Process for CO2 Capture from High-Pressure and Moderate

Article pubs.acs.org/IECR

Process for CO2 Capture from High-Pressure and ModerateTemperature Gas Streams James C. Fisher, II,†,‡ Ranjani V. Siriwardane,*,† and Robert W. Stevens, Jr.† †

National Energy Technology Laboratory, United States Department of Energy, 3610 Collins Ferry Road, Morgantown, West Virginia 26507, United States ‡ URS, 3604 Collins Ferry Road, Morgantown, West Virginia 26507, United States ABSTRACT: A novel CO2 capture process was developed using a Mg(OH)2-based sorbent suitable for warm gas CO2 removal from high-pressure coal gasification gas streams. The purpose of this study is to perform a preliminary energy study and develop a method to implement this sorbent in a power plant. The proposed CO2 removal process involves sorption of CO2 at 200−300 °C and regeneration at 385 °C. The operational temperature is ideal for CO2 capture downstream to the water−gas shift reactor in an integrated gasification combined cycle (IGCC) power plant but is applicable to other warm gas cleanup processes as well. This technology offers the ability to fully utilize the potential efficiency increases associated with warm gas clean up. Additionally, the sorbent is able to operate in the presence of steam, which distinguishes it from other technologies that require an energyintensive drying step prior to CO2 separation. Regeneration is carried out at 280 psi and 400 °C, resulting in a high-pressure CO2 product stream, which significantly reduces the auxiliary load normally associated with CO2 compression for sequestration. The chemistry of the sorption process also reduces the amount of steam traditionally required for the water−gas shift reactor by 50%, increasing the overall efficiency of the plant. The incorporation of the sorbent and the described methods resulted in an overall IGCC power plant efficiency greater than that of the competing Selexol technology.

■

INTRODUCTION Fossil fuels supply a majority of the world’s energy needs. The combustion of fossil fuels within power plants, however, is believed to be one of the major contributors to an increasing atmospheric concentration of CO2. Current carbon capture technologies impart a significant auxiliary load to the plant and hence penalize its performance, which results in a decreased net power output. Improved CO2 capture technologies are needed to minimize this impact, thereby increasing the plant efficiency with capture. By combining an advanced CO2 capture technology with a high-efficiency plant, the electricity produced may be maximized. Integrated gasification combined cycle (IGCC) is one of the most efficient coal-fueled plant configurations and thus is considered for the current application. The IGCC product streams are at high pressure and elevated temperature. The composition of the product stream is also made up of a significant concentration of CO2 (approximately 40 mol %) at high pressure, which gives rise to its separation being more straightforward than separation from a pulverized coal plant flue gas (which contains 12−15 mol % CO2). A comparison between the proposed process and current commercially available technology for CO2 removal from high-pressure gas streams, such as those in IGCC power plants, is shown in Figure 1. A coal slurry is fed into the gasification unit to produce a fuel gas in which contaminant removal (i.e., H2S, HCl) is completed immediately downstream. Scheme 1 of Figure 1 illustrates the employment of a typical pressure swing CO2 absorption process, such as the commercially available Selexol process. Used in many industries, the Selexol process is a commercial CO2 capture process that utilizes a liquid glycol ether process to remove CO2 from high-pressure streams. The © 2012 American Chemical Society

regeneration of the CO2-rich solvent is conducted by reducing the pressure. Since CO2 is recovered at near atmospheric pressure, the process results in a large pressure ratio of the CO2 compression train for sequestration. The CO2 removal process takes place below 50 °C, which requires the flue gas stream to be cooled from 200 °C. The process is also sensitive to moisture and requires energy-intensive drying. After CO2 removal, the fuel gas then requires reheating above 200 °C as feed gas for the turbines; the cooling and reheating of such a large quantity of gas results in a thermal loss that is reflected in the power plant’s overall efficiency loss when this system is applied.1 Scheme 2 of Figure 1 depicts a superior option where the CO2 separation process takes place at 200−315 °C, thus retaining a majority of the thermal energy of the stream, which contributes to improved efficiency of the IGCC process. Furthermore, if the sorbent is capable of high-pressure regeneration, the CO2 compression energy requirement may also be significantly reduced. By maintaining elevated CO2 separation and contaminant removal temperatures, the IGCC plant efficiency may be maximized. However, only a few studies related to regenerable sorbents with sufficient CO2 removal capacity at 150 to 350 °C are reported in the literature. Sorbents that capture CO2 at 450− 550 °C have been reported, but the regeneration temperatures and energies are very high for those systems.2 Zeolites have been examined and found to have very low capacities at capture temperatures above 200 °C.2−4 Lithium-based sorbents that Received: Revised: Accepted: Published: 5273

September 30, 2011 March 1, 2012 March 8, 2012 March 8, 2012 dx.doi.org/10.1021/ie2022465 | Ind. Eng. Chem. Res. 2012, 51, 5273−5281

Industrial & Engineering Chemistry Research

Article

Figure 1. Comparison of different options for precombustion CO2 removal and H2 production from IGCC power plants.

capture CO2 at 450−550 °C were reported by Nakagawa and Ohashi.5 However, the reported regeneration temperature of these Li-sorbents are nearly 800 °C, which presents a costly thermal swing regeneration, if a CO2 capture is conducted. In addition, heats of sorption and desorption are very high with Li-sorbents at 200 °C. Golden and Sircar described alkalipromoted hydrotalcite, alkali-impregnated alumina, and double salt extrudates that utilized a pressure swing process for the removal of bulk CO2 from wet high-temperature gas.6 The alkali-promoted hydrocites use a pressure swing process that recovers the CO2 at low pressures requiring significant compression energy for sequestration. Our group previously reported the development of a sodium-based sorbent that could absorb CO2 at 200−315 °C.7,8 The sorbent, however, requires a temperature above 700 °C for regeneration. This large temperature swing combined with high heat of sorption/ desorption would require a significant quantity of energy, thereby making it impractical for an IGCC application. We also reported a magnesium-based sorbent that captures CO2 at 200 °C and is regenerable at 400 °C, which is more compatible to an IGCC application.9 The current report describes the development of a warm-gas-temperature (200−400 °C) CO2 capture process that employs the previously reported magnesium-based sorbent. A novel magnesium hydroxide-based sorbent that is capable of capturing CO2 at 200−315 °C and high pressure and is regenerable at 375−400 °C was developed and patented by NETL researchers.7,9 The chemical reaction for the capture process with this sorbent is shown below in reaction 1. At high pressure, Mg(OH)2 possesses a high CO2 sorption capacity at 200−315 °C, which is significantly higher than that of the commercial Selexol process. The capture reaction/process will be hereafter referred to as “sorption”. Mg(OH)2 (s) + CO2 (g) → MgCO3(s) + H2O

Reaction 2 may also occur as two different reaction steps as shown in reactions 3 and 4: (3)

MgO(s) + H2O(g) → Mg(OH)2 (s)

(4)

The capture reaction of Mg(OH)2 with CO2 to form MgCO3 (reaction 1) is exothermic with a ΔH value of −19.7 kJ/mol of CO2. This implies that the heat required for regeneration of the reverse reaction (reaction 2) to form Mg(OH)2 is significantly lower than that required for the other carbonate decomposition sorbents.9 This is a significant advantage for the Mg(OH)2 sorbent since regeneration duty is critical for CO2 capture systems. The thermal swing requirement for the sorption and regeneration of Mg(OH)2/CO2 is also lower than that of most other systems. It is possible that the regeneration may proceed in two steps, as shown in reactions 3 and 4. The endothermic decomposition of MgCO3 (reaction 3) has a ΔH° value of 100.9 kJ/mol. The rehydroxylation (reaction 4) is exothermic with a ΔH° value of −81 kJ/mol. To utilize the low ΔH value for the overall reaction 2, the heat from reactions 3 and 4 has to be integrated into the process. A detailed thermodynamic analysis of these reactions has been previously reported.9 Each of these steps is expected to occur in different reactors. A CO2rich stream of fuel gas can enter the adsorption reator where reaction 1 will take place. The sorbent will then be transported to a second reactor where the sorbent will be heated in a steam and CO2 environment, allowing reactions 2 and 3 to occur. The third reactor will complete the rehydroxylation of any MgO that was not hydroxylated in the second reactor to Mg(OH)2 via reaction 3. An additional reaction of importance is the water−gas shift (WGS) reaction, which converts partially oxidized carbon monoxide to carbon dioxide via the reaction CO(g) + H2O(g) → CO2 (g) + H2(g)

(1)

(5)

The WGS reaction also produces valuable H2 for power production. In a typical IGCC power plant, high-quality steam is removed from the steam cycle and injected to fuel gas stream before the WGS reactor to maintain a 2:1 H2O:CO molar ratio.1 This high steam concentration ensures nearly complete consumption of the CO in the WGS reactor, maximizing the

The carbonate formed during the reaction can decompose to release CO2 and regenerate according to reaction 2 at 375−400 °C. MgCO3(s) + H2O → Mg(OH)2 (s) + CO2

MgCO3(s) → MgO(s) + CO2 (g)

(2) 5274

dx.doi.org/10.1021/ie2022465 | Ind. Eng. Chem. Res. 2012, 51, 5273−5281


Article

H2 production and overall power production of the IGCC power plant. However, the removal of the high-quality steam for the WGS reactor does have a power penality in the steam turbine that, if avoided, could increase the overall higherheating value (HHV) efficiency of the power plant. The purpose of this paper is to devlop a scheme to utilize the Mg(OH)2 sorbent in an IGCC power plant and utilize the unique chemistry to reduce the steam requirements in the overall plant. The proposed process in this paper describes a method of introducing H2O via the CO2 capture process with Mg(OH)2 into the WGS feed stream while the need for additional high-quality steam from the steam cycle is lowered or eliminated. This scheme increases the amount of steam going to the steam cycle and increases the power plants overall power production while CO2 is removed. The implementation of the proposed CO2 removal method resulted in an overall efficiency greater than that of the Selexol process.

Table 1. Sorption Capacities over Mg(OH)2 during BenchScale Flow Reactor Testing with Feed Composition of 28% CO2, 10% H2O and GHSV = 250 h−1 CO2 sorption cycle

T (°C)

P (psig)

capacity (mol/kg)

T (°C)

P (psig)

1 2 3 4 5 6

200 200 200 200 200 200

150 150 150 150 280 280

n/a 1.72 2.54 2.46 3.10 3.37

375 375 375 375 375 375

20 20 20 20 20 20

Table 2. Sorption Capacities over Mg(OH)2 during BenchScale Flow Reactor Testing with Feed Composition of 28% CO2, 10% H2O and GHSV = 250 h−1

■

CO2 sorption

EXPERIMENTAL SECTION The synthesis of the Mg(OH)2-based sorbent containing a mixture of Mg(OH)2, bentonite binder, and a promoter has been described in detail elsewhere.7 To fabricate the sorbent, the powders were combined with water to form 1−2 mm pellets followed by drying at 150 °C as previously described.7 The dried sorbent pellets were placed into a vertical tubular reactor with a 0.5 in. diameter to form a 6-in. bed. The sorbent bed was heated in nitrogen to 200 °C and then exposed to the capture gas consisting of 28% CO2, 15% H2O, and 57% N2 in a down-flow, packed-bed configuration with a gas hourly space velocity (GHSV) of 500 h−1. The capture was completed when the CO2 outlet concentration matched the inlet concentration, which was monitored via mass spectrometer (MS). After the completion of the capture, the sorbent was regenerated by flowing 15% H2O with 85% N2 at 500 h−1 GHSV through the reactor at 400 °C until CO2 was no longer present in the effluent determined by a MS CO2 profile. Subsequent to regeneration, the reactor was cooled to 200 °C and 30% H2O in N2 was introduced for 2 h to hydroxylate the MgO to Mg(OH)2. Following hydroxylation, the CO2 capture cycle was repeated. On the basis of the experimental results, a process was designed to utilize the Mg(OH)2 sorbent in an IGCC power plant with a significant reduction in steam requirements. The power plant configuration used was from a DOE report that modeled an IGCC power plant utilizing the GEE gasifier with the Selexol CO2 capture process after the WGS reactor.1 The reactors for the CO2 capture process in this work were modeled after the three process steps (i.e., capture, regeneration, and hydroxylation) where each reactor operates at similar pressures and temperatures as the experimental conditions described above. The flow rate of gases and sorbent for CO2 capture was based on a 100 kg/h fuel gas flow, which allows scaling the results to any IGCC plant, given the fuel gas flow from the WGS reactor to the capture unit. The unit was then scaled to the gas flows of the NETL report, where the GEE gasifier and utility costs were estimated.1

■

regeneration

regeneration

cycle

T (°C)

P (psig)

capacity (mol/kg)

T (°C)

P (psig)

capacity (mol/kg)

1 2 3 4

200 200 200 200

280 280 280 280

n/a 4.5 4.4 4.5

400 400 400 400

280 280 280 280

2.3 4.6 4.3 4.4

Figure 2. Bench-scale flow reactor CO2 capture data over Mg(OH)2 sorbent at 200 °C, 280 psig, GHSV = 500 h−1 with CO2/H2O/N2 = 28%/15%/57%.

Figure 3. Desorption of CO2 during regeneration of the sorbent at 280 psig and 400 °C while flowing GHSV = 500 h−1 of N2.

to determine the effect of pressure. The CO2 capture capacity improves during cycling due to better rehydroxylation due to morphology changes, as discussed in a prior paper.9 The formation of Mg(OH)2 is critical to have higher CO2 capture capacity of the sorbent. During all testing conditions, the sorbent was able to remove the CO2 down to ppm levels, yielding a removal efficiency of approximately 99%. Increasing the pressure from 150 to 280 psig resulted in a favorable increase in the sorbent’s CO2 capture capacity. Capture

RESULTS AND DISCUSSION

1. Reactor Studies. The results of a six-cycle test with a bench-scale flow reactor with feed containing 28% CO2, 10% H2O, 28% Ar, and 34% N2 are shown in Table 1. The first four cycles were conducted at 150 psig and the last two at 280 psig 5275



Article

Figure 4. Proposed CO2 capture process.

capacities on the order of 3 mol CO2/kg of sorbent were realized at 280 psig and at 200 °C. Table 2 shows the results of a four-cycle test where regeneration was conducted at 280 psig. The temperature during the high-pressure regeneration was raised from 375 to 400 °C to ensure complete regeneration. The experimental system was then modified to allow the H2O concentration to be raised from 25% to 40% during the hydroxylation step. This higher concentration of water allowed for enhanced hydroxylation, which resulted in an improved capture capacity. Figures 2 and 3 show the MS profile during a capture and regeneration after the enhanced hydroxylation with 40% H2O. Hydroxylation with 40% H2O followed by CO2 capture conducted over fresh sorbent resulted in 4.5 mol of CO2/kg of sorbent, as shown in Figure 2. Prior to capture, the CO2 concentration is shown to be at 32 vol % (dry basis) while the feed gas and injected water bypasses the reactor housing the sorbent bed. It is important to note that the MS reports a dry composition, but the actual CO2 feed concentration was 28% with 15% steam. Capture was initiated by switching the reactor online (i.e., redirecting the feed gas and injected water into the reactor/sorbent bed). After capture was initiated, the CO2 concentration in the reactor effluent decreased to ppm levels for more than 25 min, after which the CO2 concentration profile increased sharply or broke through, indicating saturation of the sorbent by CO2. The total amount of CO2 capture was 4.5 mol of CO2/kg of sorbent. The subsequent regeneration of the sorbent at 400 °C and 280 psig led to desorption of 4.6 mol of CO2/kg of sorbent (Figure 3), which suggested that the sorbent is fully regenerable under the conditions investigated. Further, Table 2 shows that the sorbent is regenerable under the temperature and pressure indicated for the regenerator in the proposed process. The rehydroxylation step of the process is critical to maintain the CO2 capture capacity during cyclic process. CO2 capture testing with MgO resulted in a low capacity of 0.25 mol CO2/ kg during the bench-scale flow reactor test at 200 °C (not shown), despite the fact that the conversion of MgO to MgCO3 reaction is predicted to be highly favorable thermodynamically.9

This result suggests that the kinetics of CO2 capture with MgO are much slower than those with Mg(OH)2. Therefore, either the presence of Mg(OH)2 or having steam with MgO is necessary for the CO2 capture process. 2. Process Description. On the basis of the experimental data, a multireactor process for CO2 capture from moderate temperatures and high pressures was developed. The advantages of this system include low-regeneration energies, high-pressure regeneration to recover a high-pressure CO2 gas stream ready for sequestration, and removal of CO2 without gas cooling. The process utilizes three reactors consisting of an absorber, a regenerator, and a polisher. The absorber and polisher units are maintained at 200 °C while the regenerator is maintained at 385 °C. The reactors can be configured as moving beds, circulating fluidized beds, or transport reactors. A unique effect of utilizing Mg(OH)2 for CO2 capture is the stoichiometric replacement of CO2 with H2O. Utilizing the Mg(OH)2 sorbent to capture the CO2 prior to the WGS reactor would produce a higher concentration of H2O in the inlet into the WGS reactor without utilizing high-quality, valuable steam. The DOE report for baseline performance of coal power plants indicates a 11.1 mol % CO2 in the WGS feed, which if captured using the Mg(OH)2 sorbent would allow for 13 kg/s of steam to be produced via the capture of CO2.1 Further savings then come from the 30% size reduction of the post-WGS capture system, which only has to remove 99 kg/s of CO2 of the total 142 kg/s of CO2 produced throughout the plant. In this work, the post-WGS shift is assumed to be an Mg(OH)2 system, but it is not required. Additionally, both capture systems are considered separate for the purpose of this study. However, more energy savings could potentially be realized by combining the regenerator and polisher from both systems into a single regenerator and polisher. However, many considerations are dependent on the specific plant and size, thereby making these considerations beyond the scope of this study. Figure 4 shows a block diagram of the proposed system with a CO2 capture reactor before and after the WGS reactor. 5276


5277

13

0.0 0.0 0.0 100.0 0

0.0 0.0 0.0 100.0

0

100

100

137 511 0 0 0 873 180

n/a n/a n/a n/a

100 n/a 200 148 76 4194 18.0 11 119 358

14

3 200.0 280 873 180 n/a 1563 n/a n/a

20 n/a 200 148 76 4194 18.0 11 119 358

2806 15 749 231 472 224 403 0

140 317 15 749 231 472 168 149 0

T (°C) P (psig) m (kg/h) Cp (J/mol K) Cp (J/kg K) MW F (mol/h) molar composition (mol %) CO2 H2 N2 H2O mass flows (kg/h) CO2

0.2 27.5 28.8 43.5

11.1 27.5 28.8 32.6

2 200.0 280 474 431 23 1402 16.5 28 671 964

1

200 280 555 687 24 1247 19.4 28 671 964

stream

T (°C) P (psig) m (kg/h) Cp (J/mol K) Cp (J/kg K) MW F (mol/h) molar composition (mol %) CO2 H2 CO H2O mass flows (kg/h) CO2 H2 CO H2O sorbent flow rate (kg/h) sorbent Mg state (mol %) (active species only) Mg(OH)2 MgCO3 MgO sorbent Mg state (wt %) (active species only) Mg(OH)2 MgCO3 MgO stream

Table 3. Process Stream Summary 4

0.0 0.0 0.0 100.0

16

0

0.0 0.0 0.0 100.0

100 n/a 1 469 500 76 4194 18.0 81 638 911

55.8 5.7 38.5

55.8 5.7 38.5 15

7806 0 0 28 127 768 796

n/a n/a n/a n/a

48.3 3.4 48.3

20 n/a 1 469 500 76 4194 18.0 81 638 911

0

5 225 280 768 796 n/a 1576 n/a n/a

48.3 3.4 48.3

7806 0 0 28 127 768 796

n/a n/a n/a n/a

385 280 768 796 n/a 1701 n/a n/a

6

93.2 4.8 0.0

96.6 3.4 0.0

17

0

0.0 0.0 0.0 100.0

7

140 317 0 0 30 919 0

65.0 0.0 0.0 35.0

400 280 171 236 43 1230 34.9 4 906 746

20 280 31 571 76 4194 18.0 1 753 965

7806 0 0 56 254 789 117

n/a n/a n/a n/a

200 280 789 117 n/a 1610 n/a n/a

50 437 32 282.7 0 168 149 0

4.8 67.6 0.0 27.6

140 317

99.4 0.0 0.0 0.6

50 280 140 664 47 2585 43.8 3 208 277

18

8 200 280 201 595 31 3698 8.4 23 891 809

0 0 0 31 571 0

0.0 0.0 0.0 100.0

0

0.0 0.0 0.0 100.0

50 280 30 572 76 4194 18.0 1 698 469

19

9 370 280 31 571 36 2006 18.0 1 753 965

20

140 317 0 0 30 919 0

65.0 0.0 0.0 35.0

335 280 873 180 n/a 1563 n/a n/a

137 511

n/a n/a n/a n/a

10 50 280 171 236 57 1625 34.9 4 906 746

11

0

0.0 0.0 0.0 100.0

650 870 495 575 42 2354 18.0 27 531 926

21

0 0 0 92 943 0

0.0 0.0 0.0 100.0

225 456 92 943 36 2006 18.0 5 163 490

12

0

0.0 0.0 0.0 100.0

435 870 495 575 42 2354 18.0 27 531 926

22

0 0 0 92 943 0

0.0 0.0 0.0 100.0

360 456 92 943 36 2006 18.0 5 163 490

Industrial & Engineering Chemistry Research Article



Article

0 0 495 575 0

100

100

0 0 495 575 0

21 20

0 0 0 873 180 0 0 30 572 0

19 18

0 0 346 0 0 0 31 571 0

17 14

0 0 200 148 0 0 0 200 148 0

13

capture exotherm cooling H2O (#13) heating required (streams 9 and 20 to 385 °C) cecomposition and rehydroxylation exotherm regenerator total regenerator duty HPHT steam flow (#9) flow from aux steam cooling (4 → 5) cooling H2O (#11) rehydroxylation exotherm (polisher) cooling H2O (#15) cooling (7 → 10) cooling H2O (#17)

50 340 kJ/s 150 kg/s 47 064 kJ/s 63 242 kJ/s 110 306 kJ/s 40 kg/s 218 kg/s 137 813 kJ/s 64 kg/s 181 101 kJ/s 540 kg/s 50 108 kJ/s 22 kg/s

Mg(OH)2 12.0 45.9 113.6 −16.6 154.9 579.1 33.0

Selexol 31.2 45.9 113.6 0.0 190.7 543.3 32.6

Denotes values adopted from the DOE report.

Analysis of the 734 MW (543 MW net) IGCC power plant modeled by DOE shows that 36 kg/s of high-quality and valuable steam is added to the fuel gas before the WGS reactor with a feed stream of H2, H2O, CO, and CO2. The addition of the steam allows for excess H2O in the WGS reactor, which promotes the combination of CO with H2O to form CO2 and H2. The proposed process introduces the cleaned effluent from the gasifier reactor into a CO2 absorber. The composition of this stream is based on the DOE report to be 11.1 mol % CO2, 27.5 mol % H2, 28.8 mol % CO, and 32.6 mol % H2O.1 Removal of the CO2 prior to the WGS reactor using the Mg(OH)2-based sorbent would increase the H2O content in the WGS inlet and reduce the requirement for additional steam. A 13.8 kg/s portion of the 36 kg/s of high-quality steam that would have been blended in the WGS reactor inlet can be utilized in the steam turbines to produce an additional 16.6 MW of power. Another effect of CO2 removal before the WGS reactor is the promotion of the kinetics of the WGS reaction by

H2 N2 H2O sorbent flow rate (kg/h) sorbent Mg state (mol %) (active species only) Mg(OH)2 MgCO3 MgO sorbent Mg state (wt%) (active species only) Mg(OH)2 MgCO3 MgO

stream

18 656 kJ/s 56 kg/s 17 399 kJ/s 26 437 kJ/s 43 837 kJ/s 19 kg/s 81 kg/s 50 928 kJ/s 23 kg/s 67,093 kJ/s 200 kg/s 18 570 kJ/s 8 kg/s

technology CO2 compression, MW base plant load, MW ASU duty,aMW steam credit, MW total auxiliary power usage, MW net power output, MW net plant efficiency (HHV %) a

Table 3. continued

capture exotherm cooling H2O (#13) heating required (streams 9 and 20 to 385 °C) decomposition and rehydroxylation exotherm regenerator total regenerator duty HPHT steam flow (#9) flow from aux steam cooling (4 → 5) cooling H2O (#11) rehydroxylation exotherm (polisher) cooling H2O (#15) cooling (7 → 10) cooling H2O (#17) Capture Unit 2 (after WGS Reactor)

Table 5. Auxiliary Loads Breakdown of an IGCC Plant Employing the Proposed Process for CO2 Separation (IGCC plant gross power generation = 734 MW)

0 0 1 469 500 0

16 15

Capture Unit 1 (before WGS Reactor)

0 0 1 469 500 0

22

Table 4. Energy Balance Summary

5278



Article

The sorbent leaves the regenerator at 385 °C nearly devoid of MgCO3 and largely rehydroxylated (stream 4; 48.3 active mol % Mg(OH)2, 48.3 active mol % MgO, and 3.4 active mol % MgCO3). The sorbent is cooled to 200 °C prior to entering the rehydroxylation unit by heat exchange with cooling water (stream 11), which is then used to preheat the sorbent entering the regenerator (stream 12). The cooled sorbent is then fed to the rehydroxylation unit near 200 °C and 280 psig, where it reacts with H2O from the recycled H2/H2O stream from the absorber unit (stream 2) to further convert the sorbent’s MgO content to Mg(OH)2. The recycling of the H2/H2O through the rehydroxylator eliminates the need for additional steam from the power island to convert the MgO to Mg(OH)2. It is important to note that no additional water feed is required for the rehydroxylation reaction within the rehydroxylation unit; the H2O consumed in the rehydroxilation reaction was generated during the capture reaction within the absorber. It is also worth noting that any trace amounts of CO2 not captured in the capture units will be captured in the rehydroxylator. The rehydroxylator unit is maintained isothermally at 200 °C during the exothermic hydroxylation reaction through heat exchange with 25 °C cooling water (stream 15). The sorbent leaves the rehydroxylator nearly fully reinstated to active Mg(OH)2 via stream 6, which feeds back to the absorber to complete the cycle. The gaseous effluent of the rehydroxylator (stream 8; 41.9 mol % H2 and 58.1 mol % H2O) is directed to the power island for power generation. The removal of H2O via MgO rehydroxylation in the rehydroxylation unit also increases the concentration of H2 in the gas stream. An important assumption employed is the sorbent utilization factor of 31%. This value stems from the conservative capture capacity value employed in our calculations, which is 4 mol CO2/kg sorbent. The theoretical limit of the sorbent, if all of the Mg(OH)2 were to react with CO2, is 13 mol of CO2/kg sorbent. Sorbent utilization of 31% was based on the measured CO2 capture capacity of 4 mol/kg and the theoretical capacity of 13 mol/kg. The sorbent utilization describes the accessibility of the active Mg(OH)2 phase within the sorbent pellet, which would likely result from diffusion limitations related to sorbent morphology. It was assumed that the remaining Mg(OH)2 was not available for reaction with CO2 and thus was treated as inert and unchanged during the capture−regeneration cycle. The process stream summary and energy balance summary are shown in Tables 3 and 4, respectively. The distribution of sorbent shown in Table 3 was found via mass balance. For instance, the molar amount of MgCO3 leaving the capture unit is assumed to be equal to the molar amount of CO2 captured. Assuming a capture capacity of 4 mol CO2/kg sorbent, found experimentally, the flow of MgCO3 can be calculated and the remainder of the sorbent is assumed to be unreacted Mg(OH)2 (i.e., inert during that cycle). The molecular weight of the sorbent was calculated by multiplying the composition of each species by its respective molecular weight, yielding the average molecular weight of the sorbent under the given composition, including the unreacted sorbent. After performing mass balances, energy balances were done to estimate the duties and flow rates of cooling water and steam required. Table 4 shows the flow rate of steam or cooling water required to maintain isothermal conditions based on the heat of reactions and temperature swings required. The formula used for the energy balances was Q = mCpΔT, were Q is the duty, m is the mass flow rate, Cp is the heat capacity, and ΔT is the

reducing the product CO2 concentration. Additionally, it would reduce the volumetric flow into the WGS reactor, allowing for a reduced reactor size. Figure 4 illustrates the proposed CO2 capture process, which captures 90% of the total amount of CO2 produced in the plant. In this figure, the fuel gas from the gasifier after impurity removal, stream 1, is introduced to the absorber. The composition of this stream is based on the DOE report to be 11.1 mol % CO2, 27.5 mol % H2, 28.8 mol % CO, and 32.6 mol % H2O.1 The stream is assumed to enter the absorber at 200 °C, which is reported in the DOE report as the temperature of the fuel gas entering the water-gas shift reactor. The sorbent flow rate is scaled to accommodate the incoming CO2 assuming a capture capacity of 4 mol CO2/kg sorbent. Due to the exothermic nature of CO2 absorption, cooling H2O (stream 13) is employed to maintain the sorbent bed at nearly isothermal conditions. CO2-free gas (i.e., CO, H2, and H2O) is removed from the top of the absorber (stream 2) and redirected into the WGS reactor. The outlet gas stream from the WGS reactor is introduced to the CO2 capture unit 2 and the effluent from the second CO2 capture unit (i.e., H2 and H2O) is later recycled into the rehydroxylation unit, eliminating the requirement for additional steam for the hydroxylation of the sorbent; this will be discussed further in detail. It is important to note that capturing a mole of CO2 with the sorbent releases a mole of H2O from the sorbent shown in reaction 1, yielding a stream with an approximate composition of 27.5 mol % H2, 28.8 mol % CO, and 43.5 mol % H2O. MgCO3 is then removed from the bottom of the capture reactor and transported to the regenerator (stream 3). The regenerator, shown in Figure 4, decomposes MgCO3 to MgO and CO2 at 385 °C and 280 psig. Heating the sorbent from 200 to 335 °C is accomplished through solid−gas heat exchange with a closed-loop steam cycle, where heat is provided by the sorbent leaving the regenerator (streams 11 and 12). A departure temperature for the solid−gas heat exchange was estimated at 25 °C; this resulted in a hot-side steam temperature (stream 12) of 360 °C (i.e., sorbent temperature of 385 °C minus the 25 °C departure temperature). The hot steam could then warm the incoming sorbent to 25 °C less than 360 °C, i.e., 335 °C. The heat duty of heating the sorbent from 335 to 385 °C, as well as overcoming the desorption energy, is accomplished through heat exchange with steam from an auxiliary steam generator. The auxiliary steam generator produces steam via coal combustion that heats the steam, which is then fed to the heat exchangers in the regeneration reactor. The steam fed to the regenerator from the auxiliary steam generator is assumed to be at 550 °C and 6 MPa. The steam operates on a closed-loop cycle to minimize the duty, and the boiler is assumed to have a conservative efficiency of 90%. Steam produced (stream 9) by heat exchange with the postregeneration gases of CO2 and H2O (stream 7) is also directly injected into the sorbent, which dilutes the desorbing CO2 and shifts the decomposition equilibrium to increase the potential for sorbent regeneration. The direct steam injection is also expected to hydroxylate about 50% of the MgCO3 entering the regenerator to produce Mg(OH)2, while the remaining MgCO3 is decomposed into MgO. The CO2 and steam leaving the regenerator is then cooled, condensing the steam and purifying the CO2. The purified CO2 (99.4 mol % or 99.8 wt %, with a balance of H2O) is then available at 50 °C and 280 psig (stream 19) for sequestration or other use. 5279



Article

temperature difference. For example, stream 20 has a Cp of 1563 J/kg °C (or 1.563 kJ/kg °C) and a flow rate of 599 kg/h (shown in Table 3); hence, the energy required to heat it from 335 to 385 °C (a ΔT of 50 °C) is 1.563 × 599 × 50 = 46 812 kJ/h. An additional duty is the heat of reactions, determined to be 19.6 kJ/mol CO2 for sorption of CO2, which was multiplied by the mass flow rate listed in Table 3. This calculation determined the total duty from the sorption reaction, which is equivalent to the desorption energies discussed above. The following assumptions were also employed to make the calculations possible: (1) CO2 capture capacity of sorbent at 200 °C, 280 psig = 4 mol CO2/kg sorbent, (2) heat of adsorption = 19.6 kJ/mol CO2 at 200 °C, (3) feed composition of 31.5 mol % CO2, 44.7 mol % H2, and 23.8 mol % H2O, (4) adiabatic reactor system, (5) perfect mixing of gas/solids for heat transfer, (6) power generation breakdown of gas turbine to steam turbine = 3:2, (7) the WGS reactor operates at 100% conversion, and (8) sorbent does not absorb water during CO2 sorption. (9) Isobaric thermal swing for CO2/H2O separation (10) Sorbent utilization = 31%; remainder of active phase is inert and has no effect on equilibrium calculations. The summary of auxiliary loads for a 734 MW (gross) IGCC plant employing the proposed CO2 separation process is summarized in Table 4. Compression costs for the sequestration of CO2 (pressurization to 2200 psia) amounts to 12.0 MW of required power (inlet gas at 50 °C, 294.7 psia and outlet at 241 °C, 2200 psia, assuming a Cp/Cv ratio of 1.3 for the gas stream). It should be noted that a significant increase in performance could be realized with a multistage compressor with cooling. The remainders of the auxiliary costs consisting of the air separation unit and baseplant load duties were assumed to be identical to the values calculated in the DOE report and were directly adopted into the current work.1 Table 5 lists the individual auxiliaries of the proposed technology, which total 154.9 MW, resulting in a net power output of 579.1 MW. These outputs are favorable when compared to the 190.8 auxiliary usage resulting in 543.3 MW of net power output in the DOE report. It should be noted that an extra 12.9 lb/s of coal required by the auxiliary steam generator may skew these numbers, but the HHV power plant efficiency takes this into account. The IGCC plant thermal efficiency was calculated on the basis of HHV of the coal fuel (Illinois #6 coal, which has a HHV of 11 666 Btu/lb). The IGCC plant thermal efficiency with CO2 separation by the proposed process was determined to be 33.0%, which compares favorably to the DOE report value of 32.6% for the same IGCC plant employing the Selexol process employed for CO2 removal. However, further improvements to plant efficiency may be realized if other warm gas contaminant removal technologies are employed (i.e., H2S, HCl). It has been reported that utilization of warm gas separation technologies may lead to an overall efficiency increase by as much as 4% for an IGCC plant,10 making the proposed CO2 capture process favorable to the Selexol process. Maintaining the thermal energy of the process streams within the plant is critical for maximizing the plant efficiency. Another significant benefit of the proposed process is that the system is insensitive to moisture content in the syngas feed.

The proposed process operates in the presence of water by design, which contrasts with typical solvent-based CO2 cleanup technologies. Typical solvent-based CO2 separation technologies, such as the commercial Selexol process, employ solvents that possess a high affinity for water. It is therefore necessary to dry the shifted syngas stream prior to entering the solventbased CO2 separation unit to less than 0.1% H2O,1 which can be very energy intensive, or the water will bind nearly irreversibly with the solvent and reduce its CO2 capture effectiveness.

■

CONCLUSIONS

■

AUTHOR INFORMATION

A process for the utilization of the Mg(OH)2 to separate CO2 at warm gas temperatures in a IGCC power plant has been designed. CO2 capture is conducted in the absorber at 200 °C and 280 psig, which benefits the overall IGCC plant efficiency by eliminating the thermal energy loss associated that would be present with typical solvent-based CO2 separation technologies, such as the Selexol process. The process in this report also produces a purified CO2 stream at 280 psig, which significantly reduces the pressure ratio (and corresponding energy use) required for the CO2 compression train for sequestration. Overall, the proposed process results in a predicted plant thermal efficiency of 33.0% (% HHV) when sized to a 734 MW gross IGCC plant, which is an efficiency improvement of 1% over the Selexol process for the same plant. Further, the proposed process is set apart from typical solvent-based CO2 separation technologies due to its insensitivity to water whereas typical solvent-based processes require an energy-intensive drying step upstream to the unit. In addition, other warm gas cleanup technologies (removal of HCl, H2S, etc.) require a warm gas CO2 removal technology, such as the proposed CO2 removal technology, to fully utilize the increase in efficiency associated with warm gas cleanup.

Corresponding Author

*E-mail: [email protected]. Tel: (304) 2854513. Fax: (304) 285-4403. Notes

The authors declare no competing financial interest.

■

REFERENCES

(1) Cost and Performance Baseline for Fossil Energy Plants; U. S. Department of Energy 2010; Vol. 1: Bituminous Coal and Natural Gas to Electricity: Revision 2. (2) Siriwardane, R. V.; Shen, M.-S.; Fisher, E. P.; Losch, J. Adsorption of CO2 on Zeolites at Moderate Temperatures. Energy Fuels 2005, 19 (3), 1153−1159. (3) Fernandez, J.; Gonzalez, F.; Pesquera, C.; Blanco, C.; Renedo, M. J. Study of the CO2/Sorbent Interaction in Sorbents Prepared with Mesoporous Supports and Calcium Compounds. Ind. Eng. Chem. Res. 2010, 49, 2986−2991. (4) Stevens, Robert W. J.; Siriwardane, R. V.; Logan, J. In Situ Fourier Transform Infrared (FTIR) Investigation of CO2 Adsorption onto Zeolite Materials. Energy Fuel 2008, 22, 3070−3079. (5) Nakagawa, K.; Ohashi, T. A Novel Method of CO2 Capture from High Temperature Gases. J. Electrochem. Soc. 1998, 145 (4), 1344− 1346. (6) Sircar, S.; Golden, C. M. A. PSA Process for Removal of Bulk Carbon Dioxide from a Wet High-Temperature Gas. U.S. Patent 6,322,612, 2001. 5280



Article

(7) Siriwardane, R. V. Regenerable Sorbents for CO2 Capture From Moderate and High Temperature Gas Streams. U.S. Patent 7,314,847, 2008. (8) Siriwardane, R. V.; Robinson, C.; Shen, M.; Simonyi, T. Novel Regenerable Sodium-Based Sorbents for CO2 Capture at Warm Gas Temperatures. Energy Fuels 2007, 21 (4), 2088−2097. (9) Siriwardane, R. V.; Stevens, R. W. Jr. Novel Regenerable Magnesium Hydroxide Sorbents for CO2 Capture at Warm Gas Temperatures. Ind. Eng. Chem. Res. 2009, 48, 2135−2141. (10) Jaeger, H. CO2 Cloud Looms Large Over IGCC. Gas Turbine World 2007, 37 (6), 21.

5281


Process for CO2 Capture from High-Pressure and Moderate

Recommend Documents