CALIFORNIA REFINERS FACE HURDLE IN FEDERAL, STATE RFG RULES

No U.S. refiners face a bigger challenge in producing reformulated gasoline (RFG) than do California's.

That's because they must cope with a change to two sets of specifications for gasoline in the 1990s: federal standards and the tougher California standards.

Dual deadlines loom in the state.

One is to meet Phase I federal RFG specifications that come into play Jan. 1, 1995. The other to is meet the more stringent Phase II California Air Resources Board (CARB) specifications that take effect only 14 months later, beginning Mar. 1, 1996.

How the state's refiners fare in this transition bears watching because California will have to meet certain challenges at least 2 years sooner than refiners elsewhere.

Among the challenges will be coping with the back to back transition itself. Huge revamps are under way by companies planning to meet both sets of rules. There also are big projects slated by refiners planning to comply only with CARB specs.

Some projects carry price tags of as much as $1 billion but include work to improve gasoline yields, going beyond RFG spec compliance.

Meeting the construction deadlines is no easy matter, even for those getting an early start. Among other things, it requires careful planning, good timing, and no small amount of good fortune to make sure the many permits required are obtained in time for construction to stay on schedule.

FEDERAL, CARB CHALLENGES

On a smaller scale, refiners elsewhere covered by the federal regulations also will face a dual transition down the road.

In 1998, the federal RFG program shifts from the simple model to the complex model and steps up performance standards by which compliance is determined. In 2000, the federal program shifts into Phase II, and there is another hike in performance standards. Even so, California is trying to do more in less time.

By the time the federal program peaks in 2000, California will have been subject to tougher requirements for 4 years. CARB's Phase II regulations exceed those in Phase II of the federal RFG program, notes Paul Michiele, chief of fuel studies and standards at the Environmental Protection Agency's Ann Arbor, Mich., division.

"California's transition is more drastic and occurs sooner," observed an industry official. "It's an extreme fuel reformulation compared with the federal level."

For the next 18 months, however, California refiners will be operating under RFG provisions of the 1990 Clean Air Act amendments.

Federal RFG is mandatory in the nation's nine worst ozone nonattainment areas.

In California, federal RFG, or EPA gasoline, is required in the Los Angeles basin, including Riverside, San Bernadino, and Ventura counties, as well as in San Diego. Elsewhere, EPA requirements must be met in eight other major U.S. urban areas. Other, less severe ozone nonattainment areas can opt into the program, as 13 states and the District of Columbia have done.

All told, the mandatory and current opt-in RFG areas account for about 30% of the U.S. gasoline market. If all ozone nonattainment areas decided to opt into the program, more than 50% of the gasoline sold in the U.S. will have to be RFG, says the American Petroleum Institute.

Starting Mar. 1, 1996, California parts ways from counterparts in other ozone nonattainment areas. The basic parameters that constitute RFG are the same: boosting oxygen, lowering vapor pressure, reducing benzene, olefins, aromatics, and sulfur, and adjusting distillation temperature. The approaches used in getting there and the magnitude of the reformulations are very different.

DIFFERENT APPROACHES

CARB takes a more prescriptive approach, setting precise specs for each fuel parameter. The federal approach, on the other hand, sets few detailed specs. EPA rules establish performance standards for reducing three criteria pollutants-volatile organic compounds (VOCs), nitrogen oxides (NOx), and air toxics.

Particularly after 1998, refiners have considerable latitude to decide how many and which of the parameters they adjust to meet the standards.

Specifically, the Phase I RFG rules adopted by EPA permit refiners to use a simple model to certify their federal spec fuels the first 3 years the program is in place.

The simple model considers four parameters to judge compliance: gasoline volatility level and oxygen content for control of VOCs and benzene and aromatics content for air toxics control. Precise specs, however, are set for only three of these parameters: oxygen content at 2 wt %, benzene at 1 vol %, and Reid vapor pressure (Rvp) at 7.2 psi maximum.

Refiners can decide how to adjust aromatics. Refiners are allowed to use average 1990 levels for olefins, sulfur, and T-90, the temperature at which 90% of the gasoline evaporates.

Starting Jan. 1, 1998, refiners covered by the federal rules must use the complex model. Under the complex model, four more parameters are used in judging compliance: oxygenate type, sulfur content, olefin content, and fuel distillation. Refiners are given a wide latitude as to how to meet higher performance standards for almost all parameters. Year-round specs apply to benzene and oxygen levels.

CARB RFG, on the other hand, covers ambitious specs that are generally well below current industry averages for each of the parameters. Refiners can choose between a flat spec limit or a 90 day average. Caps are set on each blending spec.

Refiners choosing the averaging technique must never exceed the cap and must achieve a mean that is less than the flat limit. For example, a refiner choosing to average sulfur content during 90 days must achieve a mean 30 ppm vs. the 40 ppm flat limit.

For CARB specs:

Benzene content is set at a flat limit of 1 vol %. Refiners choosing to average must meet an 0.8 vol % mean and never exceed a 1.2 vol % cap.
Aromatics are set at a maximum 25 vol % with a level of 22 vol % for averaging and a cap of 30 vol %.
Rvp is set at a flat limit of 7 psi with no averaging allowed.
Sulfur is set at a flat limit of 40 ppm/30 ppm averaging/80 ppm cap.
Olefins are set at a flat limit of 6 vol %/4 vol % averaging/10 vol % cap.
Oxygen content can vary at 1.8-2.2 wt %.
The T-90 is set at a flat rate of 300 F./290 F. averaging/330 F. cap. T-50 distillation point is set at a flat rate of 210 F./200 F. averaging 220 F. cap.

Small refiners have 2 more years to meet four of the eight specs.

In a tilt toward the federal style, CARB last June adopted a predictive model that gives the state's refiners more leeway to custom blend fuels within the framework of the rules. The model allows refiners to adjust their blends within seven of CARB's eight specs as long as overall emissions standards are met.

The predictive model can forecast changes in emissions of hydrocarbons, NOx, or toxics if adjustments are made to one or more of the seven specs where fine tuning is permitted. Changes in Rvp are not permitted.

In this way, refiners can still stay within CARB's overall guidelines for emissions without necessarily meeting all of the specs. It's an improvement, industry officials say, but far from the flexibility under federal law.

ROLLOUT BEGINS

Federal and state regulators alike hope to avoid the market disruptions that accompanied new low sulfur diesel fuel specs that took effect Oct. 1, 1993 (OGJ, Oct. 18, 1993, p. 32). While it hardly eliminates all concerns, the first delivery of RFG is sequenced under federal and state rules to provide more definite guidelines for moving the fuel to market.

Starting this month, participating refineries in California are expected to begin the rollout of RFG produced under EPA rules. The federal spec gasoline must be at terminals by Dec. 1 and at the pumps by Jan. 1, 1995. To meet this timetable, refinery deliveries to pipelines in California are scheduled to start as early as next week.

Because EPA specs required in California apply only to southern California, it was never expected that a majority of refiners throughout the state would produce federal RFG, nor would it make sense for all to do so.

Nevertheless, southern California typically imports gasoline from northern California refineries to meet about 10% of its demand, plus volumes from elsewhere, including Washington state and the Gulf Coast. Some northern California refineries say they will continue to be part of the supply mix.

WHO'S DOING WHAT

To date, seven of eight Los Angeles area refiners-ARCO, Chevron Corp., Mobil Oil Corp., Powerine Oil Co., Texaco Inc., Ultramar Inc., and Unocal Corp.-have confirmed they plan to produce EPA gasoline.

Among refiners in southern California, only independent Paramount Petroleum Corp. has indicated it won't produce federal RFG. Paramount, however, is the smallest of the group, with an estimated 7,000-8,000 b/d of gasoline output from its 39,500 b/d refinery at Paramount.

In the San Francisco Bay area, Shell Oil Co. and Tosco Corp. emphasized they will contribute to southern California's EPA gasoline needs. Both declined to give specific estimates of their EPA gasoline volumes for proprietary reasons.

"For those markets that require it, we'll have plenty of it," a Tosco official said.

Although Tosco does not yet have an approved environmental impact report (EIR) and has not started construction on its clean fuels project, the company says refinery modifications previously made will allow it to comply with federal specs. In fact, several industry officials said a number of San Francisco Bay area refineries can produce at least some EPA gasoline at present.

All told, as many as seven of the state's refineries may be sitting out the federal RFG rules. In addition to Paramount, Texaco and Kern Oil & Refining Co. in Bakersfield currently do not plan to produce federal RFG. Unocal and Chevron plan to produce EPA spec gasoline only from their Los Angeles refineries at this time. Exxon Co. and Pacific Refining Co.'s plans are unknown.

How many refineries participate, however, is not as significant as supply capabilities of those in the game.

SUPPLY/DEMAND QUESTIONS

If all participating refiners produced all EPA gasoline at estimated capacity, the southern California market likely would have a surplus.

According to one well regarded industry analyst, 52% of total crude oil capacity typically provides a conservative benchmark for estimating average refinery gasoline yields.

Using this benchmark, the seven participating southern California refiners alone conceivably could produce about 480,000 b/d of EPA gasoline. The two San Francisco refineries known to be willing to produce EPA gasoline could, in theory, add about another 160,000 b/d if they were prepared to produce all of their gasoline to EPA specs, which is not the case.

This would yield a total of 640,000 b/d, or 100,000 b/d more than the highest estimate of 1995 EPA gasoline demand in southern California.

However, weighing gross potential EPA gasoline out-put against demand in southern California by no means provides a complete picture of how the market is really supplied.

Not all gasoline produced in the state stays in the state. So even in southern California, refiners likely will produce a mix of EPA and non-EPA gasoline, sources said. The EPA spec gasoline will be sold in local markets, and some conventional gasoline may be produced for export to supply towns in adjacent states, mainly Las Vegas and Phoenix.

This is one reason anticipated EPA gasoline output numbers are so closely guarded by most refiners.

In northern California, the situation will be reversed, with some EPA spec fuel going to southern California but with the majority of output, conventional gasoline, staying in local markets or being shipped to out of state markets.

Northern California refiners are limited by antidumping provisions in how much RFG gasoline they can ship out and how much conventional gasoline they can supply locally.

Complicating the picture even further, California refiners can either bring EPA gasoline in from the Gulf Coast or substitute some of their current exports of conventional gasoline via exchanges with refiners elsewhere in order to keep more local output at home. While both of these options are more costly, they are possible.

A DELICATE BALANCE

By most accounts, potential supplies of federal RFG for California are in what might best be termed a delicate balance.

Companies with EPA gasoline projects under way report they are on track. The only possible exception is Unocal, dogged by last minute permitting problems (see Newsletter, this issue). However, even in the worst case, Unocal said it could supply its 13-14% share of the market at least at first by producing below spec gasoline and making up the difference later with above spec fuel under EPA averaging rules.

A major concern is that unanticipated start-up problems could turn an expected tight market into a temporarily short market. Last year, problems at two refineries-not problems with overall supply capabilities-caused dislocations in the diesel market.

"If no one stubs their toes, this should come off okay," summed up Tom Glaviano, supervisor of oil analysis for California Energy Commission.

CEC estimates 490,000 b/d of gasoline was sold in southern California during 1993. This is based on the region accounting for 56% of total state consumption of 877,000 b/d.

In the same year, Los Angeles refiners produced about 520,000 b/d of gasoline. While this would imply a 30,000 b/d surplus, the same refiners typically export 80,000-100,000 b/d to markets they serve in Phoenix and Las Vegas, CEC pointed out.

This translates to net gasoline supplies produced in Los Angeles and serving southern California of 420,000-440,000 b/d, or 50,000-70,000 b/d short of demand. To make up the difference, CEC said, southern California imports 60,000-70,000 b/d of gasoline, including 30,000-40,000 b/d from northern California and 20,000-30,000 b/d from elsewhere, notably Washington state and the Gulf Coast.

Richard Stellman, president of Pace Consultants, Houston, thinks the California RFG market will be "tight." He's concerned that refiners outside southern California may not be ready in time because they are not legally required to produce the new fuel.

Ultramar Planning Director Lynn Westfall projects 1995 demand in southern California at about 530,000 b/d against estimated area refinery supply of 490,000 b/d.

CEC's Glaviano said he has received reports that there are ample supplies on the Gulf Coast to meet an unexpected shortfall. The problem would be pipeline logistics, he said, and this would be temporary.

With a whole new type of fuel coming to market, there are other difficult to predict variables. Among them is how much added gasoline yield, if any, will occur from the increase in oxygenates. As a rule of thumb, oxygenates add about 10% to gasoline pool volume, but several factors temper this from translating into the equivalent in total output.

One analyst said many refiners likely will take 10% "from the bottom of the barrel" and keep their production about the same. Because oxygenates also have less energy content, there is a BTU penalty of about 2%, he noted. This means that refiners would have to produce 102% of current gasoline yields just to keep pace.

Still another factor is that benzene extraction reduces overall fuel supply.

CONSTRUCTION CRUNCH

In contrast with the situation with EPA gasoline, a major question mark with CARB fuel is the ability of refiners to complete construction in time to meet those more stringent specs.

Some refiners jumped the gun and were able to break ground a year or more ago. Others are just getting to the starting gate, with EIRs still being drafted or under review. For these refiners, the whole permitting and building process lies ahead, with only 18 months to go.

Examples of how little time 18 months may be in the scheme of things lie in how some refiners are progressing. Some of these examples also underscore the relative cost and complexity of CARB spec vs. EPA spec projects:

ARCO, at a point in late September when it was nearly ready to start producing federal spec RFG, said the overall project for CARB gasoline was about 25% complete. The company expects to finish engineering by yearend and have its first CARB unit running in July 1995. The estimated price tag for modifications to produce EPA gasoline is $135 million. The expected cost of improvements to manufacture CARB gasoline is $335 million. And this is better than it might have been. After CARB issued its predictive model in June 1994, ARCO was able to defer plans for a new C5 alkylation unit and continue to use its C4 alkylation unit built in 1989.
Shell, at about the same point in late September, said it was 30-35% of the way along on its CARB spec construction. Shell's $1 billion project appears to be the largest of all at a single site. Chevron's capital spending comes close, at $950 million, but that's for two refineries. Both companies, in addition to meeting CARB specs, are making improvements to step up light products yield.
Exxon recently received approval of its EIR for a $200 million Benecia refinery project. Asked when construction is likely to be finished, a company official said, "It'll be close," referring to the March 1996 deadline for start-up.

This doesn't necessarily mean that a refiner who hasn't started building faces trouble however.

CARB Executive Officer James Boyd, in a memo updating status of projects, said, "Each refiner's construction plans vary and require a different amount of construction time to comply with the regulation. For instance, one refinery may need more than 2 years to complete construction while another may need as little as a year."

With 18 months to go, the next 3-6 months are critical for remaining refiners to finish the EIR and permitting process and start construction.

The projects vary so much in size and scope there is no way to define the typical project. Much depends on how modern the refinery was at the outset.

Ultramar, for example, is spending a relatively modest $113 million for refinery upgrades to produce EPA and CARB gasoline. The company's extensive improvements during the 1980s paid off when it came time to make RFG. Major components of the project include a new naphtha splitter and a larger naphtha hydrotreater. The company also is converting its old naphtha hydrotreater to an olefin treater.

A summary list of Shell's modifications by contrast, consumes two pages.

Whether it's for a single unit or a refinery-wide revamp, engineering and construction contractors are challenged by the sophistication of engineering required to meet CARB specs. A process engineer with one of the big E&C companies said, "Now, we're talking about towers with 40-50 distillation trays, compared with 10 or 20 in the past, and they are five times better than they used to be."

NO MAGIC FORMULA

While projects vary tremendously, there are some fundamental process changes common in producing CARB gasoline. These include naphtha splitting to take benzene precursors out of the reformer, more desulfurization, saturization of olefins, distillation modifications, and increased hydrogen plant capacity to remove sulfur from FCC feed.

How each refiner balances the equation is a highly individual matter.

"There is no magic formula," said Jake Easton III, president of petroleum and petrochemical construction for Fluor Daniel. "Each refiner has its own strategy."

The company is the prime engineering contractor for four California refiners: Ultramar, Chevron at El Segundo, Texaco at Wilmington, and Unocal at Wilmington and Carson.

With so many new considerations in blending, making or otherwise obtaining clean blendstocks will be the key to success. Far and away preferred are alkylates. One source called alkylate the "liquid gold" of CARB gasoline.

"It will be expensive to import it, but it's also expensive building an alky unit," another observed. Still another, asked to characterize the key components of the new refining processing units, summed it up this way: "Disfavoring reformates, heavily favoring alkylates."

One of the key sources of alkylate supply would be Gulf Coast refineries, sources said. Isomerate is another preferred blendstock.

The great need for clean blendstocks, some believe, could trigger an active trading market, even globally.

One analyst suggested that Asian refineries conceivably could concentrate on producing alkylates or isomerates for the California market. In turn, California refiners could ship their reformates and even light cut FCC gasolines to Asia. The FCC gasolines are high in olefins and vapor pressure but also high in octane. "You can use them in Asia," he said.

California demand for clean blendstocks also has implications elsewhere because the state's demand could press supplies and hike prices nationwide.

CARB MARKET OUTLOOK

Eager to avoid the dislocations that occurred with introduction of low sulfur diesel, CARB has taken steps to ease the transition to RFG.

One of these actions has been to adopt a gradual rollout period for introduction of CARB Phase II gasoline. Refiners must be producing it by Mar. 1, 1996. The fuel needs to be delivered to pipelines by Apr. 15, 1996, and to the pumps June 1, 1996.

In addition, at the behest of California Gov. Pete Wilson, CARB this summer established a joint government-industry advisory committee to monitor the transition to CARB gasoline.

Consisting of almost 70 companies, agencies, and organizations, the group is charged with examining progress toward implementing the rules, anticipating performance problems, and developing contingencies to avoid a repeat of the state's experience with diesel fuel last fall.

Three subcommittees have different responsibilities. One tracks price, demand, and supply issues. A second focuses on testing the quality of RFG to determine its effect on engine performance. A third panel is responsible for increasing public awareness of the costs and benefits of the new fuel. The task force, slated to have quarterly meetings, will hold a major session Oct. 15.

CEC, one of the task force participants, will present new estimates of the supply/demand outlook to CARB in advance of the meeting. The commission's supply estimates, based on figures provided by prospective CARB refiners, predict that statewide gasoline supply capabilities will be 860,000-1.1 million b/d in 1996. Demand is projected at 840,000-920,000 b/d. Not factored into the analysis are supplies potentially destined for shipment out of state, estimated at 120,000-130,000 b/d.

CARB early this past summer released figures, also prepared from company reports, for gasoline production volumes of 880,000-1 million b/d in 1996. It then projected demand at 860,000-920,000 b/d. With ranges so broad, it is possible to show almost any scenario.

In CEC projections, if demand were on the high side of 920,000 b/d and supply met the low side projections of 860,00 b/d, that would indicate a shortfall of 60,000 b/d even before exports are accounted for. If the situation is reversed, there is a 260,000 b/d surplus-improbable, at best.

The figures indicate that the market at least has the potential to be comfortably situated if all goes well.

That's the multibillion dollar question. Where CARB gasoline is concerned, it is too early to predict.

The less complex transition to EPA gasoline should provide an early indicator of how well and quickly California's market can adapt to changing requirements.