BTEX Prediction & Removal in Amine Units Luke Burton Chad Duncan Armando Diaz Miguel Bagajewicz.

BTEX Prediction & Removal in Amine Units

Luke BurtonChad Duncan

Armando DiazMiguel Bagajewicz

Project Objective To study means of reducing incineration expenditures

associated to BTEX capture in Amine units, through Process parameter optimization Alternative/Additional Technologies to capture BTEX

Specifics: BTEX content of needs to be kept under the EPA emission limit

of 25 Ton/year . If this is achieved, a reduction in incineration temperature

from 1500 oF to 1350 oF can be accomplished with an associated savings of $303 Thousand.

Alternative Technologies, if they exist, ought to have a lower cost.

Project Methodology Discuss Existing Simulators and compare their capabilities of

predicting Acid Gas flowrate and composition BTEX capture

Determine ways of using these simulators to make approximate predictions

Assess the ability of process parameter manipulation to achieve the reduction of BTEX capture goal.

Study Alternative Technologies Adsorbents Ionic Liquids

Modeling Objective Commercial Simulators seem not to reproduce reliable

results in the case of Amine units, especially when BTEX capture is of interest.

Ideal Objective: Have a simulator that will use the right thermodynamic equation of state and liquid activity coefficients

Achievable Objectives: Use existing simulators and supersede them with additional data and make conclusions.

Amine Plant at Glance

BTEX Problems in Amine Unit

Flash Drum BTEX is emitted to the atmosphere, possible violating EPA

guidelines.

Acid gas stream BTEX present has to be incinerated at high temperatures,

therefore incurring a high fuel cost.

Sweet gas stream Some BTEX will be present, so it is removed in glycol unit.

PRO II Amine Unit Simulation

Same inlet conditions were used: Feed gas (575MMSCF), T (85°F), P (500psia), same compositions.

Results were compared to 92 wt% of CO2 usually found in acid gas stream.

AmineCalc Amine Unit Simulation

Same inlet conditions were used: Feed gas (575MMSCF), T (85°F), P (500psia), same compositions.

Results were compared to 92 wt% of CO2 usually found in acid gas stream.

CO2 Results from Pro II and AmineCalc

Feed Sweet Gas Acid GasComponents AmineCalc Pro/II AmineCalc Pro/II AmineCalc Pro/II

(mol%) Uncontrolled Controlled CO2 9.37 3.12E-09 1.33 99.95 85.54 87.05

Methane 89.57 9.83E+01 96.71 5.00E-02 1.87 0.392Ethane 0.746 0.816 0.804 0.0004 0.0277 8.52E-03

Propane 0.13 0.143 0.14 0.00006 0.00319 1.36E-03i-Butane 0.025 0.0275 2.70E-02 0 0.000326 1.71E-04n-Butane 0.025 0.0272 2.69E-02 0 0.00174 4.59E-04i-Pentane 0.046 0.0505 4.96E-02 0 0.000836 3.66E-04n-Pentane 0.005 0.0055 5.40E-03 0 7.36E-05 3.54E-05

Hexane 0.009 0.00992 9.72E-03 7.11E-07 5.32E-05 2.96E-05Heptane 0.005 0.00552 5.40E-03 0 1.40E-05 1.04E-05Octane 0.01 0.011 1.08E-02 0 2.19E-05 1.24E-05Nonane 0.008 0.00884 8.65E-03 0.00E+00 2.87E-06 2.31E-06Benzene 0.0004 2.40E-05 4.02E-04 5.00E-05 3.41E-03 2.85E-04Toluene 0.0005 0.000311 5.22E-04 1.00E-05 1.94E-03 1.66E-04

Ethylbenzene 0 0 0 0.00E+00 0.00E+00 Xylene 0.0002 9.14E-05 2.06E-04 0.00E+00 1.05E-03 8.99E-05

N2 0.05 9.14E-05 5.38E-02 0.00E+00 1.05E-03 1.04E-03H2O 0 9.14E-05 0.814 0.00E+00 1.05E-03 12.55

MDEA 0 9.14E-05 1.15E-04 0.00E+00 1.05E-03 7.15E-17

Credibility

Which simulator is correct?

AmineCalc renders 99 wt% of CO2 in the acid gas

Pro II renders 94 wt% of CO2 in the acid gas, closer to the 92 wt % reported from field data.

Thermodynamic packages in AmineCalc and Pro II might explain why.

EOS in AmineCalc Uses Peng-Robinson

equation of state.

Not as thorough as Pro II as far as the thermodynamics. Binary interaction coefficient calculated by using simple cubic mixing rule.

Mixing rule have been shown to be incapable of modeling real systems.

bVbbVV

a

bV

RTP

ˆˆˆˆ

EOS in Pro II Pro II uses SRKM equation

of state to calculate the vapor phase enthalpy and density, and liquid and vapor phase entropy.

ω, cij, kij, b, are parameters that are easily obtained.

Binary interaction coefficients mixing rule developed by Prausnitz, and shown to perform better than simple cubic mixing rule.

ˆ ˆ ˆRT a

PV b V V b

12

21

2

2

1

1 1

0.480 1.574 0.76

ijc

iij i j ij ij ji

i j

r

xa a a k k k

x x

T M T

M

BTEX Predictions Feed Sweet Gas Acid Gas

Components AmineCalc Pro/II AmineCalc Pro/II AmineCalc Pro/II(mol%) Uncontrolled Controlled

CO2 9.37 3.12E-09 1.33 99.95 85.54 87.05Methane 89.57 9.83E+01 96.71 5.00E-02 1.87 0.392Ethane 0.746 0.816 0.804 0.0004 0.0277 8.52E-03

Propane 0.13 0.143 0.14 0.00006 0.00319 1.36E-03i-Butane 0.025 0.0275 2.70E-02 0 0.000326 1.71E-04n-Butane 0.025 0.0272 2.69E-02 0 0.00174 4.59E-04i-Pentane 0.046 0.0505 4.96E-02 0 0.000836 3.66E-04n-Pentane 0.005 0.0055 5.40E-03 0 7.36E-05 3.54E-05

Hexane 0.009 0.00992 9.72E-03 7.11E-07 5.32E-05 2.96E-05Heptane 0.005 0.00552 5.40E-03 0 1.40E-05 1.04E-05Octane 0.01 0.011 1.08E-02 0 2.19E-05 1.24E-05Nonane 0.008 0.00884 8.65E-03 0.00E+00 2.87E-06 2.31E-06Benzene 0.0004 2.40E-05 4.02E-04 5.00E-05 3.41E-03 2.85E-04Toluene 0.0005 3.11E-04 5.22E-04 1.00E-05 1.94E-03 1.66E-04

Ethylbenzene 0 0 0 0.00E+00 0.00E+00 Xylene 0.0002 9.14E-05 2.06E-04 0.00E+00 1.05E-03 8.99E-05

N2 0.05 9.14E-05 5.38E-02 0.00E+00 1.05E-03 1.04E-03H2O 0 9.14E-05 0.814 0.00E+00 1.05E-03 12.55

MDEA 0 9.14E-05 1.15E-04 0.00E+00 1.05E-03 7.15E-17

USE OF EXTERNAL DATA

We used the solubility data found in Developments and Applications in Solubility. (Coquelet et. al. 2007)

In this book, the activity coefficients of benzene, toluene, ethylbenzene, and xylene are calculated experimentally for different mixtures of MDEA/DGA and Water.

Contactor Tower Results

Experimental results can be used to calculate how accurate are the simulator results; more specifically the molar composition in the liquid stream.

1

2

3

4

5

6

T1

S1

S2

S3

S4

Feed GasLiquid tray 5

Vapor tray 6

Bottoms liquid

Contactor Tower

Sweet Gas Amine

Contactor Data 575MMSCFD Feed and 702 MGal/hr MDEATray 6 T (F) 145

Tray 6 P (psia) 250 G L2 V L1

Benzene (mol%) 4.00E-04 8.67E-06 4.00E-04 9.13E-06Toluene (mol%) 5.00E-04 5.42E-06 5.02E-04 5.65E-06

Ebenzene (mol%) 0 0 0 0Xylene (mol%) 2.00E-04 3.16E-06 2.02E-06 2.97E-06

Flow Rate (g-mol/hr) 2.86E+07 8.87E+07 2.85E+07 8.88E+07


1

2

3

4

5

6

T1

S1

S2

S3

S4

GL2

L1

V

Flash

G

L2

L1

V


Flash

G

L2

L1

V

1

89.219

53.120389.6

2

1

2

12

12

10*

1

1

1

1

LP

TV

xLGyx

LPPT

V

xLGyx

xLP

PxTVxLGy

PxTPy

xLVyxLGy

TBTEX

LBTEX

GBTEX

BTEX

satBTEX

LBTEX

GBTEX

BTEX

BTEX

satBTEXBTEXL

BTEXGBTEX

satBTEXBTEXBTEX

BTEXBTEXLBTEX

GBTEX

Regenerator Tower Results

Experimental results can be used to calculate how accurate are the simulator results; more specifically the molar composition in the acid gas stream.

Regenerator Tower

Acid Gas

Vapor tray 3Liquid tray 2

Rich Amine

Lean Amine

Contactor Data 575MMSCFD Feed and 702 MGal/hr MDEA

Tray 2 T (F) 211 Tray 2 P (psia) 15.5

V' L1 V Benzene (mol%) 2.01E-05 8.67E-06 2.85E-04 Toluene (mol%) 1.17E-05 5.42E-06 1.66E-04

Ebenzene (mol%) 0 0 0 Xylene (mol%) 6.37E-06 3.16E-06 8.99E-05

Flow Rate (g-mol/hr) 3.78E+07 8.87E+07 2.66E+06

RESULTS

BTEX Concentration from Experimental Results

ComponentContactor

Calculated Pro II

xi

Benzene 1.17E-06 8.67E-06

Toluene 5.99E-07 5.42E-06

EthylBenzene 0 0

Xylene 2.78E-07 3.16E-06

yi

Benzene 2.46E-04 2.85E-04

Toluene 1.46E-04 1.66E-04

EthylBenzene 0 0

Xylene 8.10E-05 8.99E-05

BTEX Concentration from Experimental Results

ComponentRegenerator

Calculated Pro II

xi

Benzene 4.79E-05 9.13E-06

Toluene 9.96E-05 5.65E-06

EthylBenzene 0 0

Xylene 5.44E-05 2.97E-06

yi

Benzene 2.80E-04 4.00E-04

Toluene 9.55E-04 5.02E-04

EthylBenzene 0 0

Xylene 2.35E-04 2.02E-06

CONCLUSION It is our belief that Pro II produces good answers for flows and CO2

concentrations in the amine unit.

Pro II and AmineCalc overestimates the solubility of BTEX in the contactor.

• We do not have the right thermodynamics in Pro II or AmineCalc, or any simulator.

• Despite the above, we have a credible way of estimating solubilities based on experimental data.

Glycol Dehydration Units

Unit removes water from sweetened natural gas.

Glycols such as DEG or TEG usually used for these tasks.

Two commercially available simulators: GlyCalc and Pro II.

Interfaces for Glycalc and Pro II are shown.

Glycol Dehydration Units

Unit removes water from sweetened natural gas.

Glycols such as DEG or TEG usually used for these tasks.

Two commercially available simulators: GlyCalc and Pro II.

Interfaces for Glycalc and Pro II are shown.

Milagro Data:49 MMSCFD104 °F887 psig10gal/min glycol382 °F

GlyCalc Contactor Tower

In contactor tower, VLE calculations using Kremser-Brown approximation.

Approximation used to calculate K-values.

Contactor tower not rigorously modeled by using stage by stage flash calculation.

L and V is assumed to be average in every stage.

GlyCalc Regenerator For regenerator, manual notes:

“to avoid complex heat and material balances that would be needed if the regenerator were rigorously modeled, a simple empirical calculation is used”

ResultsContactor Tower on Dehydration Unit

Feed Dry GasComponents GlyCalc Pro II Milagro Data

Methane 98.8880 98.9 98.880 98.9160Ethane 0.8164 0.816 0.816 0.7994Propane 0.1605 0.16 0.160 0.1556

Isobutane 0.0263 2.630E-02 2.619E-02 2.51E-02n-Butane 0.0262 2.620E-02 2.603E-02 2.53E-02

Neopentane 0.0003 N/A 5.213E-03 3.00E-03Isopentane 0.0086 N/A 8.581E-03 8.30E-03n-Pentane 0.0053 5.290E-03 5.286E-03 5.10E-03

2,2-Dimethylbutane 0.0003 N/A 2.838E-04 3.00E-042,3-Dimethylbutane 0.0006 N/A 5.524E-04 6.00E-04

2-Methylpentane 0.0017 N/A 1.571E-03 1.60E-033-Methylpentane 0.0009 N/A 8.094E-04 9.00E-04

n-Hexane 0.0016 1.600E-03 1.584E-03 1.50E-03Heptanes 0.0051 5.070E-03 5.029E-03 5.20E-03Octanes 0.0003 5.920E-03 2.895E-03 5.00E-03Nonanes 0.0006 N/A 5.582E-04 3.00E-04

Decanes plus 0.0004 N/A 3.524E-04 1.10E-03Nitrogen 0.0569 5.690E-02 5.687E-02 5.29E-02

Carbon Dioxide 0.0000 0 0 0Oxygen 0.0000 0 0 0Water N/A 4.860E-03 5.804E-03 N/A

Benzene 3.000E-04 2.720E-04 2.189E-04 3.000E-04Toluene 5.000E-04 4.220E-04 3.281E-04 4.000E-04

Ethylbenzene 0 0 0 0Xylene 6.000E-04 3.930E-04 2.459E-04 3.00E-04

2,2,4-Trimethylpentane 1.000E-04 9.980E-05 9.912E-05 1.00E-04Cyclopentane 0 0 0 0Cyclohexane 9.000E-04 8.880E-04 8.744E-04 9.00E-04

Methylcyclohexane 1.000E-03 9.840E-04 9.851E-04 1.10E-03

Conclusions GlyCalc produces better results for BTEX in dehydration

unit.

We believe Glycalc would be able to predict the amount of BTEX present in dehydration unit.

GlyCalc would not be able to accurately predict duty in regenerator due to its simple correlation used for energy balance.

BTEX Solutions

Reduction Possibilities Two different ways to remove amine exist

Reduce absorption in amines Certain parameters can obtain this

Remove BTEX prior/post amine unit treating Solvent Alternative Technologies

First Solution

Changes in parameters such as amine flow rate, temperature and pressure of towers, etc. may reduce BTEX capture.

We performed a few simulations in Pro II to get a preliminary sensitivity analysis for the affect of temperature.

Parameter Adjustments

It is our belief that this route will not solve the emission problems.

50 70 90 110 130 150 1700.25

0.27

0.29

0.31

0.33

0.35

BTEX Emission Versus Amine Temp

BTEXCO2 Cutoff

Amine Temperature (◦F)

Bte

x E

mis

sio

n (

lb-m

ol/

hr)

Second Solution Solvents can be used:

Water

Alternative Technology Adsorbents

Activated Carbon Silica Aerogels Macroreticular Resins

Ionic Liquids

Removal by Solvent Removal By Water

0 20 40 60 80 100 120 140 160 180 2000.00000

0.00500

0.01000

0.01500

0.02000

0.02500

0.03000

0.03500

Benzene Composition Vs. Water Flow

Water Flow (MMSCFD)

Ben

zen

e M

ola

r C

om

posit

ion

CONCLUSION Manipulating the amine unit parameters (T, P, and flow

rates) will not lead to the order of magnitude changes needed to reduce the emission.

This conclusion is based both on considering results of Pro II directly and calculations based on experimental results.

Water is also not a good solvent to remove BTEX due to separation complications.

This leads to the investigation of other alternative technologies

Activate Carbon Activated Carbon has a

density of about 350 kg/m3 and surface area of 500 m2/g

Can only be used 2 cycles before 50% adsorption reduction occurs

Macroreticular Resins Macroreticular resins have an

adsorption of BTEX of about: 350 mg BTEX/1000 mg of

adsorbent *(Lin (1999))

Can adsorb and desorb BTEX for 42 cycles before a 10% reduction in adsorption

Silica Aerogels (SAG) Hydrophobic material that

has low density (0.3-0.05 g/cm3), high porosity, and high surface area (500- 1000 m2/g).

SA can be used up to 14 cycles!

Incineration Results From Pro/II, it was calculated how much fuel gas

(methane) will be needed to fully incinerate the acid gas stream at 1500°F by using a Gibbs reactor.

These calculations were based from on the following field data:

Air Fuel Acid Gas

Flowrate (ft^3/hr) 355,888 23,162 504,042

Methane 0% 100% 0.50%

Carbon Dioxide 0% 0% 84.42%

Nitrogen 78.11% 0% 0%

Oxygen 20.95% 0% 0%

Argon 0.93% 0% 0%

Water 0% 0% 15.08%

Flame Temperature Verification

We took initial and final moles from Pro II. Reaction was carried while keeping the vent gas temperature at 1500°F.

Pro II results agreed with field data within 1.3% margin of error.

Pro II results agreed with hand calculations within .64% margin of error.

)(*)(*)(*)(*

)(*)(*)(*)(*

222222

222244

,,,,

,,,,

f

o

f

o

f

o

f

o

i

o

i

o

i

o

i

o

T

TCOffCO

T

T

fN

T

T

fO

T

TOHffOH

T

TCOfiCO

T

T

iN

T

T

iO

T

TCHfiCH

HdTCpNdTCpNdTCpNHdTCpN

HdTCpNdTCpNdTCpNHdTCpN

Excess Air Limits The Limit of excess is such that the mole percent of

oxygen released to the atmosphere must be between 1-3% (Lewandowski, 2000). Lower limit due to formation of CO below 1% O2

Upper limit exist to reduce formation of NOx which occur above 3% oxygen

This data is backed by Ignacio plant data with O2 level of 2% in outlet stream

Flame Temperature VOC

The flame of incinerator must be risen to a temperature, Auto Ignition Temperature, high enough to combust VOC’s:

In order to incinerate at this temperature, long residence times in incinerator must be used A common rule of thumb for 99% incineration efficiency at .5

seconds is to add 400°F onto AIT.

Compound AIT (°F)

Benzene 1097

Ethylbenzene 870

Toluene 997

Xylene 924* (Lewandowski, 2000).

Fuel CostThermal Oxidizers Analysis

With BTEX

Amount of CH4 (MMft^3/year) Cost per year Cost per Day

221 $1,117,00 $3,000

Without BTEX (Constant Air Excess Assumed)

Amount of CH4 (MMft^3/year) Cost per year Cost per day

161 $814 $2,000

Saving per Year ($)

$303,000

**Cost of Methane at $5/MMBtu**

SAG Adsorption Process

One tank opened while the other is closed, and they will run for 12 hr periods.From the columns, the BTEX can be removed by using three different designs. These columns could be used up front of amine unit or in Acid Gas.

Column 1

Column 2

To Amine Unit/ Oxidizer

To Design

Acid Gas/Raw Gas

Comparison of Two Designs

Removing the BTEX present in the columns by blowing air through the columns.

Instead of burning the air/BTEX stream, run the stream through a condenser, and then pass it through a flash.

Activated Carbon Acid Gas

Desorb and Burn

Columns $372,000

Blower $7,000

Piping $379,000

Total FCI $636,000

Materials $257,000

Labor $38,000

Fuel $5,000

Total Operating Cost $451,000

Total Annualized Cost $493,000

Activated Carbon cost $4 per kg.

Used Pro-II Results from Milagro Type Plant

This design would have an additional cost of $191,000

In order for a saving of $100,000 to be reached price would have to be reduced to $1.15 per kg

71% discount needed

Silica AerogelsAcid Gas

Silica Aerogels cost $37 per kg from Cabot.


This design would produce a savings of $76,000

In order for a saving of $100,000 to be reached price would have to be reduced to $34 per kg

8% discount needed

Desorb and Burn

Columns $373,000

Blower $7,000

Piping $258,000

Total FCI $638,000

Materials $164,000

Labor $37,000

Fuel $5,000



Macroreticular ResinsAcid Gas

Macroreticular resins cost $43 per kg from Dow Chemical.


This design would produce a savings of $61,000

In order for a saving of $100,000 to be reached price would have to be reduced to $35 per kg

19% discount needed

Desorb and Burn

Columns $165,000

Blower $7,000

Piping $117,000

Total FCI $289,000

Materials $181,000

Labor $37,000

Fuel $5,000



Conclusions from Adsorption

There exist a saving of $303,000 in reducing the flame temperature from 1500°F to 1350°F.

This savings can then be used to design adsorption columns to remove BTEX.

Out of all the adsorbents studied silica aerogels proved to be the best adsorbent on the basis of savings and reduced cost.

Ionic Liquid Background

Ionic liquids can be used to remove carbon dioxide.

The expense of using these liquids will be examined in comparison with that of the amine unit.

Amine Unit CostAmine Unit First batch costs

First batch amine $596,000 Total $596,000

Equipment CostsAbsorption tower $1,336,000Stripping tower $210,000Heat exchangers $527,000Pump $42,000Condensers $115,000Reboilers $126,000 Total $2,355,000

Operation Costs $/year

Process water that is lost $2,873,000

Process amine that is lost $415,000

Heat at the reboiler $26,502,000

Electricity for pump $2,168,000Condenser Fan Electricity $11,000

Total $31,969,000

Hydrocarbon Losses $/yearMethane loss $329,203Ethane loss $13,290

Total Annualized Cost$32,508,000

Ionic Liquid Conclusion

0.00% 0.50% 1.00% 1.50% 2.00% 2.50% 3.00% 3.50% 4.00% 4.50% 5.00%$0

$5,000,000

$10,000,000

$15,000,000

$20,000,000

$25,000,000

$30,000,000

$35,000,000

TAC versus % Methane Loss

% Methane Lost

To

tal

An

nu

ali

ze

d C

ost

Current Design 2.37% Loss

Amine TAC

IL TAC

Ionic Liquid Conclusion

$0 $5 $10 $15 $20 $25 $30 $35 $40 $45 $50$0

$500,000

$1,000,000

$1,500,000

$2,000,000

$2,500,000

FCI versus IL cost

IL Cost, $/kg

FC

I

Amine FCI

IL FCI

QUESTIONS?

BTEX Prediction & Removal in Amine Units Luke Burton Chad Duncan Armando Diaz Miguel Bagajewicz.

Documents

Transcript of BTEX Prediction & Removal in Amine Units Luke Burton Chad Duncan Armando Diaz Miguel Bagajewicz.