1-s2.0-S0360319914027566-main

etproduction

R. Soltani*, M.A. Rosen, I. Din

Faculty of Engineering and Applied Science, U

North, Oshawa, ON L1H 7K4, Canada

a r t i c l e i n f o

ive with two other lo-

presence of CO2 in the

the ratio of hydrogen

.

Copyright 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

reserved.

In order to keep global warming to less than 2 C, some statethat the atmospheric CO2 concentration should not exceed

easures have been

proposed to mitigate global warming, including carbon diox-

ide capture and sequestration (CCS), which some propose as

an effective way to stabilise atmospheric carbon dioxide

concentrations [2,3]. Fig. 1 presents a breakdown of industrial

* Corresponding author.E-mail addresses: [email protected], [email protected] (R. Soltani), [email protected] (M.A. Rosen), [email protected]

(I. Dincer).M, Dhahran 31261, Saudi Arabia.

Available online at www.sciencedirect.com

ScienceDirect

w.

i n t e r n a t i o n a l j o u r n a l o f h yd r o g e n e n e r g y x x x ( 2 0 1 4 ) 1e1 01 Associated with Department of Mechanical Engineering, KFUPIntroduction450 ppmV CO2-equivalent [1]. Many man S/C ratio of 2.5, CO2 capture from a flue gas stream is competit

cations provided higher weighting factors are considered for the full

flue gases stream. Considering carbon removal from flue gases,

production rate and Ncc increases with rising reformer temperaturethe effects of oxygen enrichment of the furnace feed are investigated, and it is found that

this measure improves the CO2 capture conditions for lower S/C ratios. Consequently, forArticle history:

Received 11 July 2014

Received in revised form

24 September 2014

Accepted 29 September 2014

Available online xxx

Keywords:

Steam methane reforming

Hydrogen production

CO2 emission

CO2 capture

Oxygen enrichmentPlease cite this article in press as: Soltanreforming for hydrogen production,j.ijhydene.2014.09.161

http://dx.doi.org/10.1016/j.ijhydene.2014.09.10360-3199/Copyright 2014, Hydrogen Enercer 1

niversity of Ontario Institute of Technology (UOIT), 2000 Simcoe St.

a b s t r a c t

Steam methane reforming (SMR) is currently the main hydrogen production process in

industry, but it has high emissions of CO2, at almost 7 kg CO2/kg H2 on average, and is

responsible for about 3% of global industrial sector CO2 emissions. Here, the results are

reported of an investigation of the effect of steam-to-carbon ratio (S/C) on CO2 capture

criteria from various locations in the process, i.e. synthesis gas stream (location 1), pres-

sure swing adsorber (PSA) tail gas (location 2), and furnace flue gases (location 3). The CO2capture criteria considered in this study are CO2 partial pressure, CO2 concentration, and

CO2 mass ratio compared to the final exhaust stream, which is furnace flue gases. The CO2capture number (Ncc) is proposed as measure of capture favourability, defined as the

product of the three above capture criteria. A weighting of unity is used for each criterion.

The best S/C ratio, in terms of providing better capture option, is determined. CO2 removal

from synthesis gas after the shift unit is found to be the best location for CO2 capture due to

its high partial pressure of CO2. However, furnace flue gases, containing almost 50% of the

CO2 in produced in the process, are of great significance environmentally. Consequently,points in steam m hane reforming for hydrogen

Assessment of CO2 capture options from variousjournal homepage: wwi R, et al., Assessment oInternational Journal

61gy Publications, LLC. Publelsevier .com/locate/hef CO2 capture options from various points in steam methaneof Hydrogen Energy (2014), http://dx.doi.org/10.1016/

ished by Elsevier Ltd. All rights reserved.

world now use amine-based systems [12,13]. Nonetheless,

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y x x x ( 2 0 1 4 ) 1e1 02CO2 emissions based on a 2008 IEA report [3]. There, it can be

seen that the (petro) chemical sector in 2005 was responsible

for about 16% of industrial CO2 emissions, and that steam

methane reforming (SMR) accounts for a large share. Also,

from more recent IEA publications [4,5], the petrochemical

sector, after the iron and cement sector, is still the major

source of carbon dioxide emissions.

Steam reforming is the commonly used and mature tech-

nology for industrial hydrogen production. According to a life

cycle assessment of global hydrogen production [6], about 75%

of world's total hydrogen is produced by steam methanereforming. Also, a 2008 IEA report estimated the global annual

hydrogen production in 2005 at 65 Mt, with 48% from SMR [3].

Other data on the share of hydrogen production from SMR

confirm SMR to be the main process for hydrogen production.

Although SMR might be replaced in the future by other effi-

cient hydrogen production techniques, e.g. such as thermo-

catalytic decomposition [7], it is still expected to be important

in the future.

As pointed out earlier, SMR facilities emit on average 7 kg

CO2/kg H2 [3], which was equivalent to 220 Mt CO2 globally in

2005 [3]. Compared to total global CO2 emissions, the share

contributed by SMR facilities is small, at around 3% [3]. This

share is expected to increase through decarbonising the

transportation industry, partly by using fuel cell systems and

hydrogen as an energy carrier. The IEA estimates that if fuel

cell technology is applied in the transportation sector suc-

cessfully by 2050, the global hydrogen demand will be as high

Fig. 1 e Direct global CO2 emissions of the industrial sector

in Gt/yr for 2005 [adapted from 3].as 275 Mt per year [8], resulting in emissions of about 2 Gt CO2per year if SMR facilities are used. CCS would then be more

attractive for decreasing these emissions. One advantage of

introducing hydrogen to the transportation sector is that

emission control is easier at a central SMR plant than in all

cars on the road. When purified, CO2 has many uses in the

chemical industry, in solid form (dry ice), liquid form (e.g.

refrigeration equipment) and gaseous form (beverage

carbonation). It also is widely used as a reactant in chemical

processes and as an inert blanketing gas to prevent oxidation

(e.g., for food products) [9]. Recently, it has received attention

to help increase oil recovery fromdepleted or high viscosity oil

fields [10].

Numerous studies have been carried out on CO2 removal

from SMR, usually focussing on in-process capture and end-

point capture, i.e. capturing furnace flue gases or post com-

bustion capture. The former involves the capture from

Please cite this article in press as: Soltani R, et al., Assessment oreforming for hydrogen production, International Journaj.ijhydene.2014.09.161amine-based technology have been identified is not suitable

for low molar fractions of CO2 (flue gases) [14], and other

shortcomings of amine-based technology for flue gases have

been described [15], e.g. extensive solvent corrosion issues

and energy intensiveness. Removal of CO2 from flue gases by

pressure swing adsorption (PSA) has recently been investi-

gated as an alternative technology for CO2 removal by Kikki-

nides et al. [9], Park et al. [16], and Ko et al. [17]. Voss [18] has

studied the feasibility of CO2 removal from different process

locations by different technologies and compared them based

on various criteria. He concludes that PSA technology is highly

competitive to mature amine-based methods due to its

comparatively simpler procedure. Also, he states that PSA is

favourable for pressurized feed streams, namely synthesis gas

and PSA off-gas streams. Economic studies have also been

reported, e.g. assessing techno-economically CO2 capture

systems [1,19]. Overall, it is expected that CO2 avoidance costs

at SMR facilities decrease by having high pressure sources of

CO2 with high concentrations. Two main factors dominate

capture costs: CO2 partial pressure and molar concentrations

of streams. In this study, mass flows of CO2 are also consid-

ered, as these can motivate investors who seek to avoid po-

tential high CO2 emission penalties in the future.

Oxygen enrichment can enhance the performance of

steam methane reforming and other industrial processes,

especially those involving combustion. In case of SMR, oxygen

enrichment has been shown to reduce the requirement for

natural gas feed [20], mainly due to better heat transfer from

the combustion gases to the reformer. This is a consequence

of lower nitrogen concentrations in the combustion gases,

which decreases the amount of sensible heat lost with flue

gases. The other main benefit of oxygen enrichment is,

because of the relatively lower nitrogen supply, higher carbon

dioxide concentrations are attained in the combustion prod-

ucts. In this study, a fixed hydrogen production rate is

considered when analysing the impact of parameter varia-

tions and the effect of oxygen enrichment only on the com-

bustion process is examined.

In this research, we analyse an SMR process to determine

how the S/C ratio affects the criteria that play major roles in

CO2 capture: partial pressure, concentration, and overall mass

flow rate of CO2. Also, focussing on post combustion capture

from furnace flue gases, oxygen enrichment of furnace com-

bustion is investigated to see if it makes CO2 capture from flue

gases reasonable compared to synthesis gases and PSA off

gases. Finally, hydrogen production and carbon removal are

studied together and compared for varying S/C ratios and

reforming temperatures.

Methods considered

General descriptions are provided of the threemain processesprocess streams and is discussed in following sections. Sim-

beck [11] indicates that carbon capture from the flue gas

stream of an SMR furnace is possible using amine-based

capture and, in fact, most pilot plants established in theconsidered in this study: SMR, CO2 capture, and oxygen

enrichment.

f CO2 capture options from various points in steam methanel of Hydrogen Energy (2014), http://dx.doi.org/10.1016/

Steam methane reforming

A typical SMR system consists of four main sequential units:

desulfurizer, reformer, shift reactors and separation units. In

the desulfurizer, sulfur is removed from natural gas to avoid

the production of sulfur oxides and contamination of catalysts

in the reformer. To simplify the analysis, the natural gas feed

is assumed to be puremethane in this study. In the reformer, a

syngas containing H2 and CO is produced by reacting between

methane and steam (Equation (1)). There are two shift re-

actors: high temperature (HT) and low temperature (LT), both

of which convert CO produced in the reformer to CO2 and H2(Equation (2)). The main reactions involved in an SMR process

are as follows:

CH4 H2Og/CO 3H2 DHr 251 MJ=kmolCH4 (1)

COH2Og/CO2 H2 DHr 41:2 MJ=kmolCH4 (2)

Carbon dioxide capture

In externally fired steam methane reforming process, as

stated earlier, three CO2 containing streams can be identified:

1) Shifted synthesis gas upstream of the hydrogen purifica-

tion unit.

2) PSA tail gas from hydrogen purification.

3) Flue gas from steam reformer furnace system.

The three stream locations are shown in Fig. 2. Each stream

provides the potential for removing CO2. Each stream has

different specifications. The shifted synthetic gas has low CO2concentration and elevated pressure. The PSA tail gas has

high CO2 concentration and low pressure. The flue gas has low

CO2 concentration and low pressure, but it has the advantage

of containing the full mass flow of produced CO2, which

makes it of interest environmentally.

i n t e r n a t i o n a l j o u r n a l o f h yd r o g e n e n e r g y x x x ( 2 0 1 4 ) 1e1 0 3The net overall reaction is endothermic and the required

heat is normally supplied to the reformer by a furnace. Some

typical reformer operating conditions are listed as a temper-

ature of 700e1000 C, a pressure of 15e50 bar and an S/C ratio

between 2 and 5 [21]. The produced syngas is cooled before

entering the shift reactor to remove the heat of the exothermic

shift reaction (Equation (2)). The gas stream exiting the shift

reactors consists of H2, CO, CO2, H2O and the remaining

methane. After separation and removal of the water using a

condenser, the dry shifted stream enters the hydrogen puri-

fication unit from which the final product H2 exits. There are

two main technologies for hydrogen purification: PSA and

membrane separation [22]. Due to complicated nature of pu-

rification process, all separation and purification units are

assumed in this study to be simple separation steps. This

simplification is invoked because the focus of this study is on

substance concentrations, pressures and conversion ratios.

After purification of the H2 stream, the remaining gas stream

(PSA tail gas) leaves the unit at near atmospheric and with a

high concentration of CO2. This tail gas is sent to furnace as a

secondary feed stream (in addition to the main fuel),

decreasing the fuel consumption in the furnace.Fig. 2 e Schematic of SMR proce

Please cite this article in press as: Soltani R, et al., Assessment oreforming for hydrogen production, International Journalj.ijhydene.2014.09.161Two main technologies exist for CO2 removal from SMR

streams: absorption and adsorption. The former refers to

amine-based capturing method known as the monoethanol-

amine (MEA) process and the latter refers to PSA techniques.

Some other technologies for CO2 removal also can be found

[24], including vacuum pressure swing adsorption, which was

recently been introduced for this purpose [18]. MEA is a

mature technology for CO2 removal from syngas, but it has not

yet been commercialized for post-combustion capture,

mainly because the low molar fractions of CO2 are not

considered favourable [14] and the process is highly energy

intensive. This application also has solvent corrosion issues,

raising its costs significantly. The presence of contaminants

like O2 and NOx significantly increase solvent deterioration,

further increasing operational costs [25]. Consequently, the

process conditions of stream 1 allow the application of MEA

and PSA, while only PSA systems are applicable for stream 2

due to the high CO2 concentration and amine based systems

are preferred for stream 3 [18]. However MEA is not commer-

cialized due to corrosion issues.

Although many detailed specifications exist for each cap-

ture technology, in this study the focus is on the investigationss without heat integration.

f CO2 capture options from various points in steam methaneof Hydrogen Energy (2014), http://dx.doi.org/10.1016/

Second, oxygen enrichment improves heat transfer in the

inadequately accurate).

The separation of water in the condenser is complete.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y x x x ( 2 0 1 4 ) 1e1 04reformer because of the higher flame temperature and the

higher emissivity of combustion gases, associated with the

increase in CO2 and H2O concentrations (as a consequence of

decreased nitrogen) [20]. The improvement of heat transfer

through the tubes of a steam reforming furnace increases the

amount of heat received by the methane and steam mixture

and hence its temperature, so that the reforming equilibrium

is displaced toward higher hydrogen production [20]. This

phenomenon leads to higher conversion ratios and, therefore,

a lower natural gas inlet for the same hydrogen production.

For simplicity, these effects on conversion ratio are not

considered in this study. Only the effect of reducing the

furnace fuel consumption for the provision of the heat

required for steam production and reformer is of importance.

In other words, the decrease in the steam to methane ratio

due to oxygen enrichment is not studied. Overall, two positive

effects on steam methane reforming result from oxygen

enrichment in the furnace: a higher conversion ratio modified

flue gas and furnace fuel consumption. The former is not

investigated here. Different enriching technologies are

appropriate for various volumes of enriched gas, including

PSA, cryogenic refrigeration [26] andmembrane air separation

(MAS). The simplicity and modular nature of MAS makes it

well suited for retrofitting and upgrading current processes

and cycles. The usual limits in industry for enrichment are

between 25 and 35% [20]. Nonetheless, all levels of enrichment

are considered in this study, to better characterize the impact

on the CO2 partial pressure in flue gases through enrichment.

Model description

A schematic for the process investigated here is depicted in

Fig. 2. The model developed for this investigation is mainly

based on the flow diagram and industrial plant data provided

by the U.S. Department of Energy (DOE) in a project on SMR, as

reported in Ref. [23]. The processmodel is developed using the

Aspen HYSYS V7.3 process modelling and simulation soft-

ware, to permit the case study to be analyzed for various

operating conditions. Aspen HYSYS is a comprehensive pro-

cess modelling system used by many oil and gas producers,

refineries, and engineering companies around the world to

optimize process design and operations [27]. All pressures andof two dominant factors for CO2 removal: CO2 partial pressure

and molar concentration. Also, an environmental perspective

suggests that the presence of CO2 in process streams, as a

fraction of the full CO2 present in the flue gases, be considered

as another capture criterion.

Oxygen enrichment

The benefits of oxygen enrichment have been demonstrated

for many combustion and other processes industry. For

combustion, for instance, oxygen enrichment has two main

benefits. First, it decreases the concentration of parasitic ni-

trogen in the combustion gases, increasing the heat available

to the process because less heat is lost via nitrogen emissions.temperatures are based on this industrial case study, but the

hydrogen production rate is set to 1 kg/h. Then, the required

Please cite this article in press as: Soltani R, et al., Assessment oreforming for hydrogen production, International Journaj.ijhydene.2014.09.161 No excess steam is generated (even though excess steam isnormally a by-product of SMR plants for export).

The thermodynamic integrity of the simplified model is

assured by setting appropriate reactor temperatures and flow

stream temperatures exogenously. The developed model is

depicted in Fig. 3 and the pressures, temperatures and pres-

sure drops are presented in Table 1. Comparing Figs. 2 and 3, it

can be seen that there is no difference; one may ask why flue

the gases stream from the combustor is not connected to the

reformer reactor? The reason is that heat input to the

reformer is via heat, not a material stream. Therefore, the

value used as the energy input to the reformer is based on the

energy of the flue gases.

Approach

The favourability of carbon dioxide capture is assessed at

three locations using two steps. The criteria for this study are

carbon dioxide partial pressure, molar concentration and

mass flow ratio. First, the condition for typical air (21% O2) as a

furnace feed is investigated. Second, the enhancement of

capturing feasibility by oxygen enrichment is assessed and

the impact of oxygen enrichment is compared for the three

locations. Note that both the optimal and ideal (100%) levels of

oxygen enrichment are studied. The assessment is made

quantitatively in terms of a proposed CO2 capture number

(Ncc), thereby allowing the capture favourability to be evalu-

ated and compared for different conditions, as follows:

Ncc Imass Ipressure Iconcentration (3)where Imass denotes the CO2mass ratio, Ipressure the CO2 partial

pressure (in bar), and Iconcentration the CO2 molar concentra-

tion. Note that, considering environmental aspects, the car-natural gas feed rate and, by setting steam-to-methane ratio

(a design parameter), the required steam supply rate to the

process are both calculated. Several assumptions aremade for

design and analysis:

The heat interaction (transfer) among the heat exchangersis not considered, mainly because an energy analysis is not

the goal of the study.

The natural gas feed is sulfur-free, so a desulphurizationunit is not implemented.

The inlet water to the process is adequate for the process,so no treatment unit is added.

The hydrogen separation in the purifier (PSA) removes 90%of the hydrogen.

The product H2 stream is 100% pure with no othercontaminants.

The reformer reactor and the two shift reactors are equi-librium reactors.

The furnace is a Gibbs reactor (the presence of CO, H2, CO2makes a stoichiometric reactor model complicated andbon dioxide mass flow ratio is an attractive factor because

almost half of the carbon dioxide emissions is from the


Fig. 3 e Schematic of Aspen HYSYS mod

i n t e r n a t i o n a l j o u r n a l o f h yd r o g e n e n e r g y x x x ( 2 0 1 4 ) 1e1 0 5furnace for various S/C ratios (see Table 2). Thus, a higher Imassis of environmental importance in this study. In many SMR

plants, due to the following results, the preference is not to

remove CO2 fromflue gases. Moreover, in this study, a value of

unity for Ncc implies the availability of pure CO2 at 1 atm

pressure. However, the weights of all criteria are assumed the

same here (at unity).

As pointed out earlier, the hydrogen production rate is set

to 1 kg/h, so natural gas feed flow rate is adjusted. Then by

setting the S/C ratio to different values, the water flow rate is

determined. Heat integration is not incorporated into the

process modelling because an energy analysis is excludedfrom this study. In addition the fuel inlet to the furnace is

required to satisfy the heat flow required by the reformer and

Table 1 e Operating parameters considered for modelingSMR process.

Flow stream Baseline parameter value

Temperature (C) Pressure (bar)

Steam feed 510 30.0

NG feed 510 28.5

Mixed feed to reformer 649 27.0

Reformed gas 815 19.5

Cooled gas shift to HT shift 350 19.0

Cooled feed to LT shift 204 18.0

Shifted gas to purification 213 17.0

Dry syngas 38.0 16.6

Pure H2 38.0 1.60

PSA tail gas 38.0 1.00

Furnace fuel 25.0 1.00

Air inlet to furnace 25.0 1.00

Device Outlet temperature (C) Pressure drop (bar)

Reformer 815 1.72

HT shift 428 1.03

LT shift 213 1.03

Condenser 38.0 0.34

Please cite this article in press as: Soltani R, et al., Assessment oreforming for hydrogen production, International Journalj.ijhydene.2014.09.161the steam generator, and the air flow rate is summed to the

observation of 4% excess O2 in the combustion products.

So far, the effects have been studied of oxygen enrichment

and S/C ratio on carbon removal parameters. In this step,

which concentrates on carbon removal from flue gases, the

target is to analyse the hydrogen production rate compared

with carbon capture availability (Ncc) at the process end point

(flue gases) for various S/C ratios. For studying the effect on

hydrogen production of varying Ncc, another parameter needs

to be modified. For this purpose, the reformer temperature is

varied from 700 C to 1000 C for each S/C ratio, using thesimulated process. Note that, unlike for previous steps, the

el of SMR without heat integration.hydrogen production rate is fixed to permit determination of

the effect of only S/C ratio. In this parametric study, therefore,

the natural gas feed rate to the reformer is set to 1 kg/h and

consequently the hydrogen production rate changes. where

the hydrogen production rate was fixed, in this step of our

parametric study the natural gas feed rate to the reformer is

set to 1 kg/h and consequently the hydrogen production rate

changes. Other process settings are as mentioned in the cor-

responding sections. Further, for each S/C ratio, the resulting

optimum oxygen enrichment level from previous steps is

considered for the furnace air.

Table 2 e Natural gas consumption and CO2 emissionresults at selected ratios S/C.

Parameter S/C ratio

2.5 3.0 3.5 4.0 4.5

Reformer NG feed (kg/h) 2.89 2.87 2.84 2.73 2.65

Furnace fuel consumption (kg/h),

for air with 21% O2 by volume

0.00 1.50 2.30 3.05 3.50

Reactor conversion ratio (%) 73.0 77.9 81.7 84.8 87.38

Process CO2 emissiona (kg/h) 5.88 6.29 6.30 6.32 6.32

Overall CO2 emission (kg/h) 8.18 12.3 13.5 15.84 16.84

a Locations 1 & 2. Emissions from furnace are not considered.


Results and discussion

In first step the modelled process is run for S/C ratios from 2.5

to 4.5 (commonly used S/C ratios) and corresponding effects

are determined of conversion ratios in the reformer reactor

and CO2 partial pressures, concentrations and partial mass

flows. It is observed in Table 2 that, by increasing S/C, the

conversion ratio increases. This is due to an increase in the

availability of steam for reforming, which requires less natu-

ral gas feed to be used in the reactor for the same hydrogen

production and, consequently, less carbon dioxide emission

from the reactor. However, increasing the ratio S/C also raises

the energy requirement for steam generation and reforming.

Table 3 lists data for the three capture criteria considered in

typical air with 21% oxygen.

It is thus observed that the greater the ratio S/C, the lowerpartial pressure and higher molar concentration are of major

s in the furnace feed for air with 21% oxygen.

ration (%) Total pressure (bar) CO2 partial pressure (bar)

0 16.6 3.04

9 1.03 0.53

1 1.03 0.18

0 16.6 3.10

4 1.03 0.56

5 1.03 0.13

9 16.6 3.15

8 1.03 0.58

0 1.03 0.12

2 16.6 3.18

1 1.03 0.60

9 1.03 0.11

1 16.6 3.20

Fig. 4 e Effects of S/C ratio on two CO2 capture criteria at

location 1.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y x x x ( 2 0 1 4 ) 1e1 06is the natural gas usage. But this is not complete because, in

the furnace, the fuel consumption increases with S/C ratio.

Table 2 shows that for a ratio of 2.5, because of the PSA tail gas

contaminants (hydrogen, carbon monoxide and unreacted

methane), no additional external fuel is needed. For this ratio,

the PSA tail gas contains molar concentrations of around 20%

H2, 20% CO and the remainder water vapour and carbon di-

oxide. As S/C rises, the fuel injection to the furnace rises more

quickly than the rate of feed inlet to reformer decreases.

Therefore, higher steam-to-carbon ratios lead to higher car-

bon emissions to the atmosphere. Moreover, in cases where

more hydrogen production is needed, higher steam-to-carbon

ratios are advantageous. So far, it is observed that the relation

between S/C ratio and overall carbon dioxide emission is

linear. Now, we examine the effect of S/C ratio on overall

capture criteria which, for this study, are as follows:

1) CO2 partial pressure

2) CO2 molar concentration

3) CO2 mass ratio

Table 3 presents data for these parameters to identify for

which locations andwhich S/C ratios the carbon dioxide is in a

better condition to be captured. The conditions with higher

Table 3 e Characteristics of selected locations and S/C ratio

S/C ratio Location CO2 mass ratio (%) CO2 concent

2.5 1 73.0 17.9

2 73.0 51.4

3 100 17.8

3.0 1 57.0 18.1

2 57.0 52.8

3 100 12.8

3.5 1 51.0 18.1

2 51.0 53.4

3 100 11.8

4.0 1 46.0 18.2

2 46.0 53.6

3 100 11.0

4.5 1 44.0 18.22 44.0 53.39

3 100 10.78

Please cite this article in press as: Soltani R, et al., Assessment oreforming for hydrogen production, International Journaj.ijhydene.2014.09.161interests. Considering environmental aspects, higher ranges

of this criterion are better.

For better clarity, Figs. 4e7 illustrate the effect of increasing

S/C on several CO2 capture criteria. It is observed that the CO2concentration increases for process streams at locations 1 and

2 as S/C increases, yet it decreases at location 3 (flue gases).

This is due to fact that raising S/C increases the amount of

natural gas converted to hydrogen, which also produces more

carbon dioxide and raises its molar concentration at locations

1 and 2. An opposite result is observed in the furnace: the

addition of excess steam reduces the concentration of carbon

dioxide and, as a result of the higher fuel consumption and of

more parasitic contaminant nitrogen in the air inlet, more

inlet air is needed and the carbon dioxidemolar concentration

declines.

In Table 4, it can be seen that location 1 provides the best

condition for carbon dioxide removal for all five S/C ratios

tested. Based on the results, location 1 has better condition for

CO2 removal, almost twice as advantageous as location 2.

Compared to locations 1 and 2, location 3 is not appropriate for

CO2 capture, since it has the lowest value of Ncc. Another

result is that a lower S/C ratio implies better CO2 capture

possibilities. It may be asked why higher S/C ratios with lower1.03 0.62

1.03 0.11


the least likely capturing location. In this step, the effect

of oxygen enrichment of furnace feed is examined to deter-

mine its impact on Ncc of the flue gas stream. Oxygen

enrichment varies from 21% to 100% ideally for each of the S/C

ratios. Fig. 8 shows that for all S/C ratios the partial pressure of

CO2 increases but at a decreasing rate. From this behaviour

and the assumption that increasing enrichment raises costs,

the optimum oxygen enrichment level can be determined by

drawing tangent lines and perpendicular product to on the

curves. The optimum point selection is shown for one S/C

ratio (4.0) as an example (See Fig. 9). In Table 5, the oxygen

enrichment optimum levels for all S/C ratios are presented.

Again, an S/C ratio of 2.5 represents the best level due to


Table 4 e Values of Ncc at various S/C ratios.

S/C ratio Location

1 2 3

2.5 0.40 0.20 0.03

3.0 0.32 0.17 0.02

3.5 0.29 0.16 0.01

4.0 0.27 0.15 0.01

4.5 0.26 0.15 0.01

i n t e r n a t i o n a l j o u r n a l o f h yd r o g e n e n e r g y x x x ( 2 0 1 4 ) 1e1 0 7location 2.reformer natural gas feed and higher CO2 concentration and

partial pressures, leads to lower Ncc values? The answer is

that, if the environmental importance of CO2 mass ratio is

neglected, the higher S/C ratios lead to a higher Ncc because

both CO2 concentration and partial pressure increase. But,

higher S/C values, as stated earlier, concentrate CO2 emissions

in the furnace flue gases so, environmentally speaking, Imassshould be taken into consideration.

Based on results of previous section, CO2 removal

from location 3, being the source of 100% of emissions, is

permittingmore enrichment in a relatively economicmanner.

It can be seen from Table 6 that with enrichment of the

furnace feed, at optimum levels, a value of S/C of 2.5 is the best

condition for CO2 capture. This condition has the highest ef-

fect on Ncc, i.e. the optimum enrichment of the furnace feed


location 3.

Fig. 7 e Effects of S/C ratio on CO2 mass ratio at locations 1

and 2.

Please cite this article in press as: Soltani R, et al., Assessment oreforming for hydrogen production, International Journalj.ijhydene.2014.09.161Fig. 8 e Effects of oxygen enrichment of furnace feed on

CO2 partial pressure for various S/C ratios.Fig. 9 e Optimum oxygen enrichment selection.


increases Ncc from 0.03 to 0.10 (almost triple), while the effect

of enrichment on Ncc at location 3 for other S/C ratios is not

that significant compared to an S/C of 2.5. Overall, an S/C ratio

of 2.5 exhibits the best carbon capture condition and,

considering a higher weighting factor for the CO2 ratio of

location 3 (i.e., 2, 3, etc.) other than state 1, capturing would be

competitive with other locations. This higher weighting could

be a consequence of higher future carbon costs, incenting

industry to invest more on capturing carbon dioxide from flue

hydrogen in the PSA tail gases increases, which leads to higher

Fig. 10 e Effects of reformer temperature on mass flow rate

of produced H2.

Table 5 e Optimum oxygen enrichment for selected S/Cratios.

S/C ratio Optimum oxygen enrichment (%)

2.5 46

3.0 41

3.5 40

4.0 37

4.5 36

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y x x x ( 2 0 1 4 ) 1e1 08gases. Furthermore, it can be seen from Table 6 that the ideal

enrichment has a greater effect on Ncc at locations 1 and 2 at

higher S/C ratios (4.0 and 4.5). For instance, ideal enrichment

increases Ncc at location 1 at S/C 4.0 from 0.27 to 0.44, whichis significant, while it raises the carbon capture number at

location 1 for an S/C ratio of 3.0 from 0.34 to 0.47. Also, the

effect of ideal enrichment on Ncc of the flue gas stream is

almost negligible for high S/C ratios.

Fig. 10 shows that higher reformer temperatures result in

higher hydrogen production rates, directly due to the higher

corresponding conversion ratios in the reformer reactor as

result of the higher temperatures. However, at temperatures

above around 850 C, the rate of increase decreases. It can beseen that the S/C ratio has a direct effect on hydrogen pro-

duction, in that high ratios facilitate high production rates,

primarily as a consequence of the high availability of steam

for natural gas feed in the reactor. Note that the natural gas

feed rate is constant for all conditions. So far, it is observed

that high reforming temperatures and high S/C ratios are

Table 6eNcc for selected S/C ratios for optimum and idealoxygen enrichment.

S/C ratio Location Carbon capture number, Ncc

Optimumenrichment

Ideal enrichment(100%)

2.5 1 0.42 0.482 0.21 0.24

3 0.10 0.17

3.0 1 0.34 0.47

2 0.18 0.25

3 0.04 0.07

3.5 1 0.30 0.46

2 0.16 0.25

3 0.03 0.05

4.0 1 0.27 0.44

2 0.15 0.24

3 0.03 0.04

4.5 1 0.25 0.42

2 0.14 0.24

3 0.03 0.03

Please cite this article in press as: Soltani R, et al., Assessment oreforming for hydrogen production, International Journaj.ijhydene.2014.09.161more appropriate for hydrogen production. Now, the effects of

S/C ratio and reforming temperature are examined as they

relate to carbon removal, via determination ofNcc (see Fig. 11).

Fig. 11 shows that, for all S/C ratios, as reactor temperature

rises, Ncc increases, peaks at some level and thereafter de-

creases. The peak temperature as seen in Fig. 11, is at higher

temperatures for lower S/C ratios. For an S/C ratio of 2.5, for

instance, the peak occurs at 900 C, while it occurs at 800 C forS/C 4.5. It is observed in the previous sections that lower S/Cratios provide better carbon capture conditions, and similar

results are observed here by varying reactor temperatures.

Nonetheless, at an S/C ratio of 3.0, unexpected behaviour is

observed for a temperature of 850 C. Unlike product hydrogenrate, which benefits from a high S/C ratio and a high reform

temperature, lower S/C ratios are more favourable regarding

carbon capture criteria and high reforming temperatures, over

a peak level, have negative effects on carbon capture. Also, the

following observed factor is important: when temperature

increases in the reformer andmore hydrogen is produced, PSA

tail gases channelled to the furnace carry higher amounts of

H2 because the hydrogen purifier is less than 100% effective.

This effect can partially offset the higher required combustion

energy for the provision of higher reforming temperatures.

The remaining required energy can be met by adding backup

fuel to the furnace. The reason for the peak in Fig. 11 is that,

above a certain level, the ratio of additional required energy in

the reformer (more combustion needed) and additionalFig. 11 e Effects of reformer temperature on Ncc.


C ratios result in better carbon removal conditions, the ratio of

hydrogen production rate to Ncc is less for those S/C ratios.

i n t e r n a t i o n a l j o u r n a l o f h yd r o g e n e n e r g y x x x ( 2 0 1 4 ) 1e1 0 9Also, at high S/C ratiosmore hydrogen is produced, raising the

ratio of hydrogen production rate to Ncc.

Conclusions

The analysis and assessment performed in this comparative

study lead to the following main conclusions:

Higher S/C ratios provide higher CO2 partial pressures andconcentrations for synthesis gas and PSA tail gas streams.

Considering the CO2 presence at different locations, higherS/C ratios decrease the CO2 removal possibility byfuel and air intakes and consequently more nitrogen as

parasitic substance. Overall, it results in a lower value of Ncc.

One other important factor is examined: the comparative

behaviour of hydrogen production and carbon removal. The

aim of this task is to determine at what condition the process

is more appropriate in terms of hydrogen production and

what conditions result in better carbon removal. Fig. 12 shows

that reforming temperature continues to have a positive effect

on hydrogen production, except for an S/C ratio of 2.5 at

800 C, for which a slight drop is observed. Also, since lower S/

Fig. 12 e Effects of varying reformer temperature on H2/Nccratio.decreasing the value of criterion Ncc at location 3, due to

the concentration of CO2.

If less pollution is desired (i.e., less overall natural gasconsumption in the reformer and the furnace), then an S/C

ratio of 2.5 is best.

If the hydrogen production rate is of more importance thanenvironmental pollution, then an S/C ratio of 4.5 is best.

Oxygen enrichment is found to have highest positiveimpact on Ncc for an S/C ratio of 2.5, for location 3.

If a weighting system is considered for the three studiedCO2 capture criteria changes, the results change notably.

For instance, a weighting factor of more than unity for

complete CO2 mass ratio at location 3 (which depends on

the pollution regulations of a region) raises the CO2 capture

from this location relative to the other two locations.

High S/C ratios increase hydrogen production rates, as aconsequence of the high availability of steam for natural

gas feed in the reactor.

1e10; 2005 [Costa Verde, Brazil].[16] Park JH, BeumHT, Kim JN, Cho SH. Numerical analysis on the

Please cite this article in press as: Soltani R, et al., Assessment oreforming for hydrogen production, International Journalj.ijhydene.2014.09.161power consumption of the PSA process for recovering CO2from flue gas. Ind Eng Chem Res 2002;41:4122e31.

[17] Ko D, Siriwardane R, Biegler LT. Optimization of a pressure-swing adsorption process using zeolite 13X for CO2sequestration. Ind Eng Chem Res 2003;42(2):339e48.

[18] Voss C. CO2 removal by PSA: an industrial view onopportunities and challenges. Adsorption2014;20(2e3):295e9. High reformer temperatures result in higher hydrogenproduction rates, directly due to higher corresponding

conversion ratios in the reformer reactor as a result of the

higher temperature.

The highest value of Ncc results for an S/C ratio of 2.5 fortemperatures between 850 C and 950 C.

r e f e r e n c e s

[1] Meerman JC, Hamborg ES, Van Keulen T, Ramirez A,Turkenburg WC, Faaji APC. Techno-economic assessment ofCO2 capture at steam methane reforming facilities usingcommercially available technology. Int J Greenh Gas Control2012;9:160e71.

[2] IEA. Energy technology perspective. Paris: InternationalEnergy Agency (IEA); 2010.

[3] IEA. Energy technology analysis. CO2 capture and storage: akey abatement option. Paris: International Energy Agency(IEA); 2008.

[4] IEA. Global action to advance carbon capture and storage.Paris: International Energy Agency (IEA); 2013.

[5] IEA. CCS 2014. Paris: International Energy Agency (IEA); 2014.[6] Dufour J, Serrano DP, Galvez JL, Moreno J, Garcia C. Life cycle

assessment of process for hydrogen production:environmental feasibility and reduction of greenhouse gasemission. Int J Hydrogen Energy 2009;34:1370e6.

[7] Muradov N. Hydrogen via methane decomposition: anapplication for decarbonization of fossil fuels. Int J HydrogenEnergy 2001;26:1165e75.

[8] IEA. Energy technology perspective. Paris: InternationalEnergy Agency (IEA); 2008.

[9] Kikkinides ES, Yang RT, Cho SK. Concentration and recoveryof CO2 from flue gas by pressure swing adsorption. Ind EngChem Res 1993;32:2714e20.

[10] Plasynski SI, Damiani D. Carbon sequestration throughenhanced oil recovery. Washington, D.C: U.S. Department ofEnergy (DOE); 2008.

[11] Simbeck DR. Hydrogen costs with CO2 capture. In: Proceedingof the 7th international conference on greenhouse gas controltechnologies, 1059e1066. Vancouver; 2005.

[12] Folger P. Carbon capture: a technology assessment (CRSreport 7e5700). 2013. Retrieved from Online Information forDefense Community, Defense Technical Information Center(DTIC), Accession Number: ADA590346.

[13] Global CCS Institute. The global status of CCS: 2011. 2011.Canberra, Australia.

[14] Rao AB, Rubin ES. A technical, economic, and environmentalassessment of amine-based CO2 capture technology forpower plant greenhouse gas control. Environ Sci Technol2002;36:4467e75.

[15] Grande CA, Cavenati S, Rodrigues AE. Pressure swingadsorption for carbon dioxide sequestration. In: Proceedingof the 2nd Mercosur Congress on chemical engineering,[19] Damen K, Van Troost M, Faaji A, Turkenburg W. Acomparison of electricity and hydrogen production systems


with CO2capture and storage. Part A: review and selection ofpromising conversion and capture technologies. Prog EnergyCombust Sci 2006;32:215e46.

[20] Lambert J, Sorin M, Paris J. Analysis of oxygen-enrichedcombustion for steam methane reforming (SMR). Energy1997;22(8):817e25.

[21] Ding FC, Yi YF. Technology of hydrogen production andstorage. Beijing: Chemical Industry Press; 2006.

[22] Simpson AP, Lutz AE. Exergy analysis of hydrogenproduction via steam methane reforming. Int J HydrogenEnergy 2007;32:4811e20.

[23] Molburg JC, Doctor RD. Hydrogen from steam-methanereforming with CO2 capture. In: Paper presented at: 20th

annual international Pittsburgh coal conference. Pittsburg;2003.

[24] Colladi G, Wheeler F. Hydrogen production via steamreforming with CO2 capture. In: Buratti SS, editor. Chemicalengineering transactions, 19; 2010. p. 37e42.

[25] Haws R. Contaminants in amine gas treating. In: Paperpresented In: Gas Processors Association (GPA) HoustonRegional Meeting, Houston; 2001.

[26] Golan AZ, Kepler MH. Membrane based air separation. AIChESymp Ser 1986;82(250):35e7.

[27] Aspen HYSYS V7.3. 2014. Retrieved June 16, 2014, from,http://www.aspentech.com/products/aspen-hysys.aspx.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y x x x ( 2 0 1 4 ) 1e1 010Please cite this article in press as: Soltani R, et al., Assessment oreforming for hydrogen production, International Journaj.ijhydene.2014.09.161f CO2 capture options from various points in steam methanel of Hydrogen Energy (2014), http://dx.doi.org/10.1016/

Assessment of CO2 capture options from various points in steam methane reforming for hydrogen productionIntroductionMethods consideredSteam methane reformingCarbon dioxide captureOxygen enrichment

Model descriptionApproachResults and discussionConclusionsReferences

1-s2.0-S0360319914027566-main

Documents

Transcript of 1-s2.0-S0360319914027566-main