(364c) Conversion of Alaskan Natural Gas to a Liquid Fuel Using Microchanne Reqactor/Heat-Exchangersl

Conversion of natural gas to a mixture of liquid hydrocarbons (usually in the form of a synthetic light crude oil) by means of chemical reaction is known by the acronym GTL. A new GTL process is proposed with the objective of bringing to the market the energy in the proven reserve of 12E12 cubic feet in a ?stranded? field on the North Slope of Alaska near Prudhoe Bay. The liquid would be pumped through the nearby existing 800-mile-long Trans Alaskan Pipeline System (TAPS) to Valdez on Prince William Sound where it would be stored and then transported to a refinery by tanker.

A number of gas-to-liquid (GTL) processes have been proposed and a few have actually been utilized. What is unique in the proposed process is the use of modular microchannel reactor/heat-exchangers to carry out two sets of solid-catalyzed reactions: steam-methane reforming (which is highly endothermic) at 255 oF; 754 oF and ~225psig; and the Fischer-Tropsch conversion (which is highly exothermic) at 437 oF and ~480 psig.

The isolated location, the extreme environmental conditions, and the absence of the usual supporting services also influenced the design of the modified process. Prudhoe Bay, which is within the Arctic Circle, is open to shipping only six months a year. The site of the proposed plant has an extreme climate including periods of continuous night and continuous daylight. Because of the absence of any nearby habitation or industry there is not an existing supply of electricity, non-saline water, and other such materials and services.

The design was based on the conversion of 7.63E9 CFD of natural gas to 118,000 barrels of liquid hydrocarbons for the pipeline. The choice of the modular (rectangular, stacked) microchannel reactors developed by Velocys was based on the prospect of improved rates of heat transfer and reaction, and thereby a reduction in the size of the equipment and in the cost of shipping it to such a remote site..

The required raw materials are 7.74E9 CFD of natural gas, including 4.38E8 for combustion in the reactor and 9.9E6 for the generation of electricity, as well as 7.63E6 lb/hr of water for reaction, startup, and makeup for losses. It was postulated that this water would need to be purchased from a distant site. The outputs other than liquid hydrocarbons include some CO2 to be sequestrated, used in oil recovery or discharged to the atmosphere, and some waste-water with a trace of dissolved hydrocarbons.

For the reforming step, a mixture of steam and methane was passed through a set of passages in a module where it was preheated by product-to-feed exchange, further heated by exchange with hot burned gas in passage, and finally reacted. The products of reaction (Syngas) were cooled in countercurrent flow through intermediate passages of the same module. A mixture of methane and air passed through a third set of intermediate passages where it was preheated by exchange with the burned gas and after burning was cooled in passage through a fourth set of countercurrent passage by exchange with the reacting mixture and then with the entering methane and air. The modules were 1.5m high, 1m wide, and 0.6m long. The passages were 0.25 mm in height and the spacers, including the layer of catalyst were mm in height. Thermo-kinetic calculations indicated that 3298 modules operating in parallel were required. Similar modules were utilized for the Fischer-Tropsch conversion of the Syngas. In this case the reaction took place in 0.95mm passages and water was boiled in alternating 1.5mm passages. Thermal and chemical-kinetic modeling indicated that 3987 modules operating in parallel were required.

The separations can be summarized briefly as follows. CO2 is removed from the Syngas using absorption in MEA. A first flash and decanting separates gaseous, liquid hydrocarbons, and wastewater, and a second flash removes most of the residual gases. CO is separated from the gas from the first flash by the COPureTR process. The CO and the residual gas are recycled to the reformer.

The electrical requirement for compression, pumping, etc. is 782MW so a 800MW generator using natural gas was considered an integral part of the process.

Fixed and variable costs were based on a site factor of 1.4, a 25-year operation, a raw natural gas cost of $0.80/MSCF, and a tariff for TAPS of $4.87 per barrel.The total capital cost was estimated to be $7.17B, the annual fixed cost to be $572MM, the annual variable cost to be $106MM. At a discount rate of 13% an IRR of 15% and an NVP of $708MM. These cost estimates are very sensitive to the price of crude oil, which is difficult to estimate next year let alone in 25 years.

How do these predictions differ in technical and economic promise from those for (1) direct liquefaction and shipping of natural gas, (2) other gas-to-liquid reactive schemes, and (3) pumping the natural gas to the market through a pipeline?

Direct liquefaction does not appear to be feasible at this site because Prudhoe Bay is open to shipping only six months of the year. A large storage capacity would be required. The facilities for direct liquefaction would be subject to many of the same difficulties associated with isolation and climate.

Construction of a 1300-mile pipeline from Prudhoe Bay to connect in Alberta with the Canadian natural gas network has been proposed. It would have a capacity of 4.5E9 cubic feet per day and an estimated capital cost of $26B, which is far greater than that for the GTL plant on the basis of equal quantities of natural gas per day. The initial pressurizing and the intermediate repressurizing would require dedicated electrical generators driven by combustion of some of the natural gas from the pipeline.

.

Reactive conversion of natural gas to a liquid has a long history, including the following ventures. Sasol (Chevron) has been operating plants using a slurry-reactor, one with an output of 124,000 BPD, in South Africa since 1970. ORYX GTK. A joint venture of Qatar Petroleum and Sasol is said to be planning to increase the capacity of their low-temperature Fischer-Tropsch GTL plant in Qatar to 100,000 BPD. Shell Oil is planning to open a 140,000 BPD GTL plant called ?Pearl? in Qatar in 2010. In 2003 the capital cost was estimated to be $4-6B but in 2007 that was increased to $12-18B. In February 2007, ExxonMobil and Qatar Petroleum cancelled plans for a 154,000 BRD GTL plant with an estimated cost of $7B. BP operates a demonstration GTL plant in Nikishi, near Anchorage, Alaska, that produces 300BPD. It utilizes compact rather than microchannel reactors.

In conclusion, the new GTL process appears to be technically advantageous over prior ones for the North Slope site. In particular, the reductions in volume and weight of 84% and 78%, respectively, for the reactors, the reduction of the exit temperature of the reformer from 1626 to 756 oF, and the increased catalytic productivity (lbprod/lbcat/hr) of the Fischer-Tropsch conversion from 0.07 to 1.53. Whether or not these gains are sufficient to favor the process over the construction of a natural gas pipeline is uncertain because of the uncertainty of the price of oil.