The MTBE situation in the US is in flux.
The momentum to phase out MTBE seems unstoppable in California, which will likely eliminate its use by the end of 2002.
Other US states will also likely reduce or eliminate the use of MTBE, depending on the final oxygenate requirements, but not before the end of 2005.
The accompanying box shows the impacts of two possible scenarios.
While the damage to MTBE appears fatal, it may still be possible to avoid a collapse of the air benefits that MTBE has helped achieve. To date, the US government appears to support the renewable fuels industry and seems likely to encourage or even to demand the use of ethanol.
Despite significant logistical and technical disadvantages to using ethanol, ethanol use for air quality is better than no oxygenate.
Things to watch include a growing chorus of bad news on air compliance in California and elsewhere. Public awareness that bad air is just as critical as bad water is surfacing. It is encouraging to note that the US government is starting a detailed inter-departmental study of the whole air problem, looking at the various choices, costs, and problems.
What is MTBE?
Methyl tertiary butyl ether (MTBE) has become a major component of US gasoline in the past 15 years. It is used primarily in the manufacture of reformulated gasoline (RFG) following the mandates of the Clean Air Act Amendments of 1990 and similar regulations put forth by the California Air Resources Board (CARB) in California.
MTBE is a simple five-carbon molecule, classified as an alkyl ether. It blends easily with gasoline, making a fungible mixture that can be transported through pipelines as well as by barge, rail, or trucks. It has an oxygen content of about 18 wt %, a blending octane rating of 110 (average of research and motor octane numbers), and a modest Reid vapor pressure (rvp) of 8 psi.
Its oxygen content helps reduce carbon monoxide in regions exceeding national air quality standards for that pollutant.
In RFG, the oxygen content is helpful, but secondary. In RFG blending, MTBE's very high octane allows refiners to reduce the manufacture of highly aromatic reformate while maintaining the overall octane rating of the gasoline pool.
Since it is blended in RFG at about 11 vol % (to produce 2.0 wt % oxygen), it also dilutes the blend, reducing olefins, benzene and other aromatics, and sulfur. All of these are undesirable components in the RFG specifications.
US MTBE production
MTBE is made by reacting isobutylene with methanol. This basic etherification reaction occurs over a solid catalyst bed at fairly modest conditions. The equipment needed is simple and relatively inexpensive. Yields are very high, approaching theoretical limits, with essentially no side reactions or losses.
A further advantage is that when mixed butylenes are fed to the unit, only isobutylene will react; the balance passes through the unit unchanged.
There are four different schemes used today in the production of MTBE, differing only in the source of isobutylene employed.
- Catalytic cracking units in refineries. Fluid catalytic cracking units (FCCUs) produce a mixed butylene stream, which can be used to make MTBE. The remaining butylenes usually are fed to an alkylation unit.
In the US, there are 30 operating MTBE plants today that take isobutylene from the FCCU and feed MTBE directly to the gasoline stream. On average, these plants produce about 2,500 to 3,000 b/d of MTBE. Overall, they produce about 84,000 b/d of US MTBE.
- Olefins plants. Besides ethylene and other olefins, steam crackers produce a raffinate stream of crude C4s containing significant amounts of isobutylene.
There are eight such MTBE plants in operation, producing about 34,000 b/d of US MTBE. Some of these units use a mixture of raffinate and FCCU-based feeds.
- Propylene oxide manufacture. One route to propylene oxide produces a large quantity of tertiary butyl alcohol (TBA) as a byproduct. Dehydrating TBA produces isobutylene.
There are two propylene oxide plants operating in the US today. They contribute about 52,000 b/d of MTBE.
- Butane dehydrogenation. The butane dehydrogenation process passes isobutane though catalytic reactors. The reactors strip hydrogen away to produce isobutylene.
These plants are complex and fairly expensive to build and operate. There are six of them in the US, which can contribute about 87,000 b/d of MTBE.
How much MTBE is used?
US MTBE demand is about 308,000 b/d, or 3.7% of US gasoline consumption. RFG uses most of this, with 100,000 b/d going to California.
Forty-six domestic plants account for about 216,000 b/d of US MTBE supply. The US imports more than 14,000 b/d of MTBE in the form of imported RFG from Venezuela, Europe, and the Middle East. About 92,000 b/d of MTBE imports in 1999 came from Canada and other countries; much of this went to California.
The large scale of US MTBE consumption can be understood by recognizing that there are only nine countries in the world that consume more gasoline than the amount of MTBE that the US consumes.
In 1999, total net imports of finished gasoline amounted to 245,000 b/d. Replacing this volume of material will not be easy in the US, because domestic refineries are running near capacity, and construction of new refineries is discouraged.
Fig. 1 shows the uses and sources of US MTBE demand and supply in 1999.
Building up MTBE demand
In the late 1980s, as the US government was making plans for the Clean Air Act amendments, it became evident that demand for MTBE could not be met through existing propylene oxide plants nor by plants built in refineries.
Propylene oxide demand, not gasoline demand, determined the number of propylene oxide plants . The FCCU output limited the number of MTBE plants in refineries.
The overall expected MTBE demand required the construction of several large butane dehydrogenation facilities. The plants are expensive to build and operate, however. The problem was (and is) that a satisfactory return on the needed investment required a secure product price that is higher than the straightforward MTBE blending value would justify.
Suppliers developed a contract formula for MTBE to overcome this problem. They priced MTBE on the basis of the total cost of feeds (methanol plus butane) plus an additional factor to cover other costs, overheads, and profits. The major refining companies reluctantly accepted this formula.
On this pricing basis, investors financed and built six butane dehydrogenation plants. At the outset, the contract formula was not much different from spot values, and there was not much initial concern.
With the passage of the Clean Air Act Amendments late in 1990, the role of MTBE (and to a limited extent ethanol) became fixed by law.
As the 1995 deadline for the start of full RFG introduction approached, parts of the methanol industry recognized that MTBE producers could pass on the full cost of methanol directly. They took this opportunity to raise the price of methanol to unprecedented levels, causing a big spike in MTBE costs to refiners.
In retrospect, this action by the methanol industry was a serious mistake. It angered refining customers, turning them from reluctant users to enemies who could strike as opportunities arose. It also gave the ethanol industry a solid foothold in the RFG business, which they had not previously enjoyed.
While MTBE has admirable blending characteristics, it has two problems.
First, it is "moderately" soluble in water (a few percent). MTBE Material Safety Data Sheets have always called for keeping MTBE from groundwater by spills or other accidents.
Secondly, it has a very distinct odor and taste that can be detected in water at very low concentrations.
While the problem of MTBE incursion into ground water can be (and is) well controlled in industrial facilities with proper maintenance and operation, the MTBE industry clearly failed to anticipate what would happen in other circumstances.
It was well recognized in 1995 that many gasoline-service stations had underground storage tanks that leaked. Many of these tanks have been fixed at considerable expense, but many others remained unfixed or incompletely repaired.
In some of these poorly maintained tanks, gasoline containing MTBE has entered the ground water table. The water phase carries MTBE downstream much faster than the gasoline plume. Moreover, MTBE does not biodegrade as fast as other organic compounds.
With its strong smell and taste, MTBE became a clear flag that a tank was leaking nearby. Thus, the messenger became the culprit.
Matters came to a head in Southern California in 1995 when significant concentrations of MTBE were detected in community water wells. Initially, these were related only to a few places where several nearby service stations all had leaking tanks. But concern soon spread.
On the basis of these leakage reports, the California legislature introduced several bills to ban MTBE as a hazard to public health. Hearings brought out many conflicting points of view:
- Ethanol supporters wanted their oxygenate to have a larger role in gasoline. With political support, they promised that ethanol could easily take the place of MTBE.
- Major oil companies wanted the government to simplify their lives by eliminating all oxygenates requirements or by giving them more flexibility in gasoline components.
- Service station operators wanted to draw attention away from the continuing tank leakage problem.
- Local politicians sensed a hot issue.
The MTBE industry found itself with no reliable friends.
After elaborate hearings and studies by state authorities, the governor of California issued an order in March 1999 to phase out the use of MTBE by the end of 2002.
The controversy spreads
Concern about MTBE in water supplies spread quickly to other parts of the country. The EPA set up a blue-ribbon panel of executives from industries on all sides of the question. This panel met in the spring and summer of 1999 and heard testimony from numbers of presenters. The panel recommended that the use of MTBE be phased down as a threat to the water supply.
The ethanol industry attacked MTBE, brandishing itself as the answer to clean air. Television's "Sixty Minutes" devoted almost an hour to the horrors of MTBE. Finally, in March 2000, Carol Browner, the EPA administrator, held a press conference with the Secretary of Agriculture to press Congress to ban MTBE and to substitute an ethanol mandate for the RFG provisions of the Clean Air Act. The oil industry saw Browner's action as an attempt to earn votes in this election year from the normally Republican Corn Belt.
Despite differing views on the role of MTBE, all the principal actors have insisted that the advances of clean-burning fuels must be maintained if MTBE is eliminated. The CARB adjusted its Phase III gasoline regulations to make it easier to use ethanol, and the EPA has proposed special credit for ethanol to the same end.
Congressional bills vary from the elimination of the 2.0% oxygen standard in RFG, to a ban of MTBE over a period of time, or a mandate that all gasoline contain a certain amount of ethanol.
To date, Congress has taken no final action, and the RFG program has not changed much. MTBE usage is as high as ever, and it is currently commanding a high market price as refiners find that meeting the new federal summer gasoline specifications is not easy. High MTBE use in Europe has reduced shipments to the US, putting a further squeeze on supplies.
The ethanol industry seems increasingly divided on the question of oxygen mandates. Eliminating the oxygen standard is probably essential if MTBE-free gasoline is to be widespread.
If ethanol were to replace MTBE on an equal-oxygen footing, it would be necessary to almost triple the industry's daily output. While this is not an impossible task, it would require major investments, big increases in subsidies, and faith on the part of investors that their new plants would not meet the same fate as did the MTBE merchant market.
Doubling ethanol supply is achievable, however. Workable policies for maximizing ethanol use include eliminating MTBE altogether and reducing oxygen content to about 1.3 wt %, or keeping about one-third of the total MTBE usage, with the balance of the 2.0 wt % oxygen supplied by ethanol.
A recent discovery of both MTBE and ethanol in Lake Tahoe, Calif., however, may limit the use of oxygenates, in general (OGJ, June 26, 2000, p. 57).
Dexter Miller is vice-president of DeWitt & Company Inc. He is responsible for data collection and engineering interpretation in DeWitt's fuels and energy group. Previously, he served for more than 30 years in the project management and project planning groups of M.W. Kellogg and Fish Engineering & Construction. He holds BS and MS degrees in mechanical engineering from Princeton University.