IECM Technical Documentation: CO 2 Purification Unit (CPU) Models January 2019
IECM Technical Documentation:
CO2 Purification Unit (CPU) Models
January 2019
IECM Technical Documentation:
CO2 Purification Unit (CPU) Models
Prepared by:
Hari C. Mantripragada
Edward S. Rubin
The Integrated Environmental Control Model Team
Department of Engineering and Public Policy
Carnegie Mellon University
Pittsburgh, PA 15213
www.iecm-online.com
Compiled in January 2019
Integrated Environmental Control Model - Technical Documentation Table of Contents • iii
Table of Contents CO2 Purification Unit 1
Objective ............................................................................................................1 CPU Performance Models .................................................................................1
Flue Gas Compressor and Drying Model ..............................................3 Partial Condensation Model ...................................................................4 Distillation Model ..................................................................................8
CO2 Compressor Model .......................................................................10 CPU Efficiency Factor .....................................................................................10 Order of Calculations .......................................................................................11 CPU Cost Model ..............................................................................................12 User-Specified Configuration ..........................................................................13
IECM Screens ..................................................................................................14 Case Studies .....................................................................................................15
References ........................................................................................................17
Integrated Environmental Control Model - Technical Documentation List of Figures • iv
List of Figures
Figure 1. CO2 purity requirements for different applications (NETL QGESS, 2014) ......................... 2
Figure 2. Schematic of a typical CPU process (NETL, 2010) .............................................................. 3
Figure 3. Variation of CO2 purity and recovery with pressure and flash temperature ......................... 6
Figure 4. Variation of CO2 recovery with pressure and flash temperature ........................................... 7
Figure 5. CO2 recovery and purity for condensation and distillation ................................................... 9
Figure 6. Regression equations for capital cost of CPU ..................................................................... 13
Figure 7. "Set Parameters" config screen for the CPU unit ................................................................ 14
Figure 8."Set Parameters" purification screen for the CPU unit, 99.99 percent purity case .............. 14
Figure 9. "Set Parameters" purification screen for the CPU unit, high-purity case ............................ 15
Figure 10. "Set Parameters" purification screen for the CPU unit, low-purity case ........................... 15
Figure 11. Sensitivity analysis results ................................................................................................. 16
Figure 12. Sensitivity analysis for the CPU case studies .................................................................... 16
Integrated Environmental Control Model - Technical Documentation List of Tables • v
List of Tables
Table 1. Regression coefficients for the partial condensation model ................................................... 5
Table 2. Regression equations for the combined condensation and distillation model ........................ 9
Table 3. Case studies to calculate CPU efficiency factor ................................................................... 11
Table 4. Relevant flow rates and capital costs of CPU from the 2010 DOE Oxy-Combustion Report
...................................................................................................................................................... 12
Table 5. Case study input assumptions and results ............................................................................. 16
Integrated Environmental Control Model - Technical Documentation Acknowledgements • vi
Acknowledgements
This work was supported by the National Energy Technology Laboratory (NETL). The authors also
acknowledge Dr. Haibo Zhai and Karen Kietzke for their coding work in incorporating the
performance and cost models into the Integrated Environmental Control Model (IECM). Any
opinions, findings, and conclusions or recommendations expressed in this material are those of the
authors alone and do not reflect the views of any agency.
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 1
CO2 Purification Unit
Objective
This documentation reports the performance and cost models of a carbon dioxide (CO2) purification
unit (CPU) used for oxy-combustion and some post-combustion CO2 capture technologies at coal-
fired power plants.
CPU is needed to increase the purity of CO2 product streams to meet appropriate pipeline standards.
Some of these standards are prescribed in the National Energy Technology Laboratory’s (NETL)
Quality Guidelines for Energy Systems Studies (QGESS) document on CO2 purity requirements, as
shown in Figure 1 (NETL, 2013). Based on the specifications, CO2, water (H2O), oxygen (O2), argon
(Ar), and nitrogen (N2) are considered to be important components for design. Only these
components are considered in developing the CPU performance and cost models. Since sulfur oxides
(SOX) and nitrogen oxides (NOX) are removed almost completely by the CO2 capture units, in
conjunction with the sulfur dioxide (SO2) polishing unit, these components are not explicitly
considered in the models.
CPU Performance Models
The process design of CPU depends on the design constraints of the CO2 product that goes into the
pipeline. There are four major components of a CPU, as listed below (NETL, 2010; Besong et al,
2013):
1) Flue gas compression and drying: The CO2-rich flue gas that needs to be purified is first
compressed to a pressure of approximately 30 bar. Most of the water vapor in the flue gas
condenses because of this compression. Water vapor is further removed by an additional
drying unit. If no further purification is needed, this dry CO2-rich flue gas can be further
compressed for pipeline transport. Depending on the inlet flue gas, the CO2 product purity
from this step could be between 80 to 90 percent, with all the other impurities (mainly Ar,
N2, and O2) remaining in the CO2 product.
2) Partial condensation: Most pipelines require further improvement in CO2 purity. The
compressed and dry gas from the above-mentioned step is cooled using external refrigeration
to a temperature close to the critical point of CO2 (-59°C). The cooled gas is then sent
through a flash unit where the majority of the CO2 is separated from other gases. The CO2
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 2
product now reaches a purity of around 95 percent, with the other gases (Ar, N2, and O2)
contributing to the remaining 5 percent. The purity and recovery of CO2 can be adjusted
varying the operating conditions, such as the flash temperature and the number of flash
stages. A typical design includes a two-stage flash, which has been modeled in this study
(Pipitone and Boland).
3) Distillation: Though a high enough CO2 purity is achieved, the amount of O2 in the CO2
product cannot be adequately controlled with partial condensation. A further distillation step
is required to increase the CO2 purity to more than 99 percent and reduce the O2 content to
less than 100 parts per million (ppm).
4) CO2 final product compressor: The CO2 product from any of the above three steps is
compressed to pipeline pressure.
Figure 1. CO2 purity requirements for different applications (NETL QGESS, 2014)
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 3
Figure 2. Schematic of a typical CPU process (NETL, 2010)
These four steps have been modeled using Aspen Plus and the performance results are used to
develop reduced-order models (ROMs) for different equipment. The details of these models are
described below.
Flue Gas Compressor and Drying Model The first step in a CPU is the compression of flue gas (clean of impurities such as SOX and NOX) to
approximately 30 bar. In the process, most of the water vapor is condensed. The remaining water
vapor is removed in a separate drying unit.
A six-stage intercooled compressor was modeled using Aspen Plus (v9). Inlet water vapor
concentration and outlet pressure were varied to study their effect on the amount of water vapor
condensed in the compressor, the required compression energy and cooling duty for intercooling
between stages. ROMs are then developed for these variables. The isentropic efficiency of the
compressor is assumed to be 100 percent so that the Integrated Environmental Control Model
(IECM) user-defined efficiency can be used for calculation of the energy requirements. The
temperature of interstage cooling was fixed at 38°C, pressure was varied from 25 to 35 bar, and inlet
flue gas temperature was varied from 25 to 75°C.
The following regression equations were obtained:
EffH2O,condensed = 0.96362 + 0.000657*Pcompr1 + 0.08350*yH2O,fg – 0.000006*(Pcompr1)2 –
0.057983*(yH2O,fg)2 - 0.000482(Pcompr1)*( yH2O,fg) (R
2 = 98.48%) (1)
Q_cool_MJ/k = 5.182 + 0.0964*Pcompr1 – 0.0787*Tfg + 5.11*yH2O,fg + 0.003476*(Tfg)2 –
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 4
23.21*(yH2O,fg)2 (R2=95.27) (2)
EffQcool = 3.537 – 3.325*Effcompr – 0.02473*Tfg + 0.855*(Effcompr)2 + 0.000035*(Tfg)
2 +
0.02145*Effcompr*Tfg (R2 = 99.19%) (3)
Qcool,act (MJ/kmol inlet FG) = Qcool,ideal*EffQcool (4)
Wcompr,ideal (kWh/kmol inlet FG) = 1.93661 + 0.017046*Pcompr1 + 0.003314*Tfg – 2.59760*yH2O,fg
(R2=99.66) (5)
Wcompr,act = Wcompr,ideal/Effcompr (6)
Mole flow rate of condensed water is given as:
MH2O,condensed = EffH2O,condensed*MH2O,in (7)
All of the remaining water vapor in the dry gas is assumed to be dried in the dryer unit, which
utilizes activated alumina adsorbents. The adsorbents are regenerated using waste N2 streams. The
captured water vapor is assumed to go out with the waste N2 stream and is not accounted for in the
CPU model.
Mfg,dry = Mfg,in – MH2O,in (8)
This gas goes into the partial condensation unit.
Partial Condensation Model The compressed flue gas is sent to a partial condensation system. The dried flue gas is first cooled in
a heat exchanger to -27°C and then sent to a flash chamber where most of the CO2 condenses, along
with some other gases. The non-condensed gases, which consist of the remaining CO2 and other
gases, is sent to another heat exchanger where they are cooled to a much lower temperature (between
-59 and -40°C) and flashed again in a second flash chamber.
More CO2 is recovered from the gases. The condensed CO2-rich liquids from both the flash
chambers are mixed and either compressed to pipeline pressure or sent for further purification.
The partial condensation process was also modeled using Aspen Plus. To develop ROMs, the
following parameters were varied:
• Inlet flue gas composition
• Inlet pressure
• The second flash stage temperature
ROMs were developed for the purity and recovery of different gas components and for the cooling
load required for the process.
Inlet flue gas composition is expressed in terms of the ratio of molar flow rates (and hence mole
fractions) of different gas components to the CO2 mole flow. The IECM oxy-combustion module
was used to estimate the ranges of gas compositions by varying the coal type and excess air ratio.
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 5
The ranges for different components in the inlet flue gas are:
O2/CO2: 0.0077 – 0.1728
N2/CO2: 0.0998 – 0.1107
Ar/CO2: 0.0012 – 0.0603
Regression models were developed for the desired parameters. Table 1 shows the coefficients for
linear regression of mole fractions of CO2, Ar, and O2. Mole fraction of N2 is calculated as (1 – mole
fractions) of other components. The table also shows the coefficients for quadratic regression of
cooling duty and recovery of CO2 from the partial condensation unit.
yN2,out = 1 – yCO2,out – yN2,out – yO2,out.
Table 1. Regression coefficients for the partial condensation model
Coefficient yO2,out yAr,out yCO2,out recCO2 Qcool
(MJ/kmol CO2 prod)
Constant -0.02463 -0.00992 0.98136 -0.05332 54.537
Pcompr (bar) 0.000905 0.000356 0.0017 0.029831 -1.5628
Tflash (oC) -0.0003 -0.00011 0 -0.026195 0.71448
(Ar/CO2) -0.06356 0.190291 -0.17982 -1.9725 66.877
(N2/CO2) -0.07476 -0.0284 -0.09642 -2.196 72.484
(O2/CO2) 0.158934 -0.0231 -0.19083 -1.9456 67.805
(Pcompr)2 -7.40E-05 -0.000196 0.017315
(Tflash)2 0 -0.00017 0.004715
(Ar/CO2)2 0.10628 0 4.71
(N2/CO2)2 -0.22376 0.0989 4.66
(O2/CO2)2 0.10935 -0.0572 6.618
(Pcompr)*(Tflash) 0 0.000342 -0.010259
(Pcompr)*(Ar/CO2) 0.001954 0.01509 -0.8547
(Pcompr)*(N2/CO2) 0.00594 0.014913 -0.8725
(Pcompr)*(O2/CO2) 0.001843 0.01498 -0.871
(Tflash)*(Ar/CO2) 0 -0.022648 0.6409
(Tflash)*(N2/CO2) 0 -0.025147 0.7117
(Tflash)*(O2/CO2) 0 -0.022445 0.63959
(Ar/CO2)*(N2/CO2) -0.06733 0 11.36
(Ar/CO2)*(O2/CO2) 0.20252 -0.1101 12.92
(N2/CO2)*(O2/CO2) -0.0512 0 12.93
R2 96.02 94.72 99.74 99.5 98.7
From the regression equation, it was found that purity varies more with pressure than with
temperature. At each pressure, the range of variation of purity with temperature is 0.008. An average
temperature of -50°C is fixed for the purity equation. By doing this, purity is now a function of flue
gas composition and pressure only. The adjusted coefficients are given in the table.
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 6
The user fixes purity and recovery. From these two values, pressure and temperature are calculated.
Figure 3. Variation of CO2 purity and recovery with pressure and flash temperature
Range of CO2 Purity: For a given temperature, purity increases with lower pressures. Therefore, the
minimum purity occurs at 35 bar and maximum purity occurs at 25 bar. These values are used to
calculate the range of CO2 purity.
yCO2_min = yCO2 at 35 bar (9)
yCO2_max = yCO2 at 25 bar. (10)
For a specified purity, the operating pressure is calculated by solving the quadratic equation of CO2
purity.
Quadratic term of the equation is calculated as follows:
a_p=a_yCO2(7);
Linear term of the equation is calculated as follows:
b_p=a_yCO2(2) + a_yCO2(13)*(Ar/CO2) + a_yCO2(14)*(N2/CO2) + a_yCO2(15)*(O2/CO2);
Constant term of the equation is calculated as follows:
c_p=a_yCO2(1) + a_yCO2(3)*0 + a_yCO2(4)*(Ar/CO2) + a_yCO2(5)*(N2/CO2) +
a_yCO2(6)*(O2/CO2) + a_yCO2(9)*(Ar/CO2)^2 + a_yCO2(10)*(N2/CO2)^2 +
a_yCO2(11)*(O2/CO2)^2 + (Ar/CO2)*(a_yCO2(19)*(N2/CO2) + a_yCO2(20)*(O2/CO2)) +
a_yCO2(21)*(N2/CO2)*(O2/CO2) – yCO2
Finally, pressure is calculated as follows:
p = (- b_p – sqrt(b_p^2-4*a_p*c_p))/(2*a_p) (11)
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 7
Figure 4. Variation of CO2 recovery with pressure and flash temperature
Range of CO2 Recovery: At a given pressure, recovery increases with lower temperature. Therefore,
the minimum recovery occurs at -40°C and maximum recovery occurs at -55°C. These values are
used to calculate the range of CO2 recovery, for the pressure calculated above.
recCO2_min = recCO2 at -40°C. (12)
recCO2_max = recCO2 at -55°C. (13)
For a specified pressure (calculated from CO2 purity) and recovery, the operating temperature is
calculated by solving the quadratic equation of CO2 recovery.
Quadratic term of the equation is calculated as follows:
a_T_flash = a_recCO2(8);
Linear term of the equation is calculated as follows:
b_T_flash = a_recCO2(3) + a_recCO2(12)*p + a_recCO2(16)*(Ar/CO2) + a_recCO2(17)*(N2/CO2)
+ a_recCO2(18)*(O2/CO2);
Constant term of the equation is calculated as follows:
c_T_flash = a_recCO2(1) + a_recCO2(2)*p + a_recCO2(4)*(Ar/CO2) +
a_recCO2(5)*(N2/CO2) + a_recCO2(6)*(O2/CO2) + a_recCO2(7)*p^2 +
a_recCO2(9)*(Ar/CO2)^2 + a_recCO2(10)*(N2/CO2)^2 + a_recCO2(11)*(O2/CO2)^2 +
p*(a_recCO2(13)*(Ar/CO2) + a_recCO2(14)*(N2/CO2) + a_recCO2(15)*(O2/CO2)) +
(Ar/CO2)*(a_recCO2(19)*(N2/CO2) + a_recCO2(20)*(O2/CO2)) +
a_recCO2(21)*(N2/CO2)*(O2/CO2) – recCO2.
Finally, temperature is calculated as follows:
T_flash = (-b_T_flash – sqrt(b_T_flash^2-4*a_T_flash*c_T_flash))/(2*a_T_flash)
(14)
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 8
Flow rates of components in the CO2 product are given as:
MCO2,product = recCO2*MCO2,in
MN2,product = yN2,out*MCO2,product/yCO2,out
MO2,product = yO2,out*MCO2,product/yCO2,out
MAr,product = yAr,out*MCO2,product/yCO2,out
Distillation Model Distillation is used to further purify CO2. The main design parameter for the distillation column is
the amount of O2 present in the CO2 product. Depending on the application, O2 purity should be <
100 ppm or < 10 ppm.
recCO2 = 0.88616 – 0.21720*((Ar+N2)/CO2) – 0.27630*(O2/CO2) + 3.8x10-5*(O2 ppm) –
0.0001812*(Tdistill) (R2=98.97%)
If O2 ppm and Tdistill are specified, recCO2 can be calculated from the equation above. On the other
hand, if O2 ppm and recCO2 are specified, the above equation can be used to calculate Tdistill.
To simplify calculations for a combined partial condensation and distillation process, Tdistill is fixed
at its minimum value of -59°C and O2 concentration is fixed at 10 ppm.
Qcool,distill (MJ/kmol gas in) = 135.09 – 287.89*yCO2,in – 0.078406*Tdistill + 153.38*( yCO2,in)2
(R2=98.97%)
The CO2 product from the distillation column is almost 100 percent CO2 and a very small amount of
O2. The presence of N2 and Ar is negligible, compared to the concentration of O2. Hence, it is
assumed that only CO2 and O2 are present in the CO2 product and the other gases go out through the
vent stream.
Molar flow rates of gases in CO2 product and vent streams are calculated as follows:
MCO2,product = recCO2*MCO2,in
MO2,product = yO2/(1 – yO2)*MCO2,product
MCO2,vent = (1-recCO2)*MCO2,in
MO2,vent = MO2,in – MO2,product
MN2,vent = MN2,vent
MAr,vent = MAr,vent
For simplifying calculations, a combined partial condensation and distillation model was developed.
Inputs to this are the flue gas composition into CPU.
Table 2 shows the coefficients for parameters in a combined partial condensation and distillation
model.
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 9
Table 2. Regression equations for the combined condensation and distillation model
Coefficient recCO2
Qcool (distillation
only)
(MJ/kmol CO2 prod)
Coefficient T_flash (°C)
Constant -0.038805 3.11854 Constant -346.74
Pcompr (bar) 0.028721 -0.04888 Pcompr (bar) 6.113
Tflash (oC) -0.025767 0.013564 (Ar/CO2) -280.4
(Ar/CO2) -1.951910 -2.44382 (N2/CO2) -325.3
(N2/CO2) -2.160660 0.737105 (O2/CO2) -281.94
(O2/CO2) -1.933630 -2.7221 recCO2 766.8
(Pcompr)2 -0.000195 0.000554 (Pcompr)
2 -0.00918
(Tflash)2 -0.000168 4.30E-05 (Ar/CO2)
2 166.7
(Ar/CO2)2 0.005637 1.37139 (N2/CO2)
2 229.9
(N2/CO2)2 0.091358 0.12884 (O2/CO2)
2 161.6
(O2/CO2)2 -0.045878 1.73731 (recCO2)
2 -498.36
(Pcompr)*(Tflash) 0.000339 -0.00031 (Pcompr)*(Ar/CO2) -1.631
(Pcompr)*(Ar/CO2) 0.014839 0.055146 (Pcompr)*(N2/CO2) -2.179
(Pcompr)*(N2/CO2) 0.014942 -0.0169 (Pcompr)*(O2/CO2) -1.7526
(Pcompr)*(O2/CO2) 0.014618 0.062067 (Pcompr)*(recCO2) -4.888
(Tflash)*(Ar/CO2) -0.022148 -0.01531 (Ar/CO2)*(N2/CO2) 375.1
(Tflash)*(N2/CO2) -0.024796 0.004694 (Ar/CO2)*(O2/CO2) 317.63
(Tflash)*(O2/CO2) -0.021849 -0.01724 (Ar/CO2)*(recCO2) 177.3
(Ar/CO2)*(N2/CO2) 0.005347 -0.84045 (N2/CO2)*(O2/CO2) 383.01
(Ar/CO2)*(O2/CO2) -0.091822 3.08658 (N2/CO2)*(recCO2) 216
(N2/CO2)*(O2/CO2) 0.013744 -0.94617 (O2/CO2)*(recCO2)* 180.5
R2 100 98.7 R2 94.31
T_flash prediction was close to temperatures higher than -55°C. So the lower end of T_flash is
changed from -59°C to -55°C. The range of T_flash now is -55°C to -40°C.
Figure 5. CO2 recovery and purity for condensation and distillation
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 10
Recovery is higher at lower temperatures and higher pressures. For a given flue gas,
recCO2_min = recovery at 25 bar and -40°C (15)
recCO2_max = recovery at 35 bar and -55°C (16)
Fix p=30 bar. If T_flash is <-55°C, then p=35 bar. If T_flash>-40°C, then p=25 bar. (17)
CO2 Compressor Model The CO2 compressor was also modeled in Aspen Plus as a function of varying inlet and outlet
pressures. Inlet pressure was varied between 25 and 35 bar, and the outlet pressure was varied
between 120 and 160 bar. The following regression equations were developed from the results of the
Aspen models.
Work required for compression (kWh/kmol CO2 product total):
WCO2compr,ideal (kWh/kmol CO2 product total) = 1.12905 – 0.023379*Pin(bar) + 0.001975*Pout(bar)
(R2=99.57%)
WCO2compr,act (kWh/kmol CO2 product total) = WCO2compr,ideal/EffCompr (18)
WCO2compr,act (kWh/kmol CO2) = WCO2compr,act (kWh/kmol CO2 product total) / Purity
Cooling duty required for compressor inter-cooling is given by the following equations:
Qcool,CO2compr (MJ/kmol CO2 prod total) = 15.4410 – 0.14584*Pin(bar) + 0.026112*Pout(bar) –
3.3577*Eff_compr (R2=99.59) (19)
To convert it to per kmol of CO2, this has to be divided by CO2 purity. This heat is supplied by
cooling water.
CPU Efficiency Factor
The CPU refrigeration load requirement (partial condensation and distillation) calculated using the
models above were validated with the energy requirement values presented in the DOE 2010 oxy-
fuel report. The DOE reports present the total CPU energy requirement, which includes refrigeration
and compression. For these case studies, compression work is calculated from the IECM models
developed above and subtracted from the total work given in the DOE report. KeyLogic clarified
that the refrigeration system used is an auto-thermal system, which means that the refrigeration load
is met by internal heat integration without any need for external electrical input. A CPU adjustment
factor was calculated to account for modeling differences between the Aspen models and the models
used in the DOE report. The coefficients are shown in Table 3.
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 11
CPUfactor = WDoE/WIECM (20)
CPUfactor = 1.327 for compression only and 1.054 for partial condensation (21)
CPUfactor = 1.102 for partial condensation + distillation (22)
The overall CPU energy requirement is calculated as:
WCPU (kWh/kmol CO2)= (WCompr,fg + WCO2Compr)*CPUfactor (23)
Units for all the parameters above should be converted to kWh/kmol CO2.
Table 3. Case studies to calculate CPU efficiency factor
DOE case S12E S12F L12F S13F L13F
CPU configuration Partial
condensation Partial condensation + Distillation
Inlet CO2 mole fraction 0.6994 0.6994 0.708 0.6997 0.6871
Inlet H2O mole fraction 0.1726 0.1726 0.1627 0.1726 0.1726
Ar/CO2 0.0419 0.0419 0.0415 0.0419 0.0426
N2/CO2 0.1090 0.1090 0.1086 0.1089 0.1093
O2/CO2 0.0320 0.0320 0.0323 0.0316 0.0521
Temperature of flue gas (°C) 57 57 57 57 57
Purity 0.9771 0.9998 0.9998 0.9998 0.9998
Recovery 0.9102 0.909 0.909 0.9116 0.909
CO2 product total (kmol/hr) 12,285 12,049 12,707 11,543 11,601
CPU energy total (kW) 62,100 64,740 67,140 62,090 65,780
CPU energy (kWh/kmol CO2 prod) 5.17 5.37 5.29 5.38 5.67
Calculated Values
Pressure (bar) 25.0 30.0 30.0 30.0 30.0
T_flash (°C) -47.9 -46.7 -46.6 -47.3 -48.8
Refrigeration load (kWh/kmol CO2 prod) 5.08 5.29 5.29 5.27 5.31
Compression load (kWh/kmol CO2 prod) 4.91 4.91 4.91 4.90 4.98
CPU efficiency factor 1.054 1.095 1.077 1.099 1.138
Order of Calculations
The following order should be used for the overall system calculation:
1. Choose a configuration:
a. Capture with impurities (only flue gas compression and drying).
b. High CO2 purity (95 to 97 percent) (partial condensation).
c. Ultra-high CO2 purity (99.99 percent) (partial condensation + distillation).
2. Specify CO2 purity and recovery.
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 12
a. Config (a): Purity = CO2 mole fraction after drying.
b. Config (b): User specifies purity and recovery. Ranges are calculated using Equations
9, 10, 12. and 13.
c. Config (c): Purity is fixed at 99.99 percent (O2 purity = 10 ppm). Ranges for recovery
are calculated using Equations 15 and 16.
3. Determine system operating pressure.
a. Config (a): Assume 30 bar.
b. Config (b): Equation 11, using coefficients from Table 1.
c. Config (c): Equation 17, using coefficients from Table 2.
4. Determine flash temperature.
a. Config (a): No flash.
b. Config (b): Equation 14, using coefficients from Table 1.
c. Config (c): Equation 17, using coefficients from Table 2.
5. Determine compression energy using Equation 6 and Equation 18.
6. Determine cooling duty to be supplied by cooling water using Equation 4 and Equation 19.
7. Determine refrigeration duty.
a. Config (a): No refrigeration.
b. Config (b): Qcool using coefficients in Table 1.
c. Config (c): Qcool using coefficients from Table 1 + Table 2.
8. Calculate CPU efficiency using Equation 21 or Equation 22.
9. Calculate total CPU refrigeration energy required using Equation 23.
CPU Cost Model
CPU capital cost data was obtained from the 2010 DOE report on oxy-combustion power plants.
Table 4 shows the data, along with relevant flow rates.
Table 4. Relevant flow rates and capital costs of CPU from the 2010 DOE Oxy-Combustion
Report
The costs for individual components are plotted and exponential regression equations were fit as a
function of flow rates. The capital cost of compression and drying unit fits well with CO2 product
flow rate, while that of the condensing heat exchanger fits well with the total CPU inlet flow rate, as
shown in Figure 6.
The final cost functions are given as:
DOE Oxy-Fuel Report Case S12D S12E S12F L12F S13F L13F
Total CPU inlet flow (kmol/hr) 20,281 18,885 18,977 19,772 18,712 18,598
Total CO2 prod flow (kmol/hr) 15,952 12,285 12,049 12,707 11,543 11,601
CO2 Removal Cost (x $1,000, 2007)
CO2 condensing HX $ 3,424 $ 4,292 $ 4,256 $ 3,759 $ 4,169 $ 4,376
CO2 compression and drying $ 98,224 $ 77,523 $ 77,806 $ 80,579 $ 75,697 $ 78,375
Total $ 101,648 $ 81,815 $ 82,062 $ 84,338 $ 79,866 $ 82,751
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 13
CCondensingHX ($M, 2007) = 2x1012 (CPU inlet flow, kmol/hr)– 2.714
CCompr and Drying ($M, 2007) = 0.05017 (Total CO2 product, kmol/hr)0.7824
Note: CPU inlet flow contains water vapor. Total CO2 product includes CO2 as well as other
impurities.
CCPU,total = CCondensingHX + CCompr and Drying
Figure 6. Regression equations for capital cost of CPU
Other than the internal electricity consumption, there are no other variable operations and
maintenance (VOM) costs for the CPU unit. Fixed operations and maintenance (FOM) costs are the
same as the standard IECM defaults.
User-Specified Configuration
In this configuration, users can specify their own purity, recovery, energy penalty, and cost. Mass
balance equations are given below.
Mgas,in (kmol/hr) = flue gas into CPU after direct contact cooler, sulfur polisher, and drying. (after
removal of H2O and all other components, other than Ar, N2, CO2, and O2).
MCO2,out (kmol/hr) = MCO2,in * Recovery
Mgas,out = MCO2,out/purity
MAr,out = (1/3)*(1 – purity)*Mgas,out
MN2,out = (1/3)*(1 – purity)*Mgas,out
MO2,out = (1/3)*(1 – purity)*Mgas,out
User can specify energy penalty in kWh/tonne CO2 (default value is 120 kWh/tonne) and
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 14
CPU capital cost in $M/tonne/hr CO2 (default value is 0.175).
Direct cost of CPU = (CPU unit cost, $M/tonne/hr CO2)*(CO2 product, tonne/hr).
IECM Screens
Figure 7 through Figure 10 show the input (Set Parameters) and output (Get Results) screens of the
CPU model in IECM. The plant configuration is a pulverized coal plant with a membrane-based
post-combustion CO2 capture unit.
Figure 7. "Set Parameters" config screen for the CPU unit
Figure 8."Set Parameters" purification screen for the CPU unit, 99.99 percent purity case
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 15
Figure 9. "Set Parameters" purification screen for the CPU unit, high-purity case
Figure 10. "Set Parameters" purification screen for the CPU unit, low-purity case
Case Studies
Case studies were conducted using the performance and cost models developed before. NETL case
S12E was used as the base case for inlet flue gas flow rate and composition. The results are
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 16
presented in Table 5. Sensitivity analysis was conducted by varying the purity and recovery. Results
are shown in the following figures.
Figure 11. Sensitivity analysis results
Figure 12. Sensitivity analysis for the CPU case studies
Table 5. Case study input assumptions and results
Configuration Compression Only Partial Condensation Distillation
Inlet CO2 mole fraction 0.6994 0.6994 0.6994
Inlet H2O mole fraction 0.1726 0.1726 0.1726
Inlet flue gas temperature (°C) 57 57 57
Ar/CO2 0.0419 0.0419 0.0419
N2/CO2 0.1090 0.1090 0.1090
O2/CO2 0.0320 0.0320 0.0320
CPU inlet flow (kmol/hr) 18,885 18,885 18,885
Dried gas flow (kmol/hr) 15,625 15,625 15,625
Purity (%) 84.5 95.11 99.99
Recovery (%) 100 92.07 90.32
CO2 product total (kmol/hr) 15,625 12,787 11,931
Minimum purity (%) 95.11 99.99
Integrated Environmental Control Model - Technical Documentation CO2 Purification Unit • 17
Configuration Compression Only Partial Condensation Distillation
Maximum purity (%) 97.06 99.99
Pressure (bar) 35.0 30.0
Minimum recovery (%) 92.07 84.32
Maximum recovery (%) 95.80 93.91
Flash temperature (°C) -40.0 -45.3
Refrigeration duty (kWh/kmol CO2 prod) 4.69 5.69
W_compr_total (kWh/kmol CO2 prod) 4.65 4.87 4.88
W_compr_total (kWh/tonne CO2) 105.5 110.5 110.6
eff_CPU 0.000 0.053 0.095
W_CPU_total (kWh/kmol CO2) 4.65 5.12 5.42
Cost_CondensHX ($M, 2007) 4.96 4.96 4.96
Cost_Compr and Drying ($M, 2007) 95.87 81.96 77.63
Cost_CPU_total ($M, 2007) 100.84 86.92 82.59
Cost_CPU_total ($M/tonne/hr CO2, 2007) 0.1735 0.1624 0.1573
References
Besong, M.T., Maroto-Valer, M.M., Finn, A.J., (2013). Study of design parameters affecting the
performance of CO2 purification units in oxy-fuel combustion. International Journal of
Greenhouse Gas Control. 12, 441-449.
National Energy Technology Laboratory (2010). Cost and performance for low-rank pulverized coal
oxy-combustion plants, Final Report, DOE/NETL-401/093010.
National Energy Technology Laboratory (2013). Quality guidelines for energy system studies – CO2
impurity design parameters, DOE/NETL-341/011212.