SPECIAL REPORT: Transition to ULSD, ethanol goes smoothly for US refiners

Dec. 4, 2006
During the last half of 2006, the initial transition to ultralow-sulfur diesel (ULSD) and ethanol blending has gone smoothly for US refiners.

During the last half of 2006, the initial transition to ultralow-sulfur diesel (ULSD) and ethanol blending has gone smoothly for US refiners. The Oct. 15 deadline for 15-ppm (wt) sulfur ULSD at the retail outlet was ahead of schedule, which is further evidence that the transition is proceeding as planned.1

The data indicate that more must be done to reach the 80% ULSD goal; the trend indicates that US refiners will attain this goal. The US Energy Information Administration tracks the volume of ULSD that is downgraded to 500-ppm (wt) sulfur diesel. Downgrades are showing a positive trend, falling to less than 1% after from a high of 20%. This indicates the industry is adapting to ULSD production requirements.

Including ethanol in the gasoline pool has used the available supply without significant interruptions.

Two initiatives have led the requirement for ethanol blending:

  • Removal of the oxygen mandate from reformulated gasoline (RFG).
  • The renewable fuels requirement. Because the refining industry has had experience with ethanol blending, the use of additional ethanol was an incremental change.

Removing oxygenate from RFG allowed for replacement strategies of either using an alkylate-type blendstock or ethanol. The short-term solution appears to be the use of ethanol in the nonattainment areas.

This past summer had record high crude and product prices with constant or slightly increased demand for both gasoline and highway diesel. The increase in price and demand is contrary to historic market performance; it seems to indicate that neither ULSD nor ethanol blending affected supply.

Record price increases for gasoline and diesel are primarily crude related, masking the secondary effects due to supply or quality. The conclusion reached after studying these factors using the preliminary data is that gasoline and ULSD supplies were adequate.

Crude, product demand

To frame current and projected market situations, we studied the effect of worldwide crude and product demand on price. Historically, as crude prices increase, demand decreases and sets a crude price ceiling at about $26/bbl. Recent crude prices and demand have increased simultaneously.

Click here to enlarge image

Our projections are that crude demand will continue to rise and prices will reach a new equilibrium. Refinery output capacity will also rise to meet the new demand. Fig. 1 shows these effects.

Fig. 2 shows that expected increases in world demand have increased US demand.

Click here to enlarge image

KBC’s Petroleum Economics Ltd. division’s forecast for oil demand during the next decade suggests that US capacity would need to grow the equivalent of one large refinery every year to keep pace with demand.

The US has a shortage of 2 million b/d of refined products; current announced refining capacity additions will provide capacity, but the deficit will remain. The current political climate in the US is unlikely to support fiscal incentives for refinery construction. US lawmakers have little desire to give downstream handouts to refiners, while these companies are paying massive dividends to their shareholders based on upstream profits.

Despite the standoff, refiners are presumably quite satisfied with the state of US refining and feel little pressure to invest in US downstream without economic incentives. ExxonMobil Corp.’s CEO-elect Rex Tillerson has echoed his predecessor’s view that the company doesn’t need to build refining capacity. A structural shortfall in domestically refined products should provide support for future margins, even with the capacity expansions that will no doubt emerge.

The transition to ULSD diverted expansion capital into clean-fuels projects. MTBE was removed from US gasoline due to the renewable fuels mandate and MTBE’s environmental impact on groundwater. Shutting down MTBE production facilities has reduced gasoline capacity from the US markets. These two events tend to increase the US product deficit further.

Short term

The October 2006 EIA Energy Outlook indicates that gasoline and distillate inventories will be above critical levels (Fig. 3). Supply will be adequate for both gasoline and diesel in the short term. The winter draw is due to the annual heating-fuel requirements, mainly on the East Coast.

Click here to enlarge image


ULSD transition

The mandate for ULSD was implemented in the US in June 2006. The deadline for ULSD to reach the retail stations was Oct. 15, which was achieved.1

In the US, diesel-engine technology requires low sulfur levels effectively to reduce NOx and particulate matter (PM) emissions. Diesel-engine manufacturers are introducing this technology in model year 2007, starting Sept. 1, 2006. The number of vehicles sold with the engine-emission-reduction technology will be a small fraction of the total fleet in calendar year 2006 but will increase steadily.

ULSD will reduce sulfur emissions, which has been the driving force worldwide. To keep engine systems clean, the US government has mandated ULSD be sold at 80% of US diesel pumps, with 100% of pumps to be ULSD by 2010.

ULSD key success factors

To provide a mechanism to evaluate the success or failure of ULSD implementation, we developed key success factors that, if positive, would indicate a smooth implementation. Negative results indicate that serious problems exist.

ULSD key success factors are:

  • Prior planning with cooperative interaction between government and industry to set standards with a time frame to implement changes.
  • Protection of the ULSD supply by providing regulatory compliance options such as credits and a temporary compliance option (TCO) to give refiners options for ULSD production.
  • Commitment by the refining industry to invest capital for improvements to meet new demand and the regulatory community commitment to provide timely permitting to meet the construction schedules.
  • Communication between the refining industry and regulatory community.

Planning-ULSD production

Total distillate demand in the US is about 4.2 million b/d. This includes highway diesel, home heating oil, off-road diesel, and other fuels.

Of the 111 refineries operating in the US, 88 ULSD units were started up in 2006 and about 11 refineries indicated “no changes” were needed to produce ULSD for a total of 99 refineries producing ULSD. The remaining refineries produce highway diesel using sulfur credits, TCO, hardship options, or are no longer producing highway diesel.

Table 1 shows the latest update for refineries in Petroleum Administration for Defense Districts (PADDs) 2 and 3 for ULSD production.2

Click here to enlarge image

Only one third of the refineries planning to produce 100% ULSD achieved this goal by June and more were producing at the 500-ppm level than anticipated. The regulatory goal, however, is that 80% of a refineries’ supply is ULSD.

Plans for these modifications were initiated and completed in the last 5 years. Implementation required transitioning refinery, pipeline, terminals, and retail systems to ULSD. This appears to have occurred smoothly without supply interruptions.

Click here to enlarge image

Table 2 shows the implementation plan for ULSD.

Supply flexibility

Until 2010, regulatory options are available to refiners to produce greater than 15-ppm (wt) ULSD. In general, sulfur credits can be used if they were produced and available in the same PADD as the refinery using them.

During the initial phase, June 2006-May 2010, refiners can produce 20% of the ULSD as temporary compliance option (TCO) diesel (500 ppm). This production is not offset by credit use and a refiner can produce more 500-ppm diesel by purchasing credits. Two types of diesel are then produced, ULSD and EPA diesel (500-ppm sulfur). These two fuels must be segregated from the refinery to the retail pump.

The TCO is designed for the transition period from June 2006 to May 2010. Compliance is annual, which allows refiners operational flexibility. Individual refineries can use the TCO in different ways.

In July, for example, ULSD supply was constrained in the US heartland due to refinery production problems and high irrigation pump demands during the summer drought conditions. The major problem was product delivery to rural areas-truck drivers were exceeding their allowable driving hours in a given day. The governor of Nebraska issued an emergency proclamation to extend truck driver hours to allow product deliveries.3

Other incidents of reported shortages were also due to extenuating circumstances and constituted a small volume of the total US highway diesel production.

Fig. 4 shows the production of ULSD by PADD from EIA data.

Click here to enlarge image

With ULSD production starting in June, the rapid increase in production was expected. The regulation, however, is for 80% of the total highway diesel to be ULSD.

Fig. 5 shows the PADD distribution for the ULSD percentage of the total highway diesel.

Click here to enlarge image

Total diesel production is currently approaching the 80% ULSD requirement but is still slightly below the goal. Surprisingly PADD 2 is only at 66% ULSD.

Click here to enlarge image

Fig. 6 shows average ULSD in diesel for September to mid-October.

Click here to enlarge image

Another positive indication that ULSD production is smoothing out is the amount of diesel being downgraded to 500-ppm (wt) sulfur from ULSD (Fig. 7). The downgrade appears to be reaching a limit of about 1% of production.

Construction, communication

To produce ULSD, 80% of US refineries installed some type of hydrotreating equipment. The construction timeline from concept to start-up was typically 3-4 years. To achieve this goal, remarkable coordination occurred between the end users (refiners, pipeline, transportation, retailers), service providers (consultants, design contractors, construction companies), and regulators. The coordinated effort shows a mutual commitment to provide ULSD to the consumer.

Communication between the regulatory community and the industry was also good. Multiple workshops and opportunities to present ideas were made available and the final regulation incorporated many of these items.

Future growth

Future growth in highway diesel demand will be as ULSD because regulations require that 100% of diesel produced in the US be ULSD by year 2010.

Fig. 8 shows this trend.

Click here to enlarge image

The net result of ULSD implementation is less flexibility for US refiners. In the future, for example, waivers similar to those issued in the aftermath of the 2005 hurricanes will be unavailable because subsequent introductions of higher sulfur levels would cause contamination in the distribution system or motor damage.

The switch to ULSD is just the start of greater vigilance on the supply side; the US refining configuration is more susceptible to upsets than it once was because many of the ULSD projects converted existing assets. This lower flexibility means that upsets stand a greater chance of causing supply imbalances.

Until 2009, there may be some opportunities for diesel imports from Europe, but Europe’s supply is tight and, therefore, the arbitrage window might not be a satisfactory relief mechanism. There may be limited alternatives for US supply.

Also, some East Coast heating oil demand could be starting to impinge on ULSD availability. Some markets are demanding lower-sulfur heating fuel, which will eventually force refiners, especially on the East Coast where heating fuel demand is greatest, to invest in additional hydrotreating capacity.


Biodiesel may be a viable technical option in the near future and is being used in Europe due to the greater diesel demand. The American Soybean Association reports that US soy biodiesel production is 150 million gal (9,800 b/d) in 2006, will double in 2007, and will expand to 500-600 million gal (about 40,000 b/d) by 2015.

That represents 13% of the US soy-planted acreage. Other markets are going to find it difficult to add significant oilseed acreage dedicated to fuel usage.

ULSD second wave

In May 2010, the TCO periods will be over and all highway diesel will have to meet ULSD specifications. Until then, more units will be revamped or built to produce ULSD. As many as 50 new or revamped units will start up. These installations will benefit from the knowledge gained during current operations.

Ethanol as renewable

Ethanol in used in gasoline due to the renewable-fuels mandate and elimination of oxygenates in gasoline. The result was a net reduction in gasoline-production capacity due to the removal of MTBE capacity, which was partially offset by the mandate to use ethanol.

Ethanol use has many technical effects on refining, pipeline, and terminating facilities, all of which increase the cost of gasoline production. As with ULSD, we developed key success factors for ethanol use to determine if ethanol is being used as envisioned-data sources have large lag times.

Ethanol key success factors

Success factors for renewable fuels, primarily ethanol blending and production of RFG without oxygenates, are:

  • Original RFG regulations allowed for ethanol blending and most refiners had experience in this area. California Air Resources Board gasoline allowed for nonoxygenated gasoline that has achieved tailpipe emissions targets.
  • Supply logistics.
  • Refining industries’ recognition of the need for renewable fuels, with ethanol blending being the first step.
  • Wide consumer acceptance of ethanol as a clean, renewable fuel.

Ethanol blending-regulations

Three primary ethanol gasolines are produced in the US:

  • Conventional blends with no more than 10% ethanol.
  • RFG using ethanol as an oxygenate.
  • E85, which is 85% ethanol and 15% conventional gasoline.

Of these three, most ethanol is used in conventional and RFG blends. E85 requires car engines designed for multiple fuel types and the majority of the US automotive fleet is not equipped to burn this fuel. The regulatory requirement for the industry is “Regulation of Fuels and Fuel Additives: Renewable Fuel Standard Program” (EPA-OAR-2005-0161; FRL 8016-9).

Fig. 9 shows US renewable-fuels requirements.

Click here to enlarge image

Assuming that US gasoline consumption in 2006 was 142 billion gal, then the ethanol content of the pool is 2.78%. A ratio is used to determine annual industry compliance; if the 2006 volume is not achieved, a “carryover” volume into 2007 will occur.

  • Conventional gasoline containing ethanol has been produced for many years and the regulatory change does not affect production. Increases in ethanol production, however, would likely increase the use of ethanol and the ethanol-containing conventional gasoline supply.

Nonattainment areas, RFG

The Energy Policy Act removed the requirement to use oxygenates in gasoline in nonattainment areas. Removal of the oxygenate mandate in the US enabled refiners to produce RFG without oxygenates while, at the same time, specifying that US gasoline must increase its content of domestically produced ethanol. The legislation basically makes MTBE unattractive as a US gasoline-blending additive.

Ethanol began replacing MTBE in RFG starting about April 2006. Production, availability, transportation, and delivery of ethanol were some reasons for sharp increases in the price of gasoline in early summer 2006. But these were overshadowed by a run-up in crude prices.

RFG production using ethanol was part of the original Energy Policy Act regulations. Emission calculations and other requirements for compliance were also included in the regulation. Many refiners have experience with ethanol blending and producing reformulated blendstocks for oxygenate blending, the nonethanol containing refinery product that is blended with ethanol at the terminal.

Nonoxygenate routes for producing RFG use blending components such as alkylate, iso-octane, raffinate, and others. Many options exist to produce these blendstocks with a corresponding capital investment.

Refiners can gain additional gasoline-production capacity by using ethanol as a blend stock vs. investing capital in new equipment. Ethanol, for example, supplied to the East Coast by barge may be an economic option compared to other alternatives such as capital investment to produce alkylate or iso-octane.

Fig. 10 shows that ethanol has replaced ethers in RFG, based on EIA data.

Click here to enlarge image

The quantity of RFG produced without oxygenates averaged 3% before and after the oxygen mandate was removed. This supports the idea that, in the short term, ethanol is the cost-effective replacement for MTBE.

Ethanol supply logistics

Ethanol requires many steps to deliver it to the consumer:

  • Crop planting, fertilizing, and harvesting.
  • Transportation of corn to ethanol plant.
  • Ethanol production.
  • Transportation to blending facility via rail or truck.
  • Blending.
  • Distribution.

The multiple steps required to bring ethanol to the marketplace and the requirement to deliver it to each blending site greatly increase the supply complexity. Pipelines cannot ship ethanol because of its hydroscopic nature, which causes severe corrosion. Blending is therefore done locally via splash or in-line methods.

Energy supply

Ethanol requires a large investment in fossil-fuel energy to grow the feedstock, harvest, process, and deliver it to the marketplace. All of these steps require fossil fuels. The energy delivered by ethanol as a transportation fuel should exceed the energy required for production for it to be considered a viable long-term energy source.

US ethanol production is primarily from corn, which produces an ethanol that is slightly energy positive. Improvements in cellulose fermentation and energy integration will further improve the energy balance.

Future ethanol supply, demand

Ethanol demand will exceed supplies until about 2008. This will spur imports to meet demand. After 2008, US ethanol production will satisfy the country’s demands. If current projections for ethanol capacity increases are accurate, the 2012 ethanol target required in the renewable energy mandate can probably be met by 2009 without use of imports.

If Brazil continues to expand production, worldwide demand will be met.

Currently, US ethanol production is subsidized to encourage increased capacity. A phaseout of the subsidies or altering them to encourage new technologies is a possible path for regulatory consideration.

The big ethanol users will be the US and Japan, which is another significant importer from Brazil. Where gasoline demand is strong, ethanol demand will be strong. There are other producers of ethanol; possible new capacity will be built in Peru, Vietnam, and Chile.

Europe has some ethanol demand.

Fuel vs. food

The conversion of food to fuel while people are undernourished is an ethical dilemma. The benefit is the renewable nature of the fuel; however, it is produced at the expense of other options, such as food exports. If new acreage is placed into service, then the effect of fuels is minimized.

USDA predicts that 30% of US corn production will be earmarked for ethanol production by 2010. Brazilian cane dedicated to ethanol production vs. sugar peaked at about 70% in 1990 and is currently 50-60% due to sugar-market dynamics.

Biodiesel crop predictions are not available. While these industries are still growing, a balance is needed between the fuel and food production to answer the ethical dilemma.


  1. “US Switchover to ULSD Going Extremely Well,” World Refining Business Digest Weekly, Oct. 10, 2006.
  2. Hackworth, J., and Shore, J., “Ethanol, Gasoline and ULSD Supply Issues in 2006,” presented to the State Heating Oil and Propane Conference, Aug. 7-8, 2006, North Falmouth, Mass.
  3. “Diesel Problems in the Heartland,” Gasoline Column, US Federal Trade Commission, Aug. 15, 2006, http://www.ftc.gov/ftc/oilgas/archive/060815.htm.

The author

Scott Sayles ([email protected]) is a principle consultant with KBC Advanced Technologies Inc., Houston. He has 30 years of refinery and petrochemical experience, ranging from refinery plant manager to research engineer. Sayles holds a BS in chemical engineering from Michigan Technological University and an MS in chemical engineering from Lamar University, Beaumont, Tex.