Efficient Utilization of Greenhouse Gases in a Gas-to-Liquids Process

Article pubs.acs.org/est

Efficient Utilization of Greenhouse Gases in a Gas-to-Liquids Process Combined with CO2/Steam-Mixed Reforming and Fe-Based Fischer− Tropsch Synthesis Chundong Zhang,†,‡ Ki-Won Jun,*,†,‡ Kyoung-Su Ha,*,† Yun-Jo Lee,† and Seok Chang Kang† †

Research Center for Green Catalysis, Korea Research Institute of Chemical Technology (KRICT), Yuseong, Daejeon 305-600, Republic of Korea ‡ Green Chemistry and Environmental Biotechnology, School of Science, Korea University of Science and Technology (UST), Yuseong, Daejeon 305-333, Republic of Korea S Supporting Information *

ABSTRACT: Two process models for carbon dioxide utilized gas-toliquids (GTL) process (CUGP) mainly producing light olefins and Fischer−Tropsch (F−T) synthetic oils were developed by Aspen Plus software. Both models are mainly composed of a reforming unit, an F− T synthesis unit and a recycle unit, while the main difference is the feeding point of fresh CO2. In the reforming unit, CO2 reforming and steam reforming of methane are combined together to produce syngas in flexible composition. Meanwhile, CO2 hydrogenation is conducted via reverse water gas shift on the Fe-based catalysts in the F−T synthesis unit to produce hydrocarbons. After F−T synthesis, the unreacted syngas is recycled to F−T synthesis and reforming units to enhance process efficiency. From the simulation results, it was found that the carbon efficiencies of both CUGP options were successfully improved, and total CO2 emissions were significantly reduced, compared with the conventional GTL processes. The process efficiency was sensitive to recycle ratio and more recycle seemed to be beneficial for improving process efficiency and reducing CO2 emission. However, the process efficiency was rather insensitive to split ratio (recycle to reforming unit/total recycle), and the optimum split ratio was determined to be zero.

1. INTRODUCTION Gas-to-liquids (GTL) process based on Fischer−Tropsch (F− T) synthesis attracts significant attention due to high oil price since the past decade.1 GTL synthetic fuels prepared by F−T synthesis contain extremely low sulfur and aromatic compounds.2,3 Because GTL synthetic fuel also shows low emissions of carbon monoxide, nitrogen oxides, and particulate matter, it can be considered as a clean fuel.3 CO2 is a wellknown greenhouse gas, and attempts to reduce the emission of CO2 into the atmosphere have long been tried.4 In order to reduce the concentration of CO2 in the atmosphere, various strategies such as separation, storage, and utilization have been implemented. Among these strategies, CO2 hydrogenation has received much attention as a promising way of CO2 conversion, because it not only reduces the amount of CO2 emission but also produces valuable fuels and chemicals necessary for petrochemical industry.4,5 The GTL technology can also be utilized to convert wasted associated natural gas, which is known to be another greenhouse gas and usually flared due to low economic value to result in enormous amount of CO2. Another important issue is related to carbon resources. Thanks to the improved exploring, boring, and retrieving skills, the extremely abundant nontraditional natural gas resources © XXXX American Chemical Society

such as shale gas and coal-bed methane are recently being discovered and utilized. This abundance in natural gas makes the energy paradigm promptly shift from petroleum to natural gas in some regions including U.S. and China, and the shift will be worldwide. At this moment, it is very important to render the GTL technology more efficient because we do not need to disturb ourselves to change the energy-related infrastructure and the transportation vehicles already fitted to petroleum if we can effectively convert natural gas to the clean liquid fuels and useful basic chemicals. Generally, there are three steps in the GTL process. The first step is syngas generation, where reforming of methane occurs, such as steam methane reforming (SMR),6 partial oxidation of methane (POX),7 carbon dioxide reforming of methane (CDR),8 and autothermal reforming (ATR).9 The second and also the key step is syngas conversion, which produces a broad range of hydrocarbons via F−T synthesis. The third step is upgrading, which converts high molecular weight products to Received: March 2, 2014 Revised: June 11, 2014 Accepted: June 16, 2014

A

dx.doi.org/10.1021/es501021u | Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Environmental Science & Technology

Article

Figure 1. Process flowsheet diagram of CUGP (Option 1).

Figure 2. Process flowsheet diagram of CUGP (Option 2).

processes producing DME, F−T diesel, and methanol, considering the variations in the prices of natural gas (NG) and the expected product. Kim et al.22 carried out a simulation study to find the optimum reaction condition for maximum production of synthetic oil. Although various simulation works are done to seek for more efficient process, so far there are few works on the whole GTL process, considering not only process efficiency but also CO2 emission. Therefore, on the basis of our previous work based on Co catalysts for F−T synthesis,23 we are now suggesting two new processes using Fe-based F−T catalysts, which can convert CO2 to CO by not only dry reforming but also RWGS. It was found that increased process efficiency and significantly reduced CO2 emission could be realized by recycling some unreacted syngas to reforming and F−T synthesis units.

naphtha, diesel, liquefied petroleum gas, etc., through catalytic hydrocracking.2,10 Typically, in F−T synthesis unit, Co-based or Fe-based catalysts have been widely employed.11 Compared with Fe-based catalysts, Co-based catalysts have higher activity and selectivity to long chain hydrocarbons.12 However, Febased catalysts are much cheaper than Co-based catalysts and active for reverse water gas shift (RWGS) reaction. Thus, Febased F−T synthesis catalysts are beneficial for CO 2 conversion.13−15 Unfortunately, in conventional low temperature F−T process using Fe-based catalysts, the optimum H2/ CO molar ratio is around 1.7.16 In this case, about 30% CO is converted to CO2 due to WGS reaction.17 Consequently, carbon loss resulting from CO2 formation cannot be avoided and large amounts of CO2 are emitted back from the process. If syngas with high concentration of CO2 is provided in the F−T synthesis unit using Fe-based catalyst, then not CO2 formation, but CO2 consumption can occur via RWGS as follows:18,19 CO2 + H 2 → CO + H 2O , ΔH298K = 41.2 kJ/mol

2. PROCESS MODEL DEVELOPMENT In general, a GTL process is composed of a feeding unit, a gas pretreatment unit, a reforming unit, an F−T synthesis unit, an upgrading unit and a product separation unit composed of several distillation and absorption columns. However, in present work, the gas pretreatment, upgrading, and product separation units will not be investigated in detail, since they are well established in current petrochemical industry and their influence on the process performance is relatively small, as described in our previous work.23 Thus, we established two simplified but meaningful process models mainly dealing with

(1)

In recent years, many researchers have tried to find more efficient GTL process. Hao et al.20 established GTL process model to find the optimal process structures of each integrated GTL process. Bao et al.10 provided an economic analysis to find a way to optimize an integrated process for cost saving and reduction of energy usage. Lee et al.21 modeled three GTL B



Article

(See SI), based on the experimental data. External Fortran program which correlates the fractional conversion with the CO2/(CO + CO2) ratio in syngas is integrated into the Rstoic model. After F−T synthesis, a gas stream containing unreacted syngas, byproduct C2C4 light paraffins, and main gas product C2C4 light olefins discharges from the top outlet of the F−T synthesis reactor. In addition, at the bottom outlet, there is a liquid stream containing main liquid product C5+ hydrocarbons and byproduct water. Then, C2C4 light olefins and C5+ hydrocarbons are separated as main gaseous and liquid products, respectively. In the PSA unit, Cu(I)-based or Ag(I)based π-complexation adsorbent could be selected for the separation of C2C4 olefins from the paraffins.25,26 A portion of the unreacted syngas along with byproduct C2C4 paraffins is recycled to reforming unit and F−T synthesis unit to enhance the process efficiency and CO2 conversion. The remaining unreacted syngas is vented to suppress the accumulation of inert gas such as N2.

the feeding, reforming, F−T synthesis and recycling units combined with several separation vessels as a whole, which are illustrated in Figures 1 and 2. As shown in Figures 1 and 2, the main difference between the two feasible options for CUGP is described as follows: (1) for option 1, fresh CO2 is fed to the reforming unit combined with natural gas and steam to produce CO by CO2/steam-mixed reforming first and then to the F−T synthesis unit to produce targeted hydrocarbons through RWGS and F−T synthesis reactions; (2) for option 2, fresh CO2 is directly fed to the F−T synthesis unit to produce targeted hydrocarbons through RWGS and F−T synthesis reactions without entering into the reforming unit. The reforming unit includes two parts, one is the prereformer and the other is the reformer. The reaction temperature and pressure of the prereformer are 550 °Cand 5 bar (gauge), respectively. Under this condition, almost all the C2+ hydrocarbons contained in the fresh feed natural gas and in the recycled gas from the F−T unit are converted into methane in the prereformer on a nickle-based catalyst. Moreover, the RGibbs model using chemical equilibrium for C1−C4 hydrocarbons is incorporated, and in the calculation options, “calculate phase equilibrium and chemical equilibrium” option is selected. Meanwhile, the RGibbs model is also used in the reformer and two typical reactions for CDR and SMR are considered, as follows: CO2 + CH4 → 2CO + 2H 2 , ΔH298K = 247 kJ/mol H 2O + CH4 → CO + 3H 2 , ΔH298K = 206 kJ/mol

3. RESULTS AND DISCUSSION 3.1. Effects of Recycle Ratio on the Performance of the CUGP. A series of simulation cases were carried out under the prescribed process conditions listed in Table S2 (SI). During the case study, 6 different recycle ratios for each process option were implemented to investigate the effects of the recycle ratio on the syngas ratio, the efficiency, the CO2 conversion, etc. To avoid severe carbon deposition in the reforming unit, the CO2/NG/H2O molar ratio of fresh feed was set in the range of (0.25−0.31):1:(2.0−2.1). In addition, the temperature and pressure were properly selected to avoid the generation of carbon deposit. In the reformer, we can get the syngas in flexible composition due to the two competitive methane reforming reactions, CDR and SMR. The optimum ratio of H2/(2CO + 3CO2) in obtained syngas for F−T synthesis using Fe catalyst ranges from 0.99 to 1.1(see Table S2 of the SI), which is met by changing the feed ratio of fresh gas in the reasonable range mentioned above. The different recycle ratio changes the syngas ratio, thus in order to keep the H2/ (2CO + 3CO2) ratio at optimum range, CO2/NG and H2O/ NG ratio should be adjusted slightly in some cases. The CO2/(CO2 + CO) ratio in syngas can be accumulated to a certain high level (CO2/(CO2 + CO) > 0.43) by increasing the recycle ratio and the higher recycle ratio leads to higher CO2/(CO2 + CO) ratio in syngas, as can be seen in Figure 3. This is a result of the fact that CO conversion to hydrocarbons is much higher than that of CO2 in F−T synthesis unit, thus with increasing recycle ratio, more and more CO2 is accumulated in syngas stream to the F−T reactor inlet. It is worth noting that the increase of recycle ratio, in spite of the accumulation of CO2 in syngas, leads to the decrease in total amount of vented CO2 in the whole process because of the increased overall CO2 conversion (Figure 4). Furthermore, when the recycle ratio increases to about 0.65 for option 1 and 0.7 for option 2, overall CO2 conversion approaches to zero, however, at recycle ratio = 0.95, around 80% of the fresh feed CO2 is converted in both options. This can be attributed to CO2 hydrogenation in the F−T synthesis unit and CO2 reforming in the reforming unit. In the F−T synthesis unit, the high CO2/(CO2 + CO) ratio together with optimum H2/ (2CO + 3CO2) ratio makes it possible that CO2 is converted to CO by RWGS reaction and then to targeted hydrocarbons by conventional F−T synthesis reaction. In addition, higher CO2/

(2)

(3)

The reaction temperature and pressure of the reformer are 850 °Cand 5 bar (gauge), respectively. In this case, it can be assumed that the two reactions above reach chemical equilibrium, since the reaction rates are very fast at elevated temperature. In order to simulate the reformer well, “Restricted chemical equilibrium” option implemented in Aspen software was selected in the RGibbs model. After reforming, the product stream of the reformer directly enters into the F−T synthesis reactor without removing CO2, since CO2 is one of the reactants in the F−T synthesis reactor using an Fe-based catalyst at high CO2/(CO + CO2) ratio in syngas (typically higher than 0.5) combined with sufficient hydrogen condition (i.e., H2/(2CO + 3CO2) ≥ 1). The reaction temperature and pressure of F−T synthesis reactor are 300 °Cand 10 bar (gauge) respectively. The main overall reactions on an Fe-based catalyst at high CO2 content syngas can be regarded as hydrogenation of CO and CO2, which can be expressed as follows: nCO + 2nH 2 → − (CH 2)n − +nH 2O

(4)

nCO2 + 3nH 2 → −(CH 2)n − +2nH 2O

(5)

Meanwhile, several typical chain growth and RWGS reactions are considered based on the conversion of fresh feed materials and selectivity of produced hydrocarbons obtained from our experimental results.24 Five or eight representative reactions for producing paraffins and olefins are listed in Supporting Information (SI) Table S1. The F−T synthesis reactor can be successfully simulated by RStoic model where fractional conversion is designated for each reaction listed in Table S1 C



Article

be expressed in detail, which are thermal efficiency (Teff) and carbon efficiency (Ceff). In the burner-type reformer, NG is used as a heating fuel, in order to reduce the consumption of NG, the vent gas can be applied to the burner to replace some part of NG as fuel. Thus, a certain amount of NG can be saved, and thereby thermal and carbon efficiencies would be improved. In some cases of present work, usually when the recycle ratio is lower than 0.7, the heat energy derived from the vent gas is sufficient for the reformer. However, in other cases, the vent gas can not provide all the heat energy required for the burner, especially when recycle ratio increases to 0.8 or even higher. That is, generally under high recycle ratio, additional NG is needed. Thus, Teff and Ceff should be calculated in two different ways under different conditions mentioned above. The calculation methods for Teff and Ceff are defined specifically as follows: If lower heating value (LHV) of vent gas > heat duty/0.9, then,

Figure 3. Effects of recycle ratio on the CO2/(CO2 + CO) ratio in syngas.

Teff =

LHV ole. + LHV C+5 HCs LHV feed NG

Ceff =

total mol C ole. + total mol C in C+5 HCs total mol C feed NG

(6)

(7)

where “LHV ole.” is the LHV of olefins; “LHV HCs” is the LHV of C5+ hydrocarbons; “LHV feed NG” is the LHV of feed NG; “total mol C ole.” is the total moles of C atoms in olefins; “total mol C in C5+ HCs” is the total moles of C atoms in C5+ hydrocarbons; and “total mol C feed NG” is the total moles of C atoms in feed NG. Else if LHV of vent gas < heat duty/0.9, then, C5+

Teff =

Figure 4. Effects of recycle ratio on overall CO2 conversion.

LHV ole. + LHV C+5 HCs LHV feed NG + heat duty/0.9 − LHV of vent gas (8)

(CO2 + CO) ratio leads to higher per-pass as well as overall CO2 conversion. Similarly, the CO2 recycled to the reformer is also consumed through CO2 reforming. Therefore, CO2 in the both streams recycled to the reforming and F−T synthesis unit is consumed as a reactant. This is the reason why the amount of CO2 in vent gas can be significantly reduced. The productivities of targeted C2C4 light olefins and C5+ hydrocarbons increase as the recycle ratio increases, as shown in Figure S1 (see SI). It is obvious that more recycle improves the overall conversion of fresh feed materials, thus more hydrocarbons are generated. In Figure S2 (see SI), the volume flow rate at the exit of the reformer and the entrance of the F− T reactor increases as the recycle ratio increases, because the amount of recycling unreacted syngas and produced light paraffins increases. If the recycle ratio is increased, on the one hand, then more targeted hydrocarbons could be generated, higher thermal and carbon efficiencies could be obtained and less amount of CO2 is emitted. This could increase the economic benefit. However, on the other hand, the volume of the two reactors should be enlarged to meet the increased flow rates and this may significantly increase the CAPEX and OPEX for the reactors. Since we assumed that the space velocity is same for all the cases, the reactor size will be dependent on the molar flow rates. Considering both aspects, an optimum recycle ratio should be existed under certain circumstances, in order to achieve maximization of economic benefit. 3.2. Thermal and Carbon Efficiencies. Before we proceed further, two important process efficiencies need to

Ceff =

total mol C in C+5

total mol C ole. + HCs total mol C feed NG + total mol C fuel NG

total mol C fuel NG (heat duty/0.9 − LHV of vent gas) × 1.012 = LHV of fuel NG

(9)

(10)

where the number 0.9 in the denominator of eqs. 8 and 10 is the estimated overall heat efficiency of the reforming unit, and the “heat duty” refers to the total heat duty of the endothermic units in the reforming unit, such as the prereformer and reformer, as shown in Figures 1 and 2 total mole C fuel NG" is the total moles of C atoms in fuel NG. The number 1.012 in eq. 10 is the moles of C atoms in one mole of fuel NG. Figure 5 shows the thermal and carbon efficiencies at different recycle ratios. It should be noted that the thermal and the carbon efficiencies monotonously increase with the increase of recycle ratio, even in high recycle ratio region (e.g., recycle ratio = 0.75−0.95). This tendency of process efficiency is different from general case as well as our previous GTL process using Co catalyst in F−T synthesis unit in high recycle ratio region, where the process efficiency was slightly decreased.23 This is mainly attributed to the improved conversion of CO2 in F−T synthesis unit at high recycle ratio. With increasing recycle ratio, more CO2 is converted to hydrocarbons in the F−T synthesis unit. Thus, the process efficiency could increase with the recycle ratio. However, in the conventional F−T reactor, D



Article

ratio increases accordingly, which can be seen from SI Table S3. Meanwhile, the CO2 emitted in vent gas is also reduced as shown in Figure 7. However, more recycle to reformer is not

Figure 5. Effects of recycle ratio on process efficiency.

CO2 is generated rather than converted through WGS reaction, especially using Fe-based catalysts. Consequently, large amount of CO2 is emitted. In this case, with increasing recycle ratio, more and more CO2 is generated. Thus, the process efficiency could decrease due to increased CO2 formation. Figure 6 shows

Figure 7. Effects of split ratio on overall CO2 conversion (recycle ratio = 0.9).

beneficial for thermal and carbon efficiencies due to the highly endothermic property of CDR reaction, as we have mentioned above. As for the flow rates of reformer and F−T reactor as shown in SI Table S3, it is obvious that the flow rate of reformer increases and the flow rate of F−T synthesis reactor decreases with the increase of split ratio since the recycle itself to the reformer increases. 3.4. Effects of H2 Content on the Performance of the CUGP. We compared also the performance of two process options under H2/(2CO + 3CO2) = 1−1.1 (H2-sufficient) and H2/(2CO + 3CO2) ≈ 0.4 (H2-deficient, when CO/(CO + CO2) = 0.5) condition. As shown in Table 1, in order to make H2/(2CO + 3CO2) ≈ 0.4, more CO2 are needed in the reformer since different reforming process generates syngas with different H2/CO ratio. While the H2/CO ratio is around 1 in the case of CDR, the H2/CO ratio is around 3 in the case of SMR. Meanwhile, unfortunately under the H2 deficient condition, large amount of CO2 was generated in Fe-based F−T synthesis due to WGS reaction. Thus, the absence of CO2 consumption leads to increase in CO2 emission. However, less amount of CO2 is needed to meet H2/(2CO + 3CO2) = 1−1.1, and fortunately under the H2 sufficient condition, CO2 generation could be successfully suppressed or CO2 could be converted to valuable hydrocarbons via RWGS reaction in the F−T synthesis unit. Thus, according to SI Table S4, the overall CO2 conversion of H2-deficient condition is lower than that of H2-sufficient condition, especially at high recycle ratio (e.g., recycle ratio = 0.95) and the CO2 flow rate in vent gas stream under H2-deficient condition is much larger than that under H2sufficient condition. As for the thermal and carbon efficiencies, also due to the higher CO2 consumption under H2 deficient condition, more heat energy is needed in the reformer. Thus, the thermal and carbon efficiencies under H2-deficient condition are lower than those under H2-sufficient condition. It is obvious that the volume flow rates at the exit of the reformer and the entrance of the F−T reactor under H2-deficient condition are larger than those under H2-sufficient condition, since the amount of fresh feed and recycled syngas is larger under H2-deficient condition. Thus, it can be concluded that H2-sufficient condition (i.e., H2/

Figure 6. Effects of split ratio on process efficiency (recycle ratio =0.9).

the thermal and carbon efficiencies at different split ratios. As we can see, there is no big difference in thermal and carbon efficiencies at different split ratios. Meanwhile, the tendency of the thermal and the carbon efficiencies is downward with increasing split ratio, and the thermal and the carbon efficiencies reach the highest value when split ratio approaches to 0. It should be noted that the CDR is highly endothermic and more CO2 recycled to reformer consumes more energy, even though at the same time, more CO and H2 are generated as split ratio increases. The contribution of heat duty for the value of Teff and Ceff seems to be dominant when split ratio increases. On the basis of this observation, the high split ratio is not beneficial for the process efficiency. 3.3. Effects of Split Ratio on the Performance of the CUGP. A series of simulation cases including 6 different split ratios for each process option were implemented to investigate the effects of split ratio on the performance of the CUGP. The prescribed conditions and some results are listed in Table S3 of the SI. As the split ratio increases, larger amount of unreacted syngas enters into the reformer and because of the CDR reaction, more CO2 is converted to CO and H2, thus the CO2/ (CO2 + CO) ratio in syngas decreases and H2/(2CO + 3CO2) E



Article

Table 1. Performance of Process Option 1 & 2 under Hydrogen Deficient and Sufficient Conditiona CO2 (kmol/h)

H2O (kmol/h)

NG (kmol/h)

S2

76.5 78.5 45.4 45.6 CO2 (kmol/h)

229 229 229 226 H2O (kmol/h)

0.40 0.70 0.45 0.58

29.5 29.5 31.0 31.5

229 229 229 226

option

R1

S2

1 2 1 2

0.80 0.80 0.95 0.95

0.40 0.70 0.45 0.58

option

R1

1 2 1 2

0.80 0.80 0.95 0.95

reformer (cum/h)

FTS (cum/h)

Ceff

15 659 16 852 17 599 18 190 reformer (cum/h)

4414 4492 5312 5295 FTS (cum/h)

0.833 0.827 0.840 0.840

14 266 16 024 16 926 17 735

4124 4193 5244 5280

R1b

R2c

R3d

R4e

T eff.

C eff.

109 109 109 109 NG (kmol/h)

0.401 0.398 0.400 0.402

2.005 1.996 2.007 2.009

0.500 0.501 0.501 0.500

0.708 0.711 0.350 0.350

0.653 0.651 0.656 0.657

0.819 0.817 0.826 0.825

R1b

R2c

R3d

R4e

Teff

109 109 109 109

1.045 1.062 1.066 1.034

3.892 3.908 3.931 4.043

0.365 0.359 0.360 0.389

0.661 0.671 0.192 0.198

0.665 0.660 0.670 0.669

a

Note: (1) R = recycle 1/recycle; (2) S = recycle to reformer/recycle 1 (see Figures 1 and 2). bR1 = H2/(2CO + 3CO2). cR2 = H2/CO. dR3 = CO2/(CO2+CO). eR4 = CO2 in vent gas/fresh feed CO2, here, R1, R2, and R3 refer to synthesis gas composition at the inlet of F−T synthesis section.

Table 2. Comparison of Two Present CUGP Options with Known GTL Processes20,27 parameter

GTL-ATRa

GTL-CRb

GTL-Fe FRc

GTL-Fe CSTRd

CUGP 1

CUGP 2

NG feed (kmol/h) % of tail gas to reformer % of tail gas to F−T reactor carbon efficiency (%) thermal efficiency (%) total CO2 emission (g CO2/ MJ F−T product)

7770 60.0 unknown 71.0 59.4 34.4

7790 75.0

13.9 92.5

13.9 90.0

109

109

70.9 59.2 34.8

67.9 54.4 27.0

61.5 49.4 36.2

90.0 86.0 68.7 4.85

90.0 86.9 69.5 4.03

a

GTL-ATR = typical GTL process using conventional ATR.27 bGTL-CR = typical GTL process using conventional combined reforming.27 cGTL-Fe FR = typical GTL process using fixed bed F−T synthesis reactor and Fe catalyst.20 dGTL-Fe CSTR = typical GTL process using continuous stirredtank reactor and Fe catalyst.20

as shown in Figure 5, the thermal and carbon efficiencies of option 2 were computed to be lower than those of option 1 in the recycle ratio range of 0.5−0.7, but reverse results were found in the range of 0.8−0.95. The higher thermal and carbon efficiencies of option 1 at low recycle ratio (R = 0.5−0.7) are mainly attributed to the higher CO2 conversion of CDR and the sufficient heat energy of the vent gas which can fully compensate the heat needed in the reforming unit. On the contrary, the higher thermal and carbon efficiencies of option 2 at high recycle ratio (R = 0.8−0.95) was mainly due to the higher CO2 conversion in the F−T synthesis unit and the lower energy consumption in the reforming unit when additional fuel NG was needed. In addition, the thermal and carbon efficiencies of option 2 were computed to be slightly higher than those of option 1 at different split ratios, as shown in Figure 6. This also can be explained by the lower energy consumption in the reforming unit and the higher CO2 conversion in the F−T synthesis unit at the same recycle and split ratio in the case of option 2. Considering both the process efficiency and the CO2 emission, the option 2 seems to be more beneficial for CUGP especially at high recycle ratio (e.g., recycle ratio = 0.9). Table 2 shows the performance differences between previous GTL processes and our two CUGP options. As shown in Table 2, the carbon efficiency is increased by 21.1−41.3%, the thermal efficiency is increased by 15.7−40.7% and the total CO2 emission is reduced by 82.0−88.4%, compared to different literature data.20,27 Due to the efficient utilization of greenhouse gases in the reforming and the F−T synthesis units, the thermal and carbon efficiencies of the two CUGP options were shown to be successfully improved and total CO2 emissions could be significantly reduced, compared with the conventional GTL processes in literature.28−30

(2CO + 3CO2) = 1−1.1) is more beneficial for improving CO2 utilization rate and process efficiency of the CUGP. 3.5. Performance Comparison for the Two CUGP Options and Known GTL Process. As mentioned previously, CO2 can be converted to CO by CDR in the reformer or RWGS in the F−T reactor, and then to hydrocarbons via F−T synthesis. Generally, CO2 equilibrium conversion of CDR reaction in the reformer is higher than that of RWGS reaction in the F−T synthesis reactor using Fe catalyst, especially at low recycle ratio in this work. However, the CDR reaction is wellknown to be highly endothermic and the heat of reaction is about 5 times larger than the heat of RWGS reaction, as can be seen from the ΔH value given above. Meanwhile, it should be noted that the CO2 conversion of RWGS reaction can be improved by increasing recycle ratio. In addition, the RWGS reaction is conducive to suppressing the heat generated by the F−T reactor. As shown in Figure 3 and SI Table S3, the CO2/(CO2 + CO) ratio of option 2 is always higher than that of option 1 at the same recycle and split ratio. This was caused by different process structure. In option 1, CO2 is converted to CO in the reformer, so the CO2 content in the obtained syngas is reduced. In option 2, however, CO2 is directly fed to the F−T synthesis reactor without consuming CO2 in the reformer. Figure 4 shows that the amount of CO2 in vent gas of option 2 is slightly higher than that of option 1 since the consumed amount of CO2 by CDR in the reformer is found much larger than that by RWGS in F−T synthesis reactor under given conditions. However, the difference of CO2 content in vent gas between the two options was shown to decrease with increasing recycle ratio, which can be attributed to the increased CO2 conversion of option 2 in F−T synthesis reactor at higher CO2/(CO2 + CO) ratio. Apart from CO2 content in the syngas and vent gas, F



■

Article

promoted Co/γ-Al2O3 and K-promoted Fe catalysts. Catal. Commun. 2011, 12 (11), 936−939. (14) Yao, Y. L.; Liu, X. Y.; Hildebrandt, D.; Glasser, D. Fischer− Tropsch synthesis using H2/CO/CO2 syngas mixtures over an iron catalyst. Ind. Eng. Chem. Res. 2011, 50 (19), 11002−11012. (15) Riedel, T.; Schaub, G.; Jun, K. W.; Lee, K. W. Kinetics of CO2 hydrogenation on a K-Promoted Fe catalyst. Ind. Eng. Chem. Res. 2001, 40 (5), 1355−1363. (16) Dry, M. E. High quality diesel via the Fischer−Tropsch processA review. J. Chem. Technol. Biotechnol. 2002, 77 (1), 43−50. (17) Felmer, J. L.; Sllveston, P. L.; Hudglns, R. R. Steady-state study of the Fischer−Tropsch reaction. Ind. Eng. Chem. Prod. Res. Dev. 1981, 20 (4), 609−615. (18) Riedel, T.; Claeys, M.; Schulz, H.; Schaub, G.; Nam, S. S.; Jun, K. W.; Choi, M. J.; Kishan, G.; Lee, K. W. Comparative study of Fischer−Tropsch synthesis with H2/CO and H2/CO2 syngas using Feand Co-based catalysts. Appl. Catal. A: Gen. 1999, 186 (1−2), 201− 213. (19) Jun, K. W.; Roh, H. S.; Kim, K. S.; Ryu, J. S.; Lee, K. W. Catalytic investigation for Fischer−Tropsch synthesis from bio-mass derived syngas. Appl. Catal. A: Gen. 2004, 259 (2), 221−226. (20) Hao, X.; Djatmiko, M. E.; Xu, Y. Y.; Wang, Y. N.; Chang, J.; Li, Y. W. Simulation analysis of a gas-to-liquid process using aspen plus. Chem. Eng. Technol. 2008, 31 (2), 188−196. (21) Lee, C. J.; Lim, Y. S.; Kim, H. S.; Han, C. H. Optimal gas-toliquid product selection from natural gas under uncertain price scenarios. Ind. Eng. Chem. Res. 2009, 48 (2), 794−800. (22) Kim, Y. H.; Jun, K. W.; Joo, H. K.; Han, C. H.; Song, I. K. A simulation study on gas-to-liquid (natural gas to Fischer−Tropsch synthetic fuel) process optimization. Chem. Eng. J. 2009, 155 (1−2), 427−432. (23) Ha, K. S.; Bae, J. W.; Woo, K. J.; Jun, K. W. Efficient utilization of greenhouse gas in a gas-to-liquids process combined with carbon dioxide reforming of methane. Environ. Sci. Technol. 2010, 44 (4), 1412−1417. (24) Kang, S. C.; Jun, K. W.; Lee, Y. J. Effects of the CO/CO2 ratio in synthesis gas on the catalytic behavior in Fischer−Tropsch synthesis using K/Fe−Cu−Al catalysts. Energy Fuels 2013, 27 (11), 6377−6387. (25) Rege, S. U.; Padin, J.; Yang, R. T. Olefin/paraffin separations by adsorption: π-complexation vs. kinetic separation. AIChE J. 1998, 44 (4), 799−809. (26) Yang, R. T.; Kikkinides, E. S. New sorbents for olefin/paraffin separations by adsorption via π-complexation. AIChE J. 1995, 41 (3), 509−517. (27) Schanke, D.; Hansen, R.; Sogge, J.; Hofstad, K. H.; Wesenberg, M. H.; Rytter, E. U.S. patent 6696501B2, 2004. (28) Union Gas Limited. http://www.uniongas.com/about-us/ about-natural-gas/chemical-composition-of-natural-gas (accessed April 2013). (29) Steynberg, A. P.; Dry, M. E., Eds. Fischer−Tropsch Technology, Series: Studies in Surface Science and Catalysis; Elsevier: Amsterdam, 2004; Vol. 152. (30) Eilers, J.; Posthuma, S. A.; Sie, S. T. The shell middle distillate synthesis process (SMDS). Catal. Lett. 1990, 7 (1−4), 253−270.

ASSOCIATED CONTENT

S Supporting Information *

Figure S1: Effects of recycle ratio on productivities of C2C4 olefins and C5+ hydrocarbons; Figure S2: effects of recycle ratio on volume flow rate of Reformer and F−T reactor; Table S0: composition of the fresh feed and fuel NG27; Table S1: reactions in the F−T synthesis reactor model; Table S2: process conditions of simulation cases at different recycle ratio; and Table S3: performance of process options 1 and 2 at different split ratio. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*Phone: (82) 42-860-7671; fax: (82) 42-860-7388; e-mail: [email protected]. *Phone: (82) 02-3472-4892; fax: (82) 42-860-7388; e-mail: [email protected]. Present Address §

Department of Chemical and Biomolecular Engineering, Sogang University, Mapo, Seoul 121-742, Korea.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the KRICT Principal Project Program. (Project No. KK-1401-H0)

■

REFERENCES

(1) U.S. Energy Information Administration. Crude oil Price; http:// www.eia.gov/. (2) Knottenbelt, C. Mossgas “gas-to-liquid” diesel fuels-an environmentally friendly option. Catal. Today 2002, 71 (3−4), 437−445. (3) Li, X.; Huang, Z.; Wang, J.; Zhang, W. Particle size distribution from a GTL engine. Sci. Total Environ. 2007, 382 (2−3), 295−303. (4) Wang, W.; Wang, S. P.; Ma, X. B.; Gong, J. L. Recent advances in catalytic hydrogenation of carbon dioxide. Chem. Soc. Rev. 2011, 40 (7), 3703−3727. (5) Mikkelsen, M.; Jørgensen, M.; Krebs, F. C. The teraton challenge. A review of fixation and transformation of carbon dioxide. Energy Environ. Sci. 2010, 3 (1), 43−81. (6) Oliveira, E. L. G.; Grande, C. A.; Rodrigues, A. E. Methane steam reforming in large pore catalyst. Chem. Eng. Sci. 2010, 65 (5), 1539− 1550. (7) Zhou, X. W.; Chen, C. X.; Wang, F. C. Multi-dimensional modeling of non-catalytic partial oxidation of natural gas in a high pressure reformer. Int. J. Hydrogen Energy 2010, 35 (4), 1620−1629. (8) Zhang, Y. F.; Zhang, G. J.; Zhang, B. M.; Guo, F. B.; Sun, Y. L. Effects of pressure on CO2 reforming of CH4 over carbonaceous catalyst. Chem. Eng. J. 2011, 173 (2), 592−597. (9) Palma, V.; Ricca, A.; Ciambelli, P. Monolith and foam catalysts performances in ATR of liquid and gaseous fuels. Chem. Eng. J. 2012, 207−208, 577−586. (10) Bao, B. P.; El-Halwagi, M. M.; Elbashir, N. O. Simulation, integration, and economic analysis of gas-to-liquid processes. Fuel Process. Technol. 2010, 91 (7), 703−713. (11) Davis, B. H.; Occelli, M. L. Advances in Fischer−Tropsch Synthesis, Catalysts, and Catalysis: CRC Press: Boca Raton, FL, 2010. (12) Khodakov, A. Y.; Chu, W.; Fongarland, P. Advances in the development of novel cobalt Fischer−Tropsch catalysts for synthesis of long-chain hydrocarbons and clean fuels. Chem. Rev. 2007, 107 (5), 1692−1744. (13) Gnanamani, M. K.; Shafer, W. D.; Sparks, D. E.; Davis, B. H. Fischer−Tropsch synthesis: Effect of CO2 containing syngas over Pt G


Efficient Utilization of Greenhouse Gases in a Gas-to-Liquids Process

Recommend Documents