Final Draft

EnSys Energy & Systems, Inc.

make and the higher carbon to hydrogen ratio of the products which accompanies higher catalytic cracker conversion. Consequently the overall GHG emissions would not drop. This is not to be confused with beneficial environmental effects which may be associated with this example, such as reduced sulfur dioxide emissions achieved by treating of the refinery still gas prior to combustion.

Commonly employed "carbon rejection" refining schemes also do not tend to reduce GHG emissions. The operation of a coker to produce refinery coke thereby lowering the carbon to hydrogen ratio of the liquid refinery products does not avoid the eventual combustion of the coke within the refinery or downstream. Asphalt production to reject carbon runs up against a limited market for the product.

In describing the effects of the preceding examples, the envelope around the refinery was extended beyond internal fuel usage to include the ultimate combustion of all refinery products since they emit GHGs as well as fuel consumed within refinery limits.

A key point here is that GHG emissions ex the refinery (fuel) itself and GHG emissions from the refinery's fuels products are intimately interrelated. Altering one tends to alter the other.

The following table derived from the EIA PSA 1994, Volume 1, p 118 shows the makeup of the U.S. refinery fuel mix as a percentage of refinery input:

Final Draft

EnSys Energy & Systems, Inc.

With 72% of the refinery fuel comprised of gas and 17 % as catalytic coke, a necessary by-product of catalytic gas oil cracking, there is little carbon reduction to be gained from fuel switching in the U.S. However there is more potential in other OECD nations. Traditionally Europe, Japan and other world regions have tended to use internal still gas supplemented by residual fuel as the primary inputs to the refinery fuel pool. For each fuel the proportion of refinery fuel has been close to 50%.

B. 4. Fuel as a Component of Refinery Costs

U.S. refinery fixed plus variable costs total approximately $3.30 (1996 dollars) per barrel of crude oil, including the cost of purchased and internally produced fuel. Adding depreciation and capital recovery charges of a further $3.30 per barrel gives a total cost of $6.60 per barrel.

Refinery fuel and steam usage therefore accounts for approximately 65% of the variable costs and 20% of all refining costs.

B. 5 Refining Sector Employment

Employment statistics given below show a significant drop in employment in the refining sector. as well in petroleum industry upstream and downstream activities. The 33% drop for refining is related to refinery shutdowns and consolidation and in labor- saving technological change. The recent drops

Final Draft

EnSys Energy & Systems, Inc.

in the real price of crude oil, commencing in 1982 probably hastened the adoption of technological
change, impacting primarily on the skilled production workers.

Employment is the average number of employees for indicated years, including production
workers.

Sources: API, Basic Petroleum Data Book and U.S. Department of Labor, Employment, Hours and Earnings

B. 6 Opportunities for Improving Energy Efficiency in Refineries

Improvements in energy efficiency and the exploitation of alternatives to fossil fuels are driven
primarily by increases in the relative price of energy, essentially resulting in a re- optimization of
energy use versus capital investment. Petroleum is currently the dominant energy source and is likely
to remain so for at least the first half of the next century and as such will continue to influence the
energy prices of coal, natural gas and other fuels.

Some of the opportunities for improving energy efficiency in petroleum refineries include the
following:

Final Draft

EnSys Energy & Systems, Inc.

Increased recovery of refinery hydrogen from refinery fuel gases. Hydrogen generation is energy intensive and produces by-product carbon dioxide. Hydrogen is used in hydrotreating, hydrocracking de-aromatization and may be recovered by using adsorption processes to purify, gas-membrane technology and cryogenics.

Improved heat recovery from process effluents and refinery stack gases.

Improvements in heat exchanger and cogeneration technology

The use of catalytic agents to permit refinery processes to operate under conditions that require minimal energy

The use of sunlight to stimulate photosynthetic organisms to produce oxygen for the treatment of refinery wastes.

Improving electric power generation efficiency, e.g. through the use of advanced gas turbines, expansion of steam power "co-generation" projects

Pressure will be placed on technology improvements to reduce internal refinery energy consumption. However, the potential here may be somewhat limited. The 1996 EIA Annual Energy Outlook estimates the following for the out years:

With a fuel price increase falling on heavy fuel oil to a greater extent than the other refinery fuels, refining operations will be affected by pressure to reduce petroleum refinery residual fuel oil production. This would be in response to decreased demand relative to the lighter refinery products and the tendency to eliminate heavy fuel oil from the refinery fuel mix.

Final Draft

EnSys Energy & Systems, Inc. Current total U.S. residual fuel oil demand is approximately 1030 mb/d (1994) down from 2088 mb/d in 1981 and 1158 mb/d in 1991. With over 30% (314 mb/d) from imports, this represents only 5.6 % of refinery production, with only 2.1% of the refinery fuel pool being comprised of heavy fuel oil. In the same period, U.S. refinery production of residual fuel oil has decreased from 1321 to 934 to 826 mb/d, in spite of a shift towards processing heavier crude oils in the U.S. The average crude oil API gravity has dropped from 33 to 31 API in the last twelve years, the reason being an increase in the volume of heavy crude oil imports. Therefore in recent years, while the quantity of the residual fuel fractions in crude oil has increased, the output has dropped as a result of the use of energy intensive conversion processes.

Nationally, residual fuel oil comprises only 5.6% of net U.S. refinery product output compared to around 45% of raw residual-type material in the crude oil used in the U.S. The sharp difference arises because of U.S. refining's uniquely high concentration of catalytic cracking, coking and hydrocracking units which upgrade residual material to gasoline and distillate.

Currently evident in the U.S. is a progressive shift toward the total upgrading of the vacuum residuum and vacuum gas oil fractions of low sulfur crude oils. This has been achieved mainly through advances in catalytic cracking technology. Refiners also continue to invest in delayed coking for upgrading residuum from medium and heavy sour crude oils. The associated loss in residual fuel oil production is partially offset by an increase in production of petroleum coke. There is also continuing develop effort focused on hydrocracking to allow processing residuum with high metals content, often associated with the heavier crude oils.

The continuation of the trends will further decrease residual fuel oil production. The development of more tolerant catalytic cracking catalysts to handle higher sulfur and metals residuum will be a key factor in achieving progress. From an overall GHG emissions perspective, future gains will be partially offset by an increase in refinery energy usage associated with increased refinery conversion and complexity.

46-495-37