Cobalt Catalyst Turns CO2 Into Liquid Fuels Under Carefully Tuned Conditions

In a study published in Catalysis Letters, a team from the University of Johannesburg tested unpromoted and potassium-promoted cobalt-on-titania catalysts in a fixed-bed reactor and identified a narrow operating window that delivered the best results: 220 degrees Celsius, 15 bar of pressure and a potassium loading of just 1 weight percent. Under those conditions the catalyst converted 30.8 percent of the CO2 fed into the reactor and channelled an estimated 9.2 percent of the converted carbon into C5+ hydrocarbons, the long-chain molecules that resemble synthetic diesel and gasoline.

Yet when the hydrogen source is CO2 rather than syngas, cobalt has a well-documented habit of behaving as a methanation catalyst, hydrogenating the carbon all the way to methane instead of stopping at longer chains.

The catalysts themselves were made by incipient wetness impregnation, loading titania with 15 weight percent cobalt from cobalt nitrate, then adding potassium nitrate at loadings from 1 to 5 weight percent in a second impregnation step. Characterisation combined BET surface area measurements, X-ray diffraction, thermogravimetric analysis, temperature-programmed reduction and CO2 temperature-programmed desorption. Each formulation was then tested with a feed of 67.5 percent hydrogen, 22.5 percent CO2 and 10 percent nitrogen as an internal standard, at temperatures from 200 to 300 degrees Celsius and pressures from 1 to 20 bar, with every condition held for more than 24 hours and each reported value averaged over three consecutive steady-state measurements.

Raising the reaction temperature steadily increased CO2 conversion, which jumped from 22.7 percent at 240 degrees to 45.6 percent at 260 degrees, then climbed more gently to 49.7 percent at 280 degrees and 51.7 percent at 300 degrees. But the extra conversion came at a steep price. Methane selectivity rose from roughly 9.3 percent at 200 degrees to about 88 percent at 260 degrees and reached approximately 90 percent at 300 degrees, while CO selectivity collapsed from around 80 percent at 200 degrees to below 10 percent by 260 degrees. The C5+ fraction, estimated by difference from the quantified gaseous products, rose only slightly from 3.3 percent to about 4 percent between 200 and 220 degrees before falling to 1.3 percent at 240 degrees and effectively vanishing above that temperature.

At lower temperatures, CO2 is first converted to carbon monoxide through the reverse water-gas shift reaction, and that CO intermediate is then hydrogenated over the cobalt surface through Fischer-Tropsch chemistry into longer chains. The C2+ selectivity peaked at 17.7 percent at 220 degrees and the C2+ yield peaked at 3.7 percent at 240 degrees before both fell to near zero at 300 degrees, where the methane yield topped out at 46 percent. On that basis the team fixed 220 degrees as the optimum temperature for the rest of the study.

Pressure told a more nuanced story. Selectivity shifted with pressure as well: CO selectivity edged up from 55 to 60 percent between 1 and 5 bar before falling to 27 percent at 20 bar, while methane selectivity climbed from 27 to 52 percent across the same range. The C5+ selectivity peaked at 4.7 percent at 15 bar and dipped to 4.6 percent at 20 bar, and the chain-growth probability rose from 0.41 at 1 bar to 0.53 at 15 bar before dropping back to 0.42. That combination made 15 bar the study’s optimum pressure.

Potassium promotion reshaped both the catalyst and its chemistry. Adding 1 weight percent potassium cut methane selectivity from 59.9 to 55.5 percent and lifted C5+ selectivity from 7.2 to 9.2 percent, while C2-C4 selectivity rose from 29 to about 33 percent and the C2+ yield increased from 10.6 to 12.8 percent. Pushing the loading to 3 and then 5 weight percent reversed much of that benefit: C5+ selectivity fell to 6.4 and then 4.7 percent, and the C2+ yield dropped to 4.6 percent before a minor recovery to 5.2 percent.

Temperature-programmed reduction showed that every increment of potassium pushed cobalt reduction to higher temperatures: the unpromoted catalyst’s first reduction peak peaked at 371 degrees, the 1 percent K catalyst at 418 degrees, and the 3 and 5 percent K catalysts at 580 and 557 degrees respectively, signalling stronger metal-support interactions and harder-to-reduce cobalt. X-ray diffraction detected cobalt titanate phases in the 3 and 5 percent K samples, further evidence of those interactions, and complete reduction at the reactor’s 350-degree limit could not be confirmed for the higher-K catalysts. CO2-TPD, meanwhile, showed that potassium broadened and shifted the CO2 desorption peak from about 124.9 to 152.7 degrees, indicating a change in the nature and strength of the CO2 adsorption sites, although the authors caution that without quantitative uptake data this cannot be read as a conclusive increase in overall basicity.

The team is candid about the limits of its measurements. Because so little liquid product was collected, the C5+ fraction was not measured directly but calculated by difference after subtracting CO and the quantified C1-C4 products, so it may include unquantified oxygenates, analytical losses or deposited carbon, and should be treated as an estimated fraction. Replicate experiments and extended stability tests were not run for all conditions, and incomplete cobalt reduction at higher potassium loadings may have influenced the product distribution independently of the promoter itself. The authors also note that the low gas hourly space velocity of 1.2 NL per gram per hour gives long gas-catalyst contact times and should be kept in mind when comparing these numbers with literature results at higher space velocities.

A modest 1 weight percent of potassium, a moderate 220-degree operating temperature and 15 bar of pressure together gave the best balance between activity and chain growth, achieving comparatively high CO2 conversion at low temperature and pressure with minimal promoter.

Subject of Research: How process conditions and potassium loading affect CO2 hydrogenation to liquid hydrocarbons over cobalt-titania Fischer-Tropsch catalysts

Article Title: Effect of Process Conditions on Catalyst Performance During CO2 Hydrogenation to Liquid Hydrocarbons Over Co/TiO2 Fischer-Tropsch Catalysts

Article References: Tsuke, N. M., Molefe, P. S. S., Gorimbo, J., Yao, Y., Meijboom, R., & Khangale, P. R. (2026). Effect of Process Conditions on Catalyst Performance During CO2 Hydrogenation to Liquid Hydrocarbons Over Co/TiO2 Fischer-Tropsch Catalysts. Catalysis Letters, 156(11), Article 316. https://doi.org/10.1007/s10562-026-05558-3

Image Credits: AI Generated

DOI: 10.1007/s10562-026-05558-3

Keywords: CO2 hydrogenation, Fischer-Tropsch synthesis, cobalt catalyst, titania support, potassium promoter, methane selectivity, C5+ hydrocarbons, reverse water-gas shift, fixed-bed reactor, catalyst characterization, liquid fuels, South Africa