Gas-to-liquids projects gaining momentum as process list grows

June 23, 1997
A section of the 12,500 b/d Shell MDS plant at Bintulu, Sarawak, shows, from right, the Shell gasification process unit, the hydrogen manufacturing unit, and heavy paraffin synthesis reactors. Shell is nearest to full scale gas-to-liquids production. Although it produces liquid fuels, this unique niche plant's small capacity makes its financial success dependent on sale of high value waxes. Photo courtesy of Shell. David Knott Senior Editor Simplified process flow scheme [29611 bytes]
Control room at Shell's Middle Distillate Synthesis plant at Bintulu. Photo courtesy of Shell.
A section of the 12,500 b/d Shell MDS plant at Bintulu, Sarawak, shows, from right, the Shell gasification process unit, the hydrogen manufacturing unit, and heavy paraffin synthesis reactors. Shell is nearest to full scale gas-to-liquids production. Although it produces liquid fuels, this unique niche plant's small capacity makes its financial success dependent on sale of high value waxes. Photo courtesy of Shell.
David Knott
Senior Editor

Processes to convert gas to synthetic petroleum have been under development for 70 years or more, but only recently are they being seen as commercially viable.

Among reasons for a resurgence of interest are an increasing need to utilize associated gas from oil fields for environmental reasons, growing interest in developing gas fields remote from markets without large up-front costs of liquefied natural gas technology, and advances in gas-to-liquids (GTL) technology and economics.

So far, four companies are the main contenders in the race to build the first commercially viable, full-scale GTL plant:

  • Royal Dutch/Shell Group, which has already built a specialized GTL and waxes plant at Bintulu, Sarawak.

  • Sasol Ltd., Johannesburg, which has a small GTL plant operating in South Africa and has a long history of producing synthetic fuel from coal.

  • Exxon Corp., which is negotiating to build a commercial-scale GTL scheme in Qatar.

  • Syntroleum Corp., Tulsa, Okla., which has developed a gas-to-liquids process as a small pilot and is licensing the technology to petroleum majors.

  • A fifth player, Catalytica Inc., Mountain View, Calif., has received U.S. government funding to develop its own method to bypass part of the gas-to-liquids chain.

While Shell is one of two companies to have a successful GTL plant in operation, its Bintulu plant fits a particular market niche and design is not likely to be repeated elsewhere.

Besides producing some synthetic fuel from gas for the local market, Sasol provided technology for Mossgas natural gas-to-synthetic fuel plant in South Africa, which began operations in 1993. This has been a partial technological success, but a commercial flop (OGJ, Mar. 17, 1997, p. 23).

In addition, a Mobil Corp. process that converted gas to methanol and then to gasoline was used in a full-scale-albeit subsidized-plant in New Zealand during the 1980s. It now produces just methanol for Fletcher Challenge Ltd.

The four leading GTL schemes incorporate technology originally developed by Franz Fischer and Hans Tropsch in Germany during the 1920s.

The Fischer-Tropsch process was used to produce up to 600,000 metric tons/year of synthetic hydrocarbons from synthesis gas derived from coal during World War II.

Sasol's development work began in 1955, using fluidized beds to produce synfuel from synthesis gas derived from coal and more recently using a slurry phase reactor and fused iron catalysts for full GTL production.

Shell began work on its own GTL process in Amsterdam in 1973 as an offshoot of work into synfuel production from various sources that it had undertaken since the 1940s.

Shell's SMDS

The Shell Middle Distillate Synthesis (SMDS) process was the backbone of the Bintulu plant at a capital cost of $850 million.

It has capacity to produce 12,500 b/d of middle distillates but is configured to produce middle distillates, plus raffinates and waxes. The plant went on stream in May 1993.

The SMDS process involves four steps (see schematic, this page). As feedstock, the Bintulu plant takes in 100 MMcfd of natural gas from fields off Sarawak, plus 2.5 metric tons/day of oxygen from an on-site plant, the largest single oxygen unit in the world.

The first step is production of synthesis gas, a mixture of carbon monoxide and hydrogen, from natural gas and oxygen using Shell's proprietary gasification process.

The second step, heavy paraffin synthesis, is an updated version of the Fischer-Tropsch process using a proprietary Shell metallocene catalyst. Shell is the only GTL company using sophisticated metallocene catalysts, which yield highly consistent products.

Then comes hydrocracking of the waxy product to produce middle distillates, this time using a proprietary hydrocracking catalyst.

Finally comes fractionation of the middle distillates in a conventional distillation section. This stage can be adjusted to vary yields of different products.

Jack Jacometti, technical/commercial manager at Shell International Gas Ltd., said a GTL plant requires less capital than an LNG scheme and no specialized transport requirements for its output.

A GTL plant to produce 50,000 b/d of middle distillates, which Shell reckons will be the ideal size for its SMDS technology, would cost $1.5 billion to build and require dedicated gas reserves of 3 tcf.

Such a plant would be viable for a crude oil price of $15-20/bbl, said Jacometti, with the exact breakeven point in that range being fixed by availability of low cost gas and comparatively low plant construction costs.

Jacometti said a typical LNG scheme would require $1.5 billion to build a liquefaction plant, $1.5 billion to build six special ships to export the LNG, $1.5 billion to build a receiving plant, gas reserves of 7-9 tcf, and dedicated long-term customers.

About 30% of the Bintulu plant's cost was due to equipment to produce waxes-a niche product that makes the plant highly profitable, officials said-when it would not get by on small scale middle distillates production profits alone.

Because the Bintulu SMDS plant is small and comparatively uneconomic in comparison with conventional middle distillates production, Shell decided to target specialized, high-return wax markets.

Middle distillates from Bintulu are used by Shell and by customers, particularly in California, to add to their diesel fuels to bring them into compliance with stringent emissions regulations.

Because of their high quality, waxes and raffinates from Bintulu are in demand for a range of applications from lubricants and drilling fluids to candles and packaging. But the waxes market is not large enough to sustain another plant like Bintulu SMDS without decimating prices.

Sasol's SSPD

Sasol has developed the Sasol Slurry Phase Distillate (SSPD) process to convert natural gas into high quality diesel fuel, kerosine, and naphtha.

Sasol has been producing 2,500 b/d of liquids at a plant that began operation at its Sasolburg complex in May 1993.

Sasol bases SSPD on the Fischer-Tropsch process and uses iron-based catalysts developed from the original iron catalysts used by the Germans.

The SSPD process involves three phases, all of which Sasol says have been commercially proven. First step is natural gas reforming, in which gas is converted to synthesis gas.

Then follows Fischer-Tropsch pro- cess conversion of synthesis gas into waxy hydrocarbons in the slurry phase reactor developed and licensed by Sasol (see diagram, p. 17). The final stage is upgrading to middle distillates.

Since 1955, Sasol has operated the Arge process, which takes place in a tubular, fixed-bed reactor, to make wax from synthesis gas at Sasolburg.

From this, Sasol developed SSPD, completing a 100 b/d laboratory model in 1990 to enable study of hydrodynamics, heat transfer, and product separation.

The development of the slurry phase reactor has enabled Sasol to increase throughput-per-reactor, so that planned, full-scale GTL plants will consist of one or more units of 10,000 b/d capacity (see chart, this page).

The Sasol slurry phase distillate reactor is fed at the bottom with preheated synthesis gas. This is distributed in a slurry of liquid wax and catalyst particles. As gas bubbles up through the slurry, it diffuses and is converted into more wax. Produced wax is separated from the slurry in a proprietary process.

Sasol says a single SSPD module will convert 100 MMcfd of natural gas into 10,000 b/d of liquid transport fuels. Such a plant would cost $300 million to build. Additional modules would decrease the per-unit cost.

"At a gas price of $0.50/MMBTU," said Sasol, "the feedstock cost is less than $5/bbl of product. Other fixed and variable costs are estimated at a further $5/bbl of product, resulting in a direct cash cost of production of about $10/bbl."

Sasol has also developed the Sasol Advanced Synthesis (SAS) process to convert synthesis gas to gasoline and light olefins. This was said by a Sasol official to be "a brother of SSPD."

An early version of the SAS process has been used since 1955 to produce today 150,000 b/d of liquids from coal at Secunda and 7,500 b/d per reactor at the Mossgas plant.

Sasol has developed new SAS reactors that will have capacities to produce up to 20,000 b/d of syncrude. Seven of these are being installed at Secunda during 1996-99 to replace 16 existing reactors.

Exxon's AGC-21

Exxon has developed its Advanced Gas Conversion Technology 21st Century (AGC-21) process to the point where it is running a 200 b/d GTL pilot plant at Baton Rouge.

The company unveiled its process in announcing negotiations with Qatar General Petroleum Corp. (QGPC) to build a GTL plant to utilize some of the gas in North field, with estimated reserves of 250 tcf of gas.

The Exxon plan is to build a plant to convert 500 MMcfd-1 bcfd of natural gas to 50,000-100,000 b/d of middle distillates, naphtha, and catalytic cracker feedstock.

The AGC-21 process has been developed to the point where scaling up to commercial plant size is within reach, although Exxon said go-ahead for the scheme is conditional on gas prices and a satisfactory fiscal regime.

The Qatar project is expected to cost $1.2-2.4 billion, which equates to about $24,000/b/d of output. This is well below the $30,000/b/d cost reckoned to be the break-even point for GTL at current oil prices.

Salomon Bros., New York, said Exxon's initial feasibility study for a Qatar GTL plant is complete and that a commercial agreement is expected in 1997.

"The reserve size of North field makes it an obvious candidate for an LNG project," said Salomon Bros. "The fact that Exxon's AGC-21 process is chosen over an LNG plant suggests that AGC-21 is superior to, or at least competitive with, current LNG technologies.

"In our discussions with Exxon, the company emphasized that AGC-21 economics are very much case-specific. However, commerciality of the huge North field through AGC-21 tells us that the era of gas-to-liquids is now.

"Our discussion with Exxon also points out that the Qatar gas field was selected among other gas reservoir candidates. Clearly, other large gas fields are also amenable to GTL procedures.

"In Exxon's case, we believe that the AGC-21 technology can provide alternative development plans to the giant Natuna field in Indonesia, the Alaskan North Slope fields, Yemen, and other gas reserves."

Syntroleum

Syntroleum was founded in 1984 to exploit conventional Fischer-Tropsch process chemistry within a "much simpler process design than previous conversion methods."

The company took out its first patents in 1989 and the following year built a 2 b/d pilot plant that is used to test catalyst performance, reactor designs, and operating procedures.

Syntroleum says the most capital-intensive step in other synfuel processes is production of synthesis gas, typically achieved through steam reforming or a combination of steam reforming and partial oxidation with pure oxygen.

The Syntroleum Autothermal Reformer (ATR) process uses autothermal reforming with air to produce a nitrogen-diluted syngas, with a "near ideal" hydrogen/carbon dioxide ratio for Fischer-Tropsch reaction (see diagram, p. 21).

While other processes remove inert gases including nitrogen, the ATR can cope with nitrogen in the process because it is a relatively simple one-pass process with no recycle loop.

Mark Agee, president and chief operating officer of Syntroleum, told a Houston conference recently that the ATR reactor is mechanically simple, easy to start up and shut down, and relatively inexpensive to build.

"ATR does not require large scale to be cost effective," said Agee. "Its lower cost is a large contributor to cost savings realized in the Syntroleum process.

"The reactor configuration is comparable in size but less complicated than comparable systems with recycle because the recycle compressor loop, which must handle and be rated for hydrogen service, has been eliminated."

Syntroleum claims a plant using its process would be able to operate profitably at capacities as low as 5,000 b/d and in some cases 2,500 b/d: "Depending on plant size, it is anticipated that most facilities based on the Syntroleum process can be built for capital costs of $12,000-27,000/bbl of daily output."

The company also envisions its plant's relatively small size will enable barge and ship-mounted applications, so gas fields as small as 10 bcf can be depleted.

A Syntroleum official said the company has issued master licenses for its process to Texaco Inc., ARCO, and Marathon Oil Co. Syntroleum expects to conclude licensing agreements with other petroleum companies this year.

Besides these, Syntroleum has received inquiries from oil companies in Southeast Asia and South America for "something less than the master license."

Catalytica

As the company name suggests, the key to Catalytica's approach to GTL is catalysts. Rather than take the same route to synthesis gas production as other companies, Catalytica aims to develop a process to convert gas to methanol or synfuels involving direct oxidation.

Catalytica received $2 million funding from the U.S. Department of Commerce to develop its direct methane oxidation (DMO) process under a 3-year program.

The company said conversion of natural gas to methanol or syncrude typically relies on an indirect conversion process to yield synthesis gas, which is complex and requires several steps.

Many attempts have been made to develop a DMO process to yield synthesis gas, said Catalytica, but the reaction is difficult to run efficiently, and all attempts to develop a commercially viable process have failed.

The company believes its new group of catalysts, which it describes as highly selective, single-site, homogeneous catalysts, could cut methanol plant costs in half, saving more than $100 million/plant.

More significantly, it said, "Natural gas could be converted to gasoline and other fuels by way of the direct oxidation route at costs competitive with products derived from crude oil."

Catalytica said it has demonstrated feasibility of its DMO process in prototype systems, which have converted methane to a methanol derivative in one-pass yields as high as 70%.

Jacometti said that future SMDS plants will have to target middle distillate sales, because the Bintulu plant has already captured too much of the world's waxes market to justify a second similar unit.

One of the requirements for SMDS commercial viability is low-cost gas, said Jacometti, so the Middle East and the U.S. Gulf Coast are examples of likely sites for future plants.

Shell and Cairn Energy plc, Edinburgh, also announced a deal in which Shell would buy into Cairn's exploration and production assets in Bangladesh. The partners have listed SMDS as one option for future gas discoveries there.

"Shell's focus is on driving down capital costs of SMDS," said Jacometti, "through technical development, design and engineering optimization, and operational experience.

"We could easily reduce the costs of building an SMDS plant by 20-30% from what we have learned so far at Bintulu. In 10 years' time, we will have a core group of GTL experts, and this is key to driving down the costs of technology."

Jacometti said that while 50,000 b/d throughput is the optimum size for current SMDS technology, Shell is challenging its designers to come up with 20,000-30,000 b/d units: "There is a diminishing returns factor in larger plants."

Catalysts key

Saw Choo Boon, executive director of Shell Malaysia, said its metallocene catalyst for GTL is double the price of a typical refinery catalyst. Some of the cost is attributed to the fact that Shell is the sole manufacturer and does not intend to license it to other companies.

"The first 10 years of SMDS development was for the catalyst," said Saw. "Now we have a program in the Amsterdam laboratories to come up with a second-generation catalyst.

"The high cost of catalyst development and the long time required are entry barriers to the GTL market. During the last 2 years of Bintulu operation, we have learned more than we previously did in 15 years. The next Shell SMDS plant will have a modified catalyst compared with today's unit. This will be part of the natural development of the SMDS process."

Like Shell, Sasol has made catalysts the main focus of future research and development in GTL technology: "We are continually engaged in efforts to improve our catalyst technology."

Syntroleum has developed proprietary, cobalt-based catalysts to suit a number of different variations of its process and continues to put a significant amount of its resources into catalyst development.

Agee said Syntroleum began work on a "chain-limiting" catalyst in 1994, with partial funding from three major oil companies.

"The goal of the program," said Agee, "was to produce a catalyst that limits the growth of hydrocarbon chains to eliminate wax production, while at the same time minimizing the production of light hydrocarbons.

"The company has made significant progress toward this goal. Recent multi-week test runs in a fluid bed reactor at the pilot plant yielded a product profile that indicates success.

"This catalyst promises several additional efficiencies to the process configuration, including lower operating pressures, use of higher capacity fluidized bed reactors, and elimination of a hydrocracking step."

Prospects

Exxon's negotiations with QGPC are expected to lead to construction of the world's first full-scale GTL plant, but Shell, Sasol, and holders of Syntroleum process licenses are also competitively negotiating to build a second GTL plant in Qatar.

Qatar's North field has enough undedicated gas reserves to support several GTL plants. Besides Qatar, Alaska's North Slope fields are thought to have huge amounts of gas that are candidates for GTL exploitation.

A GTL plant is said to be under consideration by North Slope operators ARCO Alaska Inc. and BP Exploration (Alaska) Inc.

"The prime illustration is Prudhoe Bay in Alaska," said Agee, "whose 26 tcf of gas reserves are now considered uneconomical and, therefore, have not been booked. They have been the object of numerous unsuccessful marketing efforts via trans-continental pipeline or LNG projects, which could not be financed (see Newsletter).

"A GTL project at Prudhoe Bay would have a ready-made pipeline outlet, which, by itself, should be decisive in favor of this solution as opposed to any attempt at marketing as gas or LNG."

A Sasol official said the most likely next steps in development of SSPD technology are in Qatar and under a recently signed alliance with Norway's Den norske stats oljeselskap AS (Stat- oil).

Statoil was developing its own GTL technology based on the Fischer-Tropsch process, with a view to exploiting its vast gas reserves off Norway. Recognizing that Sasol's technology was more advanced, Statoil joined forces with Sasol.

Sasol and Statoil intend to develop a floating GTL unit, which will be used to convert associated gas into synthetic crude oil and mixed with produced crude oil for export to shore.

The GTL plant would be carried on a production ship or semisubmersible and would basically entail the first two stages of the SSPD process: gas-to-synthesis gas, followed by synthesis gas to waxy syncrude (OGJ, Apr. 28, 1997, p. 42).

The Sasol official added: "We also have a couple of other opportunities that we are exploring, in other parts of the world, but we can't say anything more for now."

Agee said Syntroleum is working with one major to develop a 2,000-2,500 b/d barge-mounted GTL plant for installation at a remote site at an estimated capital cost of $55 million.

The company is also considering a 100,000 b/d natural gas "refinery" in an industrial area.

"Between these two extremes there are myriad possibilities for tapping fields now beyond the commercial reach of gas markets," said Agee.

The Syntroleum official said Texaco, ARCO, or Marathon could announce a GTL project based on its technology "at any time," while the first commercial, full-scale GTL plant is expected to be under construction by mid-1998.

Copyright 1997 Oil & Gas Journal. All Rights Reserved.