About Chemical Innovation - Subscription Information
January 2001
Vol. 31, No. 1, pp. 21–26.
Starting the Process

Table of Contents

Tri-reforming: A new process for reducing CO2 emissions

Mick WigginsResearchers at Penn State have developed a new process for the effective conversion and use of carbon dioxide in flue gas from power plants.

The threat of global warming has fueled worldwide efforts to develop technology that reduces carbon dioxide emissions. The conversion and utilization of CO2 present an interesting paradigm to scientists and engineers because CO2 is an important source of carbon for fuels and future chemical feedstocks.

In general, CO2 can be separated, recovered, and purified from concentrated CO2 sources by two or more steps based on absorption, adsorption, or membrane separation. Even the recovery of CO2 from concentrated sources requires substantial energy input (1). The separation and purification steps can produce pure CO2 from power plants’ flue gases, but they also add considerable cost to the conversion or sequestration system (2). Current CO2 separation processes require significant amounts of energy that reduce a power plant’s net electricity output by as much as 20% (3). Although new technology developments could make this recovery easier to handle and more economical to operate in power plants, it is highly desirable to develop novel ways to use the CO2 in flue gases without going through the separation step.

The tri-reforming process we are developing at Pennsylvania State University, is a three-step reaction process. It avoids the separation step and has the promise of being cost-efficient for producing industrially useful synthesis gas.

Table 1. CO2 emissions from different sectors in the United Statesa
Emissions sources 1980 1985 1990 1995 1997
Residential sector 248.4 245.8 253.1 270.3 286.5
Commercial sector 178.3 189.7 206.8 217.9 237.2
Industrial sector 484.6 424.7 454.1 465.0 482.9
Transportation sector 378.1 384.4 432.1 458.5 473.1
End-use total 1289.4 1244.6 1346.1 1411.7 1479.6
Electric utilitiesb 418.4 439.0 476.9 495.3 523.4
a Millions of metric tons of carbon.
b Electric utility emissions are distributed across end-use sectors.
Source: References 4 and 5.

Sources of CO2 emissions

The combustion of fossil fuels is the primary source of CO2 emissions. The three major fossil fuels used worldwide are coal, petroleum, and natural gas. According to government reports, from 1980 to 1997, the end-use total from different sectors (e.g., residential, commercial, industrial, and transportation) have shown a steady increase in CO2 emissions (Table 1) (4). The top 10 major producers of CO2 in terms of total annual emissions (in alphabetical order) are Canada, China, Germany, India, Italy, Japan, Russia, South Korea, the United Kingdom, and the United States (4, 5).

Table 2 gives a history of CO2 emissions in the United States from electricity-generating units (electric utility companies and other producers) (5). Coal is the dominant fossil fuel for the electricity-generating units. It is projected that the share of natural gas–fired units will increase significantly because natural gas is perceived as a premium fuel that is cleaner than either coal or petroleum (6).

Table 2. CO2 emissions from electricity-generating units in the United States a
Emissions sources 1990 1995 1997
Electric utilities
Coal-fired 409.9 434.3 471.3
Petroleum-fired 25.3 13.0 15.0
Gas-fired 39.2 44.5 36.0
Other 1.2 0.8 1.0
Total emissions 475.5 492.7 523.4
Coal-fired 17.8 24.6 25.3
Petroleum-fired 4.3 7.3 7.4
Gas-fired 39.2 57.6 53.2
Other units 37.4 45.9 48.4
Total emissions 98.7 135.5 134.4
Total CO2 emissions
from generators 574.2 628.1 657.7
a Millions of metric tons of carbon.
Source: References 4 and 5.

Using flue gas to convert CO2

Flue gases from fossil fuel-based electricity-generating units represent the major concentrated CO2 sources in the United States (Table 2). If CO2 is separated, as much as 100 MW for a typical 500-MW coal-fired power plant would be necessary for today’s CO2 capture processes based on alkanolamines (3). It would be highly desirable to use the flue gas mixtures for CO2 conversion without the preseparation step. On the basis of our research, we believe that there is a unique advantage of using flue gases directly, rather than preseparated and purified CO2 from flue gases, for the proposed tri-reforming process.

Typical flue gases from natural gas–fired power plants may contain 8–10% CO2, 18–20% H2O, 2–3% O2, and 67–72% N2. Typical flue gases from coal-fired boilers may contain 12–14% CO2, 8–10% H2O, 3–5% O2, and 72–77% N2 (6). The typical furnace outlet temperature of flue gases is usually ~1200 °C, which decreases gradually along the pathway of heat transfer. The temperature of the flue gases around the stack is 150 °C. Current toxic emission control technologies can remove the SOx, NOx, and particulate matter effectively, but CO2, water, and oxygen remain largely unchanged.

In our proposed tri-reforming process, CO2 from the flue gas does not need to be separated. In fact, water and oxygen along with CO2 in the waste flue gas from fossil fuel–based power plants will be used to tri-reform natural gas and produce synthesis gas (syngas).

Proposed tri-reforming process

Tri-reforming refers to simultaneous reforming of oxidative CO2–steam from natural gas. It is a synergetic combination of endothermic CO2 reforming (eq. 1), steam reforming (eq. 2), and exothermic oxidations of methane (eqs. 3 and 4).

CH4 + CO2 → 2CO + 2H2 [ΔH0 = 247.3 kJ/mol] (1)
CH4 + H2O → CO + 3H2 [ΔH0 = 206.3 kJ/mol] (2)
CH41/2O2 → CO + 2H2H0 = –35.6 kJ/mol] (3)
CH4 + 2O2 → CO2 + 2H2O [ΔH0 = –880 kJ/mol] (4)

Figure 1 illustrates a conceptual design of the tri-reforming process of converting CO2 by using flue gases to produce syngas. Coupling CO2 reforming and steam reforming can yield syngas with the desired H2/CO ratios for methanol and Fischer–Tropsch (F–T) synthesis.

Steam reforming is widely used in industry for making H2 (7–9). When CO-rich syngas for oxo synthesis and syngas with a H2/CO ratio of 2 are needed for F–T synthesis and methanol synthesis, steam reforming alone cannot give the desired H2/CO ratio (9, 10). Steam reforming gives a H2/CO ratio of 3, which is too high and thus needs to import CO2 for making syngas with H2/CO ratios of 2 or lower (10).

The CO2 reforming (dry reforming) of methane has attracted considerable attention worldwide (11–13). CO2 reforming is 20% more endothermic than steam reforming as calculated in equations 1 and 2; however, it is necessary to adjust the H2/CO ratio for making MeOH or F–T syngas. Two industrial processes use this reaction: SPARG (10, 14) and Calcor (15).

Carbon formation in the CO2 reforming of methane is a major problem (eqs. 5 and 6), particularly at elevated pressures (16–18).

CH4 → C + 2 H2H0 = 74.9 kJ/mol] (5)
2 CO → C + CO2H0 = –172.2 kJ/mol] (6)
C + CO2 → 2CO [ΔH0 = 172.2 kJ/mol] (7)
C + H2O → CO + H2H0 = 131.4 kJ/mol] (8)
C + O2 → CO2H0 = –393.7 kJ/mol] (9)

When CO2 reforming is coupled to steam reforming (eqs. 5 and 6), this problem can be mitigated effectively (eqs. 7 and 8). This carbon formation in CO2 reforming can be reduced by adding oxygen (eq. 9).

Direct partial oxidation of methane to produce syngas (19, 20) and partial combustion of methane for energy-efficient autothermal syngas production (21) are being explored. These reactions are important, but the catalytic partial oxidation is difficult to control (22). The major operating problems in catalytic partial oxidation are the overheating or hot spot formation caused by the exothermic nature of the oxidation reactions. Consequently, coupling the exothermic reaction with an endothermic reaction could solve this problem (22).

The combination of dry reforming with steam reforming can accomplish two important missions: to produce syngas with desired H2/CO ratios and mitigate the carbon formation that is significant for dry reforming. Integrating steam reforming and partial oxidation with CO2 reforming could dramatically reduce or eliminate carbon formation on reforming catalyst, thus increasing catalyst life and process efficiency. Therefore, the proposed tri-reforming can solve two important problems that are encountered in individual processing. Incorporating oxygen in the reaction generates heat in situ that can be used to increase energy efficiency; oxygen also reduces or eliminates the carbon formation on the reforming catalyst. The tri-reforming can be achieved with natural gas and flue gases using the waste heat in the power plant and the heat generated in situ from oxidation with the oxygen that is present in flue gas.

Figure 2 - flow diagram of tri-generation concept
Figure 2. Proposed CO2-based tri-generation concept for making fuels, chemicals, and electricity using natural gas and flue gas from electric power plants. IGCC, integrated gasification combined cycle.
The tri-reforming process illustrated in Figure 1 is the key step in the recently proposed CO2-based tri-generation of fuels, chemicals, and electricity, as shown in Figure 2 (23, 24). In principle, once the syngas with the desired H2/CO ratio is produced from tri-reforming, it can be used to produce liquid fuels by established routes such as F–T synthesis and to manufacture industrial chemicals such as methanol and acetic acid. Syngas also can be used to generate electricity with either integrated gasification combined cycle (IGCC)-type generators or fuel cells.

The tri-reforming concept is consistent, in general, with the goals of Vision 21 EnergyPlex concept, which the U.S. Department of Energy (DOE) is developing (25, 26). The proposed goals of DOE Vision 21 for power plants include more efficient power generation (>60% with coal, >75% with natural gas), higher overall thermal efficiency (85–90%), near-zero emissions of traditional pollutants, reduction of greenhouse gases (40–50% reduction of CO2 emissions), and co-production of fuels (25).

The feasibility of tri-reforming

We have not found any previous publications or reports of using flue gases for reforming or CO2 conversion that are related to the proposed concept. Our computational thermodynamic analysis shows there are benefits of incorporating steam and oxygen simultaneously in CO2 reforming of natural gas or methane (27, 28). Some laboratory studies with pure gases have shown that adding oxygen to CO2 reforming (14, 22, 29) or to steam reforming of methane, can improve energy efficiency or synergetic effects in processing and mitigation of coking. A feasibility analysis by thermodynamic calculation showed that using CO2–water–oxygen–methane to make syngas is feasible (30). Inui and co-workers have studied energy-efficient hydrogen production by simultaneous catalytic combustion and catalytic CO2–water reforming of methane using a mixture of pure gases including methane, CO2, water, and oxygen (31). Choudhary and co-workers have reported on their laboratory experimental study on simultaneous steam and CO2 reforming of methane in the presence of oxygen at atmospheric pressure with Ni/CaO (32). Choudhary’s work shows that it is possible to convert methane into syngas with high conversion and high selectivity for CO and hydrogen. Ross and co-workers have shown that a Pt/ZrO2 catalyst is active for steam and CO2 reforming combined with the partial oxidation of methane (33).

Figure 3. Coversion yields of tri-refroming of methane
Figure 3. Tri-reforming of methane using gas mixture (CO2/H2O/O2/CH4 = 1:1:0.1:1, mol ratio) at 850 °C under 1 atm over Haldor–Topsoe R67 catalyst.
Therefore, the proposed tri-reforming of natural gas using flue gas from power plants appears to be feasible and safe, although detailed experimental studies, computational analyses, and engineering evaluations are still needed. Recent preliminary experiments in our laboratory showed that syngas with desired H2/CO ratios can be made by tri-reforming methane using simulated flue gas mixtures containing CO2, water, and O2 in a fixed-bed flow reactor. For example, we have studied the proposed tri-reforming in a fixed-bed flow reactor using gas mixtures at atmospheric pressure that simulate the cases with flue gases from coal- and natural gas-fired power plants. As an example, Figure 3 shows the results of tri-reforming methane using simulated flue gas of coal-fired plants at 850 °C for 300 min under atmospheric pressure over a commercially available Haldor–Topsoe R67 catalyst.

As mentioned earlier, the carbon formation in CO2 reforming of methane is a major concern. Our temperature- programmed oxidation results show that after a 300-min time-on-stream (TOS) for CO2 reforming at 850 °C and 1 atm, the used Haldor–Topsoe R67 catalyst contained 21.8 wt% carbon. In contrast, the same catalyst used in tri-reforming showed no sign of carbon formation after 300-min TOS; the used catalyst appears to be greenish powder compared with the black powder from CO2 reforming. Therefore, our preliminary results show that tri-reforming can be performed with stable operation on-line and no appreciable deactivation of catalyst can be observed for tri-reforming under the tested conditions.

Other technical challenges must be overcome before tri-reforming can be successfully upscaled. Flue gases contain inert nitrogen gas in high concentrations, and thus the conversion process design must consider how to dispose of nitrogen. It is possible that oxygen-enriched air or pure oxygen will be used in power plants in the future. If that becomes a reality, then the proposed tri-reforming process will be even more attractive because of much lower inert gas concentrations and higher system efficiency. Another challenge is how to deal with the small amounts of SOx, NOx, and other toxic substances that may be present in flue gases of power plants. The general status of CO2 mitigation technologies and available chemical processes for CO2 conversion has been summarized in Halmann and Steinberg’s recent book on CO2 mitigation (34).

An important feature of the proposed tri-reforming is that it is the first innovative approach to conversion and utilization of CO2 in flue gases from power plants without separating CO2. Many questions remain, and further research is needed to establish and demonstrate this new process concept.


The author thanks his co-workers S. T. Srinivas and W. Pan at Penn State and J. Armor of Air Products and Chemicals Inc. for helpful discussions on reforming. The author is grateful to DOE (UCR Innovative Concepts Program) for supporting a part of this work, H. H. Schobert and A. W. Scaroni of PSU for their encouragement of his initial efforts on CO2 conversion, and B. Miller and S. Pisupati of Penn State for helpful discussions on power plant flue gas.

cartoon of man in an old car with a "my other car has lower emissions" bumper sticker.


  1. DOE/OS-FE. Carbon Sequestration. State of the Science. Office of Science and Office of Fossil Energy, U.S. Dept. of Energy: Washington, DC, 1999.
  2. Weimer, T.; Schaber, K.; Specht, M.; Bandi, A. Am. Chem. Soc., Div. Fuel Chem. Prepr. 1996, 41 (4), 1337–1340.
  3. DOE/FE. Capturing Carbon Dioxide. Office of Fossil Energy, U.S. Dept. of Energy: Washington, DC, 1999.
  4. Energy Information Administration. International Energy Outlook; U.S. Dept. of Energy, EIA: Washington, DC, 1998 and 1999.
  5. EIA/AER. Annual Energy Review 1998, Energy Information Administration, U.S. Dept. of Energy, Washington, DC, 1999.
  6. Miller, B.; Pisupati, S. Pennsylvania State University. Personal communication on power plants and flue gases, Nov 24, 1999.
  7. Rostrup-Nielsen, J. R. Catal. Today 1993, 18, 305–324.
  8. Armor, J. N. Appl. Catal., A 1999, 176, 159–176.
  9. Gunardson, H. H.; Abrardo, J. M. Hydrocarbon Process. 1999, 78 (4), 87–93.
  10. Gunardson, H. Industrial Gases in Petrochemical Processing; Marcel Dekker: New York, 1998.
  11. Ashcroft, A. T.; Cheetham, A. K.; Green, M.L.H.; Vernon, P.D.F. Nature 1991, 352, 225–226.
  12. Rostrup-Nielsen, J. R.; Hansen, J.H.B. J. Catal. 1993, 144, 38–49.
  13. Bradford, M.C.J.; Vannice, M. A. Catal. Rev. 1999, 41, 1–42.
  14. O’Connor, A. M.; Ross, J.R.H. Catal. Today 1998, 4, 203–210.
  15. Kurz, G.; Teuner, S. Erdoel Kohle, Erdgas Petrochem. 1990, 43, 171– 172.
  16. Song, C.; Srinivas, S. T.; Sun, L.; Armor, J. N. Am. Chem. Soc., Div. Pet. Chem. Prepr. 2000, 45, 143–148.
  17. Srinivas, S. T.; Song, C.; Pan, W.; Sun, L. Am. Chem. Soc., Div. Pet. Chem. Prepr. 2000, 45, 348–351.
  18. Tomishige, K.; Himeno, Y.; Matsuo, Y.; Yoshinaga, Y.; Fujimoto, K. Ind. Eng. Chem. Res. 2000, 39, 1891–1897.
  19. Dissanayake, D.; Rosynek, M. P.; Kharas, K.C.C.; Lunsford, J. H. J. Catal. 1991, 132, 117–127.
  20. Hickman, D. A.; Haupfear, E. A.; Schmidt, L. D. Catal. Lett. 1993, 17, 223–237.
  21. Pena, M. A.; Gomez, J. P.; Fierro, J.L.G. Appl. Catal., A 1996, 144, 7–57.
  22. Ruckenstein, E.; Hu, Y. H. Ind. Eng. Chem. Res. 1998, 37, 1744–1747.
  23. Song, C. In Proceedings of the 16th Annual International Pittsburgh Coal Conference; Pittsburgh, PA, Oct 11–15, 1999; Paper 16-5.
  24. Song, C. Am. Chem. Soc., Div. Pet. Chem. Prepr. 2000, 45, 159–163.
  25. Office of Fossil Energy. Project Facts—Advanced Power and Fuel Technologies. The Vision 21 EnergyPlex Concept; DOE/FE-0364; U.S. Dept. of Energy: Washington, DC, 1999.
  26. Federal Energy Technology Center, Office of Fossil Energy. Vision 21 Program Plan–Clean Energy Plants for the 21st Century; U.S. Dept. of Energy: Washington, DC, 1999.
  27. Pan, W.; Srinivas, T. S.; Song, C. In Proceedings of the 16th Annual International Pittsburgh Coal Conference; Pittsburgh, PA, Oct 11–15, 1999; Paper 26-2.
  28. Pan, W.; Song, C. Am. Chem. Soc., Div. Pet. Chem. Prepr. 2000, 45, 168– 171.
  29. Choudhary, V. R.; Rajput, A. M.; Prabhakar, B. Catal. Lett. 1995, 32, 391– 396.
  30. Tjatjopoulos, G. J.; Vasalos, I. A. Ind. Eng. Chem. Res. 1998, 37, 1410– 1421.
  31. Inui, T.; Saigo, K.; Fujii, Y.; Fujioka, K. Catal. Today 1995, 26, 295–302.
  32. Choudhary, V. R.; Rajput, A. M. Ind. Eng. Chem. Res. 1996, 35, 3934– 3939.
  33. Hegarty, M.E.S.; O’Connor, A. M.; Ross, J.R.H. Catal. Today 1998, 42, 225–232.
  34. Halmann, M. M.; Steinberg, M. Greenhouse Gas Carbon Dioxide Mitigation: Science and Technology; Lewis Publishers: Boca Raton, FL, 1999.

Chunshan Song is director of applied catalysis in the Energy Laboratory at the Energy Institute, and associate professor of fuel science in the Department of Energy and Geo-Environmental Engineering at Pennsylvania State University (206 Hosler Building, University Park, PA 16802; 814-863-4466; csong@psu.edu). His research interests include catalytic fuel processing, reforming for syngas and hydrogen production from natural gas and carbon dioxide, shape-selective catalysis, synthesis and application of catalytic materials, conversion of hydrocarbon resources, and fuel chemistry. He has won several awards including the Wilson Award for Outstanding Research at Penn State in 2000 and the NEDO Fellowship Award from Japan in 1998. He received his B.S. degree in chemical engineering from Dalian University of Technology, Dalian, China, and an M.S. degree and Ph.D. in applied chemistry from Osaka University, Osaka, Japan.

Return to Top || Table of Contents