B. B. Kimchi, Green Gas Process Technologies Inc., Rishon LeZion, Israel; P. S. NORTHROP, Northrop Consulting, Lexington, Kentucky (U.S.); and V. PATTABATHULA, SVP Chemical Plant Services, Brisbane, Australia; and P. V. BROADHURST, Consultant, Yarm, UK
This article describes a method for the simultaneous separation of a mixture containing hydrogen (H2), methane (CH4) and carbon dioxide (CO2), which may result from certain shifted synthesis gas (syngas) mixtures. This type of stream could result from steam methane reforming (SMR) followed by water-gas shift (WGS), followed by methanation of the residual carbon monoxide (CO), for example. The resulting mixture is introduced into a distillation column to generate:
An overhead distillate stream very rich in H2
An intermediate-volatility stream off the reboiler that is very rich in CH4
A bottoms stream comprising liquid CO2.
Part 1 (December 2024) proposed a way to separate syngas mixtures and an application to the ammonia process, distillation feasibility and a thermodynamic analysis, a detailed case study and process simulation. Part 2 of this article will detail the process simulation and provide detailed lab results.
Process simulation results. The proposed technology was modeled on a process simulatora. The thermodynamic packages used were the Peng-Robinson equation of state (EOS), and the Peng-Robinson Sour-Vapor package. Tsang and Streett16 demonstrated that the prediction of H2/X [X = CO2, nitrogen (N2), CH4, Argon (Ar) and CO] is accurately achieved by the Peng Robinson EOS relative to the experimental binary vapor-liquid equilibrium (VLE) data—deviation occurs only at very high pressures near the critical pressure.
The feed stream is fed at 15 bar and –138°C with a composition as shown in TABLE 2 (Part 1, December 2024). The simulation is for a 12-tray column, with the feed on Tray 3 (numbering from the bottom).
The results of the simulation are shown in FIG. 4 and include molar flowrates of vapor (moving upward) and liquid molar flowrates (moving downward).
It can be seen that on feed Tray 3 and above, CO2 vapor is found at amounts of ≤ 8.18 kgmol/hr, which is < 2% of the CO2 introduced into the column. This indicates that virtually all of the CO2 simply moves down the column to the reboiler and is removed as bottoms liquid. The CO2-rich liquid contains about 12 mol% CH4 and traces of the other components.
The lighter components, including 2,229 kgmol/hr of H2, 2,965 kgmol/hr of CH4, 1,125 kgmol/hr of N2 and 281.8 kgmol/hr of Ar are vaporized off the feed tray. They each exceed the feed rate because the reflux from above is vaporized there, as well.
From a material balance standpoint, virtually all of the H2 and N2 comes out the top of the tower; the bulk of the CH4 comes out with the reboiler vapor and liquid streams.
The frost (freeze-out) data for CO2/CH4 was measured experimentally.15,17 For mixtures of 0.01 mol% of CO2, the frost temperature is –108°C at a pressure of 10 atm, and –103° at a pressure of 25 atm (Figure 3 in literature16). Note the relatively weak effect of total pressure on the CO2 freezing temperature (although it has a large effect on the movement of the phase envelope).
Reverting to the case in literature,5 the operating temperatures of the upper stages needed to remove CO2 to a level of 1% or less in the overhead stream are shown in FIG. 5 (Figure 3 in literature5).
Experimental data for CO2/CH4 liquid/solid equilibria is available in literature,15,17 and is shown as the liquidus curve in FIG. 5 (Figure 3 in literature5). It can be seen that the upper tray temperature profile is maintained to avoid the liquidus curve and prevent solid formation in the liquid. As discussed above, solid freeze-out in the vapor phase was prevented by the presence of high concentrations of H2, as well as by working at pressures above the critical pressure of pure CH4 to avoid the collapse of the phase envelope at high methane concentrations.
Enabling working temperatures lower than the binary CO2/CH4 SLE limit may be supported by the observation that, for high H2 + N2 concentrations relative to CO2, when cooling the mixture from gas, the diluents prevent freezing due to the reduction of the fugacity of the CO2. The presence of H2/N2/Ar diluents being flashed back into the vapor was observed in literature.13 This suggests that, “…at a given temperature, the bubble-point pressure increases with increasing mole fraction of the impurities (Ar + N2) in the mixture.”13 This makes sense as Ar and N2 (and H2) are all “lighter” than CO2.
Using the data shown in Table 2 of literature,16,13 the calculated bubble-point pressure is 5.98 MPa for the 98% CO2 + 1% Ar + 1% N2 mixture, and 8.88 MPa for the 90% CO2 + 5% Ar + 5% N2 mixture, corresponding to an increase of 17% and 74%, respectively, from the vapor pressure of pure CO2 (5.09 MPa) at 288.15 K.
The mixtures that were measured experimentally were 0.9 CO2, 0.05 N2, 0.05 Ar; and 0.98 CO2, 0.01 Ar, 0.01 N2, 0.95 CO2, 0.03 H2 and 0.02 Ar.
“Compared to the vapor pressure of pure CO2 at the same temperature, the bubble-point pressure, which is the maximum pressure that allows the formation of a homogeneous gas phase, increases significantly in the presence of a mixed gas of Ar and N2 (or Ar and H2), even though the total mole fraction of the impurity is < 0.10. Among the three impurities in the two ternary systems (i.e., N2, Ar and H2), H2 has the largest effect on the bubble-point pressure.”13,16
When the side stream is withdrawn, a complete CO2/CH4 separation is not necessary (or possible). That stream, which contains varying amounts of CO2 up to 20%, may be recycled, depending on the side stream withdrawal stage. In the example of FIG. 4, the side stream was withdrawn from the reboiler overhead. In literature,18 a revamp of a H2 plant with recycle of PSA off-gas (containing about 68% CO2, 3% CH4 and 27% H2) into the steam reformer reduced natural gas consumption. Dry reforming of CH4 with CO2 could also be contemplated if the side stream contains CO2 in various amounts.
In a modified version of the process, cold liquid reflux [mostly liquid N2, or (LiN)] is injected into the top of the column. It travels down the packing (or across the trays) and condenses CH4 while the LiN vaporizes. The liquid CH4 is withdrawn from the middle of the tower and put into a side reboiler. Residual N2 (and some CH4) is vaporized and returned to the column. The liquid CH4 from the side reboiler is sprayed into the open portion of the lower part of the column, in a manner similar to that of a single-step cryogenic distillation processb.9,12 Solid CO2 is formed as the liquid methane vaporizes. That solid falls onto a melt tray that is warmed (relatively) by CO2-rich vapor coming up from the lower part of the column. In that stripping section of the column, residual methane is removed from the liquid CO2. The bottoms product is liquid CO2 that is ready for injection.
Experimental lab results. Experimental results are discussed here for the cooling of two mixtures: the feed to the column and the mixture that is present above the feed tray (Tray 3, which is composed of Blend A and the steady-state gas vapor composition found on Tray 3). From these results, it is unclear whether the proposed process is viable. Some frosting (S + V) may occur during transient or upset conditions, which could require costly downtime to derime the tower.
In an exemplary measurement, Blend A [feed composition to the column (Part 1, TABLE 2)] was loaded to the fixed-volume equilibrium cell at ambient temperature and a pressure of 15 bar. Liquid N2 was used to cool an air bath that was in contact with the equilibrium cell. Pressure and temperature were monitored while cooling. The average cooling time to reach the inferred frosting temperature was between 15 min and 20 min.
Increased rates of pressure reduction were associated with frosting or liquid formation. Frosting was first observed at –77°C for Blend A. The pressure in the cell at that point decreased to ~10 bar. Further cooling showed a dewpoint (V + L) at –100°C, and the pressure in the cell decreased to ~7.7 bara. At this pressure and temperature (7.7 bar, –100°C), the phase diagram of pure methane shows a dewpoint of –135°C. This suggests that the liquid phase included CO2 at –100°C, though it is unclear how much of the CO2 was dissolved in the liquid phase and how much may have remained as solid.19
Follow-up measurements (not yet funded), which could include direct sampling from the equilibrium cell and faster cooling using higher bath temperatures (which may prevent nucleation), are required. Earlier attempts determined that the sample lines were plugged with solid when the bath temperature was –138°C, so modifications to the equilibrium cell may be required. Finally, it is known that for flue gas containing 0.14 molar fraction of CO2, 0.83 molar fraction of N2 and 0.03 molar fraction of O2, liquification does not occur at any pressure or temperature, and only frosting appears. Therefore, it is suggested that N2 promotes freeze-out, and by measuring a ternary CO2/CH4/H2 mixture, no frosting would be observed compared to the multicomponent mixture that contains ~35% weight N2.
Those subsequent measurements should help determine whether the original idea of utilizing an ordinary distillation column is viable. Alternatively, the process may require modification, e.g., setting the working pressure to 55 atm. This would avoid frosting and allow V + L operation across the column. It is also unclear if the single-step cryogenic distillation processb proposed previously is viable (e.g., there may not be enough CH4 present to liquefy to convert all of the CO2 to solid).
Comparison of SGFS to chemical absorbing process. TABLE 3 shows the column's energy consumption and cost. Energy costs were calculated, assuming a condenser utility cooling using a nitrogen refrigerant utility at a cost of $1/kcal/hr/yr. Lower costs may be achieved by cooling with a cascade system or other methods.
The energy demand and cost were determined for a reflux ratio of 1.9 in this example. When running the column at 15 bar, the reflux composition was primarily (>99%) liquid nitrogen.
The simulated energy consumption was compared with a double-stage a-MDEA absorbing process, which has a minimum energy consumption of 9,500 kcal/kgmole of CO2.20 TABLE 2 in Part 1 shows that 654 kgmole/hr of CO2 is the feed to the column, so the total energy for amine regeneration is 6.213 Gcal/hr, and the CO2 product is generated as a wet gas at atmospheric pressure. Additional energy is needed to compress/liquify it for sequestration or enhanced oil recovery.
When choosing a cascade system, the column pressure (or the upper trays and condenser) could be higher than 15 bar as shown in FIG. 4. For example, at a pressure of 45 bar, the liquid reflux is methane and nitrogen (vs. only nitrogen, as in FIG. 4). As a result, an overhead temperature of –166°C is required. This temperature is very close to mixed refrigerant used in LNG refrigeration cycles (–162°C).21,22,23,24
Takeaways. A three-way separation of an ammonia syngas mixture was modeled in a single cryogenic distillation column with a side stream to recover unreacted methane (along with some CO2). Energy consumption was similar to existing absorption processes. However, the CO2 was recovered as a liquid and a higher ammonia yield was obtained, resulting in an improved and more sustainable process for the manufacture of ammonia. Future work should include the integration of cascade systems for cooling, further process optimization and additional measurements. This could include measuring the phase diagram of the ternary CO2/CH4/H2. It is suspected that the presence of N2 promotes CO2 frosting at slightly higher temperatures relative to the CO2/CH4 binary, so frosting may be difficult to avoid. HP
ACKNOWLEDGEMENTS
The authors wish to acknowledge Thermo Dynamico’s staff members Rubin J. McDougal, Lane R. Gardner and Seth T. Herway for performing the lab measurements.
NOTES
Honeywell’s UniSim Design v. R370, build 13058
ExxonMobil’s Controlled Freeze Zone™
LITERATURE CITED
Berstad, D., P. Nekså and G. A. Gjøvåg, “Low-temperature syngas separation and CO2 capture for enhanced efficiency of IGCC power plants,” Energy Procedia, Vol. 4, 2011.
Air Liquide, “World premiere: Air Liquide inaugurates its CO₂ cold capture system, Cryocap™,” November 5, 2015, online: https://www.airliquide.com/group/press-releases-news/2015-11-05/world-premiere-air-liquide-inaugurates-its-co2-cold-capture-system-cryocaptm#:~:text=Cryocap%E2%84%A2%20is%20the%20first%20CO2%20capture%20technology%20using,improving%20efficiency%2C%20leading%20to%20an%20increased%20hydrogen%20production.
Honeywell, “Honeywell technology enables large U.S. carbon capture and storage project,” April 12, 2021, online: https://www.honeywell.com/us/en/press/2021/04/honeywell-technology-enables-large-us-carbon-capture-and-storage-project
U.S. Department of Energy, “Pre-combustion carbon capture research,” online: https://www.energy.gov/fecm/pre-combustion-carbon-capture-research
Eakman, J. M. and H. A. Marshall, “Separation of carbon dioxide and other acid gas components from hydrocarbon,” U.S. Patent No. 4,149,864, 1979.
Valencia, J. A. and R. D. Denton, “Method of separating acid gases, particularly carbon dioxide, from methane by the addition of a light gas such as helium,” U.S. Patent No. 4,511,382, 1985.
Malhotra, A., T. Ahmad and B. Richard, “Low-delta P purifier for nitrogen, methane and argon removal from syngas,” U.S. Patent No. 7,090,816, 2006.
Riaz, A., A. Farsi, G. Zahedi and A. Manan, “Investigation of inert gas injection in steam reforming of methane: Energy,” Sultan Qaboos University, 2011.
Northrop, P. S., A. K. Nagavarapu and J. A. Valencia, “Sour gas treating using Controlled Freeze Zone™ technology: Demonstrated commercial readiness,” Gas Processors Association Convention, San Antonio, Texas, 2014.
Dr. Ilya Polyshuk, personal communication, 2008.
Holmes, A. S. and J. M. Ryan, “Cryogenic distillative separation of acid gases from methane,” U.S. Patent 4,318,723, 1982.
Northrop, P. S. and J. A. Valencia, “The CFZ™ process: A cryogenic method for handling high-CO2 and H2S gas reserves and facilitating geosequestration of CO2 and acid gases,” Energy Procedia, Elsevier, pp. 171–177, February 2009.
Ke, J., et al., “The phase equilibrium and density studies of the ternary mixtures of CO2 + Ar + N2 and CO2 + Ar + H2, systems relevance to CCS technology,” International Journal of Greenhouse Gas Control, Vol. 56, pp. 55–66, 2017.
Eggeman, T. and S. Chafin, “Beware the pitfalls of CO2 freezing prediction,” Chemical Engineering Progress, pp. 39–44, March 2005.
Donnely, H. G. and D. L. Katz, "Phase equilibria in the carbon dioxide–methane system,” Industrial & Engineering Chemistry, Vol. 46, pp. 511–517, 1954.
Tsang, C. Y. and W. B. Streett, "Phase equilibria in the H2 / CO2 system at temperatures from 220 to 290 K and pressures to 172 MPa," Chemical Engineering Science, Vol. 36, Iss. 6, 1981.
Mraw, S. C., S.-C. Hwang and R. Kobayashi, "Vapor-liquid equilibrium of the methane-carbon dioxide system at low temperatures," Journal of Chemical & Engineering Data, April 1978.
Marcelo Tagliabue, “SMR integration with increased CO2 production,” Digital Refining: Processing, Operations & Maintenance, February 2020, online: https://www.digitalrefining.com/article/1002404/smr-integration-with-increased-co2-production
Kunjunny, A. M., M. R. Patel and N. Nath, “Revamping of CO2 removal section in Ammonia plant at IFFCO Kalol,” online: https://www.iipinetwork.org/wp-content/Ietd/content/use-enhanced-co2-removal-solvents.html
Baxter, L., “Carbon dioxide capture from flue gas,” Brigham Young University, Provo, Utah, 2011.
Kimchi, B. B., “Method and system for the separation of mixture containing carbon dioxide hydrocarbon and hydrogen,” AIChE Annual Meeting 2019.
Longman, Z., et al., “Measurement and modeling of CO2 frost points in the CO2–methane systems,” Journal of Chemical & Engineering Data, May 2011.
Fandiño, O., J. P. Martin Trusler and D. Vega-Maza, “Phase behavior of (CO2+H2) and (CO2+N2) at temperatures between (218.15 and 303.15) K at pressures up to 15 MPa,” International Journal of Greenhouse Gas Control, Vol. 36, May 2015.
Løvseth, S. W., et al., “Thermodynamics of the carbon dioxide plus argon (CO2+ Ar) system: An improved reference mixture model and measurements of vapor-liquid, vapor-solid, liquid-solid and vapor-liquid-solid phase equilibrium data at the temperatures 213–299 K and pressures up to 16 MPa,” Fluid Phase Equilibria, 2018.
Baruchi B. Kimchi is the inventor of the cryogenic distillation-based syngas process described in the paper, and the founder of Green Gas Process Technologies. His professional experience includes more than 6 yr as a patent examiner with ILPO, drafting written opinions (WOSA) for PCT applications in the field of physics, and 8 yr of R&D experience in the fields of polymers, biomass to ethanol, formulations and the proposed process as well as business development. He holds an M.Sc degree in chemistry from BIU, a B.Sc degree in chemical engineering from Technion, and a B.Sc degree in biophysics from BIU. Kimchi’s military service was with a classified unit of Air Force intelligence. For more information, contact info@thermodynamico.com. The author can be reached at kimchib@012.net.il.
Scott P. Northrop was a Senior Advisor for sour gas treating and sulfur recovery for ExxonMobil prior to his retirement in 2021. He has more than 35 yr of experience in the industry, ranging from reservoir engineering to projects to production operations to facilities research. He is named on more than 50 U.S. patents and remains active as a consulting engineer.
Venkat Pattabathula is the Director at SVP Chemical Plant Services. Prior to this position, he served as the Global Ammonia Technology Manager for Incitec Pivot Ltd., supporting manufacturing facilities in Australia and North America. His specialties include process design, project development, commissioning, plant operation, process safety management and manufacturing excellence programs. Pattabathula has been a member of the American Institute of Chemical Engineers (AIChE) since 1989 and has been an elected member of the Ammonia Safety Committee of AIChE since 2005. He is a chartered Professional Engineer of Engineers Australia and a registered Professional Engineer of Queensland, Australia. Pattabathula earned an MTech degree in chemical engineering from the Indian Institute of Technology (IIT).
PETER BROADHURST qualified in chemistry from the Universities of Bristol (BSc) and Cambridge (PhD) and then worked in the chemical industry for more than 30 yr with Union Carbide Corp., Imperial Chemical Industries (ICI) and Johnson Matthey (JM). Dr. Broadhurst gained significant experience in catalysis and syngas processes, combining these two areas for 20 yr in his time with ICI/JM. Since 2016, he has worked as an independent consultant for various organizations, remaining focused substantially on industrial catalysis and/or syngas processes.