved its performance by decreasing the overshoot significantly. In unit step change test of y1, the y2’s overshoot sharply falls from 0.342 to 0.056, or going down 84%.

Table 2 PID and DMC parameters setting in WB model (2x2 process) Wood & Berry model

(2x2 process) PID controller tuning (Proposed 2 method)

DMC controller tuning

)(

12.13

9.4

19.14

8.3

)(

)(

14.14

4.19

19.10

6.6

121

9.18

17.16

8.12

)(

)(3

8

37

3

sF

s

e

s

e

sV

sR

s

e

s

e

s

e

s

e

sY

sYs

s

ss

ss

B

D

+

++

+−

+

+−

+=

−

−

−−

−− • Loop 1: ,

,

• Loop 2: , ,

,

Ts = 1.5 P = 153 M = 35

ISSN: 1411-1284


0

0.5

1

1.5

y1

0 10 20 30 40 50 60 70 80 90-0.02

0

0.02

0.04

0.06

y2

(a) (b) (c)

Figure 1 Response of multi-loop control for WB process (step change in y1): (a) PID controller performance using Proposed 2 tuning, (b) MPC controller performance using Ts=0.1, P=10, M=2 (default setting), (c) MPC controller performance using

Ts=1.5, P=153, M=35 (tuning setting) 3.2 OR (3x3) process

The Ogunnaike and Ray (OR 3x3) process [5] as follows is considered (see Table 3). PID controller tuning uses BLT-4 method instead of proposed method by Huang et. al. [5]. In this case, responses of OR (3x3) process are unstable using their proposed method, though they are good using BLT-4.

As shown by Table 1, the tuning strategy uses FOPDT (first-order plus dead-time) model to calculate Ts, P, and M. Because of OR (3x3) model has one SOPDT (second-order plus dead-time), this transfer function has to be changed into FOPDT. By using PRC (process reaction curve) from step change testing of the SOPDT and applying a method that is developed by Cecil L. Smith [4], the FOPDT is obtained.

Figure 2 shows responses of multi-loop control for OR (3x3) process: (a), (b), and (c) PID controller performance using BLT-4 tuning (unit step change in y1, y2, y3 respectively). They have very poor performances. This disadvantage is improved by MPC controller.

Next, we are going to compare between responses using default setting and tuning setting. The results are shown by Figure 3 and Table 4. Almost performances by tuning setting are better than by the default setting. Only two point where the default setting is better than the tuning setting, they are overshoot of y2 in y1’s step change, and settling time of y2 in y2’s step change.

Table 3 PID and DMC parameters setting in OR model (3x3 process) OR model (3x3 process) PID controller tuning

(BLT-4 method) DMC controller

tuning

Loop 1: ,

, Loop 2: ,

, Loop 3: ,

,

Ts = 0.71 P = 91 M = 30

ISSN: 1411-1284


0 50 100 1500

0.5

1

y1

0 50 100 150-1

0

1

y2

0 50 100 150-10

0

10

y3

0 50 100 1500

0.1

0.2

y1

0 50 100 1500

1

2

y2

0 50 100 150-10

0

10

y3

0 50 100 150-0.01

-0.005

0

y1

0 50 100 150-0.02

0

0.02

y2

0 50 100 1500

1

2

y3

(a) (b) (c)

Figure 2 Responses of multi-loop control for OR (3x3) process: (a), (b), and (c) PID controller performance using BLT-4 tuning (unit step change in y1, y2, y3 respectively)

0

0.5

1

1.5

Response: Scenario1Output: y1Time (sec): 53.4Amplitude: 1

y1

-0.2

0

0.2

Response: Scenario1Output: y2Time (sec): 1.6Amplitude: -0.154

y2

0 10 20 30 40 50 60 70 80 90 100-0.04

-0.02

0

0.02


y3

Plant Outputs

0

0.5

1

1.5

Response: Scenario1Output: y1Time (sec): 47.5Amplitude: 1

y1

-0.2

0

0.2


y2

0 10 20 30 40 50 60 70 80 90-0.04

-0.02

0

0.02Response: Scenario1Output: y3Time (sec): 2.84Amplitude: -0.0241y3

(a) (b)

Figure 3 Responses of multi-loop control for OR (3x3) process (step change in y1): (a) MPC controller performance using Ts=0.1, P=10, M=2 (default setting); (b) MPC controller performance using Ts=0.71, P=91, M=30 (tuning setting)

Table 4 Performance of MPC controller of OR (3x3 process) model

Default Setting Tuning Setting Step Change

Controlled Variable Settling time Overshoot Settling time Overshoot

y1 53.4 0.08 47.5 0.02

y2 -0.153 -0.193 y1

y3 -0.0386 -0.0241

y1 -0.156 -0.138

y2 21.7 0.01 31 0.01 y2

y3 -0.151 -0.0552

y1 -0.0303 -0.0273

y2 -0.0852 -0.0614 y3

y3 6.52 0 3.18 0 3.3 Alatiqi case 1 (4x4) process

Consider alatiqi case 1 (A1 4x4) process (see column 1 in Table 5). Alatiqi process has 16 empirical models that consist of three FOPDT models, two models having zeros, and eleven SOPDT models. All of empirical models have to be changed in FOPDT model to calculate MPC controller tuning. We use some PRCs (process reaction curves) from step change testing of the SOPDTs and the models having

zeros, and apply a method that is developed by Cecil L. Smith [3] btain the FOPDT.

PI controller shows the poor performance (see Figure 4). Also, the controller performance as shown by Figure 5 is extremely poor, because the responses are unstable. So, in this case, the default setting can not be used in MPC controller. Inevitable, we have to use the tuning setting in this case.

ISSN: 1411-1284


Table 5 PID and DMC parameters setting in Alatiqi case 1 model (A1 4x4 process)

Alatiqi case 1 model (A1 4x4 process) PI controller tuning (Lee et al. method)

DMC controller tuning

Loop 1: ,

Loop 2: ,

Loop 3: ,

Loop 4: ,

Ts = 1.62 P = 191 M = 52

0 20 40 60 80 100 120 140 160 180 2000

0.5

1

y1

0 20 40 60 80 100 120 140 160 180 200-0.01

0

0.01

y2

0 20 40 60 80 100 120 140 160 180 200-0.5

0

0.5

y3

0 20 40 60 80 100 120 140 160 180 200-2

-1

0

y4

0

1

2

y1

-0.5

0

0.5

y2-0.05

0

0.05

y3

0 10 20 30 40 50 60 70 80 90-0.5

0

0.5

y4

Plant Outputs

(a) (b)

Figure 4 Responses of multi-loop control for alatiqi case 1 (A1 4x4) process: (a) PI controller performance using Lee et. al. tuning (unit step change in y1); (b) MPC controller performance using tuning setting Ts=1.62, P=191, M=52 (unit step

change in y1)

-5

0

5x 10

5

y1

-2

0

2x 10

4

y2

-100

0

100

y3

0 10 20 30 40 50 60 70 80 90 100-4

-2

0x 10

4

y4

Plant Outputs

Figure 5 Alatiqi’s Responses of MPC controller using default setting

IV. CONCLUSSIONS

MPC controller offers better control performances than PI/PID controller, especially in multivariable processes. Application of MPC controller in the three more complex models produces the fantastic performance. To achieve an optimum performance of MPC controller, non-adaptive DMC controller tuning can be used. This method has proved that the tuning setting has the optimum performance. In the complex processes like WB model (2x2 process) and OR

model (3x3 process), the default setting can be used in MPC controller. But, in more complex process likes Alatiqi case 1 model (A1 4x4 process), the default setting produces very poor performance. So, we must use the tuning setting.

REFERENCES

[1] D.W. Clarke. “Adaptive Predictive Control”. A

Rev. Control. Vol. 20, pp. 83-94, 1996 [2] S.J. Qin and T.A. Badgwell, (2003). “A survey of

industrial model predictive control technology”. Control Engineering Practice. (11): 733 – 764

[3] C. R. Cutler & D. L. Ramaker. “Dynamic matrix control—a computer control algorithm”. Proceedings of the JACC 1980. San Francisco, 1980.

[4] T. Marlin. Process Control: Designing Processes and Control Systems for Dynamic Performance. 2nd Edition,McGraw-Hill, New York, 2000.

[5] Huang et. al. “A direct method for multi-loop PI/PID controller design”. Journal of Process Control (13): 769–786, 2003.

[6] R. Shridhar & D. J. Cooper. “A tuning strategy for unconstrained SISO model predictive control”. Industrial & Engineering Chemical Research. (36): 729–746, 1997.

ISSN: 1411-1284


[7] R. Shridhar & D. J. Cooper. “A tuning strategy for unconstrained multivariable model predictive control”. Industrial & Engineering Chemical Research. (37): 4003–4016, 1998.

[8] Danielle Dougherty and Doug Cooper. “A Practical Multiple Model Adaptive Strategy for

Multivariable Model Predictive Control”. Control Engineering Practice. (11): 649 – 664, 2003.

[9] B. Huang., Steven X. Ding, and Nina Thornhill. “Alternative solutions to multi-variate control performance assessment problems”. Journal of Process Control (16): 457–471, 2006.

ISSN: 1411-1284


Biogrease Using Modified Palm Oil as Base Oil and Thickener Lithium Soap

Sukirno *, M.Nasikin, B.Heru, Rizqon F, Marius, Dizzi.

* Fac. of Engineering, University of Indonesia, Jl. Salemba Raya 4, Jakarta 10430 Tel. 7863516, fax. 7863515 email : [email protected]

Abstract– Biogreases without additives, have been prepared using biolubricant, as the base oil. This bio lubricant is a modified palm oil which can be synthesized from palm oil via transesterification and epoxidation reactions. The thickening agent used in this grease making study was lithium 12-hydroxystearate. The making procces of the biogrease was carried out in a closed flanged flask or autoclave. Characterizations of the biogrease were carried out using Penetration Test, Dropping Test. It was found that lithium biogrease NGLI 2 can be prepared using thickener less that was needed by mineral oil grease. Performance test by a Gear Load-carrying test showed that the biogrease products has better antiwear performance than that of lithium grease using mineral oil HVI 160 S as the base oil. Keywords– Biogrease, lithium soap, base oil. Penetration Test, Dropping Point Test, Gear Load-carrying Test.

I. INTRODUCTION ubrication is proccess of interposing a lubricating material between two surfaces in relative motions in

order to reduce friction[1]. Liquid lubricants are the most known type of lubricanting material, however semi solid lubricant or grease is widely used in many lubrication systems due to enomical consideration.

Lubricating greases are, in general, produced by dispersing a solid thickening agent in a lubricating fluid. The base oil selected in formulating a grease should have the same characteristics, as if the equipment is to be lubricated by the fluid lubricant. Hence a fundamental difference between a formulated fluid lubricant and a similarly fully formulated grease, is the presence in of the thickening agent. These thickeners are usually soaps, such as lithium, sodium and calcium salts of long chain fatty acids, and the more recently introduced thickeners, such as clays, polyureas[3]. The same as lubricating oil, greases also contain some additives in order to enhance its performance and to protect the grease and lubricated surfaces. [1]. The most common additives found in grease are antioxidants to prolong the life of a grease, anti-corrosion agents to protect metal against attack from water, sulphides or corrosive elements, extreme pressure agent to guard against scoring and galling, antiwear agent to prevent abrasion and metal to metal

contact etc. So base oil, thickener dan addives are blended to make a grease Typical grease composition are shown in table 1.

Table l Typical grease composition

(%) Base oil 75-95

Thickener 5-20 additives 0-10

Actually, the component that perform as lubricant

is the base oil[3]. The thickener gives grease its characteristic rigidity or consistency and is sometimes thought of as a “three-dimensional fibrous network” or “sponge” that holds the oil in place. During the grease in use, when the lubricant film between wearing surfaces thins, the friction heat softens the adjacent thickener, which expands and releases oil to restore film thickness. Soap thickeners not only provide consistency to grease, they affect desired properties such as water and heat resistance and pumpability.

Fluid lubricants used to formulate grease are normally petroleum or synthetic oils. Synthetic oils are higher in cost, so that the majority of greases on the market are composed of mineral oil. With growing environmental concerns, vegetable oils are also being used in applications requiring nontoxic or biodegradable greases. In America, they are developing soybean oil. For Indonesia, as the main producer of palm oil, we should develop palm oil.

In our previous research, we had successfully prepared a biolubricant via transesterification and epoxidation reaction of palm oil to a lubricant which has better oxidation stability than its feed, a RBDPO olein. This modified palm oil is called called EPOME Gliserol[2] which is biodegradable, and contain no aromatic, nitrogen, sulfur, and is expected to be used in foodgrade lubricant formulation. In this present reseach, the biolubricant is used as base oil for grease making. The biogrease product is supposed to be better in its antiwear propertiy, since its has many active functionality to protect friction surfaces.

The most important feature of a grease is its rigidity or consistency. A grease’s consistency is its resistance to deformation by an applied force. A grease that is too stiff may not feed into areas requiring lubrication, while a grease that is too fluid may leak out. The measure of rigidity or consistency is called penetration which is measured by penetration test. To measure penetration, a cone of given weight is allowed

L

ISSN: 1411-1284


to sink into a grease for 5 seconds at a standard temperature of 25 ºC). The depth, in tenths of a millimeter, to which the cone sinks into the grease is the penetration. A penetration of 100 would represent a solid grease while one of 450 would be semifluid The NLGI has established consistency numbers or grade numbers, ranging from 000 to 6, corresponding to specified ranges of penetration numbers. Table 1 lists the NLGI grease classifications along with a description the consistency of each classification.

Tabel 2 NLGI Number NLGI Worked Penetration (Grade)

at 250C (0,1mm) Virtual Assessment of Deformability

000 445-475 Very soft

00 400-430 Very soft

0 355-385 Soft

1 310-340 Creamy

2 265-295 Almost solid

3 220-250 Hard

4 175-205 Hard

5 130-160 Very hard, as soap

6 85-115 Very hard, as soap

The purpose of this research is to prepare

biogrease of NGLI 2 which is normally used for gear and wheel bearing application. In this preliminary study of making grease, the amount of the thickener is varied to observe their rigidity or consistency and dropping points in order to make a graph shown the effect of the amount of thickener to the biogrease characteristics, and its lubrication performance in comparison with similar grease using mineral oil as base oil.

II. EXPERIMENTAL

Materials used (a) Base oil

The base oil used was a bio lubricant synthesized from palm oil. It was the product of previous research and its properties are given in Table 1. The Mineral oil HVI 160 Pertamina product was also used as base oil comparison.

Table 1 Properties of base oils

Appearance Yellow pale YellowSpecific gravity [-] ASTM D-1289 0.91 0.85 0.8Viscosity @40ºC cSt ASTM D-445 35 38.9 96Viscosity @100ºC cSt ASTM D-446 6.9 8-9 11Viscosity Index [-] ASTM D-2270 >100 >100 100Pour point ºC ASTM D-97 0 8-15 -9Oxid. Test (∆ Vis) Bulk Oxidation 5.6 21.5 7.8

Test MethodBiolubricant (Modified RBDPO)

RBDPOM.O

HVI 160 S

Characterization

(b) Thickener Thickener lithium soap was obtained from the

reaction of fatty acid 12-hydroxystearate (melting point 77.5°C).and lithium hydroxide.

Preparation of greases A mixture of lithium 12-hydroxystearate and

approximately 85% of the bio lubricant to be used in the manufacture of the grease was heated in a flanged flask or autoclave, equipped with a stirrer, pressure indicator, oil heater thermometer, to 120°C and stirred, as shown in figure 1. The lithium hydroxide was added slowly until solution of the soap occurred (170°C). Heating was continued until the soap melted (200°C). Next, the solution was cooled in a cooler to approximately 70°C with vigorous stirring. atmosphere. The remainder of the base oil, is then added to give the final product. Stirring was continued for a further about 1 h, and the product, now a creamy grease, was milled in Miller..

In this experiment, the amount of the thickener is varied, and additives has not been added.

Fig.1 Autoclave

Biogrease Product Characterizations Samples of the greases were tested with

Penetrometer (ASTM D 217), Dropping Point (ASTM D 566) Water washout (ASTM D 1264). Biogrease Product Performance Test

Performance tests of biogreases were conducted using Gear Load-carrying tests. The method used to assess the load-carrying characteristics of the greases in this experiment was the gear load-carring test. In this test, a pair of gear were used. Each test was carried out at ambient temperature with 10 hours runs at 25 rev/s and the applied load of 10 kg. The wear debris is measure by AAS.

ISSN: 1411-1284


Fig.2 Gear load-carrying test

III. RESULTS AND DISCUSSION Product Appearance

Figure 3 shows a grease product prepared in autoclave, as seen under ordinary camera. It is creamy and the texture is soft and sticky.

Fig.3 Biogrease

Biogrease rigidity

Fig.4 shows the plot between penetration and composition of the thickener lithium soap obtained from the biogrease of this experiment. In the figure, there is also given a plot of a grease using mineral oil, as a comparison. From the figure can be seen that, grease consistency increase as the amount of thickener

increase, and also with the same amount of thickener the biogrease is more rigid than that of petroleum grease. This means that to produce the same level of rigidity, the biogrease need smaller amount of the thickener than that of petroleum grease.

100

150

200

250

300

0 5 10 15 20 25

% Lithium Soap

Pen

etra

tio

n (

0,1m

m)

Lithium Grease HVI 160 Lithium Biogrease

Fig.4 Penetration Value of Li Biogrease

The Biogrease Dropping Point.

Dropping point is an indicator of the heat resistance of grease. As grease temperature rises, the grease gets softer, until liquefies and the rigidity is lost. Dropping point is the temperature at which grease becomes fluid enough to drip. The dropping point indicates the upper temperature limit at which grease retains its structure, not the maximum temperature at which a grease may be used.

The dropping point measurement of the biogrease are shown in fig 5. Dropping point increases as the amount of thickener increases, and for the same amount of thickener lithium soap, the dropping point of biogrease is higher than that of mineral oil grease.

100

120

140

160

180

200

220

240

0 5 10 15 20 25

% Lithium Soap

Dro

pp

ing

po

int

C

Li Biogrease Li Grease HVI 160

Fig.5 Dropping Point of Li Biogrease This experiment also tried to make semibio grease,

where the base oils are mixture of biolubricant and mineral oil. Figure 3 shows the result of the dropping point test of semibio grease with various base oil compositions. The dropping point of semiobio grease increases as the % of biolubricant increases.

ISSN: 1411-1284


Li Grease semibio (Biolubricant +HVI 160S)

0

50

100

150

200

250

0 20 40 60 80 100

% Bio lubricant (Modified Palm Oil)

Dro

pp

ing

po

int

(der

Cel

siu

s)

Fig.6 Dropping Point of Lithium Semibio Grease

Table 6 lists the results of quality tests of biogrease in comparison with that of the mineral oil grease. Lithium biogrease is better than lithium petroleum grease, in case of their dropping point. But it is inferior in water washout tests. This test is the ability of grease to withstand the effects of water with no change in its ability to lubricate. Water may suspend the oil in the grease, forming an emulsion that can wash away or, to a lesser extent, reduce lubricity by diluting and changing grease consistency and texture. Rusting becomes a concern if water is allowed to contact iron or steel components.

Table 2 Characteristic Comparison of BioGrease and

Mineral Oil Grease

Li Soap (%) 15 15 10Penetration (0,1 mm) 220.7 148 229

NLGI number 3 5 3Dropping point (0C) 221 230 210Water washout (%) 5 <10* <15*

Li Grease ( HVI 160 S Pertamina)

Lithium Biogrease

Gear Load-carrying test The result of performance tests of the biogrease

using gear load-carrying test are shown in Figure 7. In this figure the amount of wear debris (ppm of Fe) were plotted against % biolubricant. It can be seen that, as the composition (%) of biolubricant increases the wear amount (in ppm of Fe) decreases. It means that biolubricant can improve the performance of petroleum grease when is used in grease formulation.

0

50

100

150

200

250

Lithium 15% Lithium 15% Lithium 15% Lithium 15% Lithium 10%

Biolubricant40% HVI

Biolubricant60% HVI

Biolubricant80% HVI

Biolubricant100% HVI

Biolubricant100% HVI

pp

m F

e

Fig.7 Wear Rate of Li Biogrease using Gear load-

carrying test.

IV. CONCLUSIONS

From this preliminary study of making biogrease with modified palm oil biolubricant as the base oil, several conclusions can be summarized as follow: a. The amount of lithium soap needed is ±.10 % of

total weight to produce a biogrease NGL 2 using modified palm oil as the base oil. To reach the same level of rigidity consistency, the biogrease without additives needs smaller amount of the thickener than that of petroleum grease.

b. In gear load-carrying tests, the biogrease without additives showed better antiwear performance than that of petroleum grease, with HVI 160 S as the base oil.

ACKNOWLEDMENTS

This work was funded by RUSNAS Industri Hilir Kelapa Sawit. The Authors would like to express their appreciation all who supports to this study.

REFERENCES 1. O’Brien, J.A. , “Lubricating Oil Additives”, Booser, E.R.,

editor., CRC Handook of Lubricating (Theory and Practice Tribology). Volume 2, Theory and Design, CRC Press Inc. Boca Raton, Florida, 1983.

2. Fenjerry, Y. "Preparation and Characterization of Epome

Gliserol dan Epome Monoalcohol as Foodgrade Lubricant", Bachelor Thesis, Chemical Engineering Department, University of Indonesia, Depok, 2005.

3. H. B. SILVER and I. R. STANLEY, “The Effect of The

Thickener on the Efficiency of Ioad-Carryng Additives in Greases, TRIBOLOGY international June, pp.113-118, 1974.

ISSN: 1411-1284


Model for Oxidation and Combustion of Ethanol in Wide Range of Pressures, Temperatures and Equivalence Ratios

Yuswan Muharam and Nonni Soraya Sambudi

Department of Chemical Engineering, Faculty of Engineering, University of Indonesia,

Kampus UI Depok 16424 Phone 021-7863516, fax. 021-7863515, email: [email protected]

Abstract– Ethanol as an alternative fuel can

be used either in a mixture or single component. The advantage of ethanol over gasoline as fuel is the minimum standard pollutant emitted from its burning, such as carbonmonoxide. Ethanol oxidation and combustion processes use 372 elementary reactions and yield various intermediates and products.

This computational study has several purposes to find out, such as ignition delay times, emitted pollutants and influences of pressure, temperature and equivalence ratio on ethanol oxidation and combustion. Hence, a comprehensive and representative chemical kinetic model for ethanol oxidation and combustion is needed to fulfill the purposes of this study. This model is expected for high validation and represents the real condition of ethanol oxidation and combustion. The computational results are compared to the experimental data from flow reactors and jet stirred reactors for concentration profiles, and shock tube for ignition delay times. Good agreement was found for the validation with experimental data.

Mechanism analyses were doing by applying the sensitivity and reaction flow analyses. Sensitivity analysis is used to identify rate-limiting reaction steps. Reaction flow analysis calculates the percentage of reaction contribution to the formation or consumption of chemical species. Sensitivity analysis shows that OHO + HO2 + is the most sensitive reaction for ignition delay times.

Ethanol oxidation and combustion were simulated within the temperature range of 1000-2000 K, the pressure range of 1-15 atm and the equivalence ratio (φφφφ) range of 0.5-2.0. The simulation was performed using Chemkin 3.7.1 software. Keywords– ethanol oxidation and combustion, kinetic model, ignition delay times

I. INTRODUCTION

he usage of fossil fuels suffers pollution problems that affect the environment. This has become the

main issue of energy resource exploitation. Energy alternative rises as the solution for less standard pollutant and sustainable energy. Bioethanol is an energy alternative which usage is expected to solve the energy problems.

Based on the National Energy Mix, the Government has planned the usage of biodiesel and bioethanol in amount of approximately 2% of overall national fuels usage in 2010, and will rise up to 5% in 2025 [1].

Previous chemical kinetic modeling studies of ethanol oxidation were performed by Natarajan and Bhaskaran [2]. The following modeling works focused on problems of ethanol ignition delay in shock tubes, and ethanol laminar flame speeds in burners and jet stirred reactors [2,3,4,5]. The time needed from the start of injection and the start of combustion is referred to as the ignition delay.

Egolfopoulos [4] modeling study demonstrated a good agreement with Natarajan [2] ignition delay data and found ( M)OH(HC 52 + M)OH(CHCH 23 ++ ) as the

most sensitive reaction for ignition delay. Curran [6] did not find it in his modeling study at 1100-1900 K and 2-4.5 bar. Detailed mechanism is generally validated with experimental data from certain range of operating condition [7].

So far, the modeling of ethanol oxidation and combustion had not achieved wide validation range. Hence, a comprehensive mechanism of ethanol oxidation and combustion in wide range of operating condition (pressure, temperature and equivalence ratio) is needed to be validated with various experimental data and represents the actual combustion system.

II. BASIC THEORY

Chemical reactions occur when molecules of one species collide with molecules of another species and, for some of these collisions, one or more new molecules

T

ISSN: 1411-1284


will be created. In the chemical reaction the atoms of reacting molecules are redistributed in the new molecules. To achieve this, the reacting molecules must have sufficient kinetic energy so that their chemical bonds can be broken during the impact and other bonds can be formed. As the energy of these bonds depends on the nature of the atoms and on geometrical factors, the energy content of the products of the collision may be different from the energy content of the colliding molecules. This is the basis for heat being released or absorbed in chemical reactions [8].

II.1 Temperature Dependence of Rate Coefficients

The reaction rate coefficient is given by the Arrhenius law:

−=RT

EAk aexp' (1)

where Ea is the activation energy. Not all molecular collisions will result in reaction, but only those with kinetic energy higher than the energy needed to break the bonds of the reactant molecules. This energy barrier is the activation energy. Its maximum value corresponds to the bond energies in the molecule. In dissociation reactions, the activation energy is approximately equal to the bond energy being broken. The value of activation energy may also be much smaller or even zero [9].

Since most elementary binary reactions exhibit Arrhenius behavior over modest ranges of temperature, the temperature dependence can usually be incorporated with sufficient accuracy into the exponential alone. However, for the large temperature ranges found in combustion, “non-Arrhenius” behavior of the rate coefficient tends to occur, particularly for processes that have a small energy barrier. Therefore, it is necessary to adopt a modified Arrhenius form which expresses the impact of temperature on the rate coefficient [10],

−=RT

EATk b aexp . (2)

II.2 Pressure Dependence of Rate Coefficients

The Lindemann mechanism illustrates the fact that reaction orders of complex (nonelementary) reactions depend on the conditions chosen. However, the Lindemann mechanism itself is a simplified model.

If the rate law of a unimolecular reaction is written as [A]d[P]/d akt = , then the rate coefficient k depends on

pressure and temperature. The theory of unimolecular reactions yields fall-off curves, which describe the pressure dependence of k for different temperatures. The logarithm of the rate coefficient is usually plotted versus the logarithm of the pressure [9].

Typical fall-off curves are shown in Fig. 1. At very high pressures )[M][M]/( ua-au kkkkk += tends to the

limit ∞k , i.e., the rate coefficient becomes independent

of the pressure. At very low pressure, the rate coefficient k is proportional to [M] = p/RT, resulting in a linear dependence. Similarly, if the effective activation energy is low, the reaction rate coefficient k will decrease with temperature [9].

Fig.1. Fall of curves for unimolecular reaction

3362 HCHCHC && +→ [9]

II.3 Ignition Delay Time

Ignition delay is characteristic for radical chain explosions. During the ignition delay period, the radical pool population is increasing at an exponential rate. Yet, the amount of fuel consumed, and hence the amount of energy liberated, is too small to be detected. Thus, important chemical reactions (chain branching, formation of radicals) take place during the induction time, whereas the temperature remains nearly constant [9].

Finally, the radical pool becomes large enough to consume a significant fraction of the fuel, and rapid ignition will take place. The precise definition of induction time depends on the criterion used (consumption of fuel, formation of CO, formation of OH, increase of pressure in a constant volume vessel, etc.) [9].

Due to the temperature dependence of the underlying elementary reactions, the ignition delay time depends strongly in the temperature. This is shown in Fig. 2 for several hydrocarbon-air mixtures [9].

ISSN: 1411-1284


Fig.2. Calculated (line) and measured (points) ignition delay times in hydrocarbon-air mixtures [9]


III.1 Mechanism Validation

A mechanism consisting 372 elementary reactions for ethanol oxidation and combustion has been developed in this study. The mechanism validation was performed by comparing calculation results to experimental measurements for shock tubes, jet stirred reactors and flow reactors available in literature.

10

100

1000

6 6,5 7 7,5 8

10000/T (K-1)

Ign

itio

n d

elay

tim

es(m

ikro

det

ik)

Fig. 3. Comparison between shock tube ignition delay data as

investigated by Natarajan and Bhaskaran [2] and the numerical calculations using the chemical kinetic model.

Experimental conditions: 2.5% C2H5OH, 7.5% O2, and 90% Ar, φ = 1.0,

P = 1± 0.2 and 2.0 ± 0.2 atm

As shown in Fig. 3, ignition delay time is shorter at higher pressure and temperature. This behavior was reproduced very well by the model.

Figure 4 shows temperature behavior for ignition delay time in shock tube. Pressurized wave resulted in shock tubes will increase the system pressure hence temperature. The rise of pressure and temperature will improve system reactivity and radical pool. Formation of large enough radical pool will consume significant fraction of the fuel and rapid ignition will take place.

Fig.4. Temperature behaviour from ignition delay time in shock tube

Aboussi [11] investigated ethanol oxidation in jet

stirred reactors. The experimental data were used to validate the kinetic model by comparing it with numerical calculations.

0,000001

0,00001

0,0001

0,001

0,01

0,1

0 0,05 0,1 0,15 0,2 0,25 0,3

Residence time (s)

Fra

ksi m

ol

C2H5OH sim C2H4 sim CO sim CO2 sim

C2H5OH exp C2H4 exp CO exp CO2 exp

Fig.5. Ethanol-O2-N2 (0.3%–0.9%–98.8%) oxidation in a jet stirred reactor at φ= 1.0, 1.0 atmosphere,

and a nominal temperature of 1056 K. Measurements and predictions for C2H5OH, C2H4, CO and CO2

Figure 5 shows the comparison between numerical

calculations for C2H5OH, C2H4, CO and CO2. Overall, the modeling results show good agreement with the species concentration as the residence time was varied.

Norton and Dryer [12] investigated ethanol oxidation in turbulent flow reactors. The experimental data were used to validate the kinetic model by comparing it with numerical calculations using the chemical kinetic model.

0

0,02

0,04

0,06

0,08

0,1

0,12

0 0,02 0,04 0,06 0,08 0,1 0,12

Residence time (s)

Fra

ksi

mo

l

C2H5OH exp O2 exp CO exp H2O exp H2 exp CO2 exp

C2H5OH sim O2 sim CO sim H2O sim H2 sim CO2 sim

O2 х 0,8

C2H5OH х 2

H2O

CO

H2

CO2

Fig.6. Comparison between experimental flow reactor

oxidation data for φ= 1.24 as investigated by Norton and Dryer [12] and the numerical calculations. The numerical

results were time “shifted” by 40 msec. Experimental conditions: 5.81% C2H5OH, 14.07% O2, and 80.12% N2,

ISSN: 1411-1284


Reynolds Number = 4900, P = 1atm, and Tin = 1100 K. Numerical simulations shown for C2H5OH, O2, CO, CO2, H2,

and H2O.

Fig.6. shows good agreement between experimental data and numerical results for C2H5OH, O2, CO, CO2, H2 and H2O in temperature of 1100 K and equivalence ratio of 1.24. III.2 Mechanism Analyses III.2.1 Sensitivity Analysis A sensitivity analysis is used to identify rate-limiting reaction steps. Numerical calculations for sensitivity analysis were carried out by changing the rate coefficient of a particular reaction and calculating the OH concentration.

-0,6 -0,4 -0,2 0,0 0,2 0,4 0,6 0,8 1,0 1,2

1

Φ = 0,5Φ = 2,0

Fig.7. Sensitivity coefficients of the important reactions for ignition delay time in lean and rich fuel mixture, based on

Dunphy dan Simmie[3] experimental condition in shock tube. P = 3.4 bar, T = 1430 K, φ1 = 0.5 (1.25% C2H5OH, 7.5% O2

and 91.25% Ar), φ2 = 2.0 (2.5% C2H5OH, 3.75% O2 and 93.25% Ar)

Figure 7 shows sensitivity coefficients of the

important reactions for ignition delay time. Six most sensitive reactions for lean and rich fuel mixture are: (+) OHO + HO 2 +

(+) 23 HOCH + OHOCH3 +

(+) MHCO+ MCOH ++ (+)

2OHCO+ 2HOCO+

(+) MOCHCH 23 + MHHCOCH3 ++

(-) MOCHCH 23 + MOCHCH 23 ++

A sensitivity coefficient might be positive or negative. A positive sensitivity coefficient indicates a higher OH concentration and an increased overall reaction rate, and a negative sensitivity coefficient

indicates a lower OH concentration and a decreased overall reaction rate of the system.

-0,5 0 0,5 1 1,5

1

CH3+HO2=CH3O+OH

O+OH=O2+H

H+HO2=H2+O2

CH3+CH3(+M)=C2H6(+M)

CH3+HO2=CH4+O2

C2H5OH(+M)=CH3+CH2OH(+M)

C2H5OH(+M)=C2H4+H2O(+M)

C2H5OH+OH=C2H4OH+H2O

C2H5OH+CH3=CH3CHOH+CH4

C2H5OH+HO2=CH3CHOH+H2O2

CH3CH2O+M=CH3HCO+H+M

CH3CH2O+M=CH3+CH2O+M

Fig.8. Sensitivity coefficients of the important reactions for ethanol oxidation in a flow reactor. Experimental condition: φ =0,61 (5.65% C2H5OH, 27.86% O2 and 66.49% N2), P = 1

atm and Tin = 1092 K

Figure 8 shows sensitivity analysis results for φ = 0.61 in flow reactors. Several most sensitive reactions ethanol oxidation in flow reactor are: (+) OHO + HO2 +

(+)23 HOCH + OHOCH3 +

(-)23 HOCH + 24 OCH +

(-)2HOH + 22 OH +

(-) MOCHCH 23 + MOCHCH 23 ++

(+) MOCHCH 23 + MHHCOCH3 ++

III.2.2 Reaction Flow Analysis

A reaction flow analysis calculates the percentage of reaction contribution to the formation or consumption of chemical species.

+O2 5,2%

+O2, M 83,7 %

+OH, HO2 97,8 %

+H, OH, O 64,1 %

+OH, O, H 68,3%

+OH 29,1%

+O 24%

+O2 11,1%

+O, O2 40,3%

+OH, H, O 65,1%

+O 27,8%

+M 46,8%

+O2 46%

+O 64,6%

+H 11,2 %

CO

HCO

CH2O

CH2OH CH3

CH3OH

CH2CO

CH4

C2H4

CH2HCO

CH3HCO

CH3CH2O CH3CHOH

CO2

ISSN: 1411-1284


Fig.9. The pathways of CO2 formation in flow reactor at P = 1 atm, equivalence ratio (φ) = 0.61 and T = 1092 K

Figure 9 shows carbondioxide formation from its

reaction with OH and HO2 radical. OHCO+ HCO2 + 92.6%

2HOCO+ OHCO2 + 5.2%

Carbonmonoxide is resulted from radical formyl oxidation, formyl decomposition with third body, and oxidation and decomposition of CH2HCO.

2OHCO+ 2HOCO+ 75.7%

MHCO+ MCOH ++ 8%

22 OHCOCH + OHCOOCH2 ++ 3.3%

HCOCH2 COCH3 + 1.9%

IV. CONCLUSIONS

Validation of the mechanism for oxidation and combustion of ethanol in shock tubes and flow and jet-stirred reactors shows good agreement. The elementary reaction OHO + HO2 + is shown as the most

sensitive reaction for ignition delay time at 1430 K, 3.4 and 5 bar, and the equivalence ratios of 0.5 and 2.0. It is also the most sensitive reaction for ethanol oxidation rate in flow reactor at 1092 K, 1 atm, and the equivalence ratio of 0.61. Ignition delay time in shock tube is shorter at high pressure and fuel lean condition. Lowest species concentrations in flow and jet sstirred reactor are found at low equivalence ratio, low pressure and temperature.

REFERENCES [1] Prasetyo E. “Energi Alternatif Ramah Lingkungan”.

Accesed on 22 Mei 2006, from Pikiran Rakyat. http://www.pikiranrakyat.com/cetak/2005/0705/15/otokir/utama01.htm, 2005.

[2] Marinov N.M . “A Detailed Chemical Kinetic Model for High Temperature Ethanol Oxidation”. International Journal of Chemical Kinetic, pp.183, 1999.

[3] Dunphy M.P, J.M. Simmie. Journal Chemical Society Faraday Trans 87 pp.1691–1695, 2549–2559, 1991.

[4] Egolfopolous F.N, D.X. Du, C.K. Law. Twenty-Fourth Symposium (International) on Combustion. The Combustion Institute, pp. 833, 1992.

[5] Dagaut P, M. Cathonnet, J.C. Boettner. Journal Chemistry Physic 89, pp.867–884, 1992.

[6] Curran H.J, et al. Twenty-Fourth Symposium (International) on Combustion. The Combustion Institute, pp. 769, 1992.

[7] Held T.J, F.L. Dryer. “A Comprehensive Mechanism for Methanol Oxidation”. International Journal Chemical Kinetic 30, pp.805, 1998.

[8] Muharam Y. “Detailed Kinetic Modelling of the Oxidation and Combustion of Large Hydrocarbons Using an Automatic Generation of Mechanisms”. Dissertation, University of Heidelberg, Heidelberg, pp.1-27, 2005.

[9] Warnatz J, U. Maas, R.W. Dibble. “Combustion, Physical and Chemical Fundamentals, Modelling and Simulation, Experiments, Pollutant Formation 3rd ed”. Heidelberg: Springer, 2001.

[10] Glassman I. “Combustion 3rd edition”. New York: Academic Press, 1996.

[11] Aboussi B. Dissertation, Orleans, 1991. [12] Norton T.S, F.L. Dryer. “An Experimental and Modeling

Study Of Ethanol Oxidation Kinetics in an Atmospheric Pressure Flow Reactor”. International Journal Chemical Kinetic 24, pp.319–344, 1992.

ISSN: 1411-1284


WATER AND AIR CIRCULATION SYSTEM IN TRIMARAN FISHING VESSEL LIVE FISH HOLDS

Sunaryo, Anita Muslih

Naval Architecture Study Program, Mechanical Engineering Department, Faculty of Engineering, University of

Indonesia. Kampus UI, Depok 16424, tel: 021-7270032, fax: 021- 7270033

[email protected] ABSTRACT- Indonesia has abundant of natural resources including fish and aquaculture, but so far the effort for exploiting it has not optimal yet, such as in providing sufficient fishing fleet for the national fishing industry, and the enabling of traditional fishermen community to improve their living standard. For this purpose sea fish farming is introduced to some potential fishermen communities. One type of the fish being farmed is grouper, which has high market value to be exported to the Far-East. But the fish in demand in the market is live fish, therefore triggered by this challenge as a continuation of the previous research, this research is carried out focusing in the arrangement of the water and air circulation system in the Trimaran fishing vessel live fish hold, so that the fish being transported could be kept alive until they arrive at the destination. Some requirements should be fulfilled in order to

keep the fish (in this case grouper fish) alive such as: The density of the fish per m3 of water in the fish hold, cleanliness of the water in the fish hold, salinity of the water, water temperature, mineral content and concentration in the water, oxygen and nitrogen contain, and the water circulation speed. Since the vessel was designed as a trimaran, all

the propulsion machinery are installed in the side hulls, and the middle hull is dedicated for the storing of fish, the water and air circulation system is arranged in the middle body of the middle hull. Two pumps are utilized one for circulation fresh sea water and the other for discharging the contaminated or dirty water. Air is blown in from a compressor through distribution pipes to all the fish holds. For keeping the right temperature in the fish hold sea water is sucked from the sea chest and passed through a chiller prior being flown into the fish hold.

With 16.3 ton fish hold capacity it is estimated that 0.38 ton of live grouper fish could be transported. Key words - trimaran, fishing vessel, fish hold, water and air circulation

I. INTRODUCTION

Indonesia has abundant of natural resources including fish and aquaculture, but so far the effort for exploiting it has not optimal yet and the traditional fishermen are still using traditional ways of fishing methods. In order to improve the fishermen’s standard of living various efforts have been done among others is by introducing fish farming for certain types of fish that have prospective in global market, grouper is one of such kind of fish. But grouper has to be exported live, and based on this requirement the research is focused on the method for transporting the live fish. The type of boat for carrying the fish is in the form of trimaran boat, which is considered more stable compared to the mono hull. In order to keep the fish alive the water and air circulation in the fish hold should be well maintained, this include the density of the fish per m3 of water in the fish hold, cleanliness of the water in the fish hold, salinity of the water, water temperature, mineral content and concentration in the water, oxygen and nitrogen contain, and the water circulation speed.

The objectives of the research are: to introduce the novel design of trimaran vessel, which has its propulsion engines installed in the side hulls, and the middle hull is fully used for carrying the live fish, to design the water and air circulation system in the fish holds.

II. ARRANGEMENT OF THE BOAT

Grouper has the following requirements in order

to keep them alive: 1. Salinity : 30 - 34 ppt (part per thousand) or

permil, 2. Water temperature : 24 -32 degree Celcius, 3. Water PH :7 – 9, 4. Speed of water flow : 20 – 50 cm/second 5. Content of oxygen solution minimum 3 ppm, 6. Nitrogen nitrite : 0 - 0.05 ppm, 7. Ammonia (NH3-N) < 0.02 ppm 8. Dark environment because grouper is night

active fish, 9. Density of fish in the water (kg/m3 ), 10. height of the fish hold, and 11. the wide of the hatch.

ISSN: 1411-1284


To cater the above requirements the boat is designed as a trimaran boat with the following specificationsthe side hull are the smaller parts of the boat where the propulsion engines are located.

1. Middle hull: The middle hull is the largest part of the boat

where the fish hold are located, the main particulars of the middle hull are as follows:

Table 1. Middle hull main particulars

There are three fish holds in the middle hull,

which will be filled with water so that the fish can be kept alive; one pump room located in the middle where all the machinery for maintaining the fish hold conditions is located; above the pump room is the wheel house. Arrangement of the middle hull can be seen in the figure 1.

Figure

1. Middle hull

2. Side hull The size of the side hulls are smaller than the

middle hull, their purpose is to stabilize the boat, in both of the side hulls the propulsion engines are installed. The main particulars of the side hulls are as follows:

Table 2. Side hulls main particulars

Figure 2. Side hulls The arrangement of the fish hold are as follows:

Table 3. Fish hold division:

The capacity of the boat is 380 kg of live grouper consists of 636 of fish which have economical value. The sea water needed for the fish is 15,86 m3 or 16,255 ton.

III. THE WATER AND AIR CIRCULATION SYSTEM

The water and air circulation system for the fish

holds is generated by the following equipment: 1. Circulation pump 2 HP. 2. General pump 2 HP. 3. Filter 4. Generator 5. Chiller 6. Compressor

Loa 15 m

Lwl 14,55 m

Lpp 14,2 m

D 1,31 m

H 1,67 m

B 2,84 m

Cb 0,47

Cw 0,647

Cm 0,968

Cp 0,486

Loa 10 m

Lwl 9,7 m

Lpp 9,46 m

D 0,87 m.

H 1,12 m.

B 1,9 m

Cb 0,32

Cw 0,55

Cm 0,97

Cp 0,33

Item Hold

1

Hold

2

Hold

3

total

Volume

(m3)

2,45 7,5 5,91 15,86

Berat

(ton)

2,51

1

7,687 6,057 16,25

5

ISSN: 1411-1284


7. circulation pipes 8. suction pipes for dirty water

IV. CONCLUSIONS

After being calculated for the proposed trimaran boat the water and air circulation system can be concluded as follows:

1. The total fish hold capacity is 16,225 ton, consisted of 0,38 ton grouper live fish, and 15,875 ton of sea water.

2. There are two pumps being used one for circulating the water and general pump for pumping the dirty water, each is powered by 2 hp electric motor.

3. To maintain the water temperature a chiller is being used.

REFERENCES

1. Geankoplis,C.J. Transport Processes and Unit

Operations (New Delhi: Prentice Hall of India, 1997). 2. J.O, Traung, Fishing Boat Of The World 2 ( England:

Fishing New Books Ltd, 1960), hal 87 – 89 3. Kern, D.Q., Process Heat Transfer (Tokyo: International

Student Edition, Mc Graw – Hill Book Co, 1965). 4. Mallawa. A, Sudirman, Teknik Penangkapan Ikan

( Jakarta: PT. Rineka Cipta, 2004).

ISSN: 1411-1284


Local Protein Structure Comparison

1Ford Lumban Gaol, 2Belawati Widjaja

1,2 Faculty of Computer Science, University of Indonesia,

Kampus UI Depok, Jawa Barat, Indonesia Tel. 021-7863419, fax. 021-7863415 email : [email protected]

Abstract– Genome sequencing projects are working to determine the complete genome sequence for several organisms. The sequencing projects have produced significant impact on bioinformatics research by stimulating the development of sequence analysis tools such as methods to identify genes in a genome sequence, methods to predict alternative splicing sites for genes, methods that compute the sequence homology among genes, and methods that study the evolutionary relation of genes, to name a few. Proteins are the products of genes and the building blocks for biological function. In this paper, we identify some basic background on local proteins structure, protein function and their simulation. Keywords– Genome sequence, Pattern-Based Structure Alignment, Functional Site Identification, Structure-based Functional Annotation.

I. PROTEIN STRUCTURE

1.1 Amino Acids Proteins are chains of α-amino acid molecules. An

α-amino acid (or simply an amino acid) is a molecule with three chemical groups and a hydrogen atom covalently bonded to the same carbon atom, the Cα atom. These groups are: a carboxyl group (-COOH), an amino group (-NH2), and a side chain with variable size (symbolized as R) [BT91]. The first carbon atom in a side chain (one that is connected to the Cα atom) is the Cα atom and the second one is the C atom and so forth. Figure 1 illustrates an example of amino acids.

Different amino acids have different side chains. There are a total of 20 amino acids found in naturally occurring proteins. At physiological temperatures in a solvent environment, proteins adopt stable three-dimensional (3D) organizations of amino acid residues that are critical to their function. 1.2 Four Levels of Protein Structure The levels are as follows:

Figure 1: Up: a schematic illustration of an amino acid.

Down: the 3D structure of an amino acid (Alanine) whose side chain contains a single carbon atom. The atom types are shown; unlabeled atoms are hydrogens. The schematic

diagram is adopted from [BT91] and the 3D structure is drawn with the VMD software.

• Primary structure describes the amino acid sequence of a protein.

• Secondary structure describes the pattern of hydrogen bonding between amino acids along the primary sequence. There are three common types of secondary structures: α-helix, β-sheet, and turn.

• Tertiary (3D) structure describes the protein in terms of the coordinates of all of its atoms.

• Quaternary structure applies only to proteins that have at least two amino acid chains. Each chain in a multi-chain protein is a subunit of the protein and the spatial organization of the subunits of a protein is the quaternary structure of the protein. A single-subunit protein does not have a quaternary structure.

Primary Structure In a protein, two amino acids are connected by a peptide bond, a covalent bond formed between the carboxyl group of one amino acid and the amino group of the other with elimination of a water molecule. After the condensation, an amino acid becomes a amino acid residue (or just a residue, for short). The Cα atom and the hydrogen atom, the carbonyl group (CO), and the NH group that are covalently linked to the Cα atom are the main chain atoms; the rest of the atoms in an amino acid are side chain atoms. In Figure 2, we show the primary sequence of a protein with three amino acid residues. At one end of the sequence (the left one), the residue contains the full amino group (-NH3) and is the N terminal of the sequence. The residue at the opposite end contains the full carboxyl group (-COOH) and is the C terminal of the sequence. By convention a

ISSN: 1411-1284


protein sequence is drawn left to right from its N terminal to its C terminal. Various protein sequencing techniques can determine the primary sequence of a protein experimentally. Figure 2: A schematic illustration of a polypeptide with three residues: Met, Gly and Ala. The peptide can also be described as the sequence of the three residues: Met-Gly-Ala. Secondary Structure A segment of protein sequence may fold into a stable structure called secondary structure. Three types of secondary structure are common in proteins.

• α -helix • β -sheet • turn An α -helix is a stable structure where each

residue forms a hydrogen bond with another one that is four residues apart in the primary sequence. We show an example of the α –helix secondary structure in Figure 3.

A β -sheet is another type of stable structure formed by at least two β -strands that are connected together by hydrogen bonds between the two strands. A parallel β -sheet is a sheet where the two β -strands have the same direction while an anti-parallel β -sheet is one that does not. We show examples of β -sheets in Figure 3.

A turn is a secondary structure that usually consists of 4-5 amino acids to connect α –helices or β -sheets.

Unlike the protein primary sequence, protein secondary structure is usually obtained after solving the 3D structure of the protein.

Figure 3: A schematic illustration of the α -helix and the β -sheet secondary structures. (a) the ribbon representation of the α -helix secondary structure (on the left) and the ball-stick representation showing all atoms and their chemical bonds in the structure (on the right). We also show the same representations for the parallel β -sheet secondary structure (b) and the anti-parallel β -sheet secondary structure (c). The α -helix is taken from protein myoglobin 1MBA at positions 131 to 141 as in [Fer99]. The parallel β -sheet secondary structure is taken from protein 2EBN at positions 126 to 130 and 167 to 172. The anti-parallel β –sheet secondary structure is taken from protein 1HJ9 at positions 86 to 90 and 104 to 108. Tertiary Structure and Quaternary Structure

In conditions found within a living organism, a protein folds into its native structure. The tertiary structure refers to the positions of all atoms, generally in the native structure. The process of adopting a 3D structure is the folding of the protein. Protein 3D structure is critical for a protein to carry out its function.

In Figure 4, we show a schematic representation of a 3D protein structure (myoglobin). In the same figure, we also show the quaternary structure of a protein with two chains (HIV protease).

Figure 4: Right: The schematic representation (cartoon) of the 3D structure of protein myoglobin (PDB id: 1MBA). Left: The schematic representation (cartoon) of the 3D structure of protein HIV protease (PDB id: 1MBA). HIV protease has two chains.

Two types of experimental techniques are used to determine the 3D structure of a protein. In X-ray crystallography, a protein is first crystalized and the structure of the protein is determined by X-ray diffraction. Nuclear Magnetic Resonance spectroscopy (NMR) determines the structure of a protein by measuring the distances among protons and specially labeled carbon and nitrogen atoms [PR04]. Once the inter-atom distances are determined, a group of 3D structures (an ensemble) is computed in order to best fit the distance constraints. 1.3 Structure Classification Domains A unit of the tertiary structure of a protein is a domain, which is the whole amino acid chain or a (consecutive) segment of the chain that can fold into stable tertiary structure independent of the rest of the protein [BT91]. A domain is often a unit of function i.e. a domain usually carries out a specific function of a protein. Multi-domain proteins are believed to be the product of gene fusion i.e. a process where several genes, each which once coded for a separate protein, become a single gene during evolution [PR04]. Structures are Grouped Hierarchically The protein structure space is the set of all possible protein structures. Protein structure space is often described by a hierarchical structure called protein structure classification, at the bottom of which are individual structures (domains). Structures are grouped hierarchically based on their secondary structure components and their closeness at the sequence, functional, and evolutionary level [PR04]. Here we describe a structure hierarchy, the SCOP database (Structure Classification of Proteins)

ISSN: 1411-1284


[MBHC95]. SCOP is maintained manually by domain experts and considered one of the gold standards for protein structure classification. For other classification systems see [OMJ+97]. In SCOP, the unit of the classification is the domain (e.g. multi-domain proteins are broken into individual domains that are grouped separately). At the top level (most abstract level), proteins in SCOP are assigned to a class based on the secondary structure components. The four major classes in SCOP are:

• α domain class: ones that are composed almost entirely of α -helices

• β domain class: ones that are composed almost entirely of β -sheets

• α/ β domain class: ones that are composed of alpha helices and parallel beta sheets

• α + β domain class: ones that are composed of alpha helices and antiparallel beta sheets

These four classes cover around 85% of folds in SCOP. Another three infrequently occurring classes in SCOP are: multi-domain class, membrane & cell surface domain class, and small protein domain class. Proteins within each SCOP class are classified hierarchically at three additional levels: fold, superfamily, and family. In Figure 5, we show a visualization developed by the Berkeley Structural Genomics Center http://www.nigms.nih.gov/psi/image gallery/structures.html in which globally similar structures are grouped together and globally dissimilar structures are located far away from each other. This figure shows segregation between four elongated regions corresponding to the four SCOP protein classes: α, β, α/ β, and α + β. Further details about protein structure classification can be found in [MBHC95]. Figure 5: The top level structural classification of proteins based on their secondary structure components.

II. PROTEIN FUNCTION

Proteins are the molecular machinery that perform the function of living organisms. Protein function can be described by the role(s) that the protein plays in an organism. Usually, protein function description is made at the molecular level, e.g. the role a protein

plays in a chemical reaction. Protein function can also be described at a physiological level concerning the whole organism, e.g. the impact of a protein on the functioning of an organism. We describe protein function at 3 different levels according to [OJT03]:

• Molecular function: A protein's molecular function is its catalytic activity, its binding activity, its conformational changes, or its activity as a building block in a cell. [PR04].

• Cellular function A protein's cellular function is the role that the protein performs as part of a biological pathway in a cell.

• Phenotypic function: A protein's phenotypic function determines the physiological and behavioral properties of an organism.

We need to keep in mind that protein function is context-sensitive with respect to many factors other than its sequence and structure. These factors include (but are not limited to) the cellular environment in which a protein is located, the post-translation modification(s) of the protein, and the presence or absence of certain ligand(s). Though often not mentioned explicitly, these factors are important for protein function.

In this paper, we concentrate on the molecular function of a protein. We do so since (1) it is generally believed that native structure may most directly be related to the molecular function [GSB+05], (2) determining the molecular function is the first step in the determination of the cellular and phenotypic function of a protein.

III. THE IMPORTANCE OF LOCAL STRUCTURE COMPARISON

Traditionally global structure comparison is well investigated in protein structure analysis. Recently the research focus has shifted towards local structure comparison. The transition from global structure comparison to local structure comparison is well supported by a wide range of experimental evidence. 3.1 Protein Function It is well known that in a protein there are a few key residues that if mutated, interfere with the structural stability or the function of the protein. Those important residues usually are in spatial contact in the 3D protein structure and hence form a “cluster" in the protein structure. On the other hand, much of the remaining protein structure, especially surface area, can tolerate mutations [CS99, SKBW98]. For example, in a model protein T4 Lysozyme, it was reported that single amino acid substitutions occurring in a large fraction of a protein structure (80% of studied amino acids) tend not to interrupt the function and the folding of the protein [Mat96]. Biology has accumulated a long list of sites that have functional or structural significance.

ISSN: 1411-1284


Such sites can be divided into the following three categories:

• catalytic sites of enzymes • the binding sites of ligands • the folding nuclei of proteins

Local structure similarity among proteins can implicate structurally conserved amino acid residues that may carry functional or structural significance [KWK04]. Similar Structures, Different Function It is well known that the TIM barrels are a large group of proteins with a remarkably similar fold, yet widely varying catalytic function [NOT02]. A striking result was reported in [NKGP90] showing that even combined with sequence conservation, global structure conservation may still not be sufficient to produce functional conservation. In this study, Neidhart et al. first demonstrated an example where two enzymes (mandelate racemase and muconate lactonizing enzyme) catalyze difierent reactions, yet the structure and sequence identities are sufficiently high that they are very likely to have evolved from a common ancestor. Similar cases have been reviewed in [GB01]. Different Structures, Same Function It has been also noticed that similar function does not require similar structure. For example, the most versatile enzymes, hydro-lyases and the O-glycosyl glucosidases, are associated with folds [HG99]. In a systematic study using the structure database SCOP and the functional database Enzyme Commission (EC), George et al. estimated 69% of protein function (at EC subsubclass level) is indeed carried by proteins in multiple protein superfamilies [GST+04]. In summary, we want to develop methodology to recognize protein similarity at the structure pattern level, and use such method as a complemental method to ones with similar purpose that are based on sequence and global structure similarity.

IV. CONCLUSIONS

Protein structure comparison is part of a bioinformatics research paradigm that performs comparative analysis of biological data. The overarching goal is to aid rational experiment design and thus to expedite biological discovery. Specifically, through comparison, the paradigm endeavors to transfer experimentally obtained biological knowledge from known proteins to unknown ones, or to discover common structure among a group of related proteins. Below we review some of the applications of local structure comparison including pattern based structure alignment, functional site identification, structure-based functional annotation, and protein evolution.

Pattern-Based Structure Alignment Structure alignment is vital to identifying conserved residues in protein structure, to studying the evolution of protein structures, and to facilitating structure prediction. For example, through structure alignment, domain experts discover that many diverse sequences may adopt the same global structure and such information helps significantly in structure prediction. Multiple structure alignment has been used to identify structural commonalities among a group of proteins. Two approaches have been investigated. The first uses multiple sequence alignment to solve the difficult alignment problem and then focuses on identifying the structural core of proteins. The second approach applies pairwise structure alignment iteratively in order to derive multiple structure alignment. Functional Site Identification A functional site is a group of amino acids in a protein that participate in the function of the protein (e.g. catalyzing chemical reactions or binding to other proteins). Identifying functional sites is critical in studying the mechanism of protein function, predicting protein-protein interaction, and recognizing evolutionary connections between proteins when there is no clear clue from sequence or global structure alignment. Structure-based Functional Annotation There is no question that knowing the function of a protein is of paramount importance in biological research. Correct function prediction can significantly simplify and decrease the time needed for experimental validation. However incorrect assignments may mislead experimental design and waste resources. Protein function prediction has been investigated by recognizing the similarity of a protein with unknown function to one that has a known function where similarity can be determined at the sequence level, the expression level, and at the level of the gene's chromosome location. In structure based function annotation, investigators focus on assigning function to protein structures by recognizing structural similarity. Compared to sequence-based function assignment, structure-based methods may have better annotation because of the additional information offered by the structure.

In summary, the potential to decrease the time and cost of experimental techniques, the rapidly growing body of protein structure and structure related data, and the large number of applications necessitate the development of automated comparison tools for protein structure analysis.

REFERENCES 1. [BT91] CARL BRANDEN and JOHN TOOZE.

Introduction to Protein Structure. Garland Publishing, 1991.

2. [Fer99] ALAN FERSHT. Structure and Mechanism in Protein Science. W.H. Freeman Comp., 1999.

3. [PR04] GREGORY A PETSKO and DAGMAR RINGE. Protein Structure and Function. New Science

ISSN: 1411-1284


Press Ltd, Middlesec House, 34-42 Cleveland Street, London W1P 6LB, UK, 2004.

4. [MBHC95] A.G. MURZIN, S.E. BRENNER, T. HUBBARD, and C. CHOTHIA. SCOP: a structural classification of proteins database for the investigation of sequences and structures. Journal of Molecular Biology, 247:536-540, 1995.

5. [OMJ+97] C.A. ORENGO, A.D. MICHIE, S. JONES, D.T. JONES, M.B. SWINDELLS, and J.M. THORNTON. CATH - a hierarchic classication of protein domain structures. Structure, 5(8):1093-1108,1997.

6. [OJT03] C. ORGENGO, D. JONES, and J. THORNTON. Bioinformatics: genes, proteins, and computers. BIOS Scientific Publishers ltd., 2003.

7. [GSB+05] R.A. GEORGE, R.G. SPRIGGS, G.J. BARTLETT, A. GUTTERIDGE, M.W. MACARTHUR, C.T. PORTER, B. AL-LAZIKANI, J.M. THORNTON, and M.B. SWINDELLS. Effective function annotation through residue conservation. PNAS, 102:12299-12304, 2005.

8. [CS99] MH CORDES and RT. SAUER. Tolerance of a protein to multiple polar-to-hydrophobic surface substitutions. Protein Sci., 8(2):31825, 1999.

9. [SKBW98] JM SCHWEHM, ES KRISTYANNE, CC BIGGERS, and STITES WE. Stability effects of increasing the hydrophobicity of solvent-exposed side chains in staphylococcal nuclease. Biochemistry, 37(19):693948, 1998.

10. [KWK04] E. V. KOONIN, Y. I. WOLF, and G. P. KAREV, editors. Power Laws, Scale-free Networks and Genome Biology. Springer, 2004.

11. [NOT02] N NAGANO, CA ORENGO, and JM THORNTON. One fold with many functions: the evolutionary relationships between tim barrel families based on their sequences, structures and functions. Journal of Molecular Biology, 321:741-765, 2002.

12. [NKGP90] DAVID J. NEIDHART, GEORGE L. KENYON, JOHN A. GERLT, and GREGORY A. Petsko. Mandelate racemase and muconate lactonizing enzyme are mechanistically distinct and structurally homologous. Nature, 347:692-694, 1990.

13. [GB01] JOHN A. GERLT and PATRICIA C. BABBITT. Divergent evolution of enzymatic function: Mechanistically diverse superfamilies and functionally distinct suprafamilies. Annu. Rev. Biochem., 70:20946, 2001.

14. [HG99] H. HEGYI and M. GERSTEIN. The relationship between protein structure and function: a comprehensive survey with application to the yeast genome. J Mol Biol, 288:147-164, 1999.

15. [GST+04] RICHARD A. GEORGE, RUTH V. SPRIGGS, JANET M. THORNTON, BISSAN AL-LAZIKANI, and MARK B. SWINDELLS. Scopec: a database of protein catalytic domains. Bioinformatics, Supp 1:I130-I136,2004.

Antibacterial Activity Analysis of Dillenia indica’s Kernel Extract for Escherichia coli: A Preliminary Study

Rita Arbianti, Tania Surya Utami, and Ifa Puspasari

Chemical Engineering Department, Faculty of Engineering, University of Indonesia

Kampus UI Depok 16424, Phone. 7863516, Fax. 7863515 Email: [email protected]; [email protected]

Abstract— The isolation of the crude extract from Dillenia indica’s kernel in n-hexane and chloroform obtained seven fractions called fraction A – G. The antibacterial activity test with the method of paper disc for these fractions towards the bacteria Escherichia coli found that these fractions inhibited the growth of the bacteria except fraction F which was eluted at the accommodation of 50 until 51 by 100% chloroform. While fraction D (result of accommodation of 28 until 37) and fraction G (result of accommodation 53 until 58 by 100% chloroform) proved more active inhibit the growth of Escherichia coli compared to the other, even with the antibiotic cloramphenicol. Keywords— Extraction, antibacterial activity, Dillenia indica, Escherichia coli

I. INTRODUCTION

illenia indica’s kernel produces specific scent that can indicate the existence of atsiri oil and terpenoid as one of its component, since the character of atsiri

oil is also having specific scent [1]. Triterpenoid is known having ability for antibacterial activity. This is also proved by some researches were reported that some of triterpenoid shows antibacterial activity against bacteria such as Bacillus subtilis, Escherichia coli and Micrococcus luteus. While on the other literature it was reported that Dillenia indica’s kernel has been used as a medicine to stomach illness. All these things have put a strong indication that the ability antibacterial activity is in Dillenia indica’s kernel.

Other research said that part from Dillenia indica, leaves, is also having ability to retain the growth of some specific bacteria such as Bacillus cereus [2]. On the other research it was mentioned that Dillenia papuana species specifically on the leaves contain some triterpenoid, oleanen type where this can be used as antibacterial. This ability is because it has double bonding within carbocillic and also ketene function from oleanen [3]. This research is aimed to measure antibacterial activity from Dillenia

indica’s kernel which is expected to understand the usage of Dillenia indica’s kernel specific reason.

The objective of this research is to examine

antibacterial activity of chemical compound from Dillenia indica’s kernel and to know the utilization of this kernel specifically.

II. BASIC THEORY

The testing of antibacterial activity has the same meaning with determine bacterial sensitivity of antibacterial. Basically, the method of antibacterial activity testing could be divided into Diffusion Method and Dilution Method [4, 5]. In dilution method, determining of antibacterial activity is based on diffusivity ability on gelatin slab, which have been inoculated with tested bacterial. The observation conducted with existing of inhibited area around the antibacterial.

While in dilution method, antibacterial mixing with

gelatin medium and the inoculated with tested bacterial. The observation conducted with bacterial growth in medium. Antibacterial activity determined by minimum inhibited concentration of antibacterial in retains bacterial growth.

In this research, antibacterial activity testing was

conducted by using gelatin diffusion method. Basic principle of this testing are: The compound that expected function as antibacterial put in a paper disc on the top of Petri disc, where inoculated bacterial in gelatin medium take place. After 24 hours of incubation, inhibited area formed because of diffusivity of chemical compound through gelatin, where concentration of molecules diffusion higher enough to inhibited bacterial growth. Tested bacterial that used in this research is based on approach of stomach disease caused.

D

mailto:[email protected]


III. EXPERIMENTAL RESULTS Dillenia indica’s conventionally used as medicine of

stomach disease (diarrhea). One of tested bacterial that known as pathogenic bacterial cause of infection of intestines is Escherichia coli. The testing is conducted at 25 mg/mL concentration of hexane’s crude extract and its component from isolation. Antibiotic cloramphenicol used as a positive control with same concentration, while hexane and chloroform as a solvent as negative control. The chosen of cloramphenicol is because this antibacterial has large spectrum and can diffusion well in gelatin medium [4].

Diffusion of tested solution that has ability as

antibacterial in paper disc, for several times could cause decreasing of its concentration. When the concentration is higher enough to inhibit bacterial growth, the growth would inhibit. Hence could cause the forming of area that extends over the paper disc with appropriate of concentration decreasing. The more potential of antibacterial, the more less of concentration needed. Hence assumed that the higher diameter of inhibited area, the more potential this solution as antibacterial.

Result of measurement of inhibited area diameter

can be show in Table 1, while the photograph of antibacterial testing showed in Fig. 1.

Table 1. Measurement result of inhibited area diameter

Inhibited area diameter (mm) Sample Eluted

a b c Crude extract 4 – 12 7,5 - -

Fraction A 16 – 23 15 - - Fraction B 25 – 26 11,5 - - Fraction C 28 – 37 7,5 - - Fraction D 38 – 49 27 - - Fraction E 50 – 51 11,5 - 8,5 Fraction F 53 – 58 - - - Fraction G 8 8 -

Cloramphenicol 26 Hexane -

Chloroform -

Figure 1. Photograph of antibacterial testing with cakram method

Table 1 shows that all of fraction have inhibted

effect of E.coli growth significantly, except for F fraction that eluted in 50-51 of 100% chloroform. It can be caused of existing of chemical compound that do not have ability to retain bacterial growth. While the difference between the measurement of fractions depend on the variation of components in the fraction.

D fraction has a higher inhibited area diameter (27 mm) and competitive enough with commercial antibiotic cloramphenicol (26 mm) at the same concentration. Thos fraction, that eluted at 28-37, has good ability as specific alternative antibacterial of E.coli. While the testing of used solvent i.e. hexane and chloroform, give the negative result, that indicates the using of this solvent have not influence the antibacterial activity.

There are several factors that effected antibacterial

activity, among them is incubation time. The longer incubation time can influence inhibited area diameter, because the more time needed to diffusion through gelatin medium. Hence cause higher concentration in certain spot at paper disc. Inhibited area also influenced by ability to diffusion through gelatin medium. The other factor is accuracy of inhibited area diameter measurement. For result with better accuracy, firs we must determine optical cell density of tested bacterial. Because of this complexity, this research is a preliminary study i.e. qualitative testing, so the factors mentioned above are uncontrolled. But overall, the result of this research shows that hexane’s extract of Dillenia indica’s kernel and its isolation components have same result with literatures.

IV. CONCLUSIONS

The isolation of the crude extract from Dillenia indica’s kernel in n-hexane and chloroform obtained seven fractions called fraction A – G. The antibacterial activity test with the method of paper disc for these fractions towards the bacteria Escherichia coli found that these fractions inhibited the growth of the bacteria except fraction F which was eluted at the accommodation of 50 until 51 by 100% chloroform.

Fraction D (result of accommodation of 28 until 37)

and fraction G (result of accommodation 53 until 58 by 100% chloroform) proved more active inhibit the growth of Escherichia coli compared to the other, even with the antibiotic cloramphenicol.

REFERENCES

[1] Harborne, J.B., “Metode Fitokimia: Penuntun Cara

Modern Menganalisis Tumbuhan”, diterjemahkan oleh Kosasih Padmawinata dan Iwang Soediro, Bandung: Penerbit ITB, 1987.

[2] Sornprasert, Ratapol. et al. “The Potential of Crude Extracts from Matat (Dillenia indica) against Bacteria Growth”, Department of Biology, Faculty of Science and Technology, Rajabhat Institute, Chandrakasem, Bangkok 10900, Thailand.

[3] Nick, Andre. et al. “Antibacterial Triterpenoids from Dillenia papuana and Their Structure-Activity Relationships”, Phytochemistry, Vol. 40, No. 6, P. 1691-1695, Elsevier Science Ltd., 1995.

[4] Jawetz, E., J.L Melnick, & E.A. Adelberg, “Review of

Medical Microbiology”, 11th ed., California: Lange Medical Publication, 1974.

[5] Edwards, I. E. “Antimicrobial Drug Action”, P. 18-19, Macmillan Press, 1980.

[6] Prosea, “Plant Resources of South-East Asia,” Vol. 5, Bogor: Prosea Foundation, 1995.

Antioxidant Activity Analysis for Polar Extract of Dillenia indica’s Kernel

Tania Surya Utami, Rita Arbianti, and Lina Faty

Chemical Engineering Department, Faculty of Engineering, University of Indonesia

Kampus UI Depok 16424, Phone. 7863516, Fax. 7863515 Email: [email protected] ; [email protected]

Abstract— the target of this research is to know whether compound of polar bioactive of Dillenia indica kernel own antioxidant activity. First matter taken is taking polar compound from kernel of Dillenia indica. This matter is conducted by extraction use polar solvent, which is ethanol with some water. Hereinafter, needed by dissociation of the harsh extract become some fraction by using column chromatography. Fixed phase of the column is silica gel, whereas the motion phase is ethanol and chloroform by the polarity is boosted up by gradient. The antioxidant activity of the fraction obtained tested with Carotene bleaching method. Result of extraction in the form of dilution which has brown color. After the solvent was evaporated, we obtained a harsh extract in the form of resin jell tan like caramel as much 128 gram. Result of isolation by column chromatography is yield 7 fraction, A until G, with retention time is very nearby. Result of test of antioxidant activity at environment temperature shows that good enough antioxidant activity compared to a negative control is Fraction B. While test of antioxidant activity conducted at incubation temperature of 40 and 60 oC shows Fraction of G and B own good enough antioxidant activity compared to a negative control. Keyword— Extraction, antioxidant activity, Dillenia indica, carotene bleaching method

I. INTRODUCTION

ntioxidant is a material with ability to deactivate free radical, so it can stop the oxidation process

through several mechanism i.e.: reaction with radical scavenger, singlet oxygen quencher, as a filter for UV radiation, or ion bonding. Antioxidant is a inhibitor for lipid peroxide. Antioxidant not only used as protection for the food from its expiration, but also to defend the cell, i.e.: lipid membrane failure due to oxidation. This is because the character of this antioxidant can avoid chain reaction from the radical formation. This antioxidant also can be used to as a protector from

some illness which is triggered from lipid oxidation process in human body, i.e.: heart, stroke, and cancer. One of the part of this plant that interest for learned is the kernel. From sample of kernel, it can be seen obviously that Dillenia indica’s kernel is easily to colour change when it is laid on the open air. This colour changing is most likely occured due to antioxidant compound that is available within its kernel. The target of this research is to know whether compound of polar bioactive of Dillenia indica’s kernel own antioxidant activity. First matter taken is taking polar compound from kernel of Dillenia indica. This matter is conducted by extraction use polar solvent, which is ethanol with some water. Hereinafter, needed by dissociation of the harsh extract become some fraction by using column chromatography. Fixed phase of the column is silica gel, whereas the motion phase is ethanol and chloroform by the polarity is boosted up by gradient.

II. BASIC THEORY

Antioxidant is a material with ability to deactivate

free radical, so it can stop the oxidation process through several mechanism i.e.: reaction with radical scavenger, singlet oxygen quencher, as a filter for UV radiation, or ion bonding. Antioxidant is a inhibitor for lipid peroxide. Antioxidant which is added into food must have criteria [1]:

• Efective at low concentration • No difference in colour, taste, aroma, and

nutrient of food • Easy in used • Stable at process condition and storage of the

food • None of toxicity effect in higher dosage Antioxidant is a inhibitor for lipid peroxide.

Antioxidant not only used as protection for the food from its expiration, but also to defend the cell, i.e.: lipid membrane failure due to oxidation. This is because the character of this antioxidant can avoid chain reaction

A



from the radical formation [2]. This antioxidant also can be used to as a protector from some illness which is triggered from lipid oxidation process in human body, i.e.: heart, stroke, and cancer [1]. Antioxidant effectivity is depends on several factors [2]:

• Iniciator radical • Type of antioxidant • Antioxidant concentration • Interaction with other antioxidant

Antioxidant activity testing for polar extract of

bioactive compound form Dillenia indica’s kernel conducted using Carotene bleaching method. The testing conducted with 2 condition i.e.: room temperature and incubation temperature (40, 60, and 80 oC). 10 mL of soybean oil as sample added with carotenoid and 0.2 mL polar extract of kernel. Sampel absorbance measuring with spectrophotometer UV-visibel. While soybean oil without extract compound as negative control.


Antioxidant activity testing on room temperature conducted using carotene bleaching method with absorbance measuring, for 5 days. Fig. 1 shows measurement result of absorbance. Fig. 1 shows that there is decreasing of sample absorbance caused of oxidation of soybean oil. From figure below, sample with B and D fraction at 2nd day, have a small shall compare to other. Which is meant, B and D fraction has a compound with antioxidant activity.

Perbandingan Penurunan

Absorbansi pada Temperatur Lingkungan

-0.02

0

0.02

0.04

0.06

0.08

0.1

0.12

0.14

0.16

0 2 4 6

Hari

Delta

Abs

orba

nsi

ABCDEFGKontrol

Fig. 1. Measurement result of absorbance at room temperature

Perbandingan Absorbansi pada Temperatur 40 C

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

1.4

0 2 4

Jam

Abso

rban

si

ABCDEFGKontrolEkstrak Polar

Fig. 2. Measurement result of absorbance at incubation temperature (40oC)


0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0 1 2 3 4

Jam

Abs

orba

nsi

ABCDEFGKontrol


Fig.2 and Fig. 3 above shows that better

antioxidant activity given by G fraction at incubation temperature 40oC, and B fraction at incubation temperature 60oC. When antioxidant activity testing conducted at incubation temperature 80oC, all of tested sample does not have antioxidant activity (Fig. 4).


0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0 1 2 3 4

Jam

Abso

rban

si

ABCDEFGKontrol


IV. CONCLUSIONS Result of extraction in the form of dilution which has brown color. After the solvent was evaporated, we obtained a harsh extract in the form of resin jell tan like caramel as much 128 gram. From antioxidant activity of the fraction obtained tested with Carotene bleaching method: result of isolation by column chromatography is yield 7 fraction, A until G, with retention time is very nearby. Result of test of antioxidant activity at environment temperature shows that good enough antioxidant activity compared to a negative control is Fraction B. While test of antioxidant activity conducted at incubation temperature of 40 and 60 oC shows Fraction of G and B own good enough antioxidant activity compared to a negative control.

REFERENCES [1] Gritter, Roy J., 1987, “Pengantar Kromatografi”,

Bandung: Penerbit ITB. [2] G., Scott, 1998, “Antioxidants”, Japan: Bull. Chem. Soc. [3] Robinson, Trevor, 1995, “Kandungan Organik

Tumbuhan Tinggi”, Bandung: Penerbit ITB. [4] Harina, Yuni, 2003, Tesis Magister Kimia, “Aktivitas

radical scavenger a-Mangostin dan 1,3,6-trihidroksi-8-2-(2-hidroksimetil-But-Enil)-9-Xanteron dari kulit buah Manggis”, Depok: UI.

[5] Prosea, 1995, “Plant Resources of South-East Asia”, Vol. 5, Bogor: Prosea Foundation.

[6] W, Maruti, 2005, Skripsi Sarjana Teknik, “Studi Pendahuluan Pemanfaatan Ekstrak Daging Buah Dillenia indica dengan Pelarut Non-polar sebagai Antioksidan”, Depok: Departemen TGP-FTUI.

[7] Boer, Yusnetti, 1999, “Antioksidan Kulit Buah Kandis [Garcinia parvifoloia (Miq.) Miq]”, Jakarta: Program Studi Magister Ilmu Kimia FMIPA-UI.

[8] Asri, Isri Nur, 2001, Karya Utama Sarjana Kimia, “Isolasi Dan Studi Aktifitas Antioksidan dari Rimpang Lempuyang Wangi”, Depok: Jurusan Kimia FMIPA-UI.

ISSN: 1411-1284


Understanding the Metaspaces of Street Vendors in the Cities: Temporality, Strategies and Tactics

Yandi Andri Yatmo* and Paramita Atmodiwirjo*

* Department of Architecture, University of Indonesia, Kampus UI Depok 16424

Tel. 7863512, fax. 7863514 email : [email protected] Abstract– One of the commonly emerged issues in many cities is the existence of unplanned and unexpected urban elements. The phenomenon of street vendors is a well-known example of such elements. This paper attempts to understand the phenomenon of street vendors by considering the process of its emerging and re-emerging in urban environment. The discussion of the process requires an understanding from the point of view of the street vendors as the ‘subjects’. Street vendors continuously ‘read’ the existing urban spaces; their emergence and re-emergence in urban spaces may be understood as a result of such ‘reading’ of the environment. This paper would discuss the process inherent in the process of street vendors’ ‘reading’ of the environment, based on the concept of chess as a metaspace (Bunschoten, 2003). Understanding the phenomenon of street vendors through the concept of metaspaces allows us to look beyond the spatial reality of street vendors in an actual locale in the cities. It involves the examination of temporality, as well as strategies and tactics involved within the metaspaces of street vendors. The concept of metaspaces becomes an instrument to uncover the space of knowledge of street vendors within their process of emergence and re-emergence in the cities. This may eventually help the development of planning instrument in dealing with such phenomenon in the cities. Keywords– street vendors, temporality, metaspaces, urban environment, strategies

I. TEMPORARY ELEMENTS IN THE CITIES

ost planning and design approaches emphasise on the needs to create pleasant urban

environment for its inhabitants. Urban environments should become a set of interrelated networks of people, places and events, embracing various physical and socio-cultural aspects. Nevertheless, the presence of problems and conflicts is unavoidable in urban environments. This would call the intervention of planning profession to create order of the environments.

Cities inevitably involve the existence of unplanned and unexpected elements, the majority of which become nuisance and disturbing elements. Regardless of the efforts to remove these elements from city spaces, they keep growing throughout the cities. The phenomenon of street vendors is a well-known example of such elements.

The phenomenon of temporary elements that keep

emerging and re-emerging in urban environments clearly suggest the limitation of the planning profession in determining what happens in the everyday life of the cities. Planners and architects do not have full capacity to control and prevent the possibilities of spatial conflicts and disorder. Borden et al clearly argued that “The city is not the product of planners and architects” [1]. In most cases, the acts of planners and architects merely represent a part of the broader system of power, and thus do not necessarily respond to the everyday phenomenon of urban reality. Nevertheless, architects and planners cannot neglect the existence of such unplanned and unexpected phenomenon, which is related to the instability of urban environment.

The limited ability of architects and planners in fully determining the order of the environment has also been clearly explained by Venturi in discussing the existence of “honky-tonk elements” in the cities. “The main justification for honky-tonk elements in architectural order is their very existence. They are what we have. Architects can bemoan or try to ignore them or even try to abolish them, but they will not go away. Or they will not go away for a long time, because architects do not have the power to replace them (nor do they know what to replace them with” [2].

Whilst architects and planners possess limited capability in determining the spaces of the city, there must be other players that take significant role in determining the spaces. Lefebvre suggested his theory on space as a social (re)production, and explicitly claimed the role of the society in the production of space [3]. Hence the cities should be seen as a product of the society living their everyday life in it. Furthermore, Lefebvre emphasised the use of the term

M

ISSN: 1411-1284


“subjects” that imply their active role in constructing space, instead of “users” or “inhabitants” [1].

A question may emerge on how the process of space production occurs. Especially in the context of urban environments with the phenomena of temporary elements, it becomes necessary to examine the processes that lead to the emergence of unplanned elements, things and events in the existing spaces. Furthermore, there is a need to examine the role of the “subjects” in such process of emergence, disappearance and re-emergence of temporary elements.

This paper examines the existing problems of

street vendors in Indonesian cities, especially in Jakarta. In examining the process of emerging and re-emerging of street vendors as temporary urban elements, it requires an understanding from the point of view of the street vendors as the “subjects” in the process. As “subjects”, the street vendors actively perceive opportunities in certain urban spaces and act according to that perception. The concept of chess as a metaspace and Urban Gallery suggested by Bunschoten [4] would become a framework in discussing this issue. This concept is taken as an instrument to uncover the space of knowledge of the street vendors within their process of emergence in the cities.

II. STREET VENDORS AND THE PROCESS OF THEIR EMERGENCE AND RE-EMERGENCE IN

URBAN SPACES

The practice of street vending in Indonesia has become a growing phenomenon, just like what happened in other third world cities [5,6]. Street vendors have always been positioned in a continuing debate regarding their existence in the cities, between arguments that support and reject them [5]. One of the problems of street vendors is related to their existence in urban spaces. In Indonesia, street vendors have become the target of eviction in most cities. Nevertheless, most street vendors keep emerging and re-emerging.

The prominent characteristics of street vendors are their mobility and flexibility in their everyday operation [7]. Although some street vendors may trade in permanent stalls, some others are non-permanent vendors who sell their goods in pushcarts, tents or other temporary structures, trays, boxes, or even just by laying their goods on a mat. Such variety in their modes of operation allows them to move from place to place easily, and trade in various locations at different times of the day.

Street vendors in Jakarta operate in various locations. The data indicates that the majority (56.9%) of street vendors occupy the sidewalks and the streets.

Others occupy green areas, urban parks, parking areas, shopping and market areas and other public spaces [8]. Some street vendors utilise the empty lots that can be found in the city, including the spaces under the highway flyover. Not all those locations are legally designated areas for street trading activities. Only a small number of street vendors (about 16.6%) operate their trading activities in legal areas designated to them, while the majority of street vendors (83.4%) operate in illegal location [8]. The street vendors’ activities in illegal locations have always become the primary reasons to blame them as a source of disorderliness in urban environment.

Local government in Jakarta have implemented

various policies to regulate the presence of street vendors, ranging from the registration, legalisation, designation of certain areas for vending activities and relocation of the vendors from illegal location into legal location. In many cases, eviction or ‘street cleaning’ operations were conducted against the vendors who refused to be relocated and resisted to stay in their current locations.

However, there seems to be a tendency for

inconsistent acts in dealing with street vendors. For example, in some locations, after the officers conducted street cleaning operation, they did not take proper control of the location. This can give way to the possibility for some street vendors to come back and operate there. Such inconsistent attitudes have resulted in repeated ‘street cleaning’ operations in the same places. The flexible and mobile characteristics of street vendors also make it very easy for the street vendors to dissemble their goods and leave the location to avoid the ‘street cleaning’ operations. It is also very easy for them to come back, reassemble their stuffs and start trading again only a few hours after the operation takes place.

The phenomenon of emergence and re-emergence

of street vendors in urban environment have become inseparable with everyday life of the cities. With such mobility and flexibility, street vendors becomes a kind of “ephemeral architecture” in the city as their presence “can be detached” from urban space [9]. Behind the phenomenon of emergence and re-emergence of street vendors, there may be some processes in which the street vendors perceive opportunities in certain urban spaces and act according to that perception. We will now turn into the discussion on how the street vendors ‘read’ the existing urban spaces and how their emergence and re-emergence in urban spaces may be understood as a result of their ‘reading’. We would need to first examine the concept of ‘metaspaces’ to understand the process of such ‘reading’.

ISSN: 1411-1284


III. THE METASPACES OF STREET VENDORS

Metaspace is a concept that may explain the processes that do not directly correspond to the real experience of space and at the same time relate to it. “‘Metaspace’ first was fairly vague term that allowed us to describe the space and models of games, a space that was isolated from the daily reality and yet belonged to it” [4].

Bunschoten explained how the concept of metaspace might be useful to examine the phenomenon of temporality in urban environment. “Spaces in which we can see temporality are metaspaces. … The metaspace has an innate ephemeral character. It is the site of the search of knowledge; it is the container of knowledge-management processes, but it is also a vehicle for the search of meaning in the dynamic chaos in which we live” [4]. Hence the metaspace offers a path towards understanding beyond the actual phenomena of spaces.

A clear description on how the concept of metaspace relate to the urban environment is through the analogy of chess as a metaspace. As Bunschoten described, “A chess game takes place in a metaspace: the game board, with its mix of simulated battles and real emotions” [4]. As a game, chess involve the strategies and tactics that are directed to the aim of surviving and winning the game. It involves the ‘reading’ of the situation and it would be interesting to examine how such ‘reading’, strategies and tactics might be taken as a parallel of the practice of street vendors to survive in urban environment.

In chess, strategies consist of setting and achieving long-term goals during the game, while tactics concentrate on immediate manoeuvre [10]. For example, strategies involve the evaluation of current positions, where to place different pieces and the planning of future play. Meanwhile, tactics become the means to achieve strategy. Thus these two parts of thinking – strategies and tactics – are inseparable.

In setting strategies for a chess game, there are various aspects that a player should consider thoroughly. These may include the pieces on board, pawn structure, the safety of the king, the space, and the control of squares [10]. The setting up of strategies and tactics in a chess game nevertheless involve the examination of current situation as well as reading the opportunities and possible threats. It becomes necessary to consider the actors (the pieces) involved, possible relationship between the actors, as well as the occupation and control of chess squares as the setting of the game. The examination of strategies and tactics as in chess game becomes relevant as a way to study the

phenomenon of temporality in urban environment. The practice of street vendors in finding and occupying the spaces of the city might be considered as urban tactics. Borden et al. explained, “Tactics are a more proactive response to the city; they are practices … discourses that produce objects. ... They may be attempts to solve urban problems … they may also be attempts to reconceptualize relations between the city and the self” [1]. What is important in tactics is the “intentionality”, which means that tactics are aimed towards a particular goal. In the case of street vendors, there is a goal to survive in everyday urban environment.

Many decisions made by street vendors in their relationship with the spaces of the cities could be taken as a parallel of strategies and tactics in chess game. In their practice, street vendors make decision to trade in certain places of the cities, at certain times and in certain ways. Considering the point of view of street vendors as the ‘subjects’ in such everyday practices, the decisions are based on the ‘reading’ of the situation, including the actors involved and the possibility to occupy certain settings. It also involves the examination of possible practices in time contexts – how to deal with each situation that might be different from time to time.

The examination of the setting is a crucial aspect

in a chess game. The players need to examine the status of a particular square before making any decisions to occupy it – whether the square offers opportunities to settle safely, whether the square is under threat or under control of the enemy’s pieces, and whether the square offers possibilities for further beneficial move. The practice of street vendors also involved similar evaluation. Street vendors need to examine the status of an urban spot, to what extent it is possible to trade in that location. Street vendors may consider the possibilities for attracting more customers, competition with other vendors, conflicts with formal retail, policing strategies of the government and the existence of control by certain groups of society.

Often street vendors make decision to occupy

certain spaces which are ‘ambiguous’ in their nature. Such settings do not have clear ownership and identity and thus become easier to occupy with less threat, compared to settings with clear ownership. Borden described such ambiguous setting as “zero degree architecture” [11], which often attract skateboarders to occupy. These include public spaces and streets, as well as left-over spaces. Similar with street vendors, “In parts this reflects their desire to avoid social conflict, but it is also an attempt to write anew – not to change meaning but to insert a meaning where previously there were none” [11]. Thus the acts of street vendors do not necessarily based on their perception of whether certain spaces are legal or

ISSN: 1411-1284


illegal trading locations, but more on the perception of the ownership or identity of the setting.

In some cases of street vendors in Jakarta, there

are settings which are controlled not by government officials but by certain ‘informal’ groups of people called ‘preman’ or ‘thugs’. Such practice is a representation of the control by informal structures of society but nevertheless becomes a crucial aspect to consider in understanding the practice of street vendors in the cities.

The practice of street vendors is also determined by the street vendors’ acts in ‘reading’ of the situation, including the existing threat at certain times. Threats for street vendors mostly come from the government officers in their acts to relocate or to evict the street vendors. Street vendors have been used to a ‘hide and seek’ strategy to avoid eviction. Usually they simply run away or stop their trading activities when the eviction operation is taking place, but afterwards they will come back to the location and trade as usual. This is a kind of tactics that they usually practice in case of eviction. Such tactics become possible due to the flexible and mobile characteristics of street vendors.

IV. TOWARDS AN URBAN PLANNING INSTRUMENT

It becomes clear that some aspects involved in the

process of emergence and re-emergence of street vendors in the cities might be understood by examining the street vendors’ ‘reading’ of the situation. The strategies and tactics practiced by street vendors represent the metaspaces of street vendors. They become a vehicle to uncover the knowledge of street vendors.

Following the understanding of the metaspaces of street vendors, then there is a further challenge in developing plans to deal with such phenomenon. As Bunschoten suggested, the further challenge is “to use this concept of a metaspace for the development of a planning instrument that is simultaneously a new public space for the interaction and intertwining of urban actors and their desires and interests” [4]. Bunschoten further suggested that the metaspaces would eventually lead into the concept of Urban Gallery. “When a part of a city is designated a metaspace, it becomes an Urban Gallery – a fluid form of public space that evolves in time, generating different definitions of public space and ways of participating in it” [4].

In the case of street vendors, it becomes clear that

the street vendors continuously manoeuvre within Urban Gallery. There are characteristics of instable, dynamic and ephemeral involved in the phenomenon of street vendors in the cities. Here Urban Gallery

might become “a device for the management of transient states” [4], thus as a management of such dynamic and ephemeral nature of the phenomenon.

Street vendors continuously play within

metaspaces – through their continuous ‘reading’ of the situation and through their strategies and tactics across spaces and times. To respond to the street vendors’ dynamics in metaspaces, there is a need for the development of planning instrument that is levelling into the same metaspaces. If street vendors play within metaspaces, then any efforts to deal with them should also play in the same ground of metaspaces. This suggests that the efforts to deal with street vendors in real spaces would not be successful if the government do not consider the metaspaces of street vendors.

An implementation of the planning instrument that

takes into account the dynamics of metaspaces is through the structure of surveillance that can manage the players within the metaspaces. It becomes necessary to create a condition that ‘someone is watching’, in order to redefine permissible spaces within the society. “For power over people, architecture had wielded the evil technology of the eye: spectacle and surveillance” [12]. In the case of dealing with street vendors, the government as the powerful actor needs a way to manage the movement within metaspaces, in order to control the movement of street vendors in space and time.

Understanding the phenomenon of street vendors

through the concept of metaspaces allows us to look beyond the spatial reality of street vendors in an actual locale in the cities. It involves the examination of temporality, as well as strategies and tactics involved within the metaspaces of street vendors. The concept of metaspaces becomes an instrument to uncover the space of knowledge of the street vendors within their process of emergence and re-emergence in the cities. Eventually, the concept would lead into the needs for levelling into the same ground of metaspaces in order to deal with the phenomenon. Street vendors, government and the society in general become the actors that play within the metaspaces of the phenomenon – within the Urban Gallery. “The urban gallery is in fact a system that orchestrates the conversation between actors in the urban domain” [4]. Thus a planning instrument should thoroughly consider the relationship among all actors involved in the process of emergence and re-emergence of street vendors.

REFERENCES [1] I. Borden, J. Kerr, J. Rendell, and A. Pivaro (Eds.), “The

Unknown City: Contesting Architecture and Social Space”, MIT Press, Cambridge, London, 2001.

ISSN: 1411-1284


[2] R. Venturi, “Complexity and Contradiction in Architecture”, The Museum of Modern Art, New York, 1977.

[3] H. Lefebvre, “The Production of Space”, Oxford, Blackwell, 1991.

[4] R. Bunschoten, “Stirring still: The city soul and its metaspaces”, Perspecta, 34, pp.56-63, 2003.

[5] R. Bromley, “Street vending and public policy: A global review”, International Journal of Sociology and Social Policy, vol.20 no.1/2, pp.1-29, 2000.

[6] J. C. Cross, “Street vendors, modernity and postmodernity: Conflict and compromise in the global economy”, International Journal of Sociology and Social Policy, vol.20 no.1/2, pp.30-52, 2000.

[7] I. Tinker, “Street Foods: Urban Food and Employment in Developing Country”, Oxford University Press, New York, 1997.

[8] Biro Pusat Statistik DKI Jakarta, “141.073 usaha kakilima di DKI Jakarta mampu menyerap 193.314 tenaga kerja (Hasil Sensus Usaha Kakilima 2001)”, Retrieved 8 November 2002, from the World Wide Web: http://bps.dki.go.id/P2_News/ P28_Kakilima.htm

[9] I. Maharika, “On the ephemeral: Tactility and urban politics”, Paper presented at the Indonesian Student Scientific Meeting, Manchester, 2001.

[10] Wikipedia, “Chess”, Retrieved 1 November 2007, from the World Wide Web: http://en.wikipedia.org/wiki/Chess

[11] I. Borden, “Another pavement, another beach: Skateboarding and the performative critique of architecture”, Dalam Borden, et al (Eds.), “The Unknown City: Contesting Architecture and Social Space”, pp.178-198, MIT Press, Cambridge, London, 2001.

[12] P. Tabor, “I am a videocam”, Dalam Borden, et al (Eds.), “The Unknown City: Contesting Architecture and Social Space”, pp.122-137, MIT Press, Cambridge, London, 2001.

ISSN: 1411-1284


Bringing ‘Sustainability’ into the Curriculum of Architectural Design Studio

Paramita Atmodiwirjo* and Yandi Andri Yatmo*

* Department of Architecture, University of Indonesia, Kampus UI Depok 16424

Tel. 7863512, fax. 7863514 email : [email protected] Abstract– The integration of ‘sustainability’ aspects into architectural education should be based on a strong framework. It requires a clear understanding of sustainability and related knowledge, abilities and skills that should be acquired through the education process. There is also a need for thorough development of delivery methods within the learning process. The methods should ensure the continuing practices of sustainable development as a part of lifelong learning. Without such strong framework, ‘sustainability’ would only become a jargon without clear direction on how they may be translated into learning contents and learning activities. This paper addresses the integration of the knowledge of sustainability into architecture curriculum. The discussion builds on the analysis of already established integrated studio curriculum that has been implemented at the University of Indonesia within the last few years. It includes the identification of current contents of the design studios and the possible opportunities for integrating sustainability elements into the curriculum. This paper concludes with the possible mapping of ‘sustainability’ contents and activities within the overall process of design studios throughout the curriculum. Keywords– sustainability, learning, architectural education, curriculum, design studio

I. CURRENT MAPPING OF EDUCATION FOR SUSTAINABILITY

he knowledge of sustainability in many environment-related professions has recently taken a

new dimension. It no longer becomes merely as supplementary field of knowledge, indeed it should be the basis on which everything else should be developed. It has become a key indicator to asses the appropriateness of a practice in terms of sustainable development. This re-positioning of sustainability issues also unavoidable affects built environment professional practice and education.

Such re-positioning could be identified by examining various documents outlining the criteria that must be fulfilled by every architecture courses worldwide. UIA/UNESCO Charter for Architectural Education highlights this issue in the following statements: “We, being responsible for the improvement of the education of future architects to enable them to work for a sustainable development in every cultural heritage… That the vision of the future world, cultivated in architectural schools, should include the following goals: a decent quality of life for all the inhabitants of human settlements; a technological application which respects the social, cultural and aesthetic needs of people; an ecologically balanced and sustainable development of the built environment…” [1].

The needs to integrate ‘sustainability’ into architectural education could also be traced in the validation process for architecture courses in United Kingdom. In May 2002, Royal Institute of British Architects (RIBA) together with Architects’ Registration Board (ARB) published the new Criteria for Validation document, which set that in addition to the existing syllabus, “consideration of the development of the sustainable environment” has been identified as an area to be addressed more seriously in validating schools of architecture [2].

Similarly in the United States, National Architecture Accreditation Board (NAAB) Conditions for Accreditation published in 2006 require architecture graduates to demonstrate understanding in the area of Sustainable Design. This area includes “Understanding of the principles of sustainability in making architecture and urban design decisions that conserve natural and built resources …, and in the creation of healthful buildings and communities” [3].

The statements in those documents clearly illustrate the needs for all architectural education institutions to integrate the issues of sustainability within learning processes. Nevertheless, such integration requires a clear understanding of sustainability and related knowledge, abilities and skills that should be acquired

T

ISSN: 1411-1284


through the process of education of architecture. There is also a need for thorough development of delivery methods within the learning process. The methods should also ensure the continuing practices of sustainable development as a part of lifelong learning. Otherwise, ‘sustainability’ would only become a jargon without clear direction on how they may be translated into learning contents and learning activities. Before translating the issue of ‘sustainability’ into the everyday learning processes, there is a need to clarify the possible scopes, contents and methods in learning about ‘sustainability’ in architecture.

II. LEARNING SUSTAINABILITY

‘Sustainability’ is a term that embraces various aspects. The widely accepted definition of sustainable development is “Development that meets the needs of the present without compromising the ability of future generations to meet their own needs” [4]. For the purpose of discussing sustainability in architectural education, it is important that the term is broadly defined as involving multi dimensions of environmental, social and economic [5].

Within the discussion of sustainable environment,

there seems to be a tendency to put greater weight of sustainability in terms of technology. Meanwhile, sustainability is seldom discussed as a social phenomenon [6]. This has led into education practices that tend to associate the learning of sustainability with certain areas or subjects – building technology, building physics, building services and environmental studies. Technology is definitely a significant part involved in the creation of sustainable environment, yet there are other dimensions of sustainability that need to be addressed with similar emphasises.

Sustainable, green or ecological architecture have

been defined in various ways, but at least there are some key issues involved: integrated relationship with nature; preservation and/or improvement of local ecosystems; civic action in which environmental physical and social quality is essential; benchmarks defined by experts; and the saving and conservation of energy [6]. The wide range of issues that involved within the term indicates the need for “holistic nature of sustainability thinking” [7]. As a result, architectural education should also adopt such holistic approach in developing sustainability contents in its curriculum.

Considering the need for holistic approach, then the

schools of architecture need to “promote sustainability to become part of the mainstream of architectural education” [7]. Such holistic approach is required, rather

than concentrating the teaching and learning of sustainability in certain subjects of technological or environmental studies. Such integration into the mainstream is considered as the best way to ensure the real values of education for sustainability.

The wide range of dimensions involved in

sustainability has inevitably created further complexities in the application into everyday learning process. It becomes clear that “there can be no single way to teach sustainability” [7], especially since each dimensions of sustainability may require certain learning activities and it would be impossible to equip students with the knowledge in all dimensions. A possible approach is by integrating the knowledge of sustainability into design projects [5]. In this way, students are equipped with general knowledge of sustainability and at the same time given opportunities to apply some of their knowledge into a design project.

The integration of sustainability into architectural

education does not necessarily means a dramatic change of curriculum contents and learning methods. Architectural education have been characterised by studio-based learning or project-based learning, which clearly distinguished it from other fields [8]. Studio-based learning is integrative in its nature, with the requirements to deal with certain issue by taking into account various issues of human behaviour, social, historical and physical contexts, aesthetic consideration and technology. Such integrative nature of design studio clearly offers possibilities for the integration of sustainability into learning process.

In fact, “an entirely new pedagogy need not be

developed for the learning and teaching of sustainability” [7]. The existing structure of design studio in most schools of architecture has been set up to develop the students’ abilities to deal with problems from multiple perspectives. What is particularly needed now is a careful consideration on the extent to which sustainability dimensions have or have not been introduced into current curriculum and learning methods. We will now turn into the discussion on how the integration of sustainability could be made possible in the current curriculum structure of architecture at the University of Indonesia.

III. INTEGRATION INTO DESIGN CURRICULUM:

CHALLENGES AND OPPORTUNITIES

Since 2004, the Department of Architecture at the University of Indonesia has adopted a curriculum with integrated design studios as the core of the curriculum. The integrated design studios becomes the arena for

ISSN: 1411-1284


students to acquire knowledge in various aspects related to architectural design, i.e., the knowledge of space and place, structure and construction, urban studies, environmental studies, building physics and building services. Compared to the previous curriculum in which those areas of knowledge were delivered within individual subjects, such integrated design studios facilitate students to develop understanding on how each aspect might be related to a single design project.

Throughout the curriculum, students go through five stages of design studios. Each studio has different emphasis and different levels of complexities, yet they form a continuous process of learning. Thus the knowledge acquired within one stage would form the basis for the subsequent stages.

In everyday learning activities, the emphasis of each design studio is translated into certain design projects, most of which are developed by the students based on certain trigger issues introduced by the studio staffs. Therefore, the design activities in each studio may vary in different semesters, and may even vary across individual students. Thus each student may develop different contents of knowledge from their peers. This becomes possible as long as certain performance criteria are met. The emphasis on each design studio in the current curriculum is illustrated in Table 1.

Table 1. Emphasis on each design studio Stages Emphasis Design Studio 1

The ecology of self; developing spatial ideas for self; awareness of structural logic, environment and human comfort in the actualisation of spatial ideas.

Design Studio 2

Developing spatial ideas for a primary social group; structural logic, environment and human comfort in the actualisation of spatial ideas

Design Studio 3

Advanced exploration of spatial ideas in urban contexts; technological and environmental studies to support the actualisation of spatial ideas

Design Studio 4

Advanced exploration of spatial ideas through the development of building technology, embracing aspects of environment, materials, construction and services.

Design Studio 5

All aspects of spatial design with code compliances.

The description above suggests that the current

contents of design studio are based on a holistic

approach, with the intention to integrate various bodies of knowledge related to spatial design. The existing structure of the design studios clearly allows further development to integrate more sustainable dimensions into the learning processes.

The design studios are currently structured as a

continuous learning process, with each stage put more emphasis on certain aspects. Thorough analysis of the emphasis and contents in each studio would provide the illustration on the extent to which sustainable dimensions have or have not been included. Furthermore, the description of the emphasis in each studio may guide the decision to introduce more aspects of sustainability.

As an example, Design Studio 3 and Design Studio

4 represent two stages of learning with very different emphasis. Design Studio 3 is focused on the thorough understanding of the contexts where the project would be developed. Hence the studio opens possibilities for further integration of social sustainability – including the awareness of locality, the understanding of social and cultural contexts, and anticipating the effects of project intervention to the community structure and human well-being. On the other hand, Design Studio 4 put more emphasis on the development of ideas based on the exploration of technology. This studio clearly allows more technical aspects of sustainability to be explored, including the understanding of sustainable materials, appropriate construction strategies, and the environmental studies of the projects.

Further challenge is on how the learning of sustainability become fully integrated in most design studios. In other words, each design studio should not focus on the social dimension only or technological dimension only. There is also a need to define how far the knowledge of sustainability should be delivered in each stage – from merely ‘awareness’ into the abilities to fully apply the knowledge in design projects. Some works are still currently in progress to develop the integration of further dimensions of sustainability in some design studios.

The integration of sustainability dimension into the

curriculum of design studio should also be translated into clear learning activities to acquire certain contents of knowledge. Various methods that have been previously practiced allow possibilities for learning various sustainability dimensions. The methods are primarily based on experiential learning, in which students develop knowledge by experiencing and analysing their experience. This might be done through practical works on environmental aspects related to visual, thermal and auditory experience; studying the

ISSN: 1411-1284


contexts of the projects direct site visit, observation and contacts with community; historical studies; experiments in structural logic and construction strategies; exploration of materials; computer modelling and simulation; and others.

The variety of learning activities should also

consider the availability and management of learning resources. It is important to understand the provision of equipments for practical works and reading materials that embrace a wide range of sustainability dimensions. Similarly crucial is the development of staffs’ capabilities to facilitate the learning of sustainability dimensions through exercises in the design studio.

The discussion above suggests that the integration of

‘sustainability’ into the curriculum of architecture would be possible by bringing it into the mainstream of the curriculum. This means getting it straight into the learning process of the design studios as the core of curriculum. The existing structure of integrated design studio opens many possibilities to expand the contents with more sustainability elements. Different dimensions of sustainability may become more apparent in one stage than in another. Nevertheless, as the different stages of design studios form a continuous learning process, it is possible to equip the students with a comprehensive understanding of sustainability by the time they complete their study of architecture. The learning methods that develop students’ capabilities to construct their own knowledge would also ensure the continuous process of life-long learning on sustainability far beyond higher education study.

REFERENCES [1] Union Internationale des Architectes, “UIA/UNESCO

Charter for Architectural Education”, UIA, 2005, Retrieved 1 November 2007, from the World Wide Web: http://www.uia-architectes.org/ image/PDF/CHARTES/CHART_ANG.pdf

[2] Royal Institute of British Architects, “Criteria for Validation”, RIBA, London, 2002.

[3] National Architecture Accreditation Board, “NAAB Procedures for Accreditation for Professional Degree Programs in Architecture”, NAAB, Washington DC, 2004.

[4] Virginia Environmental Endowment, “Education for Sustainable Development: An Integrated Curriculum”, Blacksburg, Virginia, 2002.

[5] B. Edwards, “The culture of sustainability in a school of architecture”, Keynote Symposium Paper, ‘Working towards a sustainable future: A Symposium on Sustainability in Architectural Education”, CEBE and RIBA, 2002.

[6] V. Canizaro and K. Tanzer, “Introduction”, Journal of Architectural Education, vol.60 no.4, pp.4-14, 2007.

[7] Centre for Education in the Built Environment, “Report of the Sustainability Special Interest Group (Architectural Education)”, CEBE, 2003.

[8] D. Hawkes, “The architect and the academy”, Architectural Research Quarterly, vol.4 no.1, pp.35-39, 2000.

ISSN: 1411-1284


LED Technology in Architectural Lighting of Building’s Façade

Siti Handjarinto and Gregorius A.G.A

Fac. of Engineering, University of Indonesia, New Campus UI Depok 16424

Tel. 7863503, fax. 7270050 email : [email protected], [email protected]

Abstract– Artificial lighting, especially exterior lighting, shows the beauty of building at night, even make it more beautiful than daytime. With artificial lighting, an architect can also create living and dramatic atmosphere on a building or cityscape at night.

On the other side, in this modern era, the using of lamps that requires a huge power to illuminate a building is considered as an energy wasting because most buildings are used in daytime only.

Light Emitting Diode (LED) is a new lighting technology that most efficient in energy saving and operational cost.

The other advantage of LED lamps in architectural lighting is ‘dynamic lighting’. With dynamic lighting, architectural lighting is no longer a static lighting, just for illuminating of building’s facade, but it can makes a building into a multimedia device.

In this paper, we discuss 3 case studies, Galleria Department Store, Uniqa Tower, and Dexia Tower; use different dynamic lighting techniques to create this multimedia device, i.e. dots, lines, and surfaces of light. Each technique has its own advantages and disadvantages, especially in image resolution and number of used lamps.

In conclusion we propose the solution of this topic covers the technical matter based on framework idea of efficient energy and maintenance artificial lighting of building’s façade and dynamic lighting trough literacy study.

I. INTRODUCTION

ight makes us feel the visual world. When the light changes, our experience also changes, this is

the main concept of architectural lighting.1 Artificial lighting of building’s facade aims to show its architectural beauty and to make it can be enjoyed by citizen at night.

The main issue of facade lighting is how to create beautiful and low energy cost façade lighting. Beside beautiful, lighting is also expected to be spectacular, dynamic, or can be a multimedia device.

1 M. David Egan, Architectural Lighting. (New York: McGraw-Hill, 2002), 2.

Now days, lamps have been more low energy cost, compact, long lifetime, and creating a wonderful architectural lighting technique. The latest artificial lighting technology is Light Emitting Diode (LED).

II. BASIC THEORY

Solid State Lighting/ Light emitting diode (LED) ight emitting diode (LED) is a semiconductor (diode) that emits incoherent monochromatic light

when it’s given direct current. LED converts electricity into light directly2 (see fig 1)

Fig.1. RGB LED (source:www.wikipedia.com)

Inside a diode, interaction between electrons and holes results a large number of photon, a basic unit of light. Thus, these photons are focused with plastic bulb and make a LED glows. (see fig 2)

Fig.2. Structure of a diode (source:www.sslc.ucsb.edu)

Performance characteristic Luminous Efficacy

The latest InGan white LED now has luminous efficacy up to 180 lm/w. This means a 5W LED bulb can replace a 60W incandescent lamp and results the same amount of lumen (light quantity). (see fig 3)

2 Vincent Laganier, ”LED Wall of Light”,International Lighting Review 2004: 72.

L

L

ISSN: 1411-1284


Fig.3. Power comparison between incandescent, fluorescent,

and LED lamps (source:www.sslc.ucsb.edu) Lifetime and maintenance

LED lamp has a long life time, i.e.: 50,000-100,000 hours, 5 times longer than fluorescent and 25 times than incandescent. Just like a fluorescent, LED’s lumen decreases along with its lifetime. LED lamp is also more durable than incandescent or discharge lamp because of its solid plastic structure. (see fig 4)

Fig.4. LED lumen depreciation graph

(source:www.sslc.ucsb.edu) Color Rendering Index

Spectral distribution graph of LED lamps is ranged in visible light wave, i.e.: 380-780 nm. (see fig 5) Thus, LED lamps have a good Color Rendering Index because can display all colors in their actual color.

LED lamp is also more efficient and safely because generates less infrared and UV than incandescent or discharge lamp. (see lamp comparison in table 1)

Fig.5. Spectral distribution of LED (source:Panasonic

lighting)

Table 1. Lamps Comparison

GLS & Halogen

Fluorescent HID LED

Dimension (inch)

3/16-18 6-96 5-44 0.2-1

Wattage 1-10,000

5-215 18-1800 0.03-5

Lumen/W 9-35 20-100 25-180 18-180 Lifetime (hr)

750-4,000

8-20,000 1,500-24,000

50,000-100,000

Start time instant rapidly 0-20 min Instant Dimming able electronic

ballast additional circuit

able

CRI 100 95 15-85 67-80

Dynamic Lighting

ynamic lighting is an architectural lighting concept and design that shows many light

compositions in a time by color changing, switching, and dimming.

Table 2. Dynamic Lighting

Category Color changing Application

Dots

Orientation Decorative Dynamic

Lines

Sign Decorative

Expressing 2D elements S

elf-

lum

ino

us

Surfaces

Uniform fill lighting

Wall washing Affect whole

room

LED spot

Accent lighting

Linear wash light

(multiple spots)

Wall washing Small room

Sur

face

Illu

min

atio

n

LED with

optical lenses

Street lighting Environment

D

ISSN: 1411-1284


III. CASE STUDY

Galleria Department Store Lighting concept

ighting concept of this building is ‘changing surface’, which façade’s color can changes

dramatically. This concept has been designed by Rogier van der Heide and Simone Collon from Arup Lighting, and its architect, Ben van Berkel and Caroline Bos from UN Studio. To create this ‘changing surface’ concept, all surfaces of building’s façade have to be renovated and redesigned to become more simple and compact. (see fig 6)

Fig.6. Galleria before (left) and after LED (right)

(source:www.lightingacademy.org) To illuminate 3278 m2 of facade, 4330 Xilver

RainDrops LED lamps (see fig 7) have been installed with aluminum frame on its all surfaces. These lamps which cover all surfaces of building’s façade make Galleria just like a nest at day time and a box of light at night time.

Fig.7. Xilver RainDrops

(source:www.lightingacademy.org)

Dynamic lighting ynamic lighting technique which is applied here is ’dots of light’. With this technique, each of

lamps can be seen as an independent light source and can be independently controlled from its color changing, switching, and dimming by DMX-512 control system.

Fig.8. Galleria at night

(source:www.lightingacademy.org)

The main advantage of this technique is its flexibility in performing light composition to create a high resolution image and writing to inform the product or special event or promotion being held there. ‘Dots of light’ makes a façade surface just like a television monitor with each lamps acts as the pixels. (see fig 8)

Nevertheless, this ‘dots’ technique still has a disadvantage. ‘Dots of light’ requires a lot of lamps to cover all surface of a façade. Therefore, it also requires a lot of expenses.

Persuasive and emotional trigger

In a competitive world of retail, efficient lighting is very important to perform the product in attracting the customer. A bright and dynamic ambience can attract, inspires, and affects the customer to come.

Creating the right trendy and stylish ambience

Color trend is affected by fashion trends and styles. Galleria Department Store as fashion retail from high braded fashion must follows this trend and style. Uniqa Tower Lighting concept

açade lighting for this building was designed by lighting designer LichtKunstLicht as an integral

part of its architecture. The lighting concept is a high quality and unique lighting and reflecting image of Uniqa Insurance to the Vienna nightscape. (see fig 9)

Fig.9. Uniqa Tower at night (source:www.uniqagrup.com)

L

D

F

ISSN: 1411-1284


Façade Uniqa Tower with 7000m2 area consists of 2 layers with 50 cm space between them (hollow glazed façade). Barco’s MIPIX-20 LED lamp with 65 IP was installed on 2900 aluminum frame modules in this hollow glazed space with 8 cm space between each lamp. (fig 10)

Fig.10. Barco’s MIPIX-20 (left) and its installation in

hollow glazed space (right) (source:www.barco.com)

Dynamic lighting ynamic lighting technique which is applied here is ’lines of light’. (see fig 11) LED lamps were

arranged according to aluminum frame modules so we can see them as lines of light at night time.

Fig.11. Uniqa Tower at night: ‘twists and turns’

sequence (source: Professional Lighting Design 51) The main use of this technique is expressing 2D

elements in a 3D element. Therefore, ‘lines of light’ is suitable to illuminate architectural details of a building. ‘Lines of light’ also uses fewer lamps than ‘dots of light’ because it doesn’t cover all surfaces of façade.

The disadvantage of line technique is resolution of images resulted is not as good as dots technique because it depends on its frames. Creating the right ambience

Up until now, there are 2 programmed lighting sequences, i.e. the first sequence that has been used since Christmas 2005 and ‘Twists and Turns’ sequence that has been used since April 2006. Lighting sequences can be easily adjusted and changed according to the special event like Christmas or New Year. As a Vienna’s landmark, lighting adjustment becomes very important because besides an office, Uniqa tower as also an integral part of a greater city’s public space, i.e. Danube Canal. Multimedia device

The integral part of lighting concept of Uniqa Tower is displaying Uniqa Insurance’s logo.(fig 12) This company’s logo is displayed with blue light and turns around the building 2-3 times in 20 minutes. Therefore, this lighting concept is one way for the company to promote itself.

Fig.12. Display of Uniqa Insurance’s logo (source:

Professional Lighting Design 51) Dexia Tower Lighting concept

he lighting concept doesn’t mean to present the fix programmed lighting sequence but more to

express architectural characteristic and urban context. Building’s characteristic, orientation, volume, and scale were used to be parameters in creating spatial concept, temporal, and interactive lighting. (see fig 13)

Fig.13. Dexia Tower (source:www.spacecannon.it)

D

T

ISSN: 1411-1284


LED lamps used to light up the tower are Ath

Luxor Custom (see fig 14) with this following specification: 1. 4.200 Ath Luxor was used to light up all windows

in 36 floors. Each lamps can be independently controlled. Lamps were installed inside the building on the bottom side of windows.

2. 22 Ath Luxor Custom with 40 pieces of 3W monochromatic blue LED in each fixtures were used to light up all columns of tower.

3. 60 Ath Luxor Custom were used to light up the perimeter and main entrance area.

Fig.14. Space Cannon’s Ath Luxor Cutom LED lamp

(source:www.spacecannon.it)

Dynamic lighting ynamic lighting technique which is applied here is ’surfaces of light’. (see fig 15) LED lamps

installed on each window create these windows glow and can be seen as surfaces of light that cover all façade of building.

Fig.15. Dexia Tower: blue, red, and yellow

(source:www.dexia-tower.com) The main advantage of surface technique is its

effectiveness in lighting up a surface (wash light). Therefore it can cuts initial cost of lighting installation.

From all techniques of dynamic lighting, ‘surfaces of light’ generates the lowest resolution images because its color changes uniformly on 1 surface.

Creating an interactive urban space Laboratory for Architecture and Urbanism,

LAb[au], introduced a new interactive lighting concept with ‘Touch’ system. In front of tower, Place Rogier, there is controller station where people can interact with tower’s lighting and appearance either individually or collectively trough multi touch screen. (see fig 16)

Static inputs (touch, hand, finger, arm, and duration) define color coordinate (hue) while direction of dynamic input (hands movement: horizontal, vertical, diagonal, positive, and negative) define color saturation (chrome). In this station, there is a monitor that sows its preview from given inputs on touch screen.

Fig.16. ‘Touch’ system, people can interact with tower

individually (left) or collectively (right) (source:www.dexia-tower.com)

IV. CONCLUSIONS

ight Emitting Diode (LED) is the latest artificial lighting technology. LED lamp is categorized as

anew lamp family, Solid State Lighting, because unlike incandescent that burns tungsten filament or discharge lamp that ionizes gas inside bulb, LED converts electricity into light directly.

To create a building into a multimedia device, there must be 2 following conditions:

1. LED is installed facing outside of building. All of 3 techniques can be used but in deciding which one and LED’s armature to be used, we need to consider the building’s size or scale to its environment or city, effectiveness of initial cost, and the impression we want to show.

2. Lighting must be given on all of a large façade surface. The façade’s surface must be uniform and flat, there are no uppermost architectural details or accent. Therefore, building’s façade can acts as a gigantic monitor television. In Galleria Department Store case study, all of old façade must be renovated and redesign to realize this concept.

Although LED’s initial cost is very expensive, its operational cost is the cheapest among any other lamps. Therefore, all of total cost in a long term is

D

L

ISSN: 1411-1284


also cheapest compared than incandescent or discharge lamps.

By using LED as a lighting technology in architecture design, it is believe that in the future there will be a tremendous architectural design variety.

REFERENCES 1) Gegana, Gregorius A. LED Technology in Architectural

Lighting of Building’s Facade. 2007. 2) Egan, M. David. Architectural Lighting. New York:

McGraw-Hill, 2002. 3) Lechner, Norbert. Heating Cooling Lighting: Design

Method for Architects. Canada: John Wiley & Sons, Inc., 1991.

4) Gary Steffy lighting Design Inc. Time Saver Standard for Architectural Lighting. United States of America: McGraw-Hill., 2000.

5) Laganier, Vincent. ”LED Wall of Light” International Lighting Review 2004: 72.

6) Cornilessen, Sjef. “Lighting Design for Healthcare” International Lighting Review 2005: 86 – 88.

7) de Kruiff, Marike, dan Martin Knoop. “Colour in Indoor Environment” International Lighting Review 2006: 100 – 103.

8) Ruffles, Paul. “Out and About” Light and Lighting, August – September 2000.

9) DenBaars, Steve. “Current Status and Future Potential of Solid State Lighting.” Style Sheet. http://www.ssldc.ucsb.edu

10) Beesbe, David A. “Light Emitting Diode (LED) Lighting Systems.” Style Sheet. http://www.ssldc.ucsb.edu

ISSN: 1411-1284


Form in The New Architecture Perspective

Yandi Andri Yatmo* and Kristanti Dewi Paramita**

* Department of Architecture, University of Indonesia, Kampus UI Depok 16424 Tel. 7863512, fax. 7863514 email : [email protected]

** Fac. of Engineering, University of Indonesia, New Campus UI Depok 16424 Tel. 7863503, fax. 7270050 email : [email protected]

BACKGROUND In daily life, human has always connected with buildings. This connection enhanced by the fact that building is a place for human activities. A building must accommodate the functional need of the user (as known as human). On the other hand, in order to develop this sense of connection, it involves the emotional side of the user. This emotional side proves that the building have its essence or what is said as form. Building’s form is a design result with high level of considerations and complexity involve on it. So many elements acquire in one kind of design. As Fritz Schumacher said, in designing there is three phases to pass. The phases consist of initial phase, middle phase, and the goal1. When planning a building, the first phase is initial phase. Middle phase is the second phase, when the architect figures the building characteristics and do some further thinking about how it would work in the future. What is simply interesting is the last phase, known as the goal. Because it has not stated as a physical goal, but rather as some principles, which was express as an informal type or general type2.This type clearly affirm the idea about the importance of a concept while designing, whether explicitly or implicitly. This concept usually evolves from the architect expression about the building environment or the imagination about the building characteristic3. Knowledge about design location is highly needed to arrange a contextual-centered analysis. As the time goes by, the community lifestyle is constantly changing. These changes affect the community comprehension about form in architecture. This is unavoidable, because human is architecture’s center4. The New Architecture is also one of the effects of those changes, which appear as the improvement of architecture’s philosophy relate to the developing of urban community. The development of form in this philosophy is unique, placed in the language that buildings provide to its user.

TOWARDS FORM IN THE NEW ARCHITECTURE Defining Form Plato said that the real meaning of form is “a quality in reality5”. Form is the idea or the essence of any object in this world. A form is stable and never changes, and the basis of the real image of the object. It is different with shape. Shape is something that can be sees by us, such as content, composition, color or texture6. It is commonly known that form and shape are usually united in one and develop the whole expression of any object exist. In architectural definition, form and shape exist in every building, just as if they exist in every living being. Soriano explain furthermore that in the tradition of humanist aesthetics, form is a bundle of what has done or contain7. Therefore, in other words that form is a package of meaning from an architecture object. The characteristics are unique, unusual, and last forever. Form could be understood permanently because everyone understands it without bordered by social class or such categories. It is not universal, but singular, which means something very extraordinary so that everyone could recognize it. Architect has the right to define the building’s form by making a subjective or personal decision. A subjective decision is usually made by using the theory or method that is significant for the design condition. On the other hand, a personal decision is made by using the designer’s experiences or memories. Charles Jencks also said about this decision process as “playing God”8, because of its random feature. Form, Community Condition and Evolution The history of the humankind elucidates the relation between building’s form and community condition. As an example, modern era appear as one of the influence of the First World War. It is said that after the war, modernism—that is before is a minority taste—highly mark out the age9. Technically, the Modern building is more economically rational to build in after war period because it has less decorative elements that give minimize the building cost. On the other hand, the

ISSN: 1411-1284


shape stability of the Modern building is a representative of the community requirement in after war condition that still in chaos. It expresses the need of a strong and powerful image. James Smith supports this idea by his statement that “…Modern design is seen through the lens of masculinity…”10 It has clearly stated that the essence of the building has connected with the community recent condition. It has said in the above sentences that community is always changing. Technology has an important part on this. The development of technology improves the human ability and accomplishment to communicate, and later influence to the development of the human’s living space. This development of human’s living space contributes to the conception of a new culture in a human life period. Chris Abel named this new culture as eco culture11. One of the characteristics of eco culture is its form that arranged for the environment, has a significant purpose and allocated for the climate condition12. The human’s living space improvement makes some influence on the flexibility of buildings development. We could build anything anywhere. It is decentralized, not centralized as the previous culture did. This decentralized pattern also affects the community condition to become plural, because there is no certain domination or power like the centralized pattern. This plural community is not limited by certain codes or references in the beginning. With plurality, the feature of the form expression would be more open and free. As Gausa said, plural could be defined as an opening for possibilities. “Plural is, by definition, open: it is not—or does not seek to be—univocal, absolute, complete...”13 Later, because there is no certain domination, it allows a building to expose its own form purely. This is the beginning of a new perspective in architecture that comes from the evolution of the community. The Thoughts of New Architecture The term of New Architecture is coined by Jeffrey Kipnis. The appearance of this term was influenced by the MOMA Decon exhibition held in 1988. Along with that, the works of several architects such as Peter Eisenman, Frank Gehry, Daniel Libeskind, Bernard Tschumi, Rem Koolhaas and Zaha Hadid also have impact for the term. Their contribution in architecture has considered

exotic and then categorized as New Architecture14. The final aim of the New Architecture is to create some real significance of Architecture in this contemporary world fully filled with commercialism. Commercialism drives the market, create barriers appear in codes or references. If the architecture world only follows these commercialism codes, it would end as the death of architecture meaning. This is because the building only pursues the profit, not the soul. Morris said “…architecture is the quintessential art, an expression of pleasure in work, an art made by people for the people as a joy for the maker and the user…” We could deduce that the profit-oriented building does not presents any art, any pleasure or even joy, because the building does not have any soul. They do not seek for their form. This architecture without meaning would not produce any bonding with the user, because they do not react emotionally. As been said before, technology has a great impact. In the New Architecture, this improvement really has an influence, especially on the communication technology and digital technology. Charles Jencks said that there is a paradigm to use these new sciences to create a new dynamic architectural expression15. The existence of the digital revolution made us think differently. Mandour support this idea by saying that digital revolution affecting the way we think about architecture. It allows the expression on a grand scale, widen the architects’ range and change the communication of architecture language16. Unger have a theory on the New Architecture product characteristics. He said that “…such an architecture must be blank, it must be point and be incongruous and incoherent…”17 Blank is because it is not connected with any classic codes. The term incongruous and incoherent, lead us to the basic need of seeking for form in New Architecture, which is the need of information. It is inescapable that information is highly required to develop the form for a building or any architecture product. This is because so many elements attached each other in one design condition, so information as many as could be obtained would support the form defining process effectively. INFORMATION AND FORM IN THE NEW ARCHITECTURE Sensing the Information in the New Architecture Gausa said that information on the New Architecture is synthetical, condensed and it consist in layers18. The layers are the key to know about how human could sense the information. As said before, technology affects the living space improvement that rise the human need about mobility. When a human moves, even though his movement coordinate is limited

ISSN: 1411-1284


because of the gravity, he could sense his surroundings widely. With this sense, human could collect information from his surrounding without struggling too much. This information also often determines the way he moves, based by his purpose or his need for the activity. De Carteau said that there is some network appears from the movement and it creates some compact stories that related one to another19. We could reckon that the movement influences the network, and it would later create certain space dimension from those kinds of circulations. The difference between the New Architecture and the previous time is primly in its movement. In the previous time, they delivered the information in united form and most of all, only seek for efficiency. They use this kind of offer because the information enlightening design in the building is not with movement but by contemplation. The contemplation way is no longer suitable, because this is an era which everything must be in speed or appear in instant. The characteristics of the New Architecture are in its kinetic forces rather than staticity20. The Influence of Information in Spaces Nowadays, information is one of the human’s basic needs. When you communicate with others and exchange information, you do not have to see them in a physical meeting like before. You could do it while doing other activities, like walking, eating or even while you are buying your groceries. This is something evolutionary if we compare it with the habits before, when speak or walk while eating could be consider not polite or taboo. These changes on communication behavior would automatically change the space meaning from static or fixed become dynamic and changeable. This meaning comes because you could use the space while continually moving or do everything you could. Soriano designated this kind of space with the term “collective space”21. He described the space as a space where the connection between the property and its own function is no longer clear22. It could be blur or even disappeared. What are constant about this space are only the individuals inside of it. The space inside is built with the imagination of the individuals. The imagination comes from the information each person collect, and everyone would react differently based on their needs or their personalities.

We could see the example in a building that has a static function like shopping mall or train station. The main user of the building is the passengers or the shoppers. However, in reality, it is not always like that. You could use the train hall or the shopping corridors for meet people, socialize and have a chat, or even playing skateboard with your mates. This is a rough example of how the building function could change and still accommodate the need of the external users. This phenomenon would create a paradox because of the cycle appears in the space—from fixed space to the dynamic one, all over again. The paradox appears because there is something contradictive happens in the space of the New Architecture. Human determine the architecture, but on the other hand, architecture determine human moves and imaginations. Tschumi described this phenomenon as “violence in architecture”23. The term violence could be analyze as the intervention between human body and space one and another. Information Kinds, the Process, and the Final The information’s main purposes are to educate, entertain, or to declare something24. The information collected for design process is related with the human life pattern. It is something about consuming, producing, or just living as a human25. It could not be served as two-dimensional form, but must contain space and time elements. The information usually presented in diagram, as compressions of information26. This information then develops some boundaries. These boundaries come from the information that establishes a dynamic plan and certain message of the building. The message builds building essence or form. The boundaries come from a set of imaginative movement from the plan that placed by the message. The message is so powerful it determined the absence of shape27. The urge to vanish the statement of shape comes from the act of avoiding any certain composition. Composition is considered as something subjective and must be avoided. The expectation is the form could create a space purely based on the message as the information we produce. Surface is a physical product of information, and the outline of the space. The definition of surface is the outer part of something28. The surfaces of space are something straightly in touch with human in their daily activities. In the New Architecture, they produce the information based on the message to influence the surfaces. Later, the human will feel that there are some certain messages from the space while they experiencing the surfaces. Surface composition could be tangible or intangible, up to the message. Both tangible and intangible surface has the same purpose to give the information to the users about the message. Then, the surfaces would establish the final form from

ISSN: 1411-1284


information process, known as the visualized form. Charles Jencks divided these form expression in to three parts. It is fractal architecture, biomorphic building, and iconic architecture29. Fractal architecture is some expression that used a repetition method of a disorganized geometric pattern30. Daniel Libeskind said that the basic idea of this form of expression is to propose disorientation consist of several components with its own meaning, just like what he did in his work Jewish Museum. Biomorphic architecture or often called as a form of organi-tech is concentrated in ecologically sustainable architecture. In the attempt to solve the urban condition with high level of pollution, the form expression that is sustainable for the environment is necessary. Besides, to prevent the face of the city from rigorous and cold image, this expression uses the nature as basis. In the end, the form appears as an organic form. The organic form is usually consists from one cell that later would “grow” based on the space-related pattern and the information contained in the form. The form could also unify with the surroundings and later would establish a landscape. The last one is iconic architecture. Its main character is super-impressive. This kind of form has several purposes that have social or public qualities. These qualities appear because of the building need to be an icon or a landmark. The iconic architecture represents the human needs to be an icon or to be different from others. It is the part from the need to be appreciated by others based to Maslow theory31. In current community with high level of plurality, the need to be ‘higher’ or monumental could no longer be applied by ornaments that show the social class. To create a monumentality expression the solution is

designing some singular and extravagant expression. In the New Architecture, this form should remain eternal and have a high level of locality concept, as a bonding form with its environment. Even though there is differences in the designer’s media, but there is some process they surpass through in the production of form. They treat the information that collected before and later produce the main expression to verify the final form for the design. CONCLUSION The thoughts of the New Architecture appear as a reaction of the degradation of architecture meaning in some era. It is an era when architecture is only market consumption and no longer has any soul in it. The New Architecture is an evolution emerges together with the development of the technology that changes human life. Human is the basic part of New Architecture. As human improve, architecture is widen. The New Architecture introduces the thoughts about building form as a distinctive issue rather than concentrated only in composition. Form in architecture is considered as an answer for human problem in reality that more complex nowadays than the time before. The final form would appear with dynamic character on it, because it is produced in the context of space and time. Information is very important in the New Architecture that divides into two sections, the collected information and the produced information. The collected information is everything about site, the user and the surrounding. The produced information is the main essence of form in the New Architecture. This information set the message of the building. The information production could be seen in the surface manipulations. These manipulations would give different information to the users and make the users to develop some perceptions. It is these perceptions that define the true meaning of form in the building. Users then could interact with the building. With form as the essence of architecture, the possibilities of exploration in a very diverse field are widely open.

REFERENCES 1 Egon Schirmbeck. Idea, form, and Architecture: Design Principles in Contemporary Architecture. Van Nostrand Reinhold Company. New York. 1987. Page 9. 2 Ibid. 3 What is meant as characteristic is the impression that the society commonly aware of. The example that usually universally accepted is that the office building must have a professional image, or a kindergarten must have a cheerful feeling. These characteristics require will have impacts on the production of the building essence. 4 Peter Noever. Architecture in Transition, Between Deconstruction and New Modernism. Prestel . Munich. 1989. Page 116. 5 http://en.wikipedia.org/wiki/Forms_in_architecture, last modification 4th January 2007.

ISSN: 1411-1284


6 Thesaurus:All reference books. 7 Manuel Gausa, et al. The Metapolis Dictionary of Advanced Architecture. City, Technology and Society in the Information Age. Actar Publisher. Barcelona. 2003. Page.236. 8 Charles Jencks dan Karl Kropf. Theories and Manifestoes of Contemporary Architecture. John Willey &Sons. Great Britain. 1997.Page.6. 9 http://en.wikipedia.org/wiki/Modernism, last modification 24th April 2007 10 ocw.mit.edu/NR/rdonlyres/Architecture/4-645Fall-2004/7BD9E846-BCA8-4235-9D09-949505AAE10B/0/responses_5_2.pdf. 11 Chris Abel. Architecture, Technology, and Process. Architectural Press. Oxford. 2004. Page18. 12 Ibid. 13 Manuel Gausa, et al. The Metapolis Dictionary of Advanced Architecture. City, Technology and Society in the Information Age. Actar Publisher. Barcelona. 2003. Page 486. 14 Jeffrey Kipnis. Towards New Architecture. Academy Edition. London. 1993. hal.41. 15 Proceedings of London Conference, "New Science, New Urbanism, New Architecture" http://www.katarxis3.com/Page3 16 Mohamed Alaa Mandour. ARCHITECTURE FORM IN THE 21ST CENTURY. Architecture Department, Helwan University, Egypt. Page 195. 17 Jeffrey Kipnis. Towards New Architecture. Academy Edition. London. 1993. Page 42. 18 Manuel Gausa, et al. The Metapolis Dictionary of Advanced Architecture. City, Technology and Society in the Information Age. Actar Publisher. Barcelona. 2003.Page 21 19 Michael De Certeau. The Practices of Everyday Life. University of California Press. California. 2002. Page 93. 20 Manuel Gausa, et al. The Metapolis Dictionary of Advanced Architecture. City, Technology and Society in the Information Age. Actar Publisher. Barcelona. 2003.Page 345. 21 Manuel Gausa, et al. The Metapolis Dictionary of Advanced Architecture. City, Technology and Society in the Information Age.Page 561. 22 Ibid. 23 Martin McQuillan. Deconstruction, A Reader. Edinburgh, University Press. Edinburgh. 1988. Page 230. 24 Manuel Gausa, et al. The Metapolis Dictionary of Advanced Architecture. City, Technology and Society in the Information Age. Actar Publisher. Barcelona. 2003. Page 343. 25 Manuel Gausa, et al.Op Cit.Page 162. 26 Ibid. 27 Ibid. 28 Thesaurus U.S, Microsoft Office 2003. 29 Proceedings of London Conference, "New Science, New Urbanism, New Architecture" http://www.katarxis3.com/Page56. 30 Thesaurus, all reference books. 31 In Architecture for People, New York, 1980, Byron Mikellides in Appendix part, there’s a theory about human need called Maslow theory, which consist of:

1. physiological needs 2. safety/security needs 3. social needs 4. self esteem/ego needs 5. actualization needs

Iconic architecture represent the fourth need, which is the need to be appreciated or considered extraordinary by their surroundings.

ISSN: 1411-1284


Utilization Effectiveness of Jakarta Municipal Asset (Case Study: East Jakarta Municipal Youth Center)

Azrar Hadi *) and Ii Karunia **)

*) Lecturer in Department of Architecture and Urban Studies University of Indonesia.. Department of Architecture, Faculty of Engineering, University of Indonesia Telp. +62-21-7863512, Fax +62-21-7863514, email : [email protected]

**) Urban Asset Management student and staff of Jakarta Municipal..

Jl Merdeka Selatan 8-9 Blok G Lt-13 Jakarta 10110 Telp. 3823255, Fax 3848850, email : [email protected]

Abstract:

This study is worked through about utilization effectiveness of East Jakarta Municipal Youth Center (EJMYC) as urban facility that developed to accommodate young generation’s activities (sport, art and culture). The purposes of this study are to know: 1) community perception of location, facility, accesibility, and neighborhood’s of EJMYC; 2) their visit motivation. This study is more quantitative research. Operational model of this study to integrate asset management theory with an eclectic models of theoritical perspectives Paul A. Bell et al. The data collected by quiestionnaire and deep interview. Two hunred and eighty four East Jakarta Municipal community that classified into student, sport teacher/trainer and public society as respondent.

Based on the data analysis and in depth interview, the result of the research as follows: 1) condition of location, facility, accesibility, and neighborhood’s of EJMYC had not yet appropriate for urban community needs, but as urban facility the EJMYC was still need; 3) utilization of EJMYC have still not optimum, ascertainable from visitor and occupancy that lower and unfitted with functionality of facilities; 4) under utilization of EJMYC as the impact of their perception and motivation. In this case the role of their perception and motivation as explainable: (a) minimum of visitor indicated that EJMYC had not yet appropriate for urban community needs; (b) minimum of visitor indicated that community motivation is weak; (c) weak motivation of community drive by bad perception about facilities condition of EJMYC and demotivated of sport teacher/trainer; 5) utilization effectiveness of EJMYC positively influenced by perception and motivation of community.

Keywords : Asset utilization, Youth Center.

1. INTRODUCTION

ast Jakarta Municipal Youth Center (EJMYC) is a public facility that developed by Jakarta Municipal to accommodate young generation’s activities (sport, art, and culture).

Its has been operating since 1971. Hasurungan Pakpahan research proved that the young generation’s infrequent utilised EJMYC (Pakpahan, 1984). According to EJMYC management report, in year 2005, incidental activities at the sprort center (30,41 %) and auditorium (31,51 %).

Mark W. Patterson said, asset management is the science or art of thedirecting the management of real estate investment to ensure that its value is maximized and enhanced over the long term for benefit of the investor (Patterson,1995:3-4). In the mean time, Rambat Lupiyoadi and A. Hamdani said, there are tree’s quality orientation that must be consistent mutually : (1) consumer perception, (2) products or services), and (3) process. Contribution from the tree’s quality orientation can be contribute to success of company or organization from customer satisfaction, employee satisfaction, dan profitability of company or organization (Lupiyoadi dan Hamdani, 2006).

Management of East Jakarta Municipality Youth Center (M-EJMYC) as organizer as useful as possible must be optimalizing existence of EJMYC. Success and achievment of M-EJMYC with optimalization of EJMYC as one of centre for community center expected can improve acceptance of retribution and invite all organizer of advertisement to install advertisement so that able to give addition of acceptance for Government of DKI Jakarta from sector of Iease of Advertisement for installation of advertisement around the city fasilitas in the range of EJYMC.

Existence of the ETMYC with all potency represent matter which require to be managed carefully and

E

ISSN: 1411-1284


socialized in an optimal manner so that able to give positive image and added value for society and all investor and also other party owning enthusiasm to do cooperation with asset exploiting. This research will to know: 1) community perception about EJMYC; 2) community motivation of visit and exploited EJMYC facility.

This Fact expected can revealing existence and potency of EJMYC as one of city facility and source of income for Jakarta Municipal. Issues in this research : 1) whether location condition, facility condition, aksesibilitas condition and neigborhod’s condition of EJMYC able to support existence of EJMYC as one of city facility that matching with the community requirement ?, and 2) whether perception about location condition, facility condition, aksesibilitas condition and neigborhod’s condition of EJMYC can influence to community motivationt to visit and exploit facility of EJMYC ?

2. BASIC THEORY

There are four factors effecting product positioning effectiveness in property marketing : 1) products characteristics (property proucts/services are usually high-valued, principally stemming from high production costs. The risk in selling these products/services are also high), 2) customer’s characteristics (proucts/services are not normally bought frequently. So, they have more sringent criteria in their purchase decisions. The most

important is their affordability. Second, the facilities to effect the demand. Third, the product characteristics), 3) competitive characteristics (channel design is influenced by competitor’s channels. Some developers want to compete in the same manner as their competitors), and 4) company characteristics (the company size determines the size of its markets and its ability to design marketing channel. Not least in importance is sales force qualities. The knowledge, experience,dedication, creativity, and team work are among the important factors determing company’s ability to withstand market challenge) (Iman, 2002:256-261).

Jeremi Dasso et all said, nany elements are considered in site selection and analysis property : 1) location or situs, 2) accessibility, 3) size and shape, 4) physical characteristics, 5) utilities and services, 6) applicable public regulations, and 7) cost or value (Dasso, 1992:222). In the mean in time, the location focus for service firms should be on the determining the volume of business and revenue. There are eight major components of volume of business and revenue fot the srvice firms : 1) purchasing power of the customer drawing area, 2) service and image compatibility with demographic of the customer drawing area, 3) competition in he area, qulaity of the competition, uniqueness of the firm’s andcompetitor’s location, 6) physical qualities of facilities and neighboring business, 7) operating policies of the firm, and 8) qulaity of management (Rendera nd Haizer, 1996:213-214).

Perception is the interpretation of received stimulus.

In other word, perception represent ways in which one senses, interprets and analyses incoming stimulus. Stimulus is something that increase bodily or mental activity. Perception has been used in developing marketing communication strategies such as the perceptual mapping, wihich is founded on the premise that individuals will seek to relate to things in term of relationship with or similarity to things with which they

are already familiar (Iman, 2002:131-132). Paul A. Bell et all said, perception of the objective physical conditions depends on the objective conditions themselves, as well as on the individual different factors and the attitudinal. If this subjective perception determines that the environtment is within an optimal range of stimulation or is congruent with intended behaviour, the rsults is homestatic, the adjectif form of homestasis, or an equalization of desired and actual input. On the other hand, if the environment is experienced as outside the

Figure -1. An eclectic model of theoritical perspectives

OBJECTIVE PHYSICAL

CONDITIONS

INDIVIDUAL DIFFERENCES

SITUATIONLA

FACTORS SOCIAL

CONDITIONS

CULTURAL FACTORS

PERCEPTION OF THE

ENVIRONMEN

AS WITHIN OPTIMAL RANGE OF STIMULATION

AS OUTSIDE OPTIMAL RANGE OF STIMULATION

HOMEOSTASIS

AROUSAL STRESS

OVERLOAD REACTANCE

COPING

ADAPTATION and/or

ADJUSTEMNT

CONTINOUED AROUSAL

and/or STRESS

POSIBBLE AFTEREFFE

CTS and/or

CUMULATIV

IF COPING UNSUCCESSFUL

IF COPING SUCCESSFUL

POSIBBLE AFTEREFFE

CT and/or

CUMULATIV

ISSN: 1411-1284


optimal range of stimulation, then one or more of the following psychological states results : arousal, stress, information overload, or reactance. The presence of one or more these states leads to coping strategies. If the attemped coping strategies are successful, adaptation or adjustmenoccurs, possibly followed by such aftereffects. Should be the coping strategies not be successful , however, arousal and stress will continue, possibly heightened by the individual’s awareness that straegies are falling. Potential aftereffects of such inability to cope include exhaustion,learned helpness, severe performance decrements, and mental disorder. Finallly, as indicated by the feedback lops, experiences with the environment influence perception of he environment for the future encounters and also contribute to individual differences for future experiences. Theoritical concepts an eclectic scheme of environment-behaviour relatishionship as presented in flowchart in figure 1. (Paul, 2001:131-133).

3. EXPERIMENTAL RESULTS EJMYC provided with facility : swimming pool,

sport’s center, auditorium, climb wall, gymnastic room, courses room’s (computer, music, and english Ianguage), cultural art gallery, mushola and park really not yet been exploited with optimum by community. Perceivable above mentioned condition remember from 284 responder bringing back kuisioner really do not all facility which is there are in EJMYC known and exploited by society and also most responder tell seldom pay a visit to EJMYC (81,3%). Some of th responden said that many responden which still own motivation pay a visit to EJMYC (55,7%).

Facility which is at most known by the responder is swimming pool ( 93%) while facility which less a lot of known by space of computer courses (20%). Meanwhile, seen from experience of responder really which is at most exploited by responder is swimming pool (72%) while facility which less a lot of exploited by responder is room of Ianguage courses (8%). Facility which is made available EJMYC to date still be needed, Tthe mentioned seen from > 70% responder express all facility in EJMYC still be needed except room courses (< 70%). The result of survey about condition indicate that condition EJMYC have disagree with requirement socialize by the reason: 1) its location uncovenience (

82,7); 2) its facility unmaintenance ( 75,1); 3) accessibilty is not good ( 83,1%); and 4) neighborhood’s less support ( 81,7%). Meanwhile, result of survey about role EJMYC really existence EJMYC still playing a part of requirement accomplishment able to go into society (60,6%), training the sports (63,4%) and the achievement improvement management in have sport (63,1%). Understandable the mentioned remember its visit to EJMYC get support from family ( 57,4%), friend’s ( 59,6%), and teacher/trainer (64,8%).

Regression output of Perception variable shown that only the facility condition which by siginifikan positive influenced to somebody perception about EJMYC. But then, location condition, accesibility, old age and the economic condition positive influenced to community perception about EJMYC but don’t significant, while environmental condition about, occupation and level of education negative influenced to community perception about EJMYC but don’t significant. In the other proses, regression output of Motivation variable shown that affiliation, family driven, friend’s driven, and teacher/trainer driven positive influenced to community motivation for visit to EJMYC. while skilled requirement and the requirement have achievement [positive influence to community motivation for visit to EJMYC but do not significant.

Unfavourable community perception about physical condition of EJMYC and lack of motivation pay a visit to EJMYC give unfavourable continuation effect to exploiting EJMYC. Anharmonic among perception, motivate and mount of visit (PB) to EJMYC as implication from anharmonic of the fact and expectation of society to benefit boundary at the price of or the expense paid its as presented in flowchart in figure-2 So, that ready and M-EJMYC not yet able to realize public service quality which the efficiency and effective. Correlation of perception and motivation with number of visited to EJMYC as presented in graphic in figure-2.

Existence EJMYC as public facilities unable to fulfill society requirement’s which immeasurable to progressively grow unfavourable perception to EJMYC and lessen community motivationto pay a visit and exploit facility of EJMYC. The Condition indicate that EJMYC which initially more destined for society have age 12-20 year, latterly claimed to able to fulfill requirement socialize in common with various [his/its]

ISSN: 1411-1284


activity. With reference to [the] mentioned require to be strived [by] settlement in so many area of[is inclusive of among other things settlement adminitrasi budget management system and retribution which the allocation utilization of EJMYC.

4. CONCLUSIONS

1. Condition of location, facility, accesibility, and

neighborhood’s of EJMYC had not yet appropriate for urban community needs, but as urban facility the EJMYC was still need.

2. EJMYC as public facility still need seeing that more than 60 % of respondent was declare that EJMYC as a needed and they used for anothers activity.

3. Utilization of EJMYC have still not optimum, ascertainable from visitor and occupancy that lower and unfitted with functionality of facilities.

4. Under utilization of EJMYC as the impact of their perception and motivation. In this case the role of their perception and motivation as explainable: a. minimum of visitor indicated that EJMYC had not

yet appropriate for urban community needs, b. minimum of visitor indicated that community

motivation is weak, c. weak motivation of community drive by bad

perception about facilities condition of EJMYC and demotivated of sport teacher/trainer.

5. Utilization effectiveness of EJMYC positively influenced by perception and motivation of community.

REFERENCES

[1] Bell, Paul A, et all, Environtmental Psychology : Principles and Practice, 5nd Edition, Thomson, Australia., 1996

[2] Dasso, Jerome et al, Real Estate-12th Edition, Prentice-Hall Inc, Englewood Cliffs -New Jersey,1992.

[3] Iman, Abdul Hamid Mar : An Introduction to Property Marketing, Universitas Teknologi Malaysia, Skudai Johor Darul Ta’zim - Malaysia. 2002.

[4] Lupiyoadi, Rambat dan Hamdani, A, Manajemen Pemasaran Jasa, Salemba Empat, Jakarta, 2006.

[5] Render, Barry, and Haizer, Jay: Principles of Operation Management, 2nd Edition, Prentice Hall, New Jersey., 1996

Figure -2. Normal P-P Plot of Regression Standardized Residual

0 . 0 0 . 2 0 . 4 0 . 6 0 . 8 1 . 0

O b s e r v e d C u m P r o b

0 . 0

0 . 2

0 . 4

0 . 6

0 . 8

1 . 0

Expected Cum Prob

D e p e n d e n t V a r i a b l e : P B

ISSN: 1411-1284


Water Quality Index as Indicator of Water Pollution. A Case Study in West Tarum Channel as Sources of

Water Supply

Djoko M. Hartono; Sulistyoweni W; Dwita Sutjiningsih M Departement of Civil Engineering Faculty of Engineering, University of Indonesia,,Depok

Tel. 7270029, fax. 7270028 email : [email protected]

Abstract– To fulfill the need for water supply, river is the main alternative of source of water among other sources of water. However, the surface water has been setting worst basically in its quality and it is over its permissible criteria standards as source water for water supply. The change in the quality of water source is caused by people activities that make pollution along the river in form of both domestic waste industry waste and solid waste. In many cases pollutant contained in the river are so hard that water supply treatment plant can not run the operation well and most of the time the service to the user being stopped. Besides, global climate change will affect raw water resources and has to consider its effect soo. Water Quality Index is one of indicator on water pollution is using to measure water quality of surface water from West Tarum Channel which are used by 3 Water Treatment Plan. Water Quality Index which has adopted from Storet Method was calculated and compare with Water Quality Index adopted from the National Sanitation Foundation USA with some modification and improvement, has been measured and the results show the raw water quality among 3 water intake has in the worst category. Keywords: criteria standard, water quality index, water treatment plant.

I. INTRODUCTION

ater is very important for human being, animals and plants and without water life on earth would

not exist. Surface water is one among other water sources as raw water to produces water supply processing by water treatment plan. The population growth, rapid urbanization and industrial growth, increasing of quality of life are some example on increasing demand for water supply. Besides, quantity and quality of surface water tend to decrease due to the fact that impact on changes in nature as well as pollution by man. The need to determine the chemical, physical and biological characteristic of natural water

resources is essential in view of their utilization particularly as a source of municipal water supply. Water quality index which was developed in the early 1970s, can be used to monitor water quality changes in a particular raw water for water supply, or it can be used to compare a raw water supply quality in other raw water supply in the region or from around the world. The result can also be used to determine if particular raw water is considered to be “healthy”. The study on raw water quality was conducted at raw water intake of Buaran, Pulo Gadung and Pejompongan Water Treatment Plan. Among other water treatment plan to fulfill demand of water supply in Jakarata, Buaran Water Treatment Plan, Pulo Gadung Water Treatment Plan as well as Pejompongan Water Treatment Plan are using same source of raw water from West Tarum Channel which is supplied form Jatiluhur Dam.

II. BASIC THEORY

Water quality status is based on State Ministry of Environment Number 115 year 2003 on which indicated several level of pollution on raw water comparing with raw water quality standard. Determination on water quality indicator can be done with 3 formulation: 1.STORET Method, which used value system as mention on Tabel 1.

Tabel 1. Determination Value System for Quality Status

Parameter Number of

sample Value Phisyc

s Chemis

t Biolog

y < 10 Maximum

Minimum Average

-1 -1 -3

-2 -2 -6

-3 -3 -9

≥10 Maxsimum Minimum Average

-2 -2 -6

-4 -4 -12

-6 -6 -18

Source: Ministry of Environment Decree No 115/2003

W

ISSN: 1411-1284


STORET method has 4 classification:

a. A Class, very good, score = 0, fulfill water quality standard

b. B Class, good, skor = -1 to -10, low pollution c. C Class, medium, skor = -11 to -30, medium

pollution d. D Class, bad, skor ≥ -31, heavy pollution

2. Pollution Index Method is used to determine level of pollution relative to parameter quality permissible. This formula also based on State Ministry of Environment Decree Number 115/2003. Pollution index can be formulated : Pij =(C1/L1j,C2/L2j,.........,Ci/L ij) (1)

Whereas, Pij = pollution indeks L ij =concentration of raw water quality parameter Ci = concentration water quality chosen Value Ci/L ij = 1,0 is critical value. Evaluation of pollution index can be categorized: a. 0 ≤ Pij ≤ 1, in good condition fit in with quality standard b.1,0 < Pij ≤ 5, low pollution c.5,0 ≤ Pij ≤ 10, medium pollutin d.Pij > 10, heavy pollution 3. Water Quality Index. One of the earlier attempt in formulating water quality index was done by National Sanitation and Foundation USA and Horton, 1970. According to them, water quality index is

QI = .....2W1W

CnWn.........2W2C1W1C

+++++

(2)

QI = water quality index C = rating W = weight M1= temperature value M2= obvious pollution value

Brown et al. and modification by Lohani, 1981, has used the approach to develop a weighted mean index of the form: n WQI = ∑ w1q1 (3) i=1

whwreas: WQI= water quality index, a number between 0 and 100

qi = the quality of the i th parameter, a number between 0 and 100

wi = the unit weight of the i th parameter, a number between 0 and 1, and

11q1wn

1i

=∑=

(4)

n = number of parameter

The Water Quality Index uses a scale from 0 to 100 to rate yhe quality of the water, with 100 being the highest possible score. Once the overall WQI score is known, it can be compared against the following scale to determine how healthy or how pollute the water is on giver time. Water quality cab be scale:

a. 91-100, excellent water quality b. 71-90, good water quality c. 51-70, average water quality d. 26-50, fair water quality e. 0-25, poor water quality

Raw water quality with rating falling in the excellent and good range would able to support a high diversity of aquatic life. In addition, the water would also be suitable for all forms of recreation, including those involving direct contact with water. Raw water quality achieving only an average rating generally have less diversity of aquatic organisms and frequently have increased algae growth. Raw water quality falling into the fair range are only be able to support a limited number of aquatic life forms, and it is expected that these water have abundant quality problem. A raw water quality with poor quality rating would not normally be considered acceptable for activities involving direct contact with the water, such as swimming. The scale with value 100, will be based on maximum permissible as mention on Government Decree Number 82-2001 for class 1 of raw quality. The scale with value 0 will base on class 4 of raw water quality. Raw water quality parameter will calculate from dominant parameter occurred during time of observation. The weight factor will use correlation from statistic.


By using multivariate analysis among number raw water parameter observed there are 9 parameter the most significance parameter among 48 parameter observed regularly by water supply enterprises. The 9 parameter are , turbidity, Total Dissolved Solid, ammonia, iron, manganese, BOD, COD, suspended solid and total coli form.

ISSN: 1411-1284


From Table 2, as can be seen raw water quality comparing with water quality standard adopted from Governor DKI Decree and Government Decree. Tabel 2. Raw Water Quality Status (Year 2003-2006)

Raw Water Quality

(Maximum) No

.

Para

meter Unit

SK Gub.DKI

No.582/ 1995

PP No.

82/2001

B P 1 P 2 PG 1. Turbidity NTU 100 - 15326 8800 9060 13246 2. TDS mg/l 500 1000 312 330 282 241 3. Ammonia mg/l 1,0 0,5 2,28 2,62 2,75 4 4. Iron mg/l 2,0 0,3 13,3 0,95 0,88 29,8 5. Manganese mg/l 0,5 0,1 6,79 5,5 1,48 5,6 6. BOD mg/l 10 2 60 42 41 60 7. COD mg/l 20 10 106 144 101 210 8. S S mg/l 100 50 5290 2834 2540 14560 9. Total coli

x 10³ Nol/ 100 ml

10 1 201 4800 680 201

Sources: Observation and calculation (2006) From above table, all of parameter exceed the quality standard, and the highest number occurred will disturb water supply treatment operation. The result of calculation each method can be seen on Table 3, Table 4 and Table 5.

Table 3. Calculation Water Quality Method Storet (Year 2003-2006)

Year Raw

Water Buaran

Raw Water Pulo

Gadung

Raw Water Pejompongan

I


II 2003 -72 -76 -64 -58

2004 -74 -74 -66 -66

2005 -72 -74 -64 -66

Source: Calculation (2006) From above table can be concluded that during year 2003 until 2006 raw water quality show in heavy pollution condition ( > -31) as a result strong effort made in produce water supply with quality standard given Table 4. Calculation Water Quality Method Pollution

Index (Year 2003-2006)

Year Raw

Water Buaran

Raw Water Pulo

Gadung


I


II 2003 11,75 11,79 10,91 10,35

2004 10,10 9,90 10,54 9,99

2005 10,70 10,34 11,09 11.03

Source: Calculation

From above table can be concluded that during year 2003 until 2006 raw water quality show in heavy pollution condition ( > 10) as a result strong effort made in produce water supply with quality standard given or treatment process stop.

Table 5. Calculation Water Quality Method WQI (Year 2003-2006)

Year Raw

Water Buaran

Raw Water Pulo

Gadung


I


II 2003 21,21 21,21 22,02 25,08

2004 27,8 15,01 34,46 32,20

2005 15,98 15,01 45,22 46,88

Source: Calculation From above table, almost all of raw water quality can be categorize as the worst raw water quality and indication have less diversity of aquatic organisms and frequently have increased algae growth. With that condition, water treatment plan is not allow to operate. Since, water supply is very essential for human being, follow up action has done and as a result more chemical put to treat the water. Water quality in Water Treatment Pejompongan better than other water intake, since in the distribution gate in Cawang , there is some addition “treatment”, by using screen, turbulence as a result on water gate and pump and additional chlorination put into transmission line.

IV. CONCLUSIONS Population growth and increasing industrial along the river which are polluting the river and erosion caused by rain fall made raw water for water supply from West Tarum Channel become worst. Special treatment by adding more chemical in processing the water has already done as a result more sludge produced. In the near future, modification on water treatment plan is needed so that water treatment plan can treat more efficiency . Many regulation in protecting raw water has made, however in implementing that regulation need more serious and coordination in between institution relevance should be made.

REFERENCES [1] American Water Works Association, 1999. “Water

Quality and Treatment, A Handbook of

ISSN: 1411-1284


Community Water Supplies”, fifth edition, McGraw Hill, 1999.

. [2] Bordalo A.A, Nilsumranchit W and Chalermwat K,

2001. Water Quality and Uses of the Bangpakong River (Estern Thailand), Water Research Volume 35 No. 15 (3635-3642). 2001.

Appendices

Appendices 1. Calculation on Storet Method (2003)

Paramet

er

Uni

t

PP 82

/2001

Raw Water Quality

Buaran Pulo

Gadung Pejompong

an 1 Pejompong

an 2

Quality

Valu

e Quality

Valu

e Quality

Valu

e

Quality

Valu

e Turb

idity

N

TU

100 1070

8

-

1

1176

8

-

1 8089

-

1 5808

-

1

76 -

1 150 -

1 135 -

1 169 -

1

4295.

57

-

3 3968.

29

-

3 2983.

43

-

3 2659 -

3 TDS m

g/l

500

288 0 241 0 330 0 282 0

92 0 145.

8 0 138 0 139.

1 0 170.

93 0 193.04 0

199.19 0

185.05 0

Ammoni

a

mg

0.5

2.28

-2 4

-2 2.62

-2 2.75

-2

0.31 0 0.89 -2 0.25 0 0.24 0

1.03 -6 2.09

-6 0.98

-6 1.00

-6

Iron m

g/l

0.3 12.6

0

-

2

19.2

4

-

2 0.95

-

2 0.46

-

2

1.17 -2 3.87

-2 0.03 0 0.03 0

5.51 -6 9.88

-6 0.22 0 0.14 0

Mangannese

mg/l

0.1

6.80

-2 4.03

-2 0.38

-2 0.29

-2

0.09 0 0.12 -

2 0.03 0 0.01 0

1.53 -

6 1.04 -

6 0.15 -

6 0.10 0

BOD mg/l

0.1

60

-2 60

-2 42

-2 35

-2

8 -2 7

-2 12

-2 14

-2

18.86

-6 18.5

-6

23.14

-6

21.64

-6

COD mg

/l

20

78.2

-2 78.2

-2 101

-2 101

-2

8.1 0 9 0 20 0 20 0 28.8

4 -6

29.84

-6

42.71

-6

44.21

-6

SS m

g

100

5180 -

2 5180 -

2 675 -

2 1107 -

2

35 0 19 0 105 0 117 0

1546 -6

1077.43

-6

311.14

-6

352.14

-6

Total Coli

mg/l

10000 2010

00 -3

201000

-3

4200000

-3

560000

-3

7000 -3

77000

-3

280000

-3 1800

-3

15342

9 -9

17928

6 -9

10200

00 -9

17482

9

-9

Total

-7

2

-7

6

-6

4

-5

8

Appendices 2. Calculation on Storet Method (2004)

Parameter Unit

PP 82/

Raw Water Quality

2001 Buaran

Pulo Gadung

Pejompongan 1

Pejompongan 2

Quali

ty

V

alue

Quality

Value

Quali

ty

v

alue

Quali

ty

Valu

e Turbidity NTU 100 1532

6 -1 9242

-1 6930

-1 5730 -1

70 -

1 138 -

1 116 -

1 101 -1

5140.1

9

-3

3830.7

1

-3

3043.1

4

-3 2771 -3

TDS mg/l 500

312 0 208 0 292 0 256 0

103 0 121.

4 0 107 0 122 0

157.82 0

156.46 0

184.79 0

179.2 0

Ammonia mg 0.5

1.21 -

2 1.75 -

2 1.67 -

2 1.21 -2

0.34 0 0.21 0 0.36 0 0.26 0

0.73 -6 0.98

-6 0.99

-6 0.79 -6

Iron mg/l 0.3 12.9

59

-

2

29.7

6

-

2 0.16 0 0.38 -2

0.931

-2 2.27

-2 0.02 0 0.02 0

7.97 -6

14.43

-6 0.08 0 0.12 0

Mangannes

e mg/l 0.1

2.77 -

2 5.6 -

2 0.90 -

2 1.48 -2

0.154

-2 0.21

-2 0.17

-2 0.01 0

0.85 -6 1.60

-6 0.43

-6 0.45 -6

BOD mg/l 0.1

18 -

2 14 -

2 34 -

2 24 -2

8 -2 8

-2 12

-2 12 -2

11.9

3 -6

11.57

-6

20.07

-6

17.57 -6

COD mg/l 20

34 -2 160

-2 82

-2 96 -2

9.6 0 16.5 0 16 0 18 0

26.7 -6

55.79

-6

38.71

-6

36.14 -6

SS mg 100

5290 -2 6210

-2 2834

-2 2378 -2

26 0 40 0 109 -2 113 -2

1873.6

1 -6

1437.5

-6

738.29

-6

622.64 -6

Total Coli mg/l 10000

201000

-3

201000

-3

4800000

-3

500000 -3

155000

-3

201000

-3

13000

-3 100 -3

194428.

57

-9

20100

0 -9

13635

71 -9

10308

0 -9

Total

-74

-74

-66 -66

ISSN: 1411-1284


Appendix 3. Calculation on Storet Method (2005) Para

meter

Uni

t

PP

82

/

Raw Water Quality

2001 Buaran Pulo Gadung Pejompongan 1 Pejompongan 2

Quality

Va

lue Quality

Va

lue Quality

Va

lu

e Quality

Va

lue

Turbidit

y

NTU 10

0 11068 -1 13146 -1 8800 -1 9060 -1

1333 -1 2754 -1 49.1 -1 1504 -1

6750.4

3 -3 7042 -3

4598.8

7 -3 5684.36 -3

TDS mg/l

50

0 179 0 192 0 207 0 179 0

91 0 121 0 107 0 115 0

128.93 0 154.31 0 153.71 0 149.21 0

Amonia mg 0.5

1.28 -2 2.52 -2 1.28 -2 1 -2

0.48 0 0.3 0 0.21 0 0.3 0

0.84 -6 1.21 -6 0.52 -6 0.69 -6

Besi mg/l

0.3

9.313 -2 24.53 -2 0.06 0 0.18 0

3.696 -2 5.25 -2 0.02 0 0.02 0

7.36 -6 13.61 -6 0.04 0 0.08 0

Mangan mg/l

0.1

1.103 -2 2.31 -2 0.79 -2 0.62 -2

0.091 0 0.23 -2 0.17 -2 0.17 -2

0.56 -6 0.73 -6 0.42 -6 0.34 -6

BOD mg/l

0.1

13 -2 20 -2 42 -2 41 -2

9 -2 8 -2 12 -2 9 -2

11 -6 12.64 -6 21.5 -6 19.86 -6

COD mg/l

20

106 -2 210 -2 61 -2 52 -2

5.6 0 14 0 20 0 20 0

48.56 -6 89.64 -6 36.43 -6 35.43 -6

SS mg 10

0 4590 -2 14560 -2 1019 -2 1101 -2

71 0 85 0 131 -2 166 -2

1803.5 -6 2815.5 -6 569.29 -6 556.29 -6

Total

Coli

mg/l

10

00

0 201000 -3 201000 -3

480000

0 -3 390000 -3

201000 -3 62000 -3 440000 -3 260 -3

201000 -9 177857 -9

147642

9 -9

86972.8

6 -9

Jumlah -72 -74 -

66 -66

Appendix 4. Calculation on Pollution Index (2003) Para

meter

Unit

PP 82

2001

Kualitas Air Baku (Maksimum, Minimum, Rata-rata)

Buaran Pulo Gadung Pejompongan 1 Pejompongan 2

Li Ci Ci/

Li

Ci/

Li baru

Ci Ci/

Li

Ci/

Li baru

Ci Ci/

Li

Ci/

Li baru

Ci Ci/

Li

Ci/

Li baru

Turbidity

NTU

100

10708

107.1

11.148

11768

117.7

11.35

8089

80.89

10.53

5808

58.08

9.82

4296

42.96

9.165

3968.29

39.68

8.99

2983.43

29.83

8.37

2659

26.59

8.12

TD

S

m

g/l

50

0

28

8

0.

58

0.

58

24

1

0.

48

0.

48

33

0

0.

66

0.

66

28

2

0.5

6

0.

56 17

1 0.34

0.14

193.04

0.39

0.39

199.19

0.40

0.4

185.05

0.37

0.37

Ammo

nia

mg

0.5

2.28

4.56

4.29

4 8 5.5

2.62

5.24

4.59

2.75

5.5 4.7

1.03

2.06

2.569

2.09

4.17

4.1

0.98

1.97

2.47

1.00

2.01

2.51

Iron

mg/l

0.3

12.6

04

42.0

1

9.11

19.2

4

64.1

3

10.0

3

0.95

3.17

3.5

0.46

1.53

1.92

5.51

18.37

7.32

9.88

32.93

8.58

0.22

0.73

0.73

0.14

0.45

0.45

Ma

nganese

m

g/l

0.

1

7 67

.95

10

.16

4.

03

40

.31

9.

02

0.

38

3.

77

3.

88

0.

29

2.8

9

3.

3

1.53

15.3

6.92

1.04

10.45

6.09

0.15

1.53

1.92

0.10

1.02

1.04

BOD

mg/l

0.1

60 600

14.89

60 600

14.89

42 420

14.11

35 350

13.72

19 190

12.39

18.5

185

12.33

23.14

231.43

12.82

21.64

216.43

12.67

COD

mg/l

20 78.2

3.91

3.96

78.2

3.91

3.96

101

5.05

4.51

101

5.05

4.51

28.8

1.44

1.79

29.84

1.49

1.86

42.71

2.14

2.65

44.21

2.21

2.72

SS mg

100

5180

51.8

9.57

5180

51.8

9.57

675

6.75

5.14

1107

11.07

6.22

1546

15.46

6.94

1077.43

10.77

6.16

311.14

3.11

3.46

352.14

3.52

3.73

Total Coli

mg/l

10000

201000

20.1

7.51

201000

20.1

7.51

4200000

420

14.11

560000

0.003

0.003

1534

29

15.3

4

6.92

1792

86

17.9

3

7.26

1020

000

102

11.0

4

1748

29

17.48

7.21

Maksimum

14.89

14.89

14.11

13.72

Sum

7.38

7.52

6.26

5.12

PIy 11.75

11.79

10.91

10.35

ISSN: 1411-1284


Appendix 5. Calculation on Pollution Index (2004)

Parame

Ter

Unit PP 82

2001

Kualitas Air Baku (Maksimum, Minimum, Rata-rata)


Li Ci Ci/Li

Ci/ Li

baru

Ci Ci/ Li

Ci/ Li

baru

Ci Ci/Li

Ci/Li

baru

Ci Ci/Li

Ci/Li

baru

Turbidity

NTU 100

15326

153.3

11.92

9242

92.42

10.82

6930

69.3

10.2

5730

57.3

9.79

5140

51.4

9.55

3830.7

1

38.31

8.91

3043.1

4

30.43

8.41

2771

27.71

8.21

TDS mg/l 500

312 0.62

0.62

208 0.42

0,42

292 0.58

0.58

256 0.51

0.51

158 0.32

0.32

156.46

0.31

0.31

184.79

0.37

0.37

179.2

0.36

0.36

Ammonia

mg 0.5 1.21

0,02

0.02

1.75

3.5 3.72

1.67

3.34

3.61

1.21

2.42

2.91

0.73

1,46

1.82

0.98

1.95

2.45

0.99

1.98

2.48

0.79

1.59

2

Iron mg/l 0.3 13 43.20

9.17

29.76

99.19

10.98

0.16

0.53

0.53

0.38

1.27

1.52

7.97

26.57

8.12

14.43

48.09

9.41

0.08

0.25

0.25

0.12

0.41

0.41

Manganese

mg/l 0.1 2.77

27.7

8.21

5.6 56 9.74

0.90

8.96

5.76

1.48

14.8

6.85

0.85

8.5 5.64

1.60

16.00

7.02

0.43

4.27

4.15

0.45

4.46

4.24

BOD mg/l 0.1 18 180 12.27

14 140 11.7

34 340 13.65

24 240 12.9

12 120 11.39

11.57

115.7

11.31

20.07

200.7

12.5

17.57

175.71

12.22

COD mg/l 20 34 1.7 2.15

160 8 5.51

82 4.1 4.06

96 4.8 4.4

16.7

0.84

0.84

55.79

2.79

3.22

38.71

1.94

2.43

36.14

1.81

2.29

SS mg 100

5290

52.9

9.61

6210

62.1

9.96

2834

28.34

8.26

2378

23.78

7.88

1874

18.74

7.36

1437.5

14.38

6.78

738.29

7.38

5.34

622.64

6.23

4.97

Total Coli

mg/l 10000

2E+06

201 12.5

201000

20.1

7.51

4800000

480 14,4

500000

50 9.49

194429

19.44

7.44

201000

20.1

7.51

136357

1

136.4

11.67

103080

10.31

6.06

Maksimum

12.5

11.7

13.65

12.9

Sum 6.92

7.70

5.99

5.78

PIy 10.1

9.9 10.54

9.99

ISSN: 1411-1284


Appendix 6. Calculation on Pollution Index (2005)

Parame

ter

Satuan PP 82

2001

Kualitas Air Baku (Maksimum, Rata-rata)


Li Ci Ci/Li Ci/Li

baru

Ci Ci/

Li

Ci/

Li

ba

ru

Ci Ci/

Li

Ci/

Li

ba

ru

Ci Ci/L

i

Ci/

Li

ba

ru

Turbidity NTU 100 11068 110.7 11.22 131

46

131

.5

11.

59

880

0

88 10.

72

906

0

90.

6

10.

78

6750 67.5 10.14 704

2

70.

42

10.

23

459

8.8

7

45.

99

9.3

1

568

4.4

56.

84

9.7

7

TDS mg/l 500 179 0.36 0.36 192 0.3

8

0.3

8

207 0.4

1

0.4

1

179 0.3

6

0.3

6

128.9 0.26 0.26 154

.31

0.3

1

0.3

1

153

.71

0.3

1

0.3

1

149

.21

0.3

0

0.3

Ammonia mg 0.5 1.28 2.56 3.04 2.5

2

5.0

4

4.5

1

1.2

8

2.5

6

3.0

4

1 2 2.5

0.84 1.68 1.13 1.2

1

2.4

2

2.9

2

0.5

2

1.0

4

1.0

8

0.6

921

1.3

8

1.6

9

Iron mg/l 0.3 9 306.7 13.43 24.

53

81.

76

10.

56

0.0

6

0.2 0.2 0.1

8

306

.67

13.

42

7.36 24.53 7.94 13.

61

45.

38

9.2

8

0.0

4

0.1

2

0.1

2

0.0

785

0.2

6

0.2

6

Mangan mg/l 0.1 1 11.03 6.21 2.3

1

23.

13

7.8

2

0.7

9

7.8

5

5.4

7

0.6

2

6.2 4.9

6

0.56 5.6 4.74 0.7

3

7.3

0

5.3

1

0.4

2

4.1

6

4.0

9

0.3

419

3.4

2

3.6

7

BOD mg/l 0.1 13 130 11.57 20 200 12.

5

42 420 14.

11

41 410 14.

06

11 110 11.2 12.

64

126

.43

11.

5

21.

5

215 12.

66

19.

857

215 12.

66

COD mg/l 20 106 5.3 4.62 210 10.

5

6.1 61 3.0

5

3.4

2

52 3.0

5

3.4

2

48.6 2.43 2.92 89.

64

4.4

8

4.2

5

36.

43

1.8

2

2.3 35.

429

1.8

2

2.3

SS mg 100 4590 45.9 9.3 145

60

145

.6

11.

81

101

9

10.

19

6 110

1

10.

19

6

1803 18.03 6.28 281

5.5

28.

16

8.2

5

569

.29

5.6

9

4.7

7

556

.29

5.6

9

4.7

7

Total Coli mg/l 10000 201000 20.1

7.51

201

000

20.

1

7.5

2

480

000

0

480 14.

4

390

000

480 14.

4

201000 20.1 7.51 177

857

17.

79

7.2

5

147

642

9

147

.6

11.

84

869

73

147

.64

11.

85

Maksimum 13.43 12.

5

14.

4

14.

4

Sum 6.99 7.6

1

6.2

4

6.9

2

PIy 10.7 10.

34

11.

09

11,

03

Non Linear Dynamic Identification of Waste Heat

Boiler Unit Based on

Adaptive Neuro Fuzzy Inference System (ANFIS)

Yuliati

Electrical Engineering Department, Widya Mandala Surabaya Catholic University.

Jl. Kalijudan 37 Surabaya,

Phone: +62-31-3891264, Email : [email protected]

Abstract- Waste Heat Boiler (WHB) unit is a

crucial plant for a fertilizer plant especially for

steam product .The water level in the its steam

drum shall be maintained at certain limit in order

to give a good steam quality. It is generally

accepted that a boiler unit is a highly nonlinear

and strongly coupled complex system. Due to the

dynamic characteristic, it is difficult to have its

model in order to give suitable control system of

the steam drum water level theoretically.

The success of model-based non linear control

technique is usually conditioned by the

availability of accurate models which reflects the

non linear process complexities. Prior to

implementing any control scheme, its dynamic

properties need to be identified properly. A real

time data driven neuro fuzzy model will be

presented for the water level of boiler unit. Neuro

fuzzy design combine architectural of neural

network and philosophical of fuzzy system aspects

of an expert resulting in an artificial brain, which

can be used as an identifier.

An experiment for identifying the water level

of boiler unit was conducted on real-time plant

operation of a fertilizer plant in Gresik, East

Java, Indonesia. Validation model was also

performed to verify the results of identification.

The performance of the procedure shows how the

neuro fuzzy algorithm produces accurate models

for non linear control design.

Keywords : system identification, waste heat boiler,

real plant, fuzzy logic, neural network.

I. Introduction

There are two types of boilers in the steam

generation i.e. fired boilers and unfired boilers (heat

recovery steam generator) is in fact a heat exchanger

extracting heat from through put gases and raising

water to steam condition. Waste heat boiler unit

representing type of unfired boiler system which

using heat transfer energy form gas turbine exhaust

gas flow. Its about 60% of 70% of the heat that

obtained by combustion in gas turbine, re-exploited

by WHB to produce the steam.

It’s generally accepted that a boiler unit is a

highly non linear and strongly coupled complex

system. Its dynamic characteristic makes it very

difficult to model and control the water level of

steam drum theoretically. The water level of High

pressure steam drum in WHB unit should be

maintained within safe limit. A too high of water

level produces wet steam that could damage the

turbine equipment and the turbine can trip off. On the

other hand, too low of water level risks uncovering

the water tubes and exposing them to heat stress and

damage. In both cases, the plant shuts down

unintentionally.

The success of model-based non linear control

technique is usually conditioned by the availability of

accurate models which reflects the non linear process

complexities. Prior to implementing any control

scheme, its dynamic properties need to be identified

properly.

Neural network have been applied successfully

in the identification and control of non linear

dynamic system. The universal approximation

capabilities of multilayer perceptron (MLP) make it

popular choice for modelling non linear systems.

However, neural network architecture is rather

complicated and difficult to be understood by human.

On the other hand, the fuzzy inference system can

initialise and learn linguistic and semi linguistic

(Sugeno) rules therefore it can be considered as

direct transfer knowledge which is the main

advantage of fuzzy inference systems but the

learning capabilities of fuzzy system is not as good

as neural network. It is then to develop to integrate

the strength of both methodologies in order to

achieve the learning capabilities and transfer

knowledge via fuzzy if-then rules, producing an

alternative approach so called neuro fuzzy.

In this paper, a development of non linear

identification for high pressure drum (HP drum) of

boiler unit will be design using neuro fuzzy

approach which the fuzzy inference system can be

tuned with a neural network algorithm during

learning based on collection of input output data

obtained from real-time experiment. By the using

real plant, it is expected that the model obtained

could capture the basic feature of water level of steam drum system and this technique could also

improve the modelling quality of the system.

II. Waste Heat Boiler Plant

Waste Heat Boiler (WHB) unit plays a crucial

role in a fertilizer plant as a part which produces

steam. The temperature of the exhaust gas from gas

turbine is about 6000C used to produce superheated

steam which has 65 kg/cm2 pressure and temperature

4600 C. Boiler Feed Water (BFW) is feed into High

Pressure Steam Drum using 2 pumps which standby

interchange each other. Control valve is used to

control flow of BFW; the level of HP steam drum is

controlled by 2 types of drum level control i.e. single

element and three element control. If the steam

product is smaller then 30 ton/hr, single element

control used to control level drum. In the other hand,

if the steam product is greater then 30 ton/hr, three

elements drum level control used to control level

drum. It combines between level drum, steam flow

and feed water flow.

WHB unit in PT. Petrokimia Gresik is an unfired

boiler completed by some sub system (Figure 1) i.e.

Super heater, high pressure (HP) evaporator, low

pressure (LP) evaporator, economizer, low header,

HP/LP down comers, HP/LP drum, duct burner, and

firing room.

Figure 1. WHB sub system

III. The Experimental Set-up

The experiment set-up is referred to Piping &

Instrumentation Diagram (P&ID) in PT. Petrokimia

Gresik. There are several conditions that must be

considered to design this experiment as follow:

a. Pressure and temperature of BFW,

Pressure and temperature of HP drum,

steam product temperature and blow down

were kept constant as much as possible.

b. Set point flow steam product is based on

the need of steam in urea plant.

c. Data have been collected in real time from

the existing Distributed Control System

(DCS) with sampling time 1 minute.

d. Each 3000 data for every corresponding

inputs and outputs have been collected. The

first 500 data will be used for

identification, and the 1500 remaining data

for model validation.

e. Due to high risk of its plant operation, the

experiment has been conducted by closed

loop manner without change the real plant

operation.

The record of input output data during the conducted

experiment can be shown in Figure 2 and will be

used for modelling and validation purpose.

(a) Flow BFW

(b) WHB Steam Flow

(c) Level WHB HP Drum

Figure 2. Representation of Input-Output Data

During identification process, it can be defined as

shown in Table 1 for input-output variable for

modelling purpose.

Table 1. Input-Output Variable for Modelling

Variable Tag

number

Description

)1(1 −kx FC.22220 Flow BFW (Ton/hr)

)1(2 −kx FC.22221 Steam Product Flow

(Ton/hr)

)1( −ky LC.22220 Level water in HP steam drum at sampling instant

k-1 as an input model (%)

)(ky LC.22220 Level water in HP steam

drum at sampling instant k

as an output model (%)

III. Neuro Fuzzy Dynamic Identification

3.1 ANFIS Structure

ANFIS implements a Takagi- Sugeno Fuzzy

Models have to be a powerful tool for the non linear

dynamic identification system. ANFIS scheme as an

estimator model can be shown in Figure 3.

Figure 3. ANFIS as an Estimator Model

The first order of ANFIS structure with three

inputs and one output can be described by the

following rule base:

Rule ith : if )1(1 −kx is Ai and )1(2 −kx is Bi and

)1( −ky is iC then

iiiii vkyrkxqkxpf +−+−+−= )1()1()1( 21 ,

mi ,...,2,1= (1)

Where m denotes the number of rules. In the other

hand, the non linear AutoRegressive with eXogenous

input (NARX) model is frequently used for non

linear identification techniques. Therefore, in order

to represent the neuro fuzzy for Multi Input Single

Output (MISO) approach can be defined as

))1(),...,(),1(),...,(()1( +−+−=+ bam nkukunkykyfky

(2)

The input vector x , of the neuro fuzzy model as:

)]1(),...,(),1(),...,([ +−+−= ba nkukunkykyx (3)

Where )1(),...,( +− ankyky and

)1(),...,( +− bnkuku are tapped delay lag (TDL)

process output and input respectively. ANFIS has

five layered architecture which has three inputs and

one output can be shown in Figure 4.

Figure 4. Structure of ANFIS

The architecture employs an adaptive network

consist of adaptive nodes in the 1st and 4

th layer. In

the 1st is for fuzzification of the input variable and T-

norm operator are deployed in the 2nd

layer to

compute the rule antecedent part (premise

parameter). The 3rd

layer normalizes the rule

strengths followed by 4th layer where the consequent

parameters are determined. The output network

computes the overall input as the summation of all

incoming signal. The description and computation in

every layer can be described as shown in Table 2.

Table 2. The Description and Computation of Each

Layer in ANFIS Method

Layer Description

1

Every nodes i is an adaptive node with a

node output :

)(,1 xO Aii µ= , and )(2,1 yO Bii −= µ

The output of this layer is the

membership values of the premise part.

The generalized bell is specified by

three parameters a,b,c:

ib

i

i

Ai

a

cx

x

−+

=2

1

1)(µ

has been chosen as membership function

2

Fixed node labelled Π which multiplies

the incoming signal and output product

represents the firing strength of a rule:

)()(,2 yxwO BiAiii µµ== , i=1,2

3

Fixed node labelled N which

normalizes the rule strengths

21,3

ww

wwO iii +== , i=1,2

4 Every node i is an adaptive node with a

node function :

( )iiiiiii ryqxpwfwO ++==,4 , i= 1,2

Referred to as consequent parameters.

5 This single node is a fixed node which

computes the overall output as the

summation of all incoming signal:

∑ ∑

∑===

i

i

i

i

ii

iiiw

fw

fwfO 5,

3.2 Hybrid Learning Algorithm

The hybrid learning algorithm has two passes for

ANFIS i.e. forward pass and backward pass. Suppose

that S1 is a set of non linear parameters and S2 is a

set of linear parameters in the forward pass, node

outputs go forward until 4 layer and the consequent

parameters are identified by the least square

estimation. In the backward pass, the error signals

propagate backward and then premise parameters are

update by gradient descent. a. Forward pass

If the values of the premise parameters are fixed,

then the overall output can be expressed as a linear

combination of the consequent parameters, the output

f can rewritten as

221

21

21

1 fww

wf

ww

wf

++

+=

( ) ( )22221111 ryqxpwryqxpw +++++=

( ) ( ) ( ) ( ) 22111111 pxwrwqywpxw +++=

( ) ( )rwqyw 222 ++ (4)

From the above equation, it can be seen that the

consequent parameters are linear respect to the

system output. If N learning data is applied to the

equation (4), it can be represented by

yA =θ

Where θ unknown parameter in S2 and y is is N

vector data. The problem is how to find the best

solution of this equation that minimize 2

yA −θ ,

is the least square estimator

( ) yAAATT 1

*−

=θ (5)

b. Backward pass

In the backward pass the error signal

propagate from the output end toward the input end

using the simple gradient steepest descent for each

training data entry. By minimizing a cost function

defined by

∑=

=T

k

keE

1

2)(

2

1 (6)

With )(ˆ)()( kykyke −= (7)

is an error function at time instant k, with y(k) is the

desired response.

For updating the changes of each parameter, the

gradients are needed to minimize E Using the

relation:

)(ˆ

)()1(ˆ)(ˆ

k

kEkk

Li

Li

Li

Θ∂

∂+−Θ=Θ η (8)

Where η is the learning rate and )(ˆ

)(

k

kELiΘ∂

∂denotes

the order derivative of the error signals (difference

between an actual output and desired output vector)

respect to the node parameter Θ .

IV. The Results of System Identification

Identification process has been conducted using

Multi Input Single Output (MISO) approach as

shown in Figure 4. For learning process and

modelling purpose, assuming that the plant under

consideration can be represented as a first order

process, then the regressor /identifier have been

chosen as

x= )]1(),1(),1([ 21 −−− kykxkx (9)

And after 50 epochs for learning process, the ANFIS

structure model can be obtained as the following: the

number of nodes: 34, the number of linear

parameters: 32, the number of nonlinear parameters:

18, the number of training data pairs: 500, and the

number of fuzzy rules: 8.

The result of system identification and its

corresponding errors of learning process can be

shown in Figure 5 and 6, respectively.

Figure 5. The result of System Identification

Figure 6. Error

It can be seen from these figure that the output

of the identified model can follow the measurement

signals satisfactorily and the error (the difference

between these two signals) are relatively small

enough.

V. Model Validation

Beside a direct observation by comparing the

plot of output measurements and the output model

based on the estimated parameter, quantitative

criteria of the Root Mean Square Error (RMSE) can

be used for validation model for a measure the

goodness of fit the model. It can be defined as

RMSE = [ ]∑=

−N

t

kykyn

1

2)(ˆ)(

1 (10)

If the RMSE 5.0≤ then the developed model can be

accepted as a good approach to represent the model

system. The 1500 remaining data was used for

validation purpose, and the result of model validation

and error of validation model can be shown in Figure

7 and 8, respectively.

Figure 7. The Result of Model Validation

Figure 8. Error of the Model Validation

The RMSE both of the identification process and the

model validation are relatively small enough and it

can be seen in Table 3.

Table 3. Root Mean Square Error

Data from :

Identification Validation

RMSE 0.0324 0.0640

VI. Conclusion

An alternative approach to identify level water

of high pressure steam drum in WHB unit using

neuro fuzzy technique has been proposed. By

comparing between the response models with the

actual output measurements revealed that the

satisfactory model matching were obtained. It means

that the non-linear characteristics of system can be

capture from the real basic feature of the WHB

system. The performance of the procedure shows

how the neuro fuzzy algorithm produces accurate

models for non linear control design. Further

application for level control system design based on

identified model is currently under study.

VII. References

[1]. Johanson, R, System Modelling and Identification,

Prentice-Hall International, Inc, NJ, USA, 1993.

[2]. Jyh-Shing Roger Jang, Chuen-Tsai Sun, Neuro Fuzzy

Modelling and Control, The proceeding of the 1995

IEEE, International Conference on Control Application, 1995.

[3]. J.S.R. Jang, C.T. Sun, E. Mizutani, Neuro – Fuzzy and

Soft Computing: A Computational Approach to

Learning and Machine Intelligent, Prentice-Hall,

International Edition, 1997.

[4]. Narendra,K.S, Parthasarathy,K, Identification and

Control of Dynamical Sistems Using Neural Network,

IEEE Transaction on Neural Network, Vol.1, No.1,

March 1990.

[5]. A.W. Ordys, A.W. Pike, M.A. Johnson, R.M. Katebi,

and M.J. Grimble, Modelling and Simulation of Power Generation Plants, Springer-Verlag, 1994.

[6]. Standard Fasel Lantjes, Waste Heat Boiler, PT.

Petrokimia Gresik, 1992.

[7]. Xiaoguang Chang, A Recurrent Fuzzy- Neuro Dynamic

Model for a Stoked-Fired Boiler, Laboratory of

Intelligent Technology and Systems Tsinghua

University, China, May 1999.

[8]., …………….ProcidiaTM Control Solutions boiler

Control Overview, Application Data by SIEMENS,

March 2006.

[9]. Yul Y.Nazaruddin, Joko W, S. Hadisupadmo, Inverse

Learning Control using Neuro Fuzzy Approach for a Process Mini Plant, International Conference on

Physics and Control, Rusia, August 2003.

ISSN: 1411-1284


LOAD PATTERN AND REALIBILITY ASSESSMENT OF POWER GENERATION SYSTEM

Suparman, Zuhal, and Rinaldy D.

Effect of Harmonic Current

Documents

Transcript of Effect of Harmonic Current