The production of syngas from natural gas is well established as an intermediate for the production of hydrogen, methanol, carbon monoxide (for acetyls), ammonia and liquid hydrocarbons via Fischer-Tropsch synthesis. While there are two main routes to convert natural gas to syngas steam reforming and Partial OXidation (POX) significant new developments and improvements in these technologies are enhancing the syngas process and changing the baseline economics of syngas production. Some of these new technologies, which were reviewed by W. Woodfin of British Petroleum plc, United Kingdom, in Hydrocarbon Engineering, November 1997, pages 76-80, and by B. MacDonald in ECN Chemscope, September 1997, pages 24-25, are summarized below.

Established Processes

The most common process for the production of synthesis gas is methane steam reforming whereby the natural gas is converted by the reaction with steam over a nickel catalyst at high temperatures (900 to 1,000ºC) and moderate pressures (16 to 20 bar). This is typically carried out in long, vertically hung tubes contained within a radiant furnace. Many world renowned engineering houses offer their own variants of this type of design, including Foster Wheeler, M.W. Kellogg, Lurgi, ICI, Kvaerner Process Technology and Haldor Topsoe, to name but a few.

A well established alternative to steam reforming is POX whereby a limited amount of oxygen is allowed to burn with the natural gas feed (creating steam and carbon dioxide at high temperatures) with the subsequent reforming reactions ‘equilibrating’ at approx-imately 1,350ºC. This type of technology is offered under license by Texaco and Shell.

A variant of this type of approach is the so-called AutoThermal Reformer (ATR). This still uses an oxidant (oxygen or air) to carry out the reactions but the hot gases equilibrate over a fixed bed of reforming catalyst. This type of design is offered by Lurgi, Haldor Topsoe and ICI.

The main trends in traditional reformer designs involve increasing the reformer temperature and pressures. However, increasing the reformer temperature or pressure increases the stresses on the reformer tubes, which is the limiting factor in most reformer designs.

The gas-heated reformer is a recent alternative to the conventional radiantly fired steam reformer. It is used in parallel with a conventional ATR (oxygen or air blown depending upon the application). The hot synthesis gas from the ATR is used to provide the heat for the reforming reactions in the gas-heated reformer. Essentially there are two variants to this type of reformer: the bayonet tube design and the open tube design.


The first commercial application of the bayonet tube reformer was by ICI in their ammonia plant in Severnside, United Kingdom, in the late 1980s. The design was later developed for a methanol application for Broken Hill Pty (BHP), Australia. The BHP plant was commissioned in 1994/1995 and has been operating successfully since then. Part of the natural gas feed is mixed with the steam and passes down through the annulus within the reformer tube filled with reforming catalyst. The reformed gas returns up through the central inner tube and is then mixed with the remainder of the natural gas and reacted with the oxidant in the ATR. The hot syngas from the ATR is then passed countercurrently around the outside of the bayonet tube; a guide tube around the bayonet provides a narrow annulus, increasing the velocity of the hot syngas and improving the convective heat transfer.

M.W. Kellogg

M.W. Kellogg has recently developed the Kellogg Reforming Exchange System (KRES) and installed the first commercial unit at the Ocelot Ammonia Company in Kitimat, British Columbia. The unit was successfully brought into service in late 1994. The KRES concept utilizes an open-tube reformer design that eliminates the need for the conventionally fired steam reformer. As in the bayonet tube designs, energy for the steam reforming step is supplied entirely through heat exchange with the hot gas from the secondary reformer. The use of open tubes rather than bayonet tubes is claimed to make this design easier to maintain.

Uhde GmbH has also developed a Combined Autothermal Reformer (CAR) process which takes place in a single vessel. A mixture of steam and part of the desulfurized natural gas feed is reformed first in the primary reforming section by catalytic reaction. The subsequent mixture of synthesis gas is discharged into the POX zone where the remaining part of the natural gas feed is introduced with the oxidant. The adiabatic temperature in the POX zone is 1,300ºC. The sensible heat of the product gas is used for the indirect heating of the primary reforming section of the CAR reactor. A demonstration unit installed by Chemco s.p., Slovakia, was started up in mid-1991. By 1994 it had accumulated more than 17,000 hours operation and is claimed to have reduced oxygen consumption by 35 percent and natural gas consumption by 15 percent (compared to their POX unit). However, to date there have been no other applications of this technology.

BP Chemicals and Kvaerner Process Technology

The BP Chemicals and Kvaerner Process Technology (KPT) have developed a syngas reformer which will be incorporated into KPT’s new methanol process. The BP/KPT route is based on conventional steam reformer technology, but with a new design for the reactor’s internal features. The technology is being developed as a more efficient process for manufacturing hydrogen, carbon monoxide, methanol and acetic acid. The process, which has reached pilot plant stage, will have a major impact on the chemical industry in potential cost savings. The companies hope to test the process in a commercial plant by the end of 1998.


Sasol has developed the Sasol Slurry Phase Distillate process which uses gas reforming in the first step to produce syngas. Its slurry phase reactor has been operating at Sasol’s Sasolburg facilities in South Africa since 1993. At the moment the company is using the process in a 10,000-barrel per day plant for liquid fuels.


As part of their development of a process to convert natural gas to liquid hydrocarbons, Exxon has developed a novel fluid-bed reaction system in which POX and steam reforming reactions are carried out simultaneously in a single large reactor containing a fluidized bed of catalyst particles. Using a catalyst to carry out these reactions simultaneously allows operation at temperatures significantly lower than POX, thereby improving thermal efficiency and reducing the oxygen requirements. Exxon claims that the unit has been successfully demonstrated at their facility in Baton Rouge, Louisiana, over a 3-year period from 1990 to 1992.


The Starchem process is based on a combination of compressors and membranes to produce an enriched air steam which is then reformed at high pressure (80 bar) with natural gas and steam. By carrying out the reforming at high pressure, the need for expensive syngas compression into the methanol synthesis stage is avoided. However, the presence of nitrogen in the syngas means that it has to be combined with a once-through methanol stage rather than a conventional methanol loop. While possible, a "reasonable cost" once-through synthesis system has not yet been commercially proven.


Shell has developed a catalyzed POX process based on a fixed-bed catalyst retained on a honeycomb monolithic carrier. They claim that the open structure allows high gas throughputs without excessive pressure drop.


In the Syntroleum process, natural gas is mixed with air in an ATR to produce syngas. In the next reactor a cobalt-based catalyst converts the syngas into hydrocarbons. As the two reactors operate in a single-pass configuration, the nitrogen contained in the air added at the beginning simply passes through the whole system. In contrast, other systems have a recycle loop in the second reactor, which means that pure oxygen must be used at the beginning.

The ability to use air, and improvements in the overall energy integration, are key to the economics of the process. Most commercial POX processes require at least the addition of oxygen-enriched air requiring extremely costly cryogenic equipment.

Emerging Technologies

Producing syngas by conventional steam reforming or POX is recognized to be a relatively expensive step. For example, approximately 60 to 70 percent of the capital cost of a methanol plant is associated with the syngas plant. While the developments outlined above could reduce the capital and operating costs to some extent, a major breakthrough in the technology is required to make any substantial reductions in the cost of producing syngas.

One of the more radical developments over recent years is a new concept that integrates oxygen separation, steam reforming and POX into a single step.

The Argonne National Laboratory, in cooperation with Amoco, has pioneered the use of membrane technology in the production of syngas. The Argonne and Amoco researchers developed an inexpensive ceramic membrane that selectively extracts pure oxygen from air. By providing oxygen at a low cost, the membrane process could lower the cost of syngas production by about 30 percent.

The ceramic is a strontium-iron-cobalt oxide. The key to its success is its stability in the hostile syngas reactor environment. In the reactor the membrane separates the air stream from the natural gas stream. The oxygen in the air combines with electrons in the membrane to create negatively charged oxygen ions which can migrate through the membrane.

The membrane eliminates the need for cryogenic oxygen and the reaction is autothermal. Once the process is started, there is enough heat generated to continue the reaction.

As well as the ongoing development work with Amoco, Argonne has become part of two much wider alliances in ceramic membrane syngas technology. In July an alliance of Amoco, BP, Praxair, Statoil and Sasol was announced to develop this technology. In a separate announcement in May this year, the United States Department of Energy announced an $84-million, 8-year project to develop membrane technology for syngas production. The project is being led by Air Products with a large number of other companies such as ARCO and Norsk Hydro, and research establishments, including Argonne, being involved.

Although the use of membrane technology is appealing for a number of reasons, the technology still has a ways to go. Apart from the continuing challenge to improve and produce even more stable ceramic membranes there is also the requirement to produce a robust commercial design for this new type of reformer. While there are still technical hurdles to overcome before the first commercial application is seen, the potential rewards will be significant with estimated capital savings of 30 percent over conventional reforming. If successful this would be the first radically new reforming technology in over 3 decades.

Return to Sythentic Fuels Report 5-2 table of contents
Return to J.E. Sinor Consultants homepage