THE ECONOMICS OF
RESIDUAL FUEL OIL DESULFURIZATION
A Study for the
Division of Air Pollution
Public Health Service
U. S. DEPARTMENT OF
HEALTH, EDUCATION, AND WELFARE
BECHTEL CORPORATION
San Francisco and New York
June, 1964
-------
FOREWORD
Under the Clean Air Act (Public::,Law 88-206, 88th Congress) the
U. S. Department of Health, Education, and Welfare is charged
with the responsibility of expediting research into the desulfuri-
zation of fuels.
The Act provides for the letting of contracts to
complement research activities performed directly by the Public
Health Service of the Depa:r:tment and by other organizations
"under research grants.
Under contract to the Public Health Service, the Bechtel Corporation
has investigated the costs of reducing the sulfur content of certain
residual fuel oils.
The Division of Air Pollution, Public Health
Service, is publishing this summary report prepared by the
"Bechtel Corporation to make these research results available to all
individuals and organizations who may find them useful.
-------
THE ECONOMICS OF
RESIDUAL FUEL OIL DESULFURIZATION
TABLE OF CONTENTS
Object and Scope 1
Summary 3
Discussion
The Nature of the Problem 8
Crude Selection and Product Slate 13
Formulating the Mathematical Model 16
Refinery Flowsheet 19
Utilities 21
Hydrogen Balance 23
Results, Case Nos. 1 to 8 25
Tables e No.
Disposition of Sulfur in Refinery Products 1
Product Specification Used in LP 2
Abridged Specifications for Fuel Oils 3
Results: Case No. 1 4
Case No. 2 5
Case No. 3 6
Case No.. 4 7
Case No. 5 8
Case No. 6 9
Case No. 7 10
Case No. 8 11
Figure No.
Figures
Refinery Flowsheet 1
Results, Case Nos. 1, 2, 3 and 4 2
Results, Case Nos. 6, 7, and ,8 3
Shadow Prices of Residual Fuel Oils 4
Comparison of Desulfurization ^Cost for California
and ;Kuwait Residual Fueil Oil 5
•New Unit 'Capacities for Existing 'Refinery -6
-------
Appendix -—*—
1. Technical Aspects of the LP Matrix A- 1 to A-3
fc
M-l to M-20
2. Glossary B-1 to B-4
3. Bibliography C-l to C-4
-------
OBJECT AND SCOPE
No. 6 Residual Fuel Oils marketed in the United States contain an
average sulfur content of about 1.6 percent, with a range encompas-
sing a minimum of about 0. 4 percent to a maximum of about 4. 5 per
cent. The emission of sulfur dioxide from furnaces burning such
fuel is recognized as a cause of significant air pollution.
The objective of this study is to establish costs for reducing the sul-
fu'r content of fuel oils manufactured from high sulfur crudes. The
desired maximum sulfur content for an acceptable fuel oil has been
taken to be 0. 5 wt.percent for this study.
Presented are eight case studies developed in detail with the aid of
a Linear Programming (LP) model of a typical 100,000 barrel per
day modern refinery. These eight cases are as follows:
Case No. Description
1 The base case - a nominal 100, 000 BPSD refinery
which, with no restrictions on sulfur, would produce
a residual fuel containing 1. 63 percent of sulfur.
Huntington Beach (California) crude is the feed stock.
With respect to residual fuel sulfur content, the prod-
uct is typical of U. S. production.
Progressively severe restrictions are placed on sulfur
content of residual fuel and the facilities required to
meet these restrictions most economically are defined.
2 Case No. 1 with coking excluded. This forces utiliza-
tion of the more expensive residual oil desulfurization
process.
3 Case No. 1 with both coking and hydro-pretreatment
of catalytic cracker feed stock excluded.
4 Case No. 1 with fuel oil production reduced to 10, 000
BPSD.
5 In this case, the residual fuel sulfur content is held
constant at 0. 5 percent while the production rate is
varied. All processing options are permitted.
-------
Case No. Description
6 Case No. 1, in which 5, 000 BPSD of Kuwait Crude Oil,
and 500 BPSD of "imported" Kuwait No. 6 residual
fuel oil are included in the refinery feed. As in Case
No. 1, progressively severer restrictions are applied
to the sulfur content of the residual fuel product.
7 The same as Case No. 6, but applied to an existing
refinery optimized for operations without restriction
on the fuel oil sulfur content. It is assumed that the
"existing" refinery is fully depreciated, so that capi-
tal charges are assigned only on new units required to
meet the restrictions on fuel oil sulfur.
8 A special case, applicable to a Gulf Coast refinery,
where cheap natural gas fuel is available for potential
conversion to hydrogen.
-------
SUMMARY
The results of the eight case studies are presented briefly in this
summary. Later in the report each of these cases is described in
detail with all of the important computer results presented in tables.
Except for Case No. 5 which deals with varying fuel oil production
rates, all of the eight cases are concerned with the increase in re-
fining costs as progressively stringent sulfur restrictions are ap-
plied. . Desulfurization costs shown are calculated as differential
costs relative to operation for that case where there is no sulfur
restriction imposed. Since the quantity of the other refinery prod-
ucts and their quality specifications are held constant for all cases,
these relative costs are realistic indications of the effect of a sulfur
restriction. The various cases may be compared with each other by
the refinery realization curves. These represent a relative profit
which is calculated by subtracting variable costs from the value of
the products. Since only variable costs are considered, the magni-
tudes of these numbers have no significance except in relation to each
other.
The area of the chart marked "viscosity give-away" indicates the
range in which the computer could not economically meet both the
viscosity and sulfur specifications for the residual fuel oil. In these
cases the product is more fluid than that required by the market.
For some applications, this low viscosity fuel oil could command a
small premium, but no such credit was taken in this study.
Case No. 1 is the base case
in which the full range of re-
finery processes are avail-
able to deal with the sulfur
problem. Desulfurization
of the residual fuel oil was
accomplished primarily by
residual oil coking and hydro-
gen treating the cutter stocks.
At low sulfur levels, the ratio
of cutter stocks to residual
oils was increased beyond the
level required for viscosity
blending alone .
40
REFINERY REALIZATION
CENTS/BARREL OF PRODUCT
CASE >
CALIFORNIA CRUDE OIL
24,300 BPSOFUELOIL
ALL REFINING OPTIONS PERMITTED
DESULFUftlZATlON CQ3T
CENTS/BARREL OF FUEL OIL
0.7 0-9 I.I 1.3 1.5 1.7
WEIGHT PERCENT SULFUR IN RESIDUAL FUEL OIL
-------
Case Nos. 2 and 3 show the
effect of excluding coking
from consideration and
thereby forcing the more
expensive alternate of
residual oil hydrodesulfuri-
zation to be used at all
levels of sulfur in the re-
sidual fuel oil.
While the differential cost
curve is slightly lower than
in Case No, 1, the refinery
realization curve is also
lower. Case No. 3 is the
same as Case No. 2 with the
added restriction that hydro-
gen treatment of catalytic
cracking feedstock was pro-
hibited. No significant dif-
ference in cost resulted.
REFINERY REALIZATION
CENTS/BARREL OF PRODUCT
——CASE 2
CALIFORNIA CRUDE OIL.
COKING EXCLUDED
24,300 BPSD FUEL OIL
___CASE 3
CALIFORNIA CRUDE OIL
COKINS AND FCCU PRETHE AT EXCLUDED
OESULFURIZATION COST
CENTS/BARREL OF FUEL OIL
O.7 0.» I.I 1.3 1.5 1.7
WEIGHT PERCENT SULFUR IN RESIDUAL FUEL OIL
REFINERY REALIZATION
CENTS/BARREL OF PRODUCT
CASE 4
CALIFORNIA CRUDE OIL
FUEL OIL REDUCED TO 10,000 BPSO
ALL REFINING OPTIONS PERMITTED
(OESULFUHIZATION COST
:ENTS/BARREL OF FUEL OIL
07 0.9 I.I 1.3 1.6 I.T
WEIGHT PERCENT SULFUR IN RESIDUAL FUEL OIL
Case Nos. 2 and 3 serve
to illustrate that coking
and resid hydrodesulfuri-
zation are competitive
processes for converting
residual stocks to more
valuable refinery products
Case No. 4 repeats Case
No. 1 except that the re-
sidual oil production is
reduced from 24, 000
BPSD to 10, 000 BPSD.
The cost of fuel oil de-
sulfurization increases
somewhat over Case No.
1, but there is also a
striking improvement in
refinery realization.
-------
Case No. 5, which is not illustrated by graph, explores the profit-
ability of fuel oil manufacture over a wide range of production
levels while maintaining a constant 0. 5 wt percent sulfur specifi-
cation. As the rate is cut from 32, 478 BPSD in a series of steps
to 10,000 BPSD, the realization rises from $0.279 per barrel of
products to $0. 531 per barrel. For the conditions of the study,
the break-even price for incremental fuel oil production at 0. 5 wt.
percent is calculated to be $2. 75. Below this price, there is
economic pressure on the refiner to reduce production of this
product.
Case No. 6 is a new base case for
supplemental studies in which a
constant 5, 000 BPSD of Kuwait
crude substitutes for part of the
California crude. The mathe-
matical data generated by the LP
permitted separate costs of de-
sulfurizing the high sulfur Kuwait
residuum to be calculated. Fig-
ure 3 shows the resulting compari
son between Kuwait and California
crudes.
Also included in Case No. 6 was
a small 500 BPSD increment of
4, 5 percent sulfur "imported"
fuel oil. The model refinery
used this as a "crude" and the
LP calculated a value or "shadow
price" for it at various sulfur
levels.
At the 0. 5 wt percent level, its
value was $. 80 lower than the
price for desulfurized product
fuel oil. This figure can be
taken as a measure of the de-
sulfurization cost.
CASE*
COMPARISON OF OE3ULFURIZATION COST FOR
CALIFORNIA AND KUWAIT RESIDUAL FUEL OILS
SOOO BPBD KUWAIT CRUDE AND
SOO OfSQ IMPORTED RESIDUAL FUEL OIL
REPLACING PART OF CALIFORNIA CRUDE
NO RESTRICTIONS ON REFINERY OPTIONS
REFINERY REAUIAT
CCNT3/BARM L TOTAL PR
Sf SIDLJii. FUEL OIL
. FROM
KUWAIT STOCKS
WEIGHT PERCENT SULFUR IN RESIDUAL FUEL OIL
-------
Case No. 7 considers the con-
sequences of a restriction on
fuel oil sulfur when it is
applied to an existing re-
finery, optimized to process
100,000 BPSD of crude with-
out limitation on fuel oil sul-
fur. It is assumed that the
refinery is fully depreciated
when the sulfur limit is ap-
plied. The incremental cost
to lower the sulfur content of
fuel oil is about 20 percent
higher than for an all-new,
optimized plant such as Case
No, 6, but is not otherwise
radically different. To con-
form to the 0. 5 wt percent
limitation, the refinery would
have to add 10, 000 BPSD of
coking capacity and 14, 000
BPSD of middle distillate hydro-
desulfurization.
REFINERY REALIZATION
CASE 6
9,000 BPSD KUWAIT CRUDE AND
SOO BPSD IMPORTED RESIDUAL FUEL OIL
REPLACING PART OF CALIFORNIA CRUDE
NO RESTRICTIONS ON REFINERY OPTIONS
— — — CASE 7
FUEL OIL SULFUR RESTRICTION APPLIED
TO AN EXISTING REFINERY.
CALIFORNIA AND KUWAIT CRUDE MIX
DESULFURIZATION COST
0.7 0.9 I.I 1.3 l.fl 1.1
WEIGHT PERCENT SULFUR IN RESIDUAL FUEL OIL
CASE 6
3.0OO BPSD KUWAIT CRUDE AND
BOO BPSO IMPORTED RESIDUAL FUEL OIL
REPLACING PART OF CALIFORNIA CRUDE
NO RESTRICTIONS ON REFINERY OPTIONS
— — — CASE 8
5.00O BPSD KUWAIT CRUDE AND
500 BPSD IMPORTED RESIDUAL FUEL OIL
REPLACING PART OF CALIFORNIA CRUDE
NATURAL GAS AT I9C/MSCF USED FOR
REFINERY FUEL
0.7 0» I.I 1.3 1.5
WEIGHT PERCENT SULFUR IN RESIDUAL FUEL OIL
Case No. 8 shows
the effect of mak-
ing natural gas fuel
available at low cost.
When Case No. 8 is
compared with Case
No, 6 it is found
that the availability
of cheap natural gas
improves the profit-
ability of the refinery
by reducing operating
costs. The refinery
configuration is not
significantly affected
by this change ex-
cept that more hydro-
gen is used in pro-
cessing.
-------
It is of particular interest that the availability of cheap hydrogen
did not reduce the cost of desulfurizing fuel oil. If anything, the
relative attractiveness of alternate uses for the stocks blended
into fuel "oil were enhanced by the availability of cheap hydrogen
with the result that fuel oil desulfurization cost was somewhat
higher than in the case with higher priced hydrogen.
The overall conclusions of this study may be summarized as follows
1. Manufacture of low sulfur residual fuel oil from
high sulfur crudes requires an incentive pricing
of $0. 40 to $0. 65 per barrel above fuel oil pro-
duced without sulfur restriction. This cost is
increased about 20 percent if applied to an exist-
ing refinery.
Z. Imported residual fuel oil can be regarded as
crude and processed in a modern refinery: Its
value is approximately $0. 80 per barrel less than
the desulfurized fuel oil product.
3. The availability of cheap natural gas reduces
refining costs in general, but it does not appear
to make fuel oil desulfurization more attractive.
-------
THE NATURE OF THE PROBLEM
The nature and extent of the petroleum sulfur problem can be seen
from Table I which is a tabulation of Bureau of Mines data in the
several 1962 Mineral Industries Surveys published by the Bureau
of Mines. The average sulfur content and API gravity of each of
the products for each geographic area was considered to be repre-
sentative of the total volume marketed in that district. Because
of this and other assumptions in the calculations, no statistical
accuracy can be claimed. However, the table does point up
clearly the fact that the greater part of the sulfur released to the
atmosphere comes from the burning of residual oils. The pro-
portion is likely to show an increase in the near future, because
competitive quality requirements are now forcing the refiners to
desulfurize larger proportions of the other products. Further,
the processes for desulfurizing these lighter stocks have been
well studied and their place in the refining scheme is well estab-
lished. For these reasons, this study has been confined entirely
to residual oils.
A thorough literature survey by the Bureau of Mines indicates that
hydrodesulfurization is the only refining process that appears prom-
ising for removing sulfur from residual petroleum oils. Several of
the major oil companies are known to be developing residual oil
desulfurization processes. However, only two commercial proc-
esses have been announced, the Gulf HDS process and the H-Oil
process developed jointly by Hydrocarbon Research, Inc. , and
Cities Service Corporation.
It should be mentioned in passing that essentially complete removal
of sulfur from residual petroleum oils by hydrogen treating is
easily accomplished in the laboratory. However, these laboratory
reactions require high hydrogen pressure, large ratios of catalyst
to oil and large quantities of hydrogen. The commercial processes
do not attempt complete desulfurization, use special catalysts and
incorporate mechanical features which attempt to overcome these
limitations.- Even so, in the present state of development residual
oil hydrodesulfurization is expensive in comparison with other
refining processes.
Published figures indicate that the total costs for the desulfurization
might be as high as $0. 70 to $1.00 per barrel of residual oil proc-
essed. If it were assumed that this is an additive cost to be applied
directly to the price of residual fuel oil, the process would be pro-
hibitively expensive for many situations. However, an analysis
(8)
-------
which considers the desulfurization process by itself cannot give a
true indication of the cost of the operation. The product oil from
the hydrodesulfurizer actually consists of a wide range of materials
no longer typical of the feed oil. From the refiner's viewpoint it
can be regarded as a synthetic "crude" with a value determined by
the products that can be prepared from it. Only a small fraction of
the hydrodesulfurized product might be blended to fuel oil if the
refinery were "optimized".
The inter-relationship between the various refinery processing
units is extremely complex, and current practice is to approach
these problems with a mathematical simulation or "model". This
study has been largely directed toward creating and manipulating
such a refinery model in which residual oil hydrodesulfurization is
only one of several processes which can contribute to reducing
sulfur content of residual fuel oil sulfur.
In approaching a study of this nature, attention must be directed to
the material itself, the residual fuel oil. ASTM Grade No. 6 fuel
oil is characterized generally by the absence of any specifications
other than flash point (for safety in storage), viscosity (to insure
pumpability), and the maximum content of water and sediments
that can be tolerated in atomizing burners. The refinery thus has
considerable latitude in the choice of components to blend in the
product. In most cases the residual fuel oil sells for less than
the cost of crude, and it is a marginal product kept to a minimum
economic production level.
The degree of freedom that exists for the refinery and the nature of
the material itself is best illustrated by some examples. For
reasons to be explained later, a 23. 8 API California crude was
selected as a basis for much of this study. If this crude were
processed in a simple topping refinery having only a crude unit
and a catalytic reformer, the product distribution might be as
follows:
Gasoline 14%
Lt. Distillates 8%
Residual Fuel Oil 76%
Operating Loss 2%
100%
Sulfur content of the residual oil would be slightly in excess of 1. 5
wt. percent.
(9)
-------
In the above example, large quantities of middle distillates and gas
oils must be retained in the residual fraction to meet the maximum
viscosity specification. Application of modern refinery processing
such as vacuum distillation and propane de-asphalting can reduce
the heavy fraction to an asphalt residue amounting to only 14.4%
on the original volume and at the same time recover oils that are
valuable as feeds to other units. The asphalt that remains is a
hard solid at room temperature. Unless it can be sold as asphalt,
its disposal is a problem for the refiner. Only two options are
generally available: the material may.be blended and sold as fuel
oil, or it may be converted to other products.
If it is to be sold as fuel oil, the refinery must blend the residual
oils with sufficient "cutter stock" to meet the viscosity specifica-
tion. Generally, this cutter stock is the lowest quality middle
distillate available. On the assumption that it is a light catalytic
cycle oil, approximately two volumes of cutter stock will be needed
for each volume of the asphalt described above. The net refinery
yield of residual fuel oil in this case is approximately 44%, the
minimum that can be attained by blending alone without a residual
oil conversion facility.
In nearly every instance where the refinery has facilities for
fractionating the various products from the crude, the most eco-
nomical fuel oil blends will consist of heavy pitch such as the
propane precipitated asphalt described above and a low viscosity
middle distillate. The fuel oil reduction that can be accomplished
by selective blending is substantial and in the above case amounts
to 32% on crude charged.
Because residual fuel oil usually sells for less than the cost of crude,
the refiner is under pressure to reduce the quantity still further
than is possible with these selective blending methods. Conversion
processes include thermal coking, viscosity breaking and the re-
sidual oil hydrodesulfurization.
Of the three processes, only coking is capable of eliminating the
fuel oil entirely. The oil is heated to approximately 850°F and
allowed to flow slowly through a large coking drum where it
decomposes to a hard carbonaceous residue and a wide boiling
range "synthetic crude" which can be fractionated into various
products. Coking, like most of the commercial non-catalytic
10)
-------
cracking processes, has the disadvantage of producing low quality
gasolines and unstable distillate fractions which must be further
processed. Further, much of the sulfur in the residual oil is
concentrated in the coke and limits its usefulness for metallurgical
purposes. It is even marginal as a fuel since the anticipated high
sulfur content would limit its applicability. Even with these limi-
tations, however, coking is a widely used process, and may be
economically attractive even when the coke product is not salable.
Viscosity breaking is a mild thermal cracking process which con-
verts part of the residual oil to distillate products. In many instances,
only the gasoline is removed, and the various gas oils and distillate
stocks resulting from the cracking are retained in the fuel oil to act
as "cutter stock". The net result is a significant reduction in the
refinery yield of fuel oil because the amount of outside cutter oils
is reduced. It is interesting to note that one Southwest refinery
for many years followed the visbreaking operation with a high
vacuum distillation step. This reduced the "black oil" to a hard
pitch which was subsequently discarded in a canyon. Where the
pitch can be disposed of, this viscosity breaking vacuum distilla-
tion scheme is an alternative to coking.
Residual oil hydrodesulfruization processes from the refiner's point
of view resemble viscosity breaking and are sometimes referred
to by the terms "hydro-visbreaking" or "resid hydrocracking". As
descriptive names, these are in some respects superior to terms
that suggest only sulfur removal. Regardless of the conditions
employed or the type of process, the desulfurization reaction is
accompanied by a substantial amount of cracking. Even simple
removal of sulfur from a heavy molecule in a way that does not
involve rupture of carbon-carbon bonds, reduces the molecular
weight and increases volatility.
If these products of the hydrocracking and hydrodesulfurization
reactions are retained in the fuel oil, the net effect is to reduce
the overall refinery yield of residual fuel oil by reducing the
need for outside cutter stock in a way analogous to viscosity
breaking. Further, if the products from the hydrodesulfurization
reaction are regarded as "synthetic crude" and fractionated to
recover all of the gas oil and distillate fractions, it will be found
that the volume of fuel oil can be reduced even further. The same
selective blending procedure applied to virgin asphalt can be used
with the residuum from the hydrodesulfurizer unit.
(11)
-------
It is important to note, however, that the yield of desulfurized
residuum from resid hydrodesulfurization processing is only a
fraction of the original charge volume if all the gas oil and other
distillable material is removed. Depending on the operating
severity, catalyst activity, and crude type, the small amount of
residuum remaining may still have a high sulfur content ap-
proaching that of the feedstock. This material may be blended
to make a reduced quantity of fuel oil or it can in some cases be
recycled to the hydrodesulfurization process so that there is no
net production of fuel oil.
Residual oil hydrodesulfurization is a relatively new development,
and only one plant is in operation. Therefore, nearly all the yield
information in the literature consists of laboratory or pilot plant
data. The yield data for this study came from two different
sources. They serve to illustrate quite well the extremes in
flexibility that exist for the design of a desulfurization unit.
The California residuum data were taken from the Galbreath
and Johnson paper. The results are typical of a mild treatment
of a refractory residuum. The sulfur content of the treated
residuum (not the total treated product) rather closely matches
the sulfur content of the original feedstock.
The Kuwait data were taken from the recent work of Beuther and
Schmid. A very active catalyst was used, together with severe
conditions. The result was a minimum yield of relatively low
sulfur residuum.
Regardless of the type of process, the primary effect of residual
oil hydrodesulfurization is a drastic reduction in volume of
residual stock for fuel oil blending.
12)
-------
CRUDE SELECTION AND PRODUCT SLATE
In'a study such as this, the selection of crude type and the product
distribution is of fundamental importance. In the early phases of
the work, some thought was given to setting up a model descriptive
of the U. S. petroleum refining industry as a whole. Such a linear
programming model .is described by Manne in connection with a
study on the capability of the industry to produce jet fuel. Unfor-
tunately, Manne's model does not include any of the hydrogen
treating processes that are essential to the current study. If
these were included along with the necessary flexibility in proc-
essing operations that Manne's model did not need, the matrix
would have several hundred equations, a simulation beyone the
scope of this project.
Also, his model was based on the crudes from the first 25 of the
major oil fields as listed in the Bureau of Mines survey for 1950.
During the 12 years from that time to 1962, 50% of the fields had
been dropped from the list to be replaced by many that had not
been discovered at that time. Assays were not available for
many of these crudes.
From a practical standpoint, residual oil hydrogen treating yield
data are available only for a few crude types. Practically all of
the work has been confined to Middle East, California or selected
West Texas crudes, none of which are really typical of the average
crude mix for the U. S. It is extremely important in an LP study
to have a consistent set of premises for the simulation.
Because of the large amount of detail required for each crude type,
the simulation had to be limited to one or two of these crudes.
California crude was selected for the base case for two reasons.
It was used in a recent study on fuel oil reduction methods by
Slyngstad and Feigelman and some of their data could be adapted
to this study. Also, the fuel oil made from California crudes
will normally run close to the national average of 1. 6 wt. per-
cent sulfur. Results based on California crude should be fairly
representative of what it will cost the industry as a whole for
desulfurization of residual fuel oil.
An assay of a particular sample of Huntington Beach crude that
closely corresponded to the average API gravity for total
California production was available. Important properties are
as follows:
(13)
-------
Gravity, °API 23. 8
Sulfur, Wt. Percent 1.3
Reid Vapor Pressure, PSI 1. 1
Viscosity, SSU @ 100°F 149
Viscosity, SSU @ 80°F 246
Price, $/BarreL Z. 46
The properties of the various fractions used in the L.P may be found
in the matrix coefficients as given in the appendix.
Product specifications and demands used for the matrix are given
in Table Z. The product distribution corresponds to the approxi-
mate market breakdown in the West Coast for 1962 as presented
by Siyngstad and FeigeLman. These provide a realistic set of
product demands corresponding to the crude gravity.
Since the primary purpose of the simulation was to study the effects
of a sulfur specification restriction on residual fuel oil, the other
products are considered to be "pools" with only average specifica-
tions typical for the material. The gasoline pool represents an
average for premium and regular gasolines with an octane number
slightly higher than the average being marketed today. The light
distillate pool includes #1 burner oil, JP~5 jet fuel, kerosine, and
Light diesel. Heavy distillates include #2 burner oil and heavy
diesel.
Specifications for these materials are limited only to boiling range
and sulfur content. The sulfur contents are typical of what a refiner
might normally set for products from this particular crude.
The LP matrix is purposely structured to eliminate any need for
product prices in determining the cost of desulfurization. The
matrix is set up to produce the required product slate for each
case irrespective of prices, so that the only variables are the
quantity and sulfur content of the fuel oil. In this way, the model
is completely independent of the effects of product market prices.
However, there is pressure on the refiner to reduce production of
fuel oil, and this pressure can be measured in terms of gross
realization relative to product prices. For this purpose the prices
estimated by Siyngstad and Feigelman as refinery net-backs for the
West Coast are used.
(14)
-------
Two products from this model refinery which are not included in
the above tabulation are coke and sulfur. Coke produced from this
type of crude will have little value for most industrial purposes and
because of its high sulfur content, it could not be burned as fuel in
the marketing area served by the refinery. For this reason, the
coke was given a negative value of $1. 00 per ton to allow for a
disposal expense.
Similarly, it was assumed that the refinery is not in the sulfur
business, and no value was assigned to-the byproduct sulfur.
Assigning a value of, say, $20/ton would be reflected as a credit
of $0, 03 to $0, 05 per barrel in favor of the resid hydrodesulfuri-
zation process in those areas where the sulfur could be marketed.
Cases 6, 7 and 8 expanded the scope of the study by replacing 5, 000
BPSD of the California crude with Kuwait. Properties are as
follows:
Gravity, °API 30. 8
Sulfur, Wt. Percent 2. 35
Reid Vapor Pressure, PSIG 8. 1
Price, $/Barrel 2.00
Since the quantity of Kuwait crude was held constant in all cases,
the price had no influence whatever on the LP matrix behavior.
The only use made of the price of Kuwait crude was in calculating
realization.
(15)
-------
FORMULATING THE MATHEMATICAL MODEL
Inter-relationships between the refinery processing units are ex-
tremely complex, and until the recent development of mathematical
programming methods, it was difficult to make meaningful compar-
isons between alternate refinery operations.
The most important and generally useful of the mathematical
programming methods is Linear Programming (LP). It has be-
come a standard technique for economic analysis in the petroleum
industry. Once an LP matrix has been set up to describe a re-
finery complex, it can be programmed into, a computer and used
to calculate the economics of alternate modes of operating the
refinery. It is possible to change product demands, specifications
and refinery processing schemes with assurance that each solution
will be accurately balanced and optimized. For this study, the
Bonner and Moore LP code and GAMMA matrix generator were
used on an IBM 7094 computer. The matrix generator has numerous
features especially provided for oil refinery problems.
Formulation of the LP matrix is by far the most time consuming
operation in a refinery simulation study. Considerable care must
be taken to make sure that all reasonable alternate processing
possibilities are provided. Also, it is extremely important that
all of the data be internally consistent if valid results are to be
obtained. For this study, the latter was taken as a guiding
principle. The base case of the study was purposely narrowed to
a single crude and marketing situation so that full attention could
be given to constructing a matrix that would incorporate most of
the process alternates that a refinery might elect to use. It was
later expanded to evaluate another crude, and to permit handling
some "imported" residual fuel oil.
In spite of its usefulness as a tool for economic analysis, linear
programming does have limitations that influence the formulation
of the problem and the interpretation of the results. The LP re-
quires that a single cost figure be assigned to each possible
operation, and this cost figure must be proportional to the "activity"
of that operation. In blending operations, this requirement is
accurately met; in others, it is approximated by restricting the
range over which a cost applies.
The development of these "cost-per-barrel" processing costs is
a matter of considerable judgment, and how they are formulated
16)
-------
is closely related to the objectives of the study. Processing costs
may be considered in three categories according to the simplifi-
cations that are made.
(!) Generjil Overhead and Operating Labor, It is
customary to ignore all costs of a general nature which
cannot be identified with a particular product or operation.
Since operating labor for a unit generally is practically
independent of the feed rate, this is also omitted from
consideration.
(2) Capital Related Costs. Where a unit already
exists, it is usually assumed that amortization, interest,
maintenance, taxes, and similar items are independent
of the quantity of material being processed. These costs
are excluded from the matrix. On the other hand, if new
facilities must be built, it is proper to assign a cost
penalty against each barrel of capacity that is needed.
(3) Utilities. Utilities, chemicals, catalyst,
running royalties, and other factors that are propor-
tional to the quantity of material processed are always
included in the cost.
These distinctions are quite important, and the method of handling
capital charges is often a source of confusion. In a study for a
"grass roots" refinery where all new facilities are to be con-
structed, the'capital related costs would be included in the matrix,
and an optimum refinery configuration would result from the
solution. After the refinery was constructed, a second LP
solution might be made, with all of the capital cost factors ex-
cluded. This L.P would find an optimum way of running the
refinery which could be quite different from the original case in
which individual unit sizes were determined. The reason for
this is the change in objective. In the first case, the refiner has
included cost factors to deliberately penalize the expenditure of
capital. The objective is to build a refinery in which the ex-
penses for capital write-off, crude and operation are balanced.
Once the facilities have been constructed, there may very well
be a more profitable way to run the refinery -- and the second
L,P solution will show this.
It is important to understand that the capital related charges, when
included in the matrix, are really penalties to force the selection
of low cost processing methods. They represent a balance of
(17)
-------
operating cost, cost of crude, and capital investment that is optimal
for the refinery.
For the purpose of this study, two limiting extremes were set up.
For most cases, each process unit considered for the LP matrix
was penalized by a capital cost which included all maintenance,
taxes and insurance, and "write-off" period of three years before
taxes. The purpose of this was to optimize the refinery for
minimum capital expenditure and at the same time establish an
upper cost boundary tha.t would reflect as severe a criterion as
the industry might employ for incremental expenditure. As an
alternate, the study also considers the case of a fully depreci-
ated refinery which requires new facilities to reduce sulfur con-
tent of residual fuel product. In this case, the capital related
charges were applied only to the new facilities.
The three year write-off period for establishing capital charges is
normal from the industry's point of view. This is a handy rule of
thumb, which corresponds to a true return of capital in about 5
years after taxes if the double declining balance method of depre-
ciation is used. About 18 months will ha.ve elapsed from the time
that a decision is ma.de until new facilities could be constructed,
The refinery accordingly is asked to commit capital for a period
of 6-1/2 years. During this time, new technologies maybe
developed that obsolete the facilities, -- or the market condi-
tions that call for the expenditure may change. If the tax laws
are amended to permit faster depreciation of new facilities re-
quired for fuel oil desulfurization, refiners would find the invest-
ment more attractive.
In the present case where fuel oil desulfurization is being con-
sidered, the possibility of unexpected developments during that
6-1/2 year period is particularly great. . Any nation-wide re-
striction that adds $0.45 or more per barrel to the cost of fuel
oil is certain to cause a re-examination of energy sources by the
industrial consumer, and it is entirely possible that the demand
for residual fuel could decrease drastically. This would certainly
prompt an investigation of alternate fuel, sources. Such an investi-
gation is beyond the scope of this report, but it would appear that
both natural gas and coal would be possible alternates in existing
fossilfueled heaters, and nuclear energy in new power plants. A
preliminary inquiry has indicated that natural gas fuel for power
plants will continue to be available for some time.
18}
-------
THE REFINERY FLOWSHEET
Figure 1 is a flowsheet showing the general flow arrangement for
the refinery model. A considerable amount of time was spent
attempting to prepare a fully descriptive flowsheet which could
show all of the connecting streams between the refinery units.
The resulting diagram was found to be unsuited to discussion
purposes because of the almost unintelligible maze of lines. In
fact, the LP matrix itself may be the most satisfactory means of
representing the various flow paths in the model.
For discussion purposes, the simplified flowsheet, Figure 1, was
prepared. The various refinery streams are grouped into five
categories corresponding to the product pools plus one additional
group for the refinery fuel gas and vapor pressure stocks.
Eleven main refinery processing units are shown, and these are
grouped according to the general function. The groupings are
self-explanatory, but it should be noted that the categories are
to some extent arbitrary. In all of the hydrogen treating proc-
esses, some "cracking" takes place. Also, all of the cracking
processes accomplish some degree of desulfurization.
The flow diagram is arranged specifically to emphasize an
important fundamental principle of petroleum refining. Almost
all of the cracking and reforming processes produce a "synthetic
crude", consisting of some products in the same classification as
the feed plu's products in all of the lighter classifications. The
catalytic reformer to which naphtha is fed produces a naphtha
product and also byproduct gas and butanes classified as "light
ends" on the diagram. Similarly, the delayed coker which
charges the heaviest residual oils produces all of the lighter
product classifications plus coke (which might be classified as
a residual product).
Each of the units shown has wide latitude in accepting nearly all
of the stocks in its feed classification that may be available. The
solid lines depict most of the flow paths actually incorporated into
the matrix. Several promising operations that were omitted for
lack of data are shown with dotted lines. Treating propane
deasphalted pitch by coking it, for example, might be quite
attractive to a refinery where the respective units were available.
Several other possibilities for processing the lighter products
were not included because they are not important to this partic-
ular study. For example, refiners occasionally find it desirable
19)
-------
to hydrogen treat and reform high sulfur naphthas from the catalytic
cracking unit.
One particular process of considerable current interest is the Gas
Oil Hydrodesulfurizer unit. Pre-treating the gas oil feed to the
catalytic cracking unit is an effective alternate to treating the dis-
tillate products, and it has the added advantage that the gasoline
yield and quality are improved at the same time. ':
The hydrocracker which converts heavier oils into gasoline and
middle distillates or middle distillates to gasoline has become
increasingly attractive to the modern refinery. A great deal of
intensive development work is currently being done to lower costs
and hydrogen consumption, as:well as to increase the flexibility in
accepting higrier end point charge stocks. For this study, it was
assumed that the maximum end point of the charge stock was 800°F.
It was also assumed that the unit would be a typical two stage design
with a "Hydro-denitrification" (HDN) pre-treatment section. The
HDN section performs a severe hydrogen treating operation which
removes both sulfur and nitrogen.
While only one "pool" is shown for the light ends, in most refineries
there are two systems for handling these materials. The unsaturated
propylene and butylenes resulting from coking and catalytic cracking
are segregated so that they may be converted to gasoline by alkylation
with isobutane or by polymerization. Saturated light ends may also be
used for hydrogen plant feed if unsaturated hydrocarbons are excluded,
20)
-------
UTILITIES
Not shown on the flowsheet are the peripheral units and utilities
which are incorporated into the matrix. An important question
which is asked in connection with hydrogen consuming processes
is the cost of hydrogen. It is frequently assumed that natural gas
must be purchased for reforming into hydrogen, and on this basis
the cost of desulfurization wouLd depend on the cost of this raw
material. As a complicating factor for this study, natural gas in
most areas of the country competes with residual fuel oil for the
bulk energy market. It is apparent that any restriction that in-
creases the cost of fuel oil will increase the demand for natural
gas, and this in turn will set a new and unpredictable price
structure for that commodity.
To ascertain the availability of natural gas fuel for power gener-
ation and other industrial uses, inquiry was made to the Federal
Power Commission who provided a copy of "Advisory Committee
Report No. 21 on Fuels for Electric Generation" dated December
1963. On page 14 of this report, the following statement is made:
"The supply of natural gas is not expected to impose any
quantitative limitations on the use of natural gas for electric
power generation between now and 1980. The use of natural
gas for this purpose, however, can be expected to vary mark-
edly among the several regions of the country. The extent to
which natural gas is used for electric generation in each
region will depend upon several factors, but primarily it will
depend upon its price relative to the prices of alternative fuels.
This, of course, will be affected by, among other things, the
supply and demand relationship. "
If the hydrogen unit is designed to do so, it can operate successfully
on waste refinery gas. Also, various hydrogen containing vent
streams from the various units can be collected and re-processed
in the hydrogen plant with a considerable saving in overall fuel
consumption. Based on these considerations, the refinery model
was structured to be completely self-sufficient in terms of fuel,
both as process energy and as hydrocarbon feed to the hydrogen
unit.
Where large amounts of catalytic cracking and coking are employed,
a refinery can approach a steam balance because these units pro-
duce byproduct steam. It was assumed that any refinery of the
(21)
-------
type depicted by this model would have a start-up stream generation
facility independent of any size that might be calculated by a refinery
balance. For this reason, no capital related charges were assessed
against steam generation.
The matrix also includes a balance on electric power, and some
simplification had to be made in this regard. It was assumed that
power would be generated by an efficient "public utility" type in-
stallation, but that fuel for this purpose would be supplied by the
refinery. The use of electricity was accordingly penalized at a
constant 3 znil/Kwh rate for generation and distribution charges,
but excluding the fuel value.
Considerable attention was given to structuring the problem so that
it would operate in a closed environment. All fuel used for its own
internal purposes for refinery fuel or electric power was assumed
to meet the same sulfur specifications as that marketed. In other
words, the refinery was required to burn a No. 6 fuel oil having the
same specifications as imposed on the material for sale. This led
to interesting behavior for those problems where the sulfur re-
striction was severe. At the high price developed for desulfurized
fuel, the model found it profitable to burn iso-butane and propylene
rather than convert them to gasoline as would normally be the case.
(22
-------
HYDROGEN BALANCE
The cost of hydrogen is the single largest item of expense connected
with desulfurization processes. However, there is no simple direct
relationship between the amount of hydrogen consumed and the result
ing sulfur level in a specific product. The reason, of course, is
that refinery products are blends of many components, and the
hydrogen requirements for removing sulfur from the various blend
components is variable. As has been pointed out, the residual fuel
oil blends developed by the linear program largely consist of high
sulfur residuum cut back with light distillate. Most of the sulfur
reduction comes from treating the cutter stocks. As compared
with direct resid hydrodesulfurization, this procedure results in
substantial savings in hydrogen. While the resid HDS process
requires in the order of 1000 SCF hydrogen per barrel of product,
the requirements for desulfurization of cutter stock were generally
in the range of 100 to 300 SCF per barrel.
The tabulated run data for.the Case Studies (Tables Nos, 4 to 11)
includes a figure for hydrogen plant capacity, and the differential
hydrogen consumption figures can be easily computed for many of
the cases. There is no particular pattern in the data, which simply
indicated that hydrogen utilization was optimized along with other
economic factors.
It must be stressed that the hydrogen plant sizes shown in the tab-
ulations are only relative. Hydrogen cannot be stored and the
hydrogen plant must have considerable reserve capacity for peaks.
This reserve should be more or less constant for various situations
and should not greatly influence the economics. The refinery
matrix includes a complete hydrogen balance in which all by-
product sources are utilized up to the limit of their availability.
A practical refinery based on any of the processing schemes
selected by the computer would undoubtedly require a sub-
stantially larger hydrogen plant than the balances indicate.
Hydrogen is a relatively expensive refinery raw material, and a
considerable amount of effort can be justified in conserving its
use. It is assumed in this problem that such steps would be taken.
For example, the hydrogen consumption figures for the middle dis-
tillate hydrogen treater are taken from an actual design in which
two stages of pressure let down are used so that the flash gas can
be recovered and recycled.
-------
Proper attention to design details can greatly improve the hydrogen
utilization in a "hydrogen refinery". As Gwin points out, the purity
requirements for the various processes does vary, and vent gases
from one unit can serve as hydrogen feed to another. Even further,
the various low pressure hydrogen rich flash gas streams that are
available can be collected, compressed, and reprocessed through
the hydrogen plant. The methane and ethane impurities are cracked
to more hydrogen, while the hydrogen present simply is carried
through unchanged. A substantial saving in fuel can be demon-
strated by such a procedure.
Byproduct hydrogen from catalytic reforming is important as a
source of hydrogen in a refinery. In this particular refinery model,
the hydrogen from this source is relatively constant because the
quantity of gasoline did not vary between cases. California crudes
in general have high naphthene contents and give large byproduct
hydrogen yields when the naphtha fractions are reformed. For
other crudes, the overall size of the hydrogen plant would have to
be increased, but this again would be a more or less constant
factor.
The costs for hydrogen in this study assume a modern high pressure
steam reforming unit, and are as low as can be reasonably expected
at the present state of development. Because of the current interest
in hydrogen processing in general, attention is being focused on
hydrogen manufacturing process. However, it is doubtful that any
significant "break-through" can occur which would greatly influence
the results of this study. The reason is the large influence of fuel
and other utility costs. Under the conditions of Case 1 to 7, the
cost of hydrogen plant feed was equated to fuel produced by the
refinery on the basis of its heating value. This resulted in a
hydrogen cost of about $0, 30 per MSCF, of which only $0. 13 per
MSCF is capital write-off. A large-scale, community type plant
would, in effe\ct, be able to reduce only this part of the cost.
24
-------
RESULTS
The general conclusions have been given in the summary. In this
section the results of each case study will be examined in greater
detail.
THE CASE STUDIES
Case No. 1 24, 300 BPSD Fuel Oil Reference;
Variable Sulfur Specification Table No. 4
For this case, all of the products were held constant according to the
demands and specifications in Table No. 4. Only the sulfur specifi-
cation on No, 6 fuel oil was varied. In this way the effect on refinery
costs attributable to the one single specification could be evaluated.
The economic "optimum" sulfur content where there is no restriction
is 1, 63 wt. percent for this case. This was used as a base in com-
puting the differential cost as the sulfur content is progressively
reduced to a lower limit of 0. 5 wt. percent. These differential
costs are plotted in Figure 2,
The refinery configuration for this case is particularly interesting.
Except for iteration No, 7-52 at 0. 5 wt. percent sulfur, the Resid
HDS unit was not used, and all desulfurization was obtained by a
combination of coking, FCCU pre-treating {the Gas Oil HDS), and
treatment of cutter stocks. Below approximately 0. 95 wt. percent
sulfur, the amount of low sulfur cutter stocks exceeds the viscosity
blending requirements with the result that there is a give-away of
viscosity. The viscosity of this fuel oil is lower than the specifi-
cation requirement (45 SSF @ 122°F) and in some markets could
command a small premium as a No, 5 fuel oil.
How the computer blended the final fuel oil product to arrive at
0. 5% sulfur is interesting and shows typically how a refiner would
deal with this problem.
Component Wt. %S Volume
Resid HDS byproduct middle dist. 0.5 106
Vacuum pitch. 2.03 5, 149
Resid HDS product pitch 1. 7 163
Virgin heavy middle dist. (desulfurized) 0. 12 10, 289
FCCU Jot. Cycle oil (desulf. feed) 0.01 5, 195
FCCU Hvy. Cycle oil (desulf. feed) 0. 2 2, 946
FCCU Decant oil (desulf. feed) 0.05 452
24, 300
(25)
-------
The-"Refinery Realization" as plotted in Figure 2 (and also in
Figure 3 and 5} requires definition and qualification. The term
as used in the context of this report has only relative significance
because a number of costs facing an actual refinery operation
have not been deducted. Typical of these are the capital charges
associated with a number of off-site facilities such as start-up
steam generation, crude and product storage, crude receiving
and product shipment facilities and other costs, such as operating
labor, factory overhead, laboratories and other controls, etc.
These were omitted from consideration because they are affected
only to a very minor extent if at all by the variants considered in
the study.
Case Nos. 2 and 3 24, 300 BPSD Fuel Oil Reference:
Va riable rSuif ur Spe c^ficj-tion Tab Ie s 5 JU 6
In Case No. 2, coking was excluded from the processing scheme.
This forced the use of the resid hydrodesulfurization process as the
only alternative to meet the increasingly severe restrictions on fuel
oil sulfur. Comparison of Case No. 2 with Case No. 1 would then
provide an assessment of the relative economics of coking and resid
hydrodesulfurization.
In Case No. 3, both coking and pretreatment of fluid catalytic cracker
feed were excluded. In this case, the only means available for middle
distillate desulfurization would be the middle distillate hydrogen
treater. Comparison with Case No. 2 would, therefore, provide an
appraisal of the economics of catalytic cracker feed pretreatment
versus middle distillate treatment.
Data for Case No. 2 are summarized in Table 5. It will be noted
that the gross realization for Case 2 is uniformly lower than Case 1.
By examination of the "Refinery Configuration" portions of Tables 4
and 5, it will be noted that the coking capacity called for in Case 1
has been replaced by resid hydrodesulfurization. The conclusion is
that as far as California crude is concerned, coking is economically '
more attractive than resid hydrodesulfurization for processing
California crude at any level, of sulfur content. However, the
incremental cost of producing low sulfur fuel in Case 2, as com-
pared to its base case (Machine Iteration 10-73), is lower than in
Case I. (See also Figure 2)
(26
-------
Data for Case No. 3 are summarized in Table 6. If Case 3 is
compared to Case 2, there is a rather small advantage in gross
realization for Case 2, but the difference is very minor in the
case of California crude. Again looking at the Refinery Con-
figurations of Cases 2 and 3, the exclusion of Cat cracker feed
pretreatment in Case 3 required a' substantial increase in middle
distillate hydrogen treatment capacity for Case 3. The slightly
improved gross realization for Case 2 indicates that Cat cracker
feed pretreatment is slightly more economical than middle dis-
tillate hydrogen treatment. The difference is very small and
could reverse under other conditions. The incremental cost of
fuel desulfurization of the Case 3 refinery compared to its base
case (which is essentially the same as the Case 2 base case) is
essentially the same as Case 2.
Case No. 4 10,000 BPSD Fuel Oil Reference:
Variable Sulfur Specification Table No, 7
All Processing Options Permitted
In Case No. 4, the product slate was altered from that used in Case
Nos. 1, 2 and 3 by reduction of residual fuel oil production from
24, 300 BPSD to 10, 000 BPSD. All refining options available to
Case No. 1 were permitted; thus, comparison of Cases 1 and 4
would provide an assessment of the profitability of manufacturing
fuel oil.
Comparison of the data in Table 7 with those in Table 4 show that
gross realization at all levels of sulfur in the fuel is higher for Case
4 than for Case 1. Fuel oil production is clearly unattractive, and
there is substantial pressure to reduce production.
The incremental cost of desulfurizing residual fuel in the Case 4
refinery over its base case {Machine Iteration 8-66) is somewhat
higher than for Case 1 (Machine Iteration 7-77). This is possibly
related to the larger percentage of FCCU decant oil which is blended
to fuel in the smaller volume of Case 4. Decant oil is a very low
quality slop oil, high in sulfur, which usually has to be blended to
fuel as the only practically available means of disposal.
Case No. 5 Constant Sulfur Specification of 0. 5 wt, %
Variable Fuel Oil Quantity Reference:
All Processing Options Permitted Table No. 8
For this case, the residual oil sulfur specification was held at a
27)
-------
constant value of 0. 5 wt, percent while the total quantity of fuel oil
was varied. Note that Iteration No. 8-52 of Case No. 4 represents
a point at 10, 000 barrels of residual fuel that can be considered a
part of Case No. 5. Several points are worth noting: (1) As the
production of fuel oil is increased, the amount of gas oil cracking
needed decreased (FCCU and hydrocracker). Larger amounts of
the feed stocks to these untis are consumed in the fuel oil blends.
(2) At one point, the Resid Hydrodesulfurization unit entered the
solution showing that it and coking are competitive processes, the
choice of which will be governed by other factors than residual oil
sulfur specification, (3) At the low residual oil yield levels, the
hydrocracker becomes important as a means of using up distillate
oils that otherwise would have gone to fuel oil blending. In all of
the runs, the hydrocracker is important where middle distillates
are in excess.
The significance of Case No. 5 lies in gross realization per barrel
of refinery product figures. These vary greatly with the total fuel
oil production.
Fuel Oil, BPSD 10, 000 13, 138 22, 380 28, 284 32,478
$ Realization 43, 742 41,694 36,059 32,404 29, 629
$ Loss/SD -- 2, 048 7, 683 11, 338 14, 113
$ Loss/Barrel -- 0.65 0.62 0.62 0.62
At the product values used for calculating the realization, each barrel
of fuel oil produced above 10, 000 BPSD represents a loss of gross
realization of $0. 60/barrel or more. The break-even point for 0. 5
wt. percent sulfur No. 6 fuel oil is accordingly close to $2. 75/barrel
for the conditions of this study.
No computer runs were made to explore effect of fuel oil volume at
high sulfur levels. A comparison between Case Nos. 1 and 4 shows,
though, that at the 1. 6 wt. percent sulfur level, there is a gain in
refinery realization of about $2, 300 in reducing fuel oil from 24, 300
to 10, 000 BPD. This corresponds to an incremental loss of realization
of $0. 16 for each barrel of fuel oil produced. Even without a sulfur
restriction, reducing fuel.'oil appears attractive to the refiner.
Consideration of the above cases shows clearly what is already well
recognized in the oil industry; production of No. 6 fuel oil is not a
profitable business. It also explains why nearly one half of the No. 6
fuel oil consumed in the United States is imported.
28)
-------
Under the pricing used in this study ($2, 14 per barrel for No. 6 fuel
and $2.46 per barrel for crude), the refiner is under pressure to
reduce fuel oil production without any restriction on sulfur content;
he is, in effect, losing $0. 16 for each barrel of No. 6 fuel he pro-
duces, if his production rate is about 25 percent of crude run. If
a specification restricting fuel sulfur to 0. 5 percent maximum were
applied, the loss would rise to $0,60 as compared to alternate
processing schemes which would reduce the production of No. 6
fuel. To prevent such a loss, a refiner would find strong justifi-
cation for spending capital funds to purchase the process equipment
required to reduce production of No. 6 fuel, and may even be able
to justify going out of the No. 6 fuel business altogether.
It is also interesting to note that although the JLP was given the
option of propane deasphalting, this alternate was never selected.
Case No. 6 Substitution of 5000 BPSD Kuwait Crude
Oil in Crude Slate Plus Addition of 500
BPSD "Imported" Kuwait No. 6 Fuel Oil
Variable Sulfur Specification Reference:
Table No. 9
In Case Nos. 6, 7 and 8, the effect of incorporating a second crude
was studied. In addition, provision was made for including means of
desulfurizing a relatively small amount of "imported" high sulfur
No. 6 fuel oil made from Kuwait crude.
The LP matrix was also structured to provide in the products, an
extra 1, 000 BPSD of No. 6 fuel oil made entirely from Kuwait com-
ponents. By following the shadow prices calculated by the LP pro-
gram, the costs for refining the Kuwait crude, and desulfurizing
the "imported" fuel oil could be determined on an incremental basis.
By holding the Kuwait components to incremental quantities, the
basic structure of the refinery would not be so drastically changed
as to make comparison with the non-Kuwait cases impractical.
The refinery models did take into account the special characteristics
of Kuwait crude. Thus, special models were incorporated in the
matrix for Kuwait crude distillation, catalytic reforming including
the pretreat, catalytic cracking, residual oil hydjr-adeS'Ulfurization
and coking. To prevent excessive complication of the LP matrix,
the Kuwait models were simplified toy limiting the reformer to only
one level of octane improvement,, and the cat cracker to one con-
version level. These simplifications were possible because with
(29).
-------
the relatively small amount of Kuwait crude in the feed, the approx.
imation would introduce a negligible error; the advantage of the
simplification was an appreciable reduction in the machine time
for reaching a solution.
The product slate for Case No. 6 {and also for Case Nos. 7 and 8)
was as follows:
BPSD
Motor Gasoline 40, 300
Light Middle Distillate 8,000
Heavy Middle Distillate 16, 900
Residual Fuel Oil Z4, 300
Special Kuwait Re-
sidual Fuel Oil 1, 000
90,500
The pertinent results of Case No. 6 are summarized in Table No. 9,
and in Figures 3, 4 and 5.
The gross realization per barrel of product given in Table No. 9 for
Case No. 6 is generally comparable to Case No. 1 - they differ only
by about $0. 01 per barrel. The incremental cost of fuel oil desulfur
ization is, as might be expected, higher than in Case 1, reflecting
the higher quantity of sulfur in the feed.
Of much greater interest are the data plotted in Figures 4 and 5. In
Figure 4 are given the "Shadow Prices" for Kuwait fuel oil, Cali-
fornia fuel oil, and the "imported" high sulfur fuel oil, together
with the shadow cost of removing 1 percent sulfur from one barrel
of California and Kuwait crude. These data are all plotted as a
function of the percentage of sulfur and each of these fuels.
Referring first to the shadow prices of Kuwait, California and
"imported" fuel oil, the following conclusions may be drawn:
1. Kuwait fuel oil is relatively expensive material to refine to
low sulfur content. This is apparently due to the increased
requirement for resid hydrodesulfurization and middle dis-
tillate hydrogen treatment as compared to California crude,
Interestingly, the shadow prices of Kuwait and California
crude become equal at 0. 5 percent sulfur, indicating that
at this sulfur level, any incremental production could be
made interchangeably from either crude, as far as cost is
concerned, where the production rate is 25, 300 BPSD of
total fuel.
(301
-------
2. Imported high, sulfur fuel oil is an undesirable feed. At 0. 5%
sulfur specification for fuel oil, the shadow price of this
material is $2.04, compared to an assumed market price of
$2. 14 and a shadow cost of manufacturing specification fuel
of $2. 84 per barrel.
The data in Figure 5 were obtained by integrating the shadow prices
given in Figure 4 over the whole range of sulfur reduction. Figure
5, therefore, gives the total cost of desulfurization of Kuwait and
California residual fuel, from their natural sulfur contents down
to 0. 5 percent sulfur. Thus, Kuwait fuel would cost an incremen-
tal $0. 658 per barrel for desulfurization to 0. 5 percent compared
to $0. 54 for the balance of the fuel. This has been labelled "Cali-
fornia" in Figures 4 and 5 - actually it includes Kuwait components
not accommodated in the 1, 000 BPSD of special Kuwait fuel. The
difference in incremental desulfurization cost for "California" crude
in Case 6 from that of Case 1 is partly due to this contamination
with Kuwait.
The shadow cost of desulfurizing residual fuel given in the lower
portion of Figure 4 requires further comment. It will be noted
that the shadow cost for removal of 1% sulfur from a barrel of
fuel oil is higher for California crude over the range from about
1, 3% to 0. 5%. In interpreting these curves, it must be remembered
that the shadow cost is a derivative or slope of the total cost at some
point. In the above range, the refiner would be installing the bulk
of his desulfurization facilities for California crude, whereas he
would have to install them much earlier for Kuwait.
Case No. 7 Existing Refinery Operation Under
Progressively Lowered Maximuni
Limits on Sulfur Content of No. 6 Reference:
Fuel Table No. 10
Case 7 simulates a modern 100,000 barrel refinery built with no
regard for residual fuel oil sulfur and develops the various process
ing capacities the owner of such a refinery would have to install in
order to comply with progressively more restrictive limits on the
sulfur content of No, 6 fuel oil.
The matrix includes Kuwait crude and Kuwait fuel in the feed as in
Case No. 6 with the additional feature that a base capacity was
established for each of the major refinery processing units. This
(31)
-------
basic refinery was set up after examination of several preliminary
computer runs. It is a computer optimized refinery and is typical
of-the type of modern refinery that might be built for the product
mix and crude slate used in this study. The fluid catalytic crack-
ing unit, the catalytic reformer, and the naphtha hydrotreater
units were purposely set at figures somewhat larger than the
optimum called for by the computer solution. This is also true
of the crude unit which was set at a nominal capacity of 100, 000
barrels per day. The adjustments in all cases were relatively
small and intended to give the refinery the operating flexibility
that the refinery designers would specify for a real situation.
Furthermore, all units in this basic refinery are assumed to be
fully depreciated so that no capital charges apply. An LP matrix
was then structured to permit the construction of new facilities as
required to meet increasingly severe sulfur specifications in the
residual fuel. For these new facilities a capital investment charge
was included.
The original refinery showed a satisfactory hydrogen balance
sufficient for the modest size hydrocracker and middle distillate
hydrotreater. By-product hydrogen from the reformer was
adequate for these purposes. The matrix was given the option to
build hydrogen manufacturing facilities as needed. Results of
Case No. 7 are not radically different from those obtained in the
other cases presented. The cost of sulfur removal as shown in
Figure 3 is about 10-20% higher than for the new refinery cases.
This indicates that the position of the oil company having an
existing refinery is not radically different from the "grass roots"
cases previously studied. Figure 6 summarizes the results of
this particular study by showing the incremental additions to each
unit plotted against the weight percent fuel oil sulfur specification.
Further information is given in Table No. 10. These results are
entirely consistent with the others that have been presented. As
can be seen, delayed coking and middle distillate hydrogen treating
dominate. At 0. 5 wt, percent sulfur specification the refinery
must add a 10, 000 barrel delayed coker and almost 14, 000 barrels
of middle distillate hydrogen treating. A nominal amount of new
catalytic cracking is required to balance gasoline production. This
does not appear to be particularly significant and simply indicates
that possibly the base refinery may not have had sufficient crack-
ing capacity for adequate flexibility.
-------
As has been explained in Cases 2 & 3, residual oil hydrotreating and
coking are competitive processes. Which one is most economical
depends on the particular circumstances of crude type, hydrogen
cost and the importance of capital expense factors. Both processes
can be used to effect either a reduction in total fuel oil production
or to reduce its sulfur level. This is illustrated nicely by the
different behavior of California and Kuwait residual stocks. Coking
was preferred in almost all cases with California stocks, while
residual oil desulfurization was found to be most economical for
Kuwait stocks. There are two basic reasons for this:
1. When the residual hydrodesulfurization process is carried
out on Kuwait residuum, desulfurization reactions assume
a much more important position than in the case of
California residuum, probably because of the large
amount of sulfur present in Kuwait, Each cubic foot
of hydrogen consumed removes a greater weight of
sulfur from Kuwait than from California stock.
2. The coker distillates from undesulfurized Kuwait residuum
are much higher in sulfur than comparable materials from
California, and require more severe processing to convert
into salable products.
Taken together, these two factors favor residuum hydrodesulfurization
for Kuwait stocks.
With only one exception, there is no indication that the type of crude
would have any significant effect on the way in which units would have
to be added to an existing refinery such as is considered in Case 7.
This single exception is the hydrogen plant. The availability of highly
naphthenic naphthas from California crude produces relatively large
yields of by-product hydrogen from the reformer. Thus, with a
more paraffinic crude input, hydrogen plant sizes would be larger
than indicated here,
Case No. 8 Cheap Natural Gas Available Reference:
Table No. 11
To cover the situation where cheap natural gas fuel is available, and
hence, the possibility of cheap manufactured hydrogen, the refinery
of Case No, 6 was re-optimized on the basis that natural gas would
be available at a cost of $0. 19 per thousand cubic feet.
(33)
-------
There is no indication that the availability of low cost hydrogen
makes it any more attractive for the refinery to desulfurize No, 6
.fuel oil. In fact, if Case Nos. 6 and 8 are compared, it is evident
that the principal effect of cheap natural gas is to improve the
refinery profitability by reducing operating costs, mostly by re •
ducing crude intake.
The pertinent data are summarized in Table 11 and Figure 3,
Examination of these data, and comparing them with Case No. 6
leads to the following conclusions:
1. Availability of cheap natural gas does not make any significant
change in the optimum size of a coker.
2. Hydrocracking is substantially more favorable when low cost
hydrogen is available, and tends to displace an equal quantity
of fluid catalytic cracking. .This applies even though there
has been no change in the product slate.
3, Low cost hydrogen does not make a significant difference
in the optimum sizing of resid hydrodesulfurization. A
more drastic reduction in hydrogen cost might tip the
balance in favor of resid hydrodesulfurization over coking
however.
(34)
-------
TABLES
-------
TABLE NO. 1
DISPOSITION OF SULFUR IN REFINED PRODUCTS
YEAR OF 1962
Sulfur
Product - 1000 BPD
Approx. %
Total
Refined Net Sulfur
in Area Imported Exported Consumed #/Bbl Tons/Day %Content Tons/Day Burned
Composite Average - Total U. S. (Excluding Rocky Mountain Region)
Gasoline 4,146 38 - 18 4,166 Z54 529,083
Kerosine (Incl.
. 043
228
5.0
Comm. Jet)
Military Jet Fuel
Distillate Fuel Oil
Residual Fuel Oil
Asphalt
All Other
422
261
1,900
769
281
787
18
30
32
722
18
95
- 1
- 23
- 35
- 2
- 84
439 283 62,119 .079 49 1.1
291 275 40, 013 . 067 27 0.6-
1,909 294.5 281,100 ,213 599 13.2
1,456 348.7 253,854 1.428 3,625 80.1
297
798
Totals
8,566
953
-163
9,356
1,166,169
4f528 100.0
-------
TABLE 2
PRODUCT SPECIFICATIONS USED IN LP MODEL
Motor Gasoline - Total Pool
Research Octane (F-l) 95. 0
Reid Vapor Pressure, PSI 10.0
Light Middle Distillate
Max. Wt. % Sulfur 0. 25
Minimum B. P. , °F 330
Max. B. P. , "F 540
No Coker of FCCU Oils Permitted
Heavy Middle Distillate
Max. Wt. % Sulfur 0. 50
Minimum B. P. , °F 330
Max. B. P., °F (Cracked) 650
Max. B. P., "F (Virgin) 700
Residual Fuel Oil
Max. Wt. % Sulfur 0.5-1.7
Max. Viscosity, SSF at 122°F 175
Minimum I. B. P. , of Blend Stock °F 330
PRODUCT DEMANDS
Based on approximately 100,000 BPD of Crude (Note 2)
Value
BPSD $/Barrel
Motor Gasoline 40,300 $ 4.95
Light Middle Distillate ' 8,000 $ 3.02
Heavy Middle Distillate 16,900 $ 3.02
Residual Fuel Oil (Note 1) 24f300 $ 2.14
Note 1 - Quantity of residual fuel oil is varied in the same LP runs.
Other products are always kept constant.
Note 2 - Estimated values in refinery storage.
-------
TABLE NO. 3
ABRIDGED SPECIFICATIONS FOR FUEL OILS
GRADE NO.
Test
Flash Pt., °F Min.
Pour Pt., °F Max.
Water fa Sediment, % Max.
Carbon Resid, 10%BTMS,
% Max.
Ash, Wt. %, Max.
Viscosity
Kinematic, CS, 100F, Max.
Kinematic, CS, 100F, Min.
SUS, Sec., 100F, Max.
SUS, Sec., lOOF, Min.
SFS, Sec. , 122F, Max.
SFS, Sec., 122F, Min.
100
tr.
100
20
130
20
130
150
0. 10
0. 50
1.00
2.00
0. 15 0.35
0.10 0.10
Distillation
10% Max.
90% Max.
90% Min.
420
550 640
540
2.2 (3.6) (26.4)
1.4 (2.0) ( 5.8) (32. 1)
37.93 125
32.6 45 150
40 300
45
Source:
ASTM Tentative Standard D-396-61T
Note:
Numbers in parenthesis are equivalent kinematic viscosities.
-------
CASE NO. 1 EFFECT OF VARYING SULFUR CONTENT OF RESIDUAL FUEL OIL TABLE NO. 4
Machine Iteration Number ?-5Z 7-54 7-58 7-66 7-71 7-77
Wt. % Sulfur in Fuel Oil 0.5 1.04 1.lZ 1. ~6 1.40 1.63
Value of Products - $/SD 326,685 3Z6,685 3Z6.6B5 326, 685 326.685 326,685
Less: Cost of Crude 239,351 ~35,835 ~35, 768 235,206 235,011 235,582
Capital Related Coata 39,201 35,788 34,625 33,993 33,503 n, 764
Variable Operating Coata 13,107 11,90b II, biZ H, 744 H, 793 11,414
Grosa Realization - $/SD 35,OZb 43,156 44.680 45,742 46,378 46,925
Crude Run. BPSD 97.297 95,868 95,841 95,612 95.533 95.765
Product Sla.te. BPSD 89,500 89.500 89,500 89.500 89.500 89,500
Realization/Crude Run - $/B 0.360 0.450 0.466 0.478 0.485 0.490
RealiMUon/Total Products $/B 0.391 0.482 0.499 0.511 0.518 0.524
Crude Cost/Bbl PrOduct 2.614 2.635 Z.634 2. 6Z8' ~.6Z6 2.632
Capital t; Variable Cost/Bbl ~r()duct 0.584 0.533 0.517 0.511 0.506 0.494
A Crude Cost/Bbl Product + 0.042 ... 0.003 ... 0.002 - 0.004 - 0.006 Bade
A Capital t; Variable Cost/Bbl Product .. 0.090 .. 0.039 + 0.023 .. 0.017 .. 0.012 Base
To-tal A Cost/Bbl Product .. 0.132 t 0.042 .. 0.025 .. 0.013 .. 0.006 Base
Total ACost/Bbl Fuel Oil t 0.486 t 0.155 + 0.092 + 0.048 .. o. all Base
REFINERY CONFIGURATION
Crude Unit, BPSD 97,297 95,868 95,841 95.612 95,533 95.765
Reformer Pretreate1" , BPSD z.z,848 21,062 20,316 20,221 20,217 19,564
Reformer , BPSD 20,097 19.345 19,626 19,623 19,562 19,932
FCCU Feed Pretreater, BPSD 29,686 18,524 7,116 7,673 8.850 0
Fluid Catalytic CrackinJ! Unit, BPSD z.z.613 22,925 21.028 21,278 2.2.052 19,993
Middle Dis tilla.te HYdr02en Treater , BPSD 16,865 16,569 19.969 16,103 12,82.0 11,923
Delayed Cokin2 Unit. BPSD 2.3,712 17,82.4 17,846 17,948 18.233 19.605
Alkylation Unit, BPSD 2,981 2,993 2,839 2,986 3.084 3,003
Resid. HDS Unit, BPSD 815 0 0 0 0 0
Hydrocracking Unit. BPSD 1.007 2,679 3,981 ".010 3,891 4.960
Hydrogen Plant. MMSCF ISD 9.2:03 4.424 0 0 0 0
Sulfur Plant, M LB/SD Z61 ZJO 183 18/ 189 14Z
-------
CASE NO.2 EFFECT OF VARYING SULFUR CONTENT OF RESiDUAL FUEL OIL TABLE NO. ,
Delayed Coking Excluded
Machine Iteration Number 10-51 10-56 10-59 10-62 10-68 10-7Z 10-73
W t. C1J'e Sulfur in Fuel Oil 0.5 0.639 0.717 1.022 1.135 1.258 1.276
Value of Products - $/SD 326.685 3Z6.685 326.685 326.685 326,685 326,685 326,685
Less: Cost of Crude 230,310 227,506 227,302 U6,974 226,846 ZZO,719 226,371
Capital Related Cos ts 51,767 50,087 i8,285 46,433 46,003 46,031 46,248
Variable Operating Costs 13, 282 ~ -.&.ill ~ -1b.ill -1b...!1! 12.298
Gross Reali~ation - $/SD 31,326 36,354 38,685 40,9n 41,591 41,761 41,762
Crude Run, BPSD 93,6ZZ 92,482 92,399 92,266 92,214 92,162 92,023
Product Slate, BPSD 89,500 89,500 89,500 89,500 89,500 89,500 89,500
Realization/Crude Run - $/B 0.335 0.393 0.419 0.444 0.451 0.453 0.454
Realization/Total Products - SIB 0.350 0.406 0.432 0.458 0.465 0.467 0.467
Crude Cost/Bbl Product 2.573 2.542 2.540 l.536 2.535 Z.533 2.52.9
Capital & Variable Gost/Bbl Product 0.72.7 0.702 0.678 0.656 0.651 0.650 0.654
A Crude Coat/Bbl Product + 0.044 + 0.013 + O. all + 0.007 + 0.006 + 0.004 Base
ACapital & Variable C08t/Bbl Product + 0.073 + 0.048 + ~+0.002. ~ - 0.004 BaBe
Total ACoBt/Bbl Product + O. 117 + 0.061 + 0.035 + 0.009 + 0.003 + 0.0 Base
Total b.Coflt/Bbi Fuel Oil + 0.431 ... 0.225 + 0.129 + 0.033 + 0.011 + 0.0 B11f1e
REFINERY CONFIGURATION
Crude Unit. BPSD 93,622 9Z. 482 92,399 92.266 92.2.14 92.162 92.,023
Reformer Prctreater. BPSD 20.718 20,420 19,971 19,470 19,29Z 19.242. 19,343
Reformer, BPSD 17,511 17,173 16,793 16,717 16.710 16,739 16.482
FCeU Feed Pre treater, BPSD 5,323 667 0 0 0 0 0
Fluid Cata lytic Crackin2 Unit, BPSD 28,335 29,217 29.373 29,082 28,864 29.647 31,637
Middle Distillate Hydrol!:en Treater. BPSD 22..912. 23,805 2.2,847 18.405 15.656 12,2.29 10.951
Alkylation Unit, BPSD Z,079 2,041 2,1B2 2,309 Z,336 2:,32:0 2,445
Hesid. HDS Unit, BPSD Z8.555 Z7. J19 24,996 2,3,2.82 23,2.37 2.3.947 l5,l10
Hydrocrackinl!: Unit. BPSD 0 0 387 1.041 1,276 1.351 n6
Hvdrollen Plant, MMSCF /SD 20.848 16,92.3 15,2.43 13,671 13. Z 77 13.414 B.313
Sulfur Plant. M LB/SD 233 ~17 207 190 182 185 190
-------
CASE NO.3
EFFECT OF VARYING SULFUR CONTENT OF RESIDUAL FUEL OIL
TABLE NO.6
Delayed Coking and FCCU Pretreating Excluded
Machine Iteration Number
W t. 'Yo Sulfur in Fuel Oil
Value 01 Products - $/5D
Less: Cost 01 Crude
Capital Related Costs
Variable Operating Costa
Gross Realization - S/SD
Crude Run. BPSD
Product Slate. BPSD
Realization/Crude Run - $/B
Rea.lization/Total Products SIB
Crude COB t/Bbl Product
Capital &: Variable Cost/8bI Product
Ii:. Crude Cost/Bbl Product
0. Capital &: Variable Cost/Bbl Product
Total /.lCost/Bbl Product
Total .6.CoBt/Bbl Fuel on ".
REFINERY CONFIGURATION
Crude Unit. BPSD
Reformer f'retreater. BPSD
Reformer, BPSD
Fluid Catalytic Crackinll Unit, BPSD
12.-50
0.5
326,685.
2,2.9,882.
52..491
13.408
30.904
93,448
89,500
0.331
0.345
2.568
0.736
+ 0.039
+ 0.082
+ 0.121
+ 0.446
93,448
19,82.3
16.971
30.141
l'Aiddle Distillate Hvdrol!en Treater, BPSD 31.049
2.,077
Alkvlation Unit, BPSD
Reaid. HDS Unit. BFSD
Hvdrocrackinl! Unit, BPSD
HyQrogen Plant. MMSCF /SD
Sulfur Plant. M LBISD
.
2.8~ 502
o
18.189
226
12.55
0.660
326,685
2.2.7.489
49,770
12..647
36. 179
92.,475
89,500
0.398
0.411
2.542.
0.697
+ 0.013
+ 0.043
+ 0.056
t (}.2D6
92.475
2.0.310
16,920
29, 334
23.984
2,069
2.6,869
o
16.456
214
Base is machine iteration No. 10-73 of Case No. 2.
12,-58
0.794
326.685
22.7.2.74
49,073
12,38Z
38.956
92.,388
89,500
0.4lZ
0.435
2.539
0.676
t 0.010
+ 0.02.2
+ 0.0:32
t O. U8
92..388
19.931
J6. 782'
Z9,369
ZZ,686
2.,197
2.4,72.9
15.069
IZ-60
1. OZZ
326,685
ZZ6.974
46.433
lZ,306
40,97Z
9Z, 266
89, 500
0.444
0.458
2.536
0.656
+ 0.007
+ 0.002
+ 0.009
.. 0.033.
92,2.66
19.470
16,117'
29,082
18.405
2, 309
23,282
445
1,041
13.671
206
'90
At higher
SuIIuJ" Ieveb,
re8ulte are
identic,,1 to
Cue Z
-------
CASE NO.4 EFFECT OF VARYING SULFUR CONTENT OF RESIDUAL FUEL OIL TABLE NO. 7
Conelant Product Slate, 10,000 BPSD Residual Fuel Oil
Machine Iteration Number 8~5Z 8.53 8-57 8~58 8-60 8-6Z 8-66
Wt. " SulCur in Fuel Oil 0.5 0.939 1.069 1.2.62, 1.384 1.5oIZ I. 610
Value of Products - $/50 2,96,083 Z96,083 2,96,083 2,96,083 Z96,083 2,96,083 2,96,083
Lees: C08t of Crude 2,OZ,559 2,01,181 2,01,174 2,01,2,01 2,01,031 2,01,036 2,01,075
Capital Related Costs 37,079 35,955 35,398 34,565 34,2,44 33,930 33,794
Variable Operating C08tS 12,.703 ~ ~ --!.l.lli ...&Q.!! lZ.030 ...!!...ill.
CiroSl Realization. $/50 43,74Z 40,64Z 47,343 48,356 48.796 49,087 49. Z2,5
Crud!! Run. BPSD 8Z,341 81.781 81,778 81,789 Bl,7Z0 8I,7ZZ 81,738
. Product Slate , BPSO 75,200 75,200 75,200 75,ZOO 75,2,00 75,200 75,ZOO
Realization/Crude Run - $/B 0.531 0.570 0.579 0.591 0.597 0.601 0.602
Rea.Hzation/Total Product. $/B 0.582 0.62,0 ' 0.630 0.643 0.649 0.653 0.655
Crude Co&t/Bbl Product 2.693 z.675 Z.675 Z.675 Z.673 Z.673 Z.674
Capita.l Ie Variable Co.t/Bbl Product 0.66z O,64Z 0.632 0.619 0.615 0,611 0.608
A Crudoc COllt/Bbl Product + 0.019 + 0.001 + 0.001 t 0.001 . 0.001 . 0.001 Ba.e
A Capital Ic Variable Co~ l/ Bbl Product + 0.054 + 0.034 i....!h.lli :!:.JL..Ql.!. + 0.007 ~ --.!!!!L
Total .ACollt/Bbl Pruduct + 0.073 + 0.035 + 0.025 + 0.012, + 0.006 + O.OOZ Ballc
Total AC08t/Bbl Fuel Oil ... 0.549 ,to.2.U ... 0.188 ... 0.090 + 0.045 ... 0.015 Baec
REFINERY CONFIGURATJON
Crude Unit, BPSO 82.,341 81.781 81,778 81,789 81, no 81,122. 81,738
Reformer Prctrcater. BPSn 19,32,1 18,42,8 18,064 17,518 17,465 17,484 17,445
Reformer. BPSD 19,062. 18,686 18.82.9 19,063 18,92,6 18,957 18,990
FCCU Feed Pretreater, apse 19,797 14.32.6 8,730 0 0 0 0
Fluid Catalytic Cnckinll Unit. apsn 2.1,985 2,2,,2.58 2.1,316 19.836 2.1,461 2.1,197 2,1,02.9
Middle Dilltillate Hydroe-en Treater. BPSD 12,,191 12.,2.98 13,964 16,762 15.466 14,397 13,273
Delaved Cokinll Unit. BPSD 22.,887 2,0,570 2,0,62.9 ZO.769 2.0.82,2 2.1,2.37 21,401
Alkylation Unit. BPSD 3.42.3 3.435 3,360 3.2.39 3,317 3,384 3,388
Hydrocrackinll Unit, BPSD 4,507 5,z'79 5,920 6.904 6,734 6.734 6.824
HydrOflen Pla.nt. MMSCF /SD 8,607 6.781 4.613 1.2.48 0.741 0.531 0.343
Sulfur Plant, M La/SO 236 217 204 184 186 183 179
-------
CASE NO.5 TABLE NO. 8
EFFECT OF VARYING RESIDUAL FUEL OIL PRODUCTION AT CONSTANT SULFUR CONTENT
Varying Fuel Oil Production at 0.5 Wt. ,. Sulfur
Machine Iteration Number 8-52 9-73 9-74 9-75 9-76
Heavy Fuel Oil - BPSD 10,000. 13,138 2Z,380 28,264 n. 478
Value of Products - $/SD 296,083 302,798 322,576 335.168 344,186
Less: Cost of Crude 202,559 210,701 234,698 248,9S6 261.63J
Capita.l Related Costs 37,079 37,484 38.640 40,471 39,494
Variable Operating C08h 12.703 lZ,919 13,179 13,337 13,432
Gross Realization - $/SD 43, 742 41,694 36,059 32,404 29.629
Crude Run - BPSD 82,341 85.651 95,406 101,202 106,354
Product Slate - BPSD 75,200 78,338 87,580 93,464 97,678
Realization/Crude Run - $/B 0.,531 0.487 0.378 o. no 0.279
Realization/Total Products $/B 0.582 0.532 0.412 0.347 0.303
REFINERY CONFIGURATION
Crude Unit, BPSn 82,341 85,651 95,406 101,202 106,354
Reformer Pretreater. BPSD 19,321 20.270 22,456 23,657 24,558
Reformer. BFSD 19,062 19,360 20,048 20,203 21,217
FCCU Feed Pre treater, BPSD 19,797 22,964 29,370 30,953 28,749
Fluid Catalytic CrackinSl Unit, BPSD 21,985 22,OJ7 22,341 23,176 20,920
Middle Dis tillate H'(droR'en Treuer , BPSD 12,191 13,073 16,040 18,566 22,290
Delayed Coker, BPSD 22,887 2.3,187 24,182 22.943 26,599
Alkylation Unit, BPSD 3,423 3,319 3.063 2,813 2,650
Rcsid. HDS Unit. BPSD 0 0 0 Z.497 0
Hvdrocrackin2 ~nit, BPSD 4.507 3,617 1,495 0 0
Hydro£Cn Plant, MMSCF /SD 8.607 8.824 8.915 9.798 6.015
Sulfur Plant, M LB/SD 236 242 259 267 264
-------
CASE NO.6 EFFECT OF INCLUDING 5000 BPSD OF KUWAIT CRUDE AND TABLE NO. 9
500 BPSD OF KUWAlT NO.6 FUEL OIL IN REFINERY FEED
MAchine Itera.tion No. 18.67 18.71 18.73 18-75 18.77 18-85 18-89 18-94 18-98 18-1004 18-105 18-106 18-108
Wt. CVo SuUur in Fuel Oil 0.500 0.6Z5 0.840 0.967 1. 070 1.2.13 1.421 1. 530 1.640 1.842 3.184 ~. ~95 ~. 9B6
Value of Products - S/SD 328.325 32,8,3ZS 328.325 328,325 328,325 328.325 nB.3Z5 3Z8. 325 328,325 328,325 32,8,325 328,325 328,32,5
Less; Cost oi Crude 239,144 237,835 Z35.S6Z 234,2.04 2,33,568 232,,998 Z3Z,520 232..793 232, 849 2,32,889 232. 917 2,32,982, 233,114
Capital Related Costs 38,30Z. 37.664 ~6. 640 36.050 35,126 34,374 33.667 33,182. 32,,930 32.699 n.485 32.,396 32,217
Variable Operating
Costs 14.047 13.751 13.256 IZ.961 12.710 12,528 lZ.412 12.343 U:Z61 12.398 12. 359 12.349 12.328
Gross Realiza.tion 36,832 39.075 42.867 45,110 46,921 48,425 49,666 50,007 50,2.85 50,339 50,564 50,598 50.666
Crude Run. BPSD 98,148 97,616 96.69Z 96,140 9S.8Bl 95.649 95.455 95. %6 95. 5B9 95.605 95,617. 95,643 95,697
Product Slate. BP5D 90.500 90,500 90.500 90.500 90,500 90,500 90,500 90,500 90,500 90,500 90,500 90,500 90,500
Realiution. S/Bbl Ct'ude 0.370 0.400 0.443 0.469 0.4B9 0.506 0.5Z0 0.52.3 O. SZ6 0.5Z7 0.52.9 0.5Z9 0.5Z9
ReAliza.tion. $/:8bl Producu; 0.40Z 0.43Z 0.474 0.498 0.51B 0.5~5 0.549 0.553 0.556 0.556 0.559 0.559 0.560
Crude Cost/Bbl Product Z.04Z Z.6Z8 Z.603 Z.588 Z.5B1 Z.575 Z.569 2..512 Z.573 Z.573 Z.574 Z.574 Z.576
Ga.P. it Va-r. Cost/8bl Product 0.578 0.570 0.551 0.542 C.5Z9 0.518 0.510 o.50~ 0.499 0.498 0.496 0.494 0.492.
.aCrude Cost/B'bl Product 0.066 0.05Z 0.02.7 0.012 0.005 M 0.001 M 0.007 M 0.004 ~ 0.003 ~0.003 - O. 002 - 0.001.
OCap. & Var. Coet/Bbl Product 0.086 0.018 0.059 0.050 0.037 O.OZO 0.018 0.011 0.001 . 0.006 0.004 a.aOl
TOUlI 0. Cost/Bb! Product O.lSZ. 0.130 0.086 0.Q62 0.042 0.025 0.01l 0.007 0.004 ~ 0.002
Total ACos t/Bbl Fuel Oil 0.544 0.405 0.30B O.ZZZ 0.150 0.089 O. 039 O.OZ.S 0.014 O.Oll 0.007
REFINERY CONFIGURA nON
Crude Unit. BPSD 98,148 97,616 96,692 96,140 95,881 95,649 95,455 95.566 95,589 95.605 95.617 95.643 95,697
Reformer Pretreater. BPSD 21,945 2.1,552. 2.0.961 2.0.640 19.890 19.778 19.378 19.456 19.361 19.249 19,2.52. 19.Z70 19.306
Reformer. BPSD 20,981 20,578 19,859 19.42.0 19;383 19.42.1 19.2.S1 19.52.9 19,5B5 20.432 20.431 20,448 20.481
FCCU Feed Pretreater, BPSD 9.915 9,821 10,849 11,866 4,150 -1,752 3,020 0 0 0 0 0 0
FCC Unit. BPSD 18,773 19.636 20,896 21.5B2. 2.3,414 Z3.96Z Z3.560 23,077 22..821 2.0,Ol8 20,021 19.99Z 19.93~
Mid.distillate H2.-Treater. BPSD 2.4, 787 2.3.516 ZO.8Z4 19.160 ZI.Z45 ZO.760 14.899 14.5Z3 12,647" IZ.60Z 12,644 IZ.604 12.704
De1aved Coker. BPSD 2.5.508 23.256 19,324 16,948 16.140 15,383 15,091 15,845 16.104 15,593 15.686 15. B23 16,098
Alkvla.tion Unit, BPSD 2, 746 2.803 2.911 2,979 2,935 Z.B91 2.998 Z.967 Z.979 2.,984 2.985 2,981 2,975
Resid. HOO Unit, BPSD 1,430 1,430 1,430 1,402 1.375 1. Z6B 1.2.2.0 1.138 1, 023 930 56Z 375 0
Hvdrocra.ckinll Unit. BPSD Z.093 2.,795 2,841 2,820 3,724 3,955 4,2.70 4.488 4,678 6,080 6,072 6.064 6,048'.....-
Hydroli!'en Plant. MSCFD 2,676 2.,741 3,262. 3.715 1.025 0 0 0 0 0 0 0 0
Sulfur Plant. M LS PSD z65 Z56 Z40 Z31 ZI6 Z03 IB9 175 168 16B loB 169 170
-------
CASE NO.7 EFFECT OF FUEL OIL SULFUR RESTRICTIONS ON AN E}QSTING REFINER Y TABLE NO. 10
Machine Iteration No. 17-69 17-71 17-72 17-74 17-83 17-89 17 ~94 17"98 17-109 17-114 17-115 17-117
Wt. " Sulfur in Fuel Oil 0.500 0.602 0.843 1.08Z 1.186 I.Z92 1.499 1. 631 1. 741 Z.124 3.Z71 3.276
Va.lue of ProductS - S/SD --. 328,825 328,825 328.825 328, BZ5 32.8,825 328.BZ5 328,825 328,825 328,825 328,82.5 32.8,825 328,825
Less: Cost of Crude .; 240, 330 239,211 236,533 234.059 233,506 233, n2. 233,435 2.33,170 Z32,969 233,112. 2.33.098 233,097
Capital Rela.ted Costs ,"' 1,128 1,142. 1,155 1.010 6lZ 404 23 339 388 163
'".\ ,'.
Varia.ble Operating Cosu -.:.......::::: 19,622 18,917 17.Z87 15.683 15,230 14.184 13,37Z lZ.650 12.519 12..494 12.,471 12..456
Gross Realization 67,745 69,555 73.850 78,073 79,477 80,515 81.995 82,666 82,949 83.056 83,256 83.272
Crude Run. BPSD 98.630 98,175 97.087 96,081 95.856 95,944 95,827 95,719 95,638 95,696 95.690 95,690
Product Slate. BPSD 90.500 90,500 90.500 90.500 90,500 90,500 90.500 90,500 90.500 90,500 90,500 90.500
Jt.ealiza.tion, $/Bbl Crude 0.687 0.708 0.761 0.813 0.82.9 0.839 0.856 0.864 0.867 0.868 0.870 0.870
R..""Hzation, S/Bbl Products 0.749 0.769 0.816 0.863 0.878 0.890 0.906 0.913 0.917 0.918 0.920 0.920
Cnlde Cost/Bbl Product 2.656 2..643 2.613 2..586 2.580 2.582 2.579 2..576 2..574 2..576 2.575 Z. 57S
Ca.'!). & Var Cost/Bbl Product 0.ZZ9 0.Z2Z 0.204 O. 184 0.175 0.161 0.148 0.144 0.143 0.140 0.138 0.138
4 Crude Cost/Bbl Product 0.081 0.068 0.038 0.011 0.005 0.007 0.004 0.001 - 0.001 0.001
.e.Ca'P. &. Vu. Cost/Bbl Product 0.091 0.084 0.066 0.046 0.037 0.023 0.019 ~ 0.005 0.002
Total L.Cost/Bbl Product O.ln 0.152 0.104 0.OS7 0.042 0.030 0.014 0.007 0.004 0.003
Total ~Cost/Bbl Fuel Oil 0.615 0.544 0.372 0.204 0.]50 0.107 0.050 0.025 0.018 0.011
REFINERY CONFIGURATION NEW CAPACITY (UNUSED EXISTING CAPACITY IN PAR.£:NTHESES)
!!!!
Crude Unit, BPSD 100,000 (1,370) (1,8Z5) (2,913) (3,919) (4,144) (4,056) (4,173) (4.281) (4,362) (4,304) (4.310) (4,310)
Reformer Pretrea.ter. BPSD 23.000 (1.050) (1,444) (2,361) (3,133) (2,905) (2,320) (2,377) (2.509) (2.530) (2,514) (2,514) (2.518)
Reformer. BPSD 23,000 ( 755) (1.179) (2.167) (3.037) (3.048) (3,135) 13.382) (3,631) (3,969) (3.911) (3.946) (3,926)
FCCU Feed Pretrea.tcr , BPSD
FCC Unit. BPSD 20,000 378 1,147 2.937 4,292 3,213 Z48
Mid-Distillate HZ -Treater. BPSD 13,000 13.547 12.447 9,973 8,323 9.361 9.203 4.0l0 72 ( 109)
Delaved Coker , BPSD 16,000 10.268 8.377 3.972 ( 294)
Alkylation Unit, BPSD 3.000 (664) (592) (424) (Z59) (Zl8) (140) (90) (40)
Resid. HDS Unit. BPSD 1.180 1.180 1,151 1.123 926 611 34 513 587 247
. Hvdroc:ra.c:k.inll Unit. BPSD 7,000 (694) (1,770) (1,436) (355) (174) (686) (751) (712) (730)
Hvdrollen Plant MSCFD 2,162 l.l49 2.447 1.163 ~
Sulfur Plant. M LB PSD 170 90 82 64 45 40 35 Z5 10 2 2 3 2'
'-
'- -
-------
CASE NO.8 EFFECT OF INCLUDING 5000 BPSD KUWAIT CRUDE AND
500 BPSD OF KUWAlT NO.6 FUEL - CHEAP NATURAL GAS AVAILABLE TABLE NO. 11
:Machine Iteration No. 19-67 19-71 19-72 19-76 19-82 19-89 19-90 19.97 19-104 19-105 19-106 19-108 19-1 IZ
Wl. % Sulfur in Fuel Oil 0.500 0.625 0.803 1. 061 1.167 1.303 1.451 1.609 1. 716 2.838 3,458 3.986 4.103
Value of Products - $/SD 328,825 328;825 328,825 328,825 328,825 328,825 328,825 328,825 328,825 328,825 328,825 328,825 328, 825
Less: Cost of Crude 235,818 234,126 232,057 229,067 230,529 230,131 230,096 230,433 230,327 230,200 230,160 230,167 230,170
Capital Related Costs 38,706 38, JOI 37,253 36,040 34,616 34,008 33,451 n,897 32.599 32,441 32, 364 32,206 32,204
Variable Operating
Costs 16,624 16,623 16.379 ....!hlli. 14,484 14,332 14,087 13,757 13,989 14,058 14,058 14,130 14,129
GroSt> Realization 37,677 39.975 43,136 47,706 49,196 50,354 51,191 51,738 51,910 52,126 52,243 52.3ZZ 52, 3Z2
Crude Run, BPSD 96, 796 96.108 95,267 94,052 94,646 94,484 94,470 94.607 94,564 94,512 94,496 94,500 94. 455
PToduct Slate, BPSD 90,500 90.500 90,500 90,500 90.500 90,500 90,500 90,500 90,500 90,500 90.500 90,500 90,500
Realization, $/Bbl Crude 0.389 0.415 0.452 0.507 0.519 0.532 0.541 0.546 0.548 0.551 0.552 0.553 0.570
Rea.lization, $/Bbl Products 0.416 0.441 0.476 0.527 0,543 0.556 0,565 0.571 0.573 0.575 0.577 0,578 0.595
Crude Cost/Bbl Product 2.605 2.587 2.564 2.531 2.547 2.542 2.542 2.546 2.545 2,543 2.543 2.543 2.543
Caf,l. & Var. Cost/Bbl Product 0.611 0,604 0.592 0,575 0.542 0.534 .0.525 0.515 0.514 0,513 O. S12 0.512 0.512
~ Crude Cost/Bbl Product . 0.063 0.043 0.021 0.012 0.004 0.001 0.001 0.003 0.002 0.000 0.000 0.000 8ase
kCap. & Var. Cost/Bbl Product 0.099 0.092 ~ ~ 0.030 0.022 0.013 0.003 0.0(12 0.001 ~ 0,000 Bas~
"Iotal bCost/Bo1 Product 0.162 C.I3> 0.101 0,051 0.0}4 0.021 0.012 ~ 0.004 0.001 0.000 0.000 Bast:
Total "'-Cost/Bbl Fuel Oil 0.578 0.488 0.363 0.182 0.123 0.077 0.044 0.02,3 0.016 0.007 0.003 0,000 Ea.se
REFINER Y CONFIGURATION
Crude Unit, BPSD 96,796 96,108 95,267 94,052 94, 646 94,484 94,470 94, 607 94,564 94,512 94,496 94,500 94,500
Reformer Pretreater, Bf'SD 21,717 21,247 20, 729 19,978 19,704 19,362 19,214 19,123 19,059 19,031 19,049 19,056 19,056
Reformer. BPSD 22,285 22,077 21,574 20,846 20,702 20.476 20, 386 20,377 20.726 20.8Z4 20.853 20,957 20.958
FCCU Feed Pretreater, !\PSD 11,066 10,336 10,985 11,910 0 0 0 0 0 0 0 404 404
FCC Unit. BPSD 13,152 13.410 14,144 15,210 19,70Z 20,02.4 20,160 20,098 18,897 18,553 18,487 18,035 18.035
Mid-distillate Hz-Treater, BPSD 24, 059 22,996 20,964 17,853 22,747 19,859 16,470 13,51S 12,589 12,613 12,735 12, S61 12,561
Delaved Coker, BPSD 25,816 23,706 20,498 15,878 IS,850 IS,141 15,095 15,551 IS,979 15,960 IS,94Z 16,137 16,137 . "-
Alkvlation Unit, BPSD 2,882 2,947 3,041 3,178 2,911 2.969 2,993 3,008 2,960 2,944 2,935 2,929 2,929 -..
Resid. HDS Unit, BPSD 1,430 1,430 1,387 1,378 1,300 1,248 1,214 1.032 671 447 332 0 0
Hvdrocrackinrz Unit, BPSD 4,887 5,358 5.562 5,858 5.871 6,046 6,110 6.183 6, 787 6.982 7,007 7,152 7,152
Hvdrollen Plant, MSCFD 6.833 7,138 7,720 8,558 2,052 1,483 980 0 0 0 0 0 0
Sulfur Plant. M LB PSD 267 256 244 224 204 194 183 In 170 171 171 173 173
-------
FIGURES
-------
GENERALIZED
REFINERY
FLOWSHEET
eJ1.WOAJ.
SEPARATION
e!!2WID
J!IM2WI
TREATING
-=
ClUCKING AND
~
~
i,.J\SOlIN£
1I"lli!ll.!I
UGHT ENOS CLI
"'--' ".-, 4UIYLATION
..,
POI..TlIERI%ATIOtI
P- ..h. !.., 4
" I""PHTHA " . . CATALYTIC ..!'..
"OS REFORMER
4 ,J!-o
"APHTHAS 1'"
I, . 1
L '
"'--' ". "'---'
, .L - 1-
MIDDLE " 1- ~~ - - -
DISTILLATE . - "'" f----o H'rOROCRACXER .!!..-
"OS - ,---.
~ CRUDE r ,-<
UN,T '
I : -
MJDDLEOISTlLLATES(III, '
I, L ' l
i .
,
,
,
"'--' '
,
,
, ' ,
" 1- ...!!..- -
'" FLUID
,,, ' - CATALYTIC .
,..... "" , CRACKER ~ - -
-
GAS OILS (GJ
I, - JLJ
"'--'
, I-"--
VACUUM RESID . f----< f-'!-
RESIDUAL DELATED
'" " f-'!-
, COKER
r-. "" , , - r-'-
'
l ' I----COKE
1 '
,
: ,
, RESIDS!R) ,
: ' 1
POA GAS OIL tG1 '
, ,
PROPANE ' ,
DEASPHALTER POA PITCH If 11 ' ,
FIGURE 1
LIGHT ENDS Tt!
GASOLI"! AND FUEL
NAPHTHAS TO
GASOLINE
IIIIrDOLE DISTILLATES TO
K[ROSINE, HEATING OIL
ANDRES'DUAL FUEL
(lLENDING
GAS OILS TO
RESIDUAL FUEL
BLENDING
RESIDUAL OILS TO
FUEL OIL
BLENDING
-------
..
:!~,
,n
-'
...
a:
a:
..
.. ..
....
"
...
z
...
u
20
VISCOSITY
GIVE-AWAY
'0
o
0.'
0.7
WEIGHT PERCENT SULFUR IN RESIDUAL FUEL OIL
CASE 1
CALIFORNIA CRUDE OIL
24,300 apso FUEL OIL
ALL REFINING OPTIONS PERMITTED
FIGURE 2
REFINERY REALIZATION AND
INCREMENTAL COST OF FUEL OIL
RESULTING FROM SULFUR RESTRICTION
--
'0
RtFINERY REALIZATION
CENTS I BARREL OF PR~OUCT \
--\--
'0
-.-.
-CASE 2
CALIFORNIA CRUDE OIL
COKING EXCI.UOED
24,300 apse FUI':LQIL
---CASE 3
CALIFORNIA CRUDE OIL
COKING AND Fcev PRETREAT EXCLUDED
..
VISCOSITY
GIVE-AWAY
0
,., ,.. '.7 0.' 07 0.' ...
0.'
,.,
'0
20
L'
,..
WEIGHT PERCENT SULFUR IN RESIDUAL FUEL OIL
JL~
REFINERY REALIZATION
CENTS/BARRI::L OF PRODUCT
--
.0
'\
..
20
VISCOSITY
GIVE-AWAY
..
'.7
o
0.'
0.7
WEIGHT PI~RCENT SULFUR IN RESIDUAL FUEL OIL
L'
CASE 4
CALIFORNIA CRUDE OIL
FUEL OIL REDUCED TO 10,000 BPse
ALL REfiNING OPTIONS PERMITTED
0.'
L'
'.7
,..
-------
120
liD
REFINERY REALIZATION
100
90
80
--
---
.J
.... 70
0::
0::
i
40
30
20
VISCOSITY
GIVE-AWAY
VISCOSITY
GIVE-AWAY
10
DESULFURIZATION COST
o
0.5
0.7
WEIGHT PERCENT SULFUR IN RESIDUAL FUEL OIL
WEIGHT PERCENT SULFUR IN RESIDUAL FUEL OIL
FIGURE 3
REFINERY REALIZATION AND
INCREMENTAL COST OF FUEL OIL
RESULTING FROM SULFUR RESTRICTION
60
REFINERY REALIZATION
-----
----
---
--
--
50
....---
-----r
40
- CASE 6
5,000 8PSD KUWAIT CRUDE AND
50D 8PSD IMPORTED RESIDUAL FUEL OIL I
REPLACING PART OF CALIFORNIA CRUDE
NO RESTRICTIONS ON REFINERY OPTIONS I
,I - - ~ CASE 8 I j
5,000 8PSD KUWAIT CRUDE AND
50D BPSD IMPORTED RESW'UAL FUEL 01 L
REPLACING PART OF CALIFORNIA CRUDE i
NATURAL GAS AT 19~/MSCF USED FOR I
REFINERY FUEL I
NC' RESTRICTIONS ON REFINERY OPTIONS j
I
I
I
- CASE 6
5,000 BPSD KUWAIT CRUDE AND
500 BPSD IMPORTED RESIDUAL FUEL OIL
REPLACING PART OF CALIFORNIA CRUDE
NO RESTRICTIONS ON REFINERY OPTIONS
30
VISCOSITY
GIVE-AWAY
- CASE 7
FUEL OIL SULFUR RESTRICTION APPLIED
TO AN EXISTING REFINERY,
CALIFORNIA AND KUWAIT CRUDE MIX
20
VISCOSITY
GIVE-AWAY
10
--
--
DESULFURIZATION COST
.....----
1.1
0.9
1.3
o
0.5
0.7
1.3
1.5
1.7
0.9
1.5
1.7
1.1
-------
3.00
...J 2.00
w
ex:
ex:
-------
FIGURE 5
CASE 6
COMPARISON OF DESULFURIZATION COST FOR
CALIFORNIA AND KUWAIT RESIDUAL FUEL OILS
5000 BPSD KUWAIT CRUDE AND
500 BPSD IMPORTED RESIDUAL FUEL Oil
REPLACING PART OF CALIFORNIA CRUDE
NO RESTRICTIONS ON REFINERY OPTIONS
70
60
REFINERY REALIZATION
CENTS/BARREL TOTAL PRODUCTS
50
..J
w40
0::
0::
eX
m
.....
'"
...
r5 30
u
RESIDUAL FUEL OIL
FROM
KUWAIT STOCKS
RESIDUAL FUEL OIL
FROM
CALIFORNIA STOCKS
I
20
10
o
0.5
2.0
3.0
4.0
1.0
WEIGHT PERCENT SULFUR IN RESIDUAL FUEL OIL
-------
14
FIGURE 6
STUDY OF
EXISTING 100,000 BARREL REFINERY
REFINERY ADDITIONS AND EXPANSION
REOUIRED TO MEET LOWER RESIDUAL
FUEL OIL SULFUR LEVELS
>-
«
c
.......
VJ
-l
W'O~
0::" ':.
0:: ~
« ~
ID ~
\~
~~ 8
C
W
W
Z
>-
t:6
u
«
Cl.
«
U
I-
Z4
::>
==
w
Z
2
12
.
\\
~ \
'\ \
\ \
.
.
.
.
.
.
,
.
.
.
~
.
\ I
'.
. ,)\
"""~ ,
....,K ':.
," ,
.
I
,
"
o
0.5
MIDDLE DISTILLATE
HYDROGEN TREATING
DELAYED COKING
I
SULFUR RECOVERY,
TENTHOUSANDLBS/DAY
I
CATALYTIC CRACKING
. I I
HYDROGEN MANUFACTURE
MMSCF PER DAY
"
..\
1.0
2.0
2.5
FUEL OIL WT % SULFUR SPECIFICATION
1.5
CASE 7
CONSTANT PRODUCT
DEMANDS
40,300 BPSD GASOLINE
B,OOO BPSD LIGHT
OISTILLATE
16,900 BPSD HEAVY
DISTILLATE
25,300 BPSD RESIDUAL FUEL
BASE REFINERY
UNIT
CAPACITY
100,000 BPSa
CRUDE
FLUID CATALYTIC
CRACKER
HYDROCRACKER
MID DIST. HYDRO-
TREATER I!,OOO"
CATALYTIC RE-
FORMER
NAPHTHA HYDRO-
TREATER 23,000"
DELAYED COKER 16,000"
ALKYLATION 3,000"
SULFUR RECOVERY 170,000 LBSID
20000 "
7:000 "
23,000 "
BASE CASE CRUDE MIX
WITH
NO SULFUR RESTRICTION
HUNTINGTON BEACH
CALIFORNIA 90,680 epSD
KUWAIT ~.ooo BPSQ
THE KUWAIT CRUDE RATE WAS HELD
CONSTANT AT ALL SULFUR LEvELS
AND THE CALIFORNIA CRUDE RATE
WAS ALLOWED TO VARY.
30
-------
APPENDIX
-------
TECHNICAL ASPECTS OF THE LP MATRIX
This appendix presents the data used in the LP matrix. It is be-
lieved that no useful purpose would be accomplished by giving a
matrix tableau or a listing of the input data cards. The exact
arrangement of the matrix and the form of the input data depends
on the particular LP code that is being used, and undoubtedly
would be modified for any extension of this study. The data in
this appendix is sufficiently complete and detailed that the prob-
lem can be reconstructed with little effort for any LP code.
Even with modern high speed computers, there are definite economic
limitations on the size of the matrices to be solved. The time for
solution varies roughly as the cube of the number of equations, so
that there is considerable incentive for keeping the matrix small.
In this particular problem, a large number of parallel solutions
were desired, and the effort involved in creating a compact matrix
was justified. The following is a brief description of some of the
techniques used:
1. Stream combinations Many of the by-product streams from
various refinery units are produced in small amounts. Very
little is lost in simply combining these by-product streams
with others of similar nature. As an example, the by-
product naphthas from the heavy oil hydrodesulfurizer
units were all considered to be equivalent to virgin naphtha
from crude. This same combining technique was employed
for most of the by-product distillates and gas oils from the
hydrodesulfurizers which were considered equivalent to the
virgin material from crude. One consequence of this is
that an exact sulfur balance was not possible. As the sulfur
by-product was assigned no value, this discrepancy is of
little importance.
2. Non-restricting constraints In most LP problems many of
the balance equations do not represent real constraints from
a mathematical viewpoint. For example, no restriction was
placed on the availability of crude. Practically, its cost can
be merged into the cost for the crude unit operations. The
Bonner and Moore LP matrix generator has features which
facilitate this kind of operation. A large fraction of the
equations were made non-restricting in this way.
A-l
-------
3, Compositing When two or more units or operations are
related by a single feed-product connecting stream, this
stream can be eliminated as a constraint by creating a com-
posite operation for each permutation that exists. Here
again, the Bonner and Moore code facilitated this operation.
The code also includes facilities for reconstructing the re-
sults in terms of the original matrix
By applying these techniques, the reduced machine matrix was held
to 57 equations and 95 non-basic variables. Actual machine solution
time was about Z minutes for Cases 1 to 5 with approximately an
equal amount of time for the report writers. Cases 6 to 8 took
slightly longer.
Most of the data presented in this report was generated by the
technique of parametric programming. Starting with an optimum
solution, the computer program was instructed to explore the
effect of a change in sulfur specification until some basic change ,
occurred in the structure of the computer solution. Because 20
or more steps were involved in most cases, printed output was
taken only when the step involved a total change of 0. 1 wt. percent
sulfur or more. This accounts for the seemingly random selection
of sulfur specifications in the tabulations.
The results from the parametric programming steps are particu-
larly instructive in showing the relationships between factors of
interest. Part of the printed results from the machine is a tab-
ulation showing how each basic variable reacts to the change in
parametrized specification {the sulfur spec.) The rate of hydrogen
consumption change with each "barrel-percent" sulfur specification
change varied widely with each change in refinery structure. This
simply confirmed the observation that there is no basic hydrogen
consumption rate that can be related to a given degree of desulfuri-
zation of the fuel oil. Hydrogen is simply one of many economic
factors that are optimized for total minimum cost.
One particularly useful feature of the LP model was the exact
liquid volume balance. Debugging and error checking was greatly
facilitated by the fact that the crude run minus the total of the
products had to equal a net volume loss calculated by the com-
puter. The volume loss or gain associated with each operation
is given in the data listings that follow. These were determined
according to the following rules:
A-2
-------
1. The volume equivalent fuel oil (6. 2 million BTU per barrel)
for by-product fuel gas was counted as a product from each
operation. The net volume loss or gain is the difference
between feed and products for the operation including the
fuel gas. Coke was not counted.
2. A separate constraint was set up for refinery fuel. When
fuel gas is transferred to this constraint, a volume loss is
taken. Similarly, burning product fuel oil, butanes, etc.,
results in a volume loss.
3. Transferring excess hydrogen to the fuel gas constraint
resulted in a volume gain. In turn when this was trans-
ferred to refinery fuel, a loss was taken.
In the following tables are listed the various parameters used in the
development of the matrix including the following:
Capital Related Charges
Variable Operating Cost Factors
Blending Properties of Product Streams
Process Yields & Utility Demands for each Process
Utility Cost in terms of Fuel
Miscellaneous Refinery Transfers
A-3
-------
CAPITAL RELATED CHARGES USED IN LP MODEL
Gross
InvestInent Maintenance Insur.. Taxes TotaL Capital
Description of Base Total Charge Cost &: Misc. Cost Rela ted Char gr.:
Refinerv Units Capacity Investment ($/Bbl) ($/Bbl) ($/Bbl) -11@bl)
(Note 1) (Note Z) (Note 3) (Note 4) (Note 4)
Comb. Crude Unit 80,000 BPSD $4,700,000 0.0576 0.0069 0.0035 0.068
Propane Deasphalter ZO, 000 BPSD $4,000,000 0.1960 0.OZ64 0.0118 0.234
Reforming Pretreater 20,000 BPSD $2,200,000 0.1078 0.0146 0.0065 O. lZ9
Catalytic Reformer ZO,OOO BPSD $3,300,000 0.1618 0.OZ19 0.0097 O. 193
Delayed Coker ZO, 000 BPSD $5,000,000 0.2451 0.0332 0.0147 0.293
Resid. Hydrodesul£urizer - 10,000 BPSD $5,300,000 - 0.5196 0.1091 0.0312 0.661
Hydrocracker 8,500 BPSD $4,000,000 0.4614 0.0692 O.OZ77 0.558
Mid. Vi.l. HZ Treater ZO, 000 BPSD $Z,400,OOO 0.1176 0.0159 0.0071 0.141
FCCU Feed Pretreater ZO,OOO BPSD $3,251,000 0.1593 0.OZ16 (J.0096 0.191
Fluid Catalytic Cracker ZO, 000 BPSD (5) $5,800,000 0.Z843 0.0383 0.0170 0.340 (5)
Alkylation Unit 4,000 BPSD $2,600,000 0.6373 0.0860 0.0382 0.762
Sulfur Plant Z,OOO M Lbo/SD $2,700,000 0.0013/M Lb.. O. 0002/M Lbo. O. 0001/M Lb.. O. 002/M Lb..
Hydrogen Plant 3Z,OOO MSCF /SD $4,300,000 O.1309/MSCF 0.OZZ7/MSCF 0.0079/MSCF 0.162/MSCF
(I)
(Z)
(3)
(4)
(5)
Nominal stream day unit charge rates assumed for estimating capital investment.
Battery limits construction costs,
Based on 1020 operating days (3 years) pay-out period before taxes, interest and amortization.
Estimated a.nnua.l cost for these items divided by the total feed for 340 operating days.
Variable capacity factors were used in the LP model to account foI' difference in coke yield of different feedstock.s.
(M-l)
-------
Qpera.ti.Q.p
FCCU (600-9500 gas oil, treated or un-
treated)
FCCU (700-9500 gas oil, treated or un-
treated)
FCCU (Heavy coker gas oil blended,
treated or untreated)
FCCU (PDA gas oil blend, treated
or untr eated)
FCCU (Increase conversion from 620/0
10 64.5')',)
Hydrocracker (Virgin Stocks, Max. dist.
production)
Hydrocracker (Virgin Stocks, Max. gaso-
line Production)
Hydrocracker (FCCU LCO/coker LGO,
max. .,gasoline production)
Crude Unit (Atm. & Vac.)
Propane Deas phalte r
CataLytic Reformer, (5. R. Gasoline 98 RON
reformate)
Catalytic Reformer, (95 & 9Z RON
reforma.te)
Ca.talytic Reformer, (Hydrocrackate,
98 RON Reformate)
Coking, Delayed
Residual Oil Hydrodesulfurizer - ~
Hydrogen Plant (Values per MSCF Hydrogen)
Sulfur Plant (Values per lb. of elemental
Sulfur)
AlkyLation. HF
Reformer Pretreater
Mid-Distillate Hydrotrea.ter
FCCU Feed Pre treater
Cooling Water Facilities (Values per 1000
gal. Cooling Water)
Boiler .& Steam Plant (Values per 1000 Ibs.
steam)
Electric Power (Values per KWH)
Amine Plant (Values per lb. Sulfur extracted)
.(1)
(Z)
(3)
(4)
VARIABLE OPERATING COST FACTORS FOR LP MODEL
$JJ.L
0.1575
0.1575
0.1575
0.1575
4.0
4.0
4.0
Catalyst
lhUJl
0.Z05
0.Z05
0.Z05
0.Z05
O.OZZZ
0.0111
0.0111
WL
0.03Z3
0.03Z3
0.03Z3
0.on3
0.068
0.137
O. 146
0.0888
0.0444
0.0444
0.95 - 0.0083Z-0.079Z
0.00Z5
0.75
0.005
0.010
- 0.0038
0.010
O.OZ77(I)
0.002036(1)
.:I:1=-
BFW
BFW
BFW
BFW
BFW
BFW
CW
BFW
Water (3)
Qa1L!L
10.3
8.91
II. Z
IZ. I
0.Z89
I. ,85
0.0438(4)
O.IZ
UlL
CI.0072
0.0063
0.0078
0.0085
O.OOOZOZ
O.OOIZ
0.0044
0.084
Estimated total including an allowance for maintenance and incremental investment
Electrical cost estimated as $0.003/KWH + 9000 BTU/KWH fuel consumption
Boiler feedwater valued at $0. 70/Mgal. Cooling tower make-up valued at $0. lO/Mgal.
Cooling tower water make-up ratio, Mgal/Mgal.
Chemicals
&: Misc.
$.L.!L
0.050
0.050
0.050
0.005
0.010
0.010
0.040
0.00175
0.100
0.0038
(M-Z)
In.ta.L
$0.0395
$0.0386
$0.0401
$0.0408
$O.OOOZOZ
$0. [18
$0.187
$0.196
$0.005
$0.010
$0.0888
$0.0444
$0.0444
$0.010
$0.083Z
$0.0037
$0.00175
$0.100
$0.010
$0.0038
$0.010
$0.0359
$0.084
$0.003(Z)
$0.00Z036
-------
Blending
Component
Iso-Butane (Note 1)
N - Butane
Butylenes
Reformate 92 RON
Reformate 95 RON
Reformate 98 RON
Saturated CS-1750
Coker Light Gasoline
Light Hydrocrackate
400 EP Cat. Gasoline
330 EP S. R. Naphtha
C4 Alkylate
C3 Alkylate
Pol ymer Gas cline
Kuwai t Light Gasoline
Kuwait Reformate
(1)
BLENDING PROPERTIES OF PRODUCT STREAMS
Research Octane No. With
1 ce Pb lee Pb 3cc Pb
98. Z
98. Z
97.2
95.7
98.6
100.6
88.5
86. 1
89.2
95.6
78.6
100.6
97.6
96.6
84.6
94.8
100.0
100.0
99.0
97.7
99.9
101.8
91.9
89.4
93.3
97.3
82.5
102.5
99.5
98.1
88.6
96.6
Octanes assumed to be same as N-butane
to avoid preferential use of iso-butane
for gasoline blending.
101. 4
101.4
100.0
99.0
100.8
102.6
93.8
91. 3
95.6
98.4
84.8
103. ?
100.7
99.0
91.0
97.7
(M-3)
RVP
(PSIA)
80.0
59. I
60.0
5.0
5.0
6.0
5.4
10.5
12.0
7. I
3.0
5.0
5.0
1.0
12.0
4.00
Blending
Com.ponent
Viscosity
Blending
Number
330/400 - 540 Kerosine -100
540-650 Hvy. Vir. DIst. + 30
400 -650 Raw Coker Dis t. 0
600 -950 FCCU Feed +390
700 -950 FCCU Feed +4 75
650 -800 Hydrocracker Feed +335
800 -900 Gas Oil +520
Vacuum Pitch, 950+ + 1110
Propane Deasphalter Pitch + 1750
Resid. HDS Pitch + 1750
Light FCCU Cycle Oil -5
Hvy. FCCU Cycle Oil +310
540-680 H-Oil +210
Hydrocracked Distillate -100
FCCU Decant Oil +875
Desulf. FCeu Cycle Oil -5
Desulf. 400 -650 Coker Di,st. 0
Desulf. 540 -650 Hvy. Vir,. Dis t. + 30
Desulf. Kerosine -100
Light Cycle Oil, Trtd. Feed -5
Hvy. Cycle Oil, Trtd. Fel~d +310
Decant Oil, Trtd. Feed +875
Coker HGO, 650-950 +350
380-540 Kuwait Distill. -150
540 -650 Kuwait Distill. + 40
Kuwait FCC & Coker Cycle - 53
Kuwait Heavy Cycle Oil +334
Kuwait Low S. Fr. Distill. + 50
,Kuwait Desulf. Resid. +700
Kuwait Vacuum Resid. +870
Treated Kuwait Lt. Distill. -150
Treated Kuwait Hvy. Distill. + 40
Treated KW Coker &: Cycle - 53
Kuwait 650-1100 +380
Kuwait Coker Heavy G. O. +400
Imported Fuel Oil 569
Sulfur
(Wt. 0/0)
0.52
1. 22
1.25
1.47
1.51
I. 42
1. 54
2.03
2.57
I. 70
1. 15
1.30
0.50
o
1. 40
0.12
0.13
0.12
0.05
0.01
0.02
0.05
1.55
0.61
I. 75
2.40
3.20
o
0.75
5.40
0.061
0.175
0.24
3.00
4.00
4.50
o API
36.8
30.3
35.0
ZI.7
19.6
22.9
18.6
, 6.2
-5.1
-7.4
23.6
16.5
25.8
38.0
3.8
24.6
36.0
31.3
37.8
26.5
24.0
10.2
20.0
41.6
32.9
25.0
15.4
33. I
10.0
3.5
41.6
32.9
25.0
21.2
18.0
12.0
-------
FEED
CONSTR.
1.0
0.0
1.0
0.600
0.025
0.013
0.400
FEED
CONSTR.
1.0
0.0
1.0
0.600
0.025
0.013
0.400
.1
CRUDE DISTILLA TION
CRUDE OIL DISTILLATION - BASE
CASE FOR 330 EP GASOLINE A~D
HYDROCRAC~ER FEED GAS OIL
PRODucr
STREAMS
0.001
0.001
0.021
0.125
0.IY5
0.102
0.125
0..125
0.305
0.064
HUNTINGTON BEACH CRUDE
ISO-BUTANE'
NORMAL BUTANE
PENT ANES-I'15
115-330 NAPHTHA, SR
LT DISTILLATE 540 EP
HVY DISTILLATE 650 EP
HYOROCRACKER FEED 650-800
HEAVY GAS OIL BOO-950
VACUUM PITCH 950'
VOLUME GAIN
CAPACITY CONSTRAINT FACTOR
WATER, MGAL/B
SHAM, "LBS/B
FUEL, FDE/B
ELECTRICAL, KwH/B
GASDLiNEI OIST SWING
CRUDE DISTILLATION - 330 EP GASO
'AND 700-950 FCCU FEED
PRODUC T
STREAMS
0.001
0.001
0.021
0.125
0.IY5
0.142
0.210
0.305
0.064
HUNTINGTON BEACH CRUDE
ISDBUTANE
NORMAL BUTANE
~ENTANES-175
REFORMER FEEC 330 EP
LT DIsrlLLATE 540 EP
HVY DISTILLATE 700 EP
H[GH BP CAT FEED 700-950
VACUUM PITCH 950'
vOLUME GAIN
CAPACITY CONSTRAINT FACTOR
WATER MGAL/B
STEAM MlBSI B
FUEL FOE/B
ELECTRICAL KWH/B
GASOLINE/OIST SWING
\
CRUOE U[STILLATION - 330 E~
GASOLIrlE AND FULL RINGE FCCU FEEO
FEED
CONSTR.
1.0
0.0
1.0
0.600
0.025
0.013
0.400
PRUDUCT
STREAMS
0.001
0.001
0.021
0.[25
0.195
0.0,7
0.295
0.305
0.064
HUNTINGTON BEACH CRUOE
ISOBUTANE
NORMAL "UTANE
PENTANES-175
175-330 ~APIITHA, SR
LT OISTILLATE 540 E~P
HVY OIST[LLATE 600 EP
FULL ~ANGE CAT FEED 600-950
VACUU" PITCH 950'
VOLUME GA [N
CA~ACITY CONSTRAINT FACTOR
wA TE R "GALIB
STEAM "LBS/B
FUEL FOEI B
ELECT,[CAL KWH/B
GASOLINE/DIST SWING
SWING O[STILLATE Td GASOLINE
FEED
CONSTR.
0.064
O. [25
0.0
0.064
NOTE:
PROUUCT
STREAMS
0.IB9
540 EP
L[GHT DISTILLATE
175-330 NAPHTHA, SR
175-400 NAPHTHA, SR
VOLUME GA[N
LIMIT DIST/GASO SwiNG
COLUMN LABELED "FEED CONSTRAINTS"
REPRESENTS POSITIVE MATRIX COEFFICIENTS
COLUMN LABELED "PRODUCT STREAMS"
REPRESENTS NEGA TIVE MA TRIX COEFFICIENTS.
(M-4)
-------
VACUUM RESIDUE OPERATIONS
RESIDUAL OIL HYDRO-OESULFURIZATION
FEED
CONSTR.
1.0
1.000
0.098
1.0
0.0312
0.965
8.4
0.0218
. PRODUCT
STREAMS
0.131
0.130
. 0.369
0.200
0.058
4.000
0.01
0.07
0.13
VACUUM PITCH 950+
HYDROGENtMSCF/8
400-540 CUT
540-680 CUT
680-975 FCCU CUT
/-I-DIL RESID
CI-C3,FOE
SULFUR AS H2S,L8S/8BL
N-BUTANE
C5-175 GASOLINE
175-400 GASOLINE
VOLUME GAIN
CAPACITY CONSTRAINT FACTOR
STEAM, MLBS/B
COOLING WATER, MGAL/8
ELECTRICITY, KWH/B
FUEL FOE B/B
PROPANE OE-ASPHALTING
FEED
CONSTR.
1.0
1.0
0.050
0.0532
1.500
PRODUCT
STR::AMS
0.528
0.472
0.0
VACUUM PITCH 950+
DECARBONIZED OIL
PROPANE DEASPHALTEO
VOLUME LOSS
CAPACITY CONSTRAINT
STEAM MLBS/B
PRDCESS FUEL, FOE/B
COOLING WATER MGAL/B
PITCH
FACTOR
(M-5)
FEED
CONSTR.
1.0
1.0
0.038
1.440
0.0329
1.300
DELAYED COKING
PRODUCT
STREAMS
3.13
0.1081
0.0124
0.0055
0.0163
0.0162
0.0711
0.1349
0.3636
.q.1414
0.1107
0.1245
VACUUM PITCH 950+
SULFUR IN GAStLBS/BBL,
FUEL GAS (FOE)
PROPYlENE
ISO-BUTANE
BUTYl ENE
NORMAL BUTANE
COKER LT GASOLINE
CDKER HVY GASOLINE 200-400
COKER LGO 400-650
COKER HGO 650-950
OELAYEO COKE PRODUCTtMLBS/B
VOLUME LOSS.
CAPACITY CONSTRAINT FACTOK
STEAM, MLBS/8
ELECTR1CITYt KWH/B
PROCESS FUEL, FOE/B
COOLING WATER, MGAL/B
-------
REFORMER FEED PRETREATING
REFORMER FEED PRETREATING. ViRGIN
FEED
CONSTR.
1.0
0.050
1.0
0.013
0.75
PRODUCT
STREAMS
0.995
0.,4
0.005
330 EP VIRGIN NAPHTHA RAw
HYOROGEN, MSCF/B .
330 EP VIRGIN NAPHTHA TRTD
H2S, SULFUR LBS/BBL
VOLUME LOSS
CAPACITY CON~TRAINT FACTOR
PROCESS FUEL, FOE/B
ELECTRICITY, KWH/B
REFORMER FEED PRETREATING
400 EP NAPHTHA
FEED
CONSTR.
1.0
0.050
1.0
0.013
0.75
PRODUCT
STREAMS
0.9'15
0.39
0.005
400 EP VIRGIN NAPHTHA RAW
HYDROGEN, MSCF/B
400 EP VIRGIN NAPHTHA TRIO
H2S. SULFUR LBS/B
VOLUME LOSS
CAPACITY CON~TRAINT FACTOR
FUEL OIL FOE 8/8
ELECTRICITY KWH/B
REFORMER FEED PRETREATING, COKER
HEAVY NAPHTHA
FEED
CUNSTR.
1.0
0.120
1.0
0.013
0.75
PRODUCT
STREAMS
0.995
O.6~
0.005
CUKER HEAVY GASOLINE
HYDROGEN. MSCF/8
PRETREATED CRACKED NAPHTHA
H2S, SULFUR LBS/BRL
VOLUME LOSS
CAPACITY CONSTRAINT FACTOR
FUEL OIL B8L/RBL
ELECTRICITY, KWH/R
CA TAL YTIC REFORMlNG
REFORM HEAVY HYOROCRACKER GASOLINE
TO 98 RON
FEED
CONSIR.
1.0
1.0
0.0647
0.0010
3.8
0.745
FEED
CONSTR.
1.0
1.0
0.0647
0.0010
3.B
0.145
(M-6)
PRODUCT
STRt:AMS
0.7890
0.0 :!24
0.0132
0.0212
0.9110
0.0222
REFORMER
PROuuCT
STREAMS
0.6500
0.0123
0.0277
0.0443
0.8260
0.0297
HEAVY HYOROCKACKATE
HYDROGEN MSCF/B
FUE
I-C4
N-C4
REFORMA IE, ~8
VOLUM,E LOSS
CAPACITY CONSTRAINT
FUEL FOE
STEAM MLBS/8
ELEC IR KWH/8
COOLING wATER MGAL/8
COKER NAPHIHA
92 ON
COKER HVY GASOLINE
HYDROGEN MSCF/B8L
FOE
ISO-BUTANE
N-8UTANE
REFORMAIE, 92
VOLUME LOSS
CAPACITY CONSIRAINT
FUEL, FOE
SIEAM ML8S/B
ELECTRICITY KWH/B
COOLING wATER HGAL/8
FACTOR
FACTOR
-------
FEED
CONSTR.
1.0
1.0
0.0647
0.0010
3.8
0.745
FEED
CONS TR.
1.0
1.0
0.0647
0.0010
3.8
0.745
REFORMING
C LE AR
PRODUCT
STREAMS
0.925
0.0202
0.0086
0.0136
0.9260
0.0316
REFORM ING
REFORMING
TO 95 RON
PRODUCT
STREAMS
0.9750
0.0364
0.0146
0.0232
0.8820
0.0438
CA TAL YTlC REFORMING
330 EP NAPHTHA TO 92 RON
175-330 NAPHTHA, SR
HYOROGEN,MSCF/B
FUEL GAS, FOE
I SO BUTANE
NORMAL BUTANE
REFORMATE, 92RON
VOLUME LOSS
CAPACITY CONSTRAINT FACTOR
FUEL, FOE
STEAM, MLBS/B
ELECTRICITY, KWH/B
COOLING WATER, MGAL/B
AT 95 RON CLEAR, 330
330 EP VIRGIN NAPHTHA
CLEAR
175-330 NAPHTHA, SR
HYDROGEN, MSCF/B
FUEL GAS, FOE
ISO-BUTANE
NORMAL BUTANE
REFORMATE, 95 RUN
VOLUME LOSS
CAPACITY CONSTRAINT
FUEL FOE
STEAM MLBS/B
ELEC TR. KWH/B
COOLING WATER MGAL/B
FACTOR
REFORMING 330 EP VIRGIN NAPHTHA
TO 98 RON CLEAR
FEED
CONSTR.
1.0
1.0
0.0641
0.0010
3.8
0.745
(M-7)
PRODUCT
STREAMS
1.0000
0.0394
0.0159
0.0255
0.B140
0.0452
175-330 NAPHTHA SR
HYDROGEN MSCF/B
FUEL GAS,fOE
ISO-BUTANE
NORMAL BUTANE
REFORMATE, 98 OCTANE
VOLUME LOSS
CAPACITY CONSTRAINT FACTOR
IFUEL,FOE
STEAM, MLBS/B
HEC TR. KWH/B
COOLING WATER, MGAL/B
-------
CATALYTIC REFORMING
REFORMING 400 EP VIRGIN NAPHTHA
TO 92 RON CLEAR
FEED
CONSTR.
1.0
1.0
0.0647
0.0010
3.8
0.745
PRODUCT
STREAMS
0.915
0.0285
0.0120
0.0188
0.9120
0.0287
175-400 NAPHTHA, SR
HYDROGEN,MSCF/B
FUEL GAS,FOE
ISO BUTANE
NORMAL BUTANE
REFORMATE, 92RON
VOLUME LOSS
CAPACITY CONSTRAINT FACTOR
FUEL, FOE.
STEAM, MLBS/B
ELECTRICITY, KWH/B
COOLING WATER, MLBS/B
REFORMING 400 EP VIRGIN NAPHTHA
TO 95 RON CLEAR
FEED
CONSTR.
1.0
1.0
0.064.7
0.0010
3.8
0.745
PROUUCT
STREAMS
0.9680
0.0296
0.0120
0.0193
0.9065
0.0326
175-400 NAPHTHA, SR
HYDROGEN MSCF/B
FOE
I-8UTANE
N-BUTANE
REFORMATE, 95 RON
VOLUME lOSS
CAPACITY CONSTRAINT
FUEL,FOE
STEAM MLBS/B
ELEC TR. .KWH/8
COOLING WATER MGAL/B
FACTOR
FEED
CONSTR.
1.0
1.0
0.0647
0.0010
3.8
0.145
(M-8)
REFORMING 400 EP VIRGIN NAPHTHA TO
98 RON CLEAR
PRODUCT
STREAMS
0.9900
.0.0319
0.0130
0.0208
0.8990
0.0353
175-400 NAPHTHA
HYDROGEN, MSCF/B
FOE
I-C4
N- BU TANE.
REFORMATE 98 ON
VOLUME lOSS
CAPACITY CONSTRAINT FACTOR
fUEL, FOE
STEAM MlBS/B
ELECTRICITY KWH/B
COOLING WATER MGAl/B
-------
MIDDLE DISTILLATE HYDROGEN TREATING
MIDDLE DIST HYD~DGEN TREATER,
LIGHT VIRGIN DISTILLATES
FEED
CONSTR.
1.0
0.100
0.012
1.0
0.017
0.33
1.9
0.0183
PRODUCT
STRtAMS
0.013
0.91ib
0.002
0.0110
MIDDLE DISTILLATE 350-560V
HYDRDGEN,MSCF/O,
C5+ GASOLINE
MID DIST HTR LVGO PRODUCT
N-BUTANE
FUEL GAS FOE
VOLUME GAIN
CAPACITY CONSTRAINT FACTOR
STEAM,MLBS/B
COOLING WATER MGAL/BBL
ELECTRICITY KwH/B
PROCESS FUEL, FOE/B
MIODLt DISTILLATE HYDROGEN TREATER
HEAVY VIRGIN DISTILLATES
FEED
CONSTR.
1.0
0.200
0.0164
1.0
0.017
0.33
1.9
0.0183
PRODUCT
STREAMS
0.02B
0.970
0.005
0.0134
MIO DIST 560-650 V
HYDROGEN,MSCF/B
C5+ GASOLINE
MID DIST HTR HVGO PRODUCT
N-BUTANE
FUEL GAS FOE
VOLUME GAIN
CAPACITY CONSTRAINT FACTOR
STEAM, MLBS/B 170PSIG
CODLING WATE~ MGAL/BBL
ELECTRICITY KWH/B
PROCESS FUEL, FUE/B
FEED
CONSTR.
1.0
0.300
0.0254
1.0
0.017
0.33
1.9
0.01B3
NOTE:
(M-9)
MIDDLE DISTILLATE HYDRUGEN TREATER
COKER DISTILLATE AND FCCU LIGHT
CYCLE OIL
PRODUCT
STREAMS
0.041
0.961
0.004
0.0134
MIDDLE DISTILLATE 4DO-650LC
HYDROGEN,MSCF/B
CS+ GASOLINE
MIO DIST HTR LCO PRODUCT
N-.OU TANE
FUEL GAS FOE
VOLUME GAIN
CAPACITY CONSTRAINT FACTO~
STEAM, MLBS/B 170PSIG
COOLING wATER MGAL/BOL
ELECTRICITY KWH/B
PROCESS FUEL, FOE/a
THESE COEFFICIENTS WERE USED BY THE
MA TRIX GENERA TOR TO CREA TE A SEPARA
"DESULFURIZER" FOR EACH STREAM TO BE
TREA TED. HZS YIELDS FROM THESE OPERf
TrONS WERE ENTERED SEPARATELY.
-------
HYDROCRACKlNG
HYDRDCRACKER - CONVERT VIRGIN LIGHT
DISTILLATE TO GASOLINE
HYDRDCRACKER - CONVERT 650-800
VIRGIN GAS OIL TO GASOLINE
FEED PRODUCT FEED PRoDUC T
CoNSTR. STReAMS CONS TR. STREAMS
1.0 540-650 VIRGIN DIST 1.0 HY DR OCR A CK E R FEED 650-800
1.9 H2 MSCF/B 2.05 H:~ MSCF/B
3.14 H2S SULFUR L8S/B 4.5 H;~S t SULFUR LBS/BBL
0.034 FUEL GAS FOE 0.04B FUEL GAS rOE
0.144 I SO BUTANE 0.152 ISO-BUTANE
0.060 N-BUT ANE 0.063 N--BU TANE
0.364 LT HYDRDCRACKATE C5-1 BO 0.3B2 LT HYDRDCRACKATE C 5- 180
0.666 HY HYDRoCRACKATE .400EP 0.6n HY HyoRoCRACKATE 400EP
D.26B VOLUME GAIN 0.337 VOLUME GAIN
1.0 CAPACITY CONSTRAINT FACTOR 1.0 CAPACITY CONSTRAINT FACTOR
0.02B3 STEAM ML BS/B 1.IB COOLING WATER MGALIB
1.IB CODLING WATER MGAL/B 0.02B3 srEAM MLBS/B
10.0 ELECTRICITY KWH/B 12.6 ELECTRICITY KWH/B
0.098 FUEL FOE BIB 0.098 FUEL FOE BIB
HYDRoCRACKER 651)-800 V I RG IN GAS
OIL FEED FOR MAXIMUM LIGHT
DISTILLATE PRODUCTION
FEED
CONS TR.
1.0
1.65
PRODUCT
S TP EAI'S
4.5
0.030
0.016
0.025
0.061
0.209
0.815
(M-I0)
0.156
1.0
0.S49
0.02S3
8.95
0.0415
HYORoCPACKER FEED 650-S00
H2 MSCF/BBL
H25. SULFUR LBS/B
FUEL GAS FOE
I SO-BUTANE
N-IIUTANE
LT HYDRoCRACKATE C5-180
HY HYDRoCRACKATE 400EP
HYORoCRACKED DISTILLATE
VOl.UME GAIN
CAPACITY CONSTRAINT FACTOR
COOLING WATER MGAL/S
STEAM MLBS/B
ELECTRICITY KWH/S
FUEL FOE BIB
-------
HYDROCRACKING
HYDROCRACKER - LIGHT COKER GAS
OIL F~ED FOR MAXIMUM GASOLINE
HYDROCRACKER - CONVE~T CATALYTIC
LIGHT CYCLE OIL TO GASOLINE
FEED PRODUCT FEED PROUUCT
CONSTR. STREAMS CONSTR. STREAMS
1.0 COKER LGO 400-650 1.0 LT CYCLE OIL
3.1 H2 MSCF/B 3.3 112, MSCF/B
4.2 H2S SULFUR LBS/B 3.9 H2S, SULFUR LBS/BflL
0.042 FUEL GAS FOE 0.042 FUEL GA,>,FOE
0.117 ISO-BUTANE 0.117 ISO-BUTANE
0.052 N-BUTANE o.on N-BUTANE
0.317 ~T HYDROCRACKATE C 5- 1 80 0.317 LT HYDROCRACKATE C 5-180
0.780 HY HYOROCRACKATE 400EP 0.780 HY HYDROCRACKATE 400tP
-0.308 VOLUME GAIN 0.3011 VULUME GAIN
1.0 CAPACITY CONSTRAINT FACTOR 1.0 CAPACITY CON~TRAINT FACTO~
1.180 COOLING WATER MGAL/B 0.0283 STEAM MLSS/B
0.0283 STEAM MLBS/B 1.180 CUOLING WATER MGflLlB
14.0 ELEC TR IC!TY KWH/B 14.0 ELEC TRIC I TY KWH/S
0.098 FUEL FOE B/B 0.098 FUEL FOE B/B
(M-II)
-------
FEED
CONSTR.
L.O
0.02B12
0.018L
L.O
0.96200
0.3LO
0.00292
FCCU
PROUUC T
STREAMS
2.140
0.0836
0.OB35
0.04BO
0.0100
0.0040
0.3B9
0.32~
0.03B
0.020
0.1~82C
FLUID CATALYTIC CRACKING UNIT - UNTREATED FEED
600-950 GAS.O[L 62 CONV.
600- 950 FCCU FEEO
FCCU REGEN COKE MLBS/BBL
H2S, L8S 5ULFURI BBL.
FUEL GAS. FOE
PROPYLENE
I SO-BUTANE
N- BUHNC
8UTYlENE .
400 EP GASOLINE
LT CYCLE OIL 400/650
hVY.CYCLE OIL
DECANT OIL
VOLUME GALN
CAPACITY CONSTRAINT FACTOR
~TEAM PROOUCTION, MLBS/B
COOLING WATER, MGAL/B
ELECTRIC[TY, KWH/B
FUEL FOE, BIB
CATALYTIC CRACKING OF PROPANE
OE-A$PHALTEO GAS OIL IN BLEND WITH
VIRGIN GAS OIL
FEEO
CONSTR.
0.353L
0.646'1
0.0330
0.06 LL
L.14903
1.0460
0.392
0.00355
PROOUCT
STREAMS
2.46
0.OBB5
0.OB06
0.0340
0.0090
O..Otl4Q
0.3B50
0.3450
0.0150
0.0200
0.12L70
POA GAS OIL
600-950 FCCu FEEO
FCCU REGEN CUKE, MLBS/B
H2S, LBS SULFUR/B
FUEL GAS, FOE
PROPYLENE.
ISO-BUTANE
N-BUTANe
BUTYl ENE
400 EP GASOL[NE
LT CYCLE OIL
HVY CYCLE 01 L
DECANT OIL
VOLUME GAIN
CAPACITY CONSTRA[NT FACTOR
STEAM PRODUCTION, MLBS/B
COOLING WATE~, MGAl/B
ELECTRICITY, KWH/B
FUEL FOE BIB
FEED
CONSTR.
L.O
0.03L2L
0.0699
L.OB610
0.96000
0.340
O.0031B
FCCU
100-950 GA $ 0 I L
&2
CONV.
PROOUC T
STREAMS
2.41
0.OB22
0.OB01
0.0400
O.OLOO
O.OBLO
0.39&0
0.211
0.OB3
0.020
0.09420
100-9$0 Fceu FEED
FCCU REGEN COKE MLBS/BBL
H2S, LBS SULFUR/BBl
FUEL GAS ,FIJE
PROPYLENE
I SO-BUTANE
N -BUTANE
BUTYL ENE
400 EP GASOL[NE
LCD
HCO
DECANT OIL
VULUME GAIN
CAPACITY CONSTRAINT FACTOR
STEAM PROOUCTION. MLBS/B
COOLING WATER, MGAL/B
ELECTRICITY, KWH/B
FUEL FOE BIB
CATALYT[CALLY CRACK HEAV. COKER
GA$ OIL [N BLENU WITH VIRG[N GAS
OIL
FEEO
CONSTR.
O. L212
0.B12B
0.029OB
0.0722
1. 03343
0.96100
0.322
0.00296
PROOUCT
STREAMS
2.44
0.OB26
0.OB36
0.0450
O.OLOO
0.OB30
0.3BBO
0.3220
0.03BO
0.G200
0.L1320
COKER HOO 650 - 950
600 - 950 FCCU FEED
FCCU REGEN COKE. MLSS/B
H2$, LBS $ULFUR/B
FUEL GAS. FOE
PROPYLENE
[SO- BUTANE
N-BUTANE
BUTYLENE
400 EP GASOL[NE
LT CYCLE OIL
HVY CYCLE OIL
OECANT OIL
VOLUME GAIN
CAPACITY CUNSTRA[NT FACTOR
STEAM PRODUCTION, MLBS/B
COOLING WATER, MGAllB
ELECTR[CITY, KWH/R
FUEL FOE OIB
(M-12)
-------
HYDROGEN TREA TING OF FCCU FEED
HYDROGEN TREAT FCCU FEED, 600-950
VIRGIN GAS \IlL
FEED
CONSTR.
1.0
0.675
0.0746
1.0
0.017
0.50
3.0
, 0.0183
PRODUCT
STREAMS
4.5
0.0286
0.026
0.0~7
0.123
0.800
600-950 VIRGIN GAS OIL
HYDROGEN,MSCF/B
H2S AS SULFUR,LBS/B
FUEL GAS FOE
N-BUTANE
GASOLINE, 48 API
DIESEL, 30 API-OESULF
FCCU FEED. 600-950 TREATED
VOLUME GA IN
CAPACITY CONSTRAINT FACTOR
STEAM, MLBS/B
COOLING WATER, MGAL/B
ELECTRICITY, KWH/S
FUEL, FOE
HYDROGEN TREAT PDA GAS OIL FCCU
FEED
FEEO
CONS TR.
1.0
0.675
0.0746
1.0
0.017
0.50
3.0
0.0183
PRODUCT
STREAMS
4.5
0.0286
0.026
0.091
0.123
O.BDO
PDA GAS OIL
HYDROGEN, MSCF/B
H2S AS SULFUR, LBS/B
FUEL GAS, FOE
N-BUTANE
GASOLINE, 48 API
DIESEL, 30 API
FCCU FEED PDA GAS OIL TRTD
VOLUME GAIN
CAPACITY CONSTRAINT FACTOR
STM MLBS/B
CODLING WATER. MGAL/B
ELECTRICITY, KWH/B
FUEL. FOE
FEEO
CONSTR.
1.0
0.675
0.0679
1.0
0.017
0.50
3.0
0.0183
HYDROGEN TREAT
OIL FCCU FE EO
PRODUCT
STREAMS
4.7
0.0252
0.0227
0.OU7
0.363
0.650
700-950 VIRGIN GAS
700-950 VIRGIN GAS OIL
HYDROGEN, MSCF/B
H2S AS SULFUR,LBS/B
FUEL GAS,FOE
N-SUTANE .
GASOLINE, 4B API
DIESEL, 30 API
FCCU FEED 700-950 TREATED
VOLUME GAIN
CAPACITY CONSTRAINT FACTOR
STM, MLBS/B
COOLING WATER, MGAL/B,
ELECTRICITY, KWH/B
FUEL, !FOE
HYDROGEN TREAT HEAVY COKER GAS OIL
FCCU FEED
'FEED
CONS TR.
1.0
0.625
0.0734
1.0
0.017
0.50
3.0
0.01B3
PRODUCT
STREAMS
4.8
0.0332
0.0302
0.044
0.21>1>
0.700
650-950 COKER GAS OIL
HYDROGEN, MSCF/B
H2S AS SULFUR,LBS/B
FUEL GI.S, FOE
N-BUTANE
GASOLINE, 4B API
OIESEL, 30 API
FCCU FEED TREATED COKER G 0
VOLUME GAIN
CAPACITY CONSTRAINT FACTOR
STEAM ~.LBS/B
CODLING WATER, MGAL/B
ELECTRICITY, KWH/B
FUEL, FUE
(M-13)
-------
FLUID CATALYTIC CRACKING UNIT -.PRETREATED FEED
CATALYTIC CRACKING OF PRE-TREATED
600-950 GAS Oil
FEED
CoNSTR.
1.0
0.0188T
0.0936
0.65703
0.650
0.195
0.00193
PRooUCI
STREAMS
0.146
0.0793
0.0953
0.0670
0.0130
0.0830
0.3760
0.2380
0.1220
0.0200
0.0566
CATALYTIC CRACKING OF PRE-TREATEO
HEAVY COKER GAS Oil IN 8lENo WITH
600-950 VIRGIN GAS 011.
FCCU FEED. 600-950 TREATED
FCCu REGEN CoKE.MlBS/B.
HZS, LBS SULFUR/B
FUEL GAS FOE 8/B
PROPYLENe.
I SO-8UTANE
N-8UTANt:
8UTYlENES
400 EP GASoli NE
IT CYCLE OIL-DESULFUKllED
HVY CYCLE Oll-DESUlFURllEO
DECANT Oil - oESUlFURIZEo
VOLUME GAIN
CAPACITY CONSTRAINT FACTOR
STEAM PRODUCTION, MlBS/B
COOLING WAT~R, HGAt/S
ELECTRICITY, KWH/B
FUEL FOE BIB
FEEO
CONSIR.
0.1272
0.8728
0.01983
0.0877 .
0.69046
0.655
0.207
0.00197
CATALYTIC CRACKING OF PRE-TREATED
BLEND OF PoA GAS OIL AND 600-950
VIRGIN GAS OIL.
FEED
CONSTR.
0.3531
0.6469
0.02315
0.0766
0.B0606
0..734
0.217
0.00256
PRODUCT
STREAMS
0.466
0.0842
0.0924
0.053
0.012
0.OB3
0.372
0.261
0.099
0.02
0.0701
PRODUCT
STREAMS
0.44b
0.0783
0.0954
0.064
0.013
0.082
0.315
0.238
0.122
0.02
.0.0616
FCCU FEED. IREAIEo COKER GO
FCCU FEED. 600-950 TREATED
FCCU REGEN CUKE, MLBS/B
HZS, LBS, SULFUR/S
FUEL GAS FOE BIB
PROPYLENE
1 SO- BUTANE
N-BUTANE
8UTYlENE5
400 EP GASOLINE
IT CYCLE OIL-oESULFURIZEo
HVY CYCLE 0IL-OESUlFuR1ZEo
OECANT OIL
VOLUME (i,AlN
CAPACITY CONSTRAINT FACTOR
STEAM PRODUCTION, MLBS/S
COOLING WATER. MGAL/S
ELECTRICITY. KWH/B
FUEL FOE BIB
CATALYTIC CRACKING OF PRE-TREATEO
700-950 VIRGIN GAS OIL
FCCU FEED,TRTO POA CAS OIL
FCCU FEEO,600-950 TREATEC
FCCu REGEN COKE, MLBS/B
H2S. LBS SULFUR/B
FUEL GAS FOE BIB
PROPYLENE
ISO-BUTANE
N-BUTANE
BUTYLENES
400 EP GASOLINE
. IT CYCLE OIL-DESULFuRIZED
HVY CYCLE OIL-DESULFURIZEO
DECANT OIL - DESULFURIZED
VOLUME GArN
CAPACITY CONSTRAINT FACTOR
STEAM- PRODuCTION - ML8S/B
COOLING WATER, MGAL/B
ELECTRICITY, KWH!B
FUEL FOE BIB
FEEO
CONSIR.
1.0
0.02136
0.0854
0.74373
0.648
0.225
..0.00219
PRODUCT
STREAMS
0.476
0.0779
0.0925
0.OS90
0.013G
o.OBOO
0.3830
0.1930
0.1670
0.0200
0.0426
FCCU FEED, 700-950 TREATED
FCCU REGEN COKE, MLBS/B
H2S L8S SULFUR/B
FUEL GAS FUE BIB
PROPYLENE
ISO-BUTANE
N-BUTANf
BUTVlENES
400 EP GASOLINE
LT CYCLE OIL-oESULFURIZEo
HVY CYCLE 0Il-oE5ULFURIZEO
DECANT OIL - DESUlFURIZED
VOLUME GAIN
CAPACITY CONSTRAINT FACTOR
STEAM PRODUCTION, MLBS/O
COOLING WATER, MGAL/B
ELECTRICITY, KWH/B
FUEL FOE 8/8
(M-14)
-------
FEED
CONSTR.
0.0510
1.53
0.110
I. 77577
0.53
0.644
0.0052
FEED
CONS T~.
0.0510
1.53
0.110
1.77577
0.53
0.644
0.0052
INCREASE FCCU CONVERSION
INCREASE FCCU CONVERSION FGR
PRETREATED STOCKS
VALUES ARE SCALED BY 100 FGR
CONVEN I ENCE
PROGUCT
STREAMS
0.159
0.156
0.25
0.547
0.528
0.264
, COKE
FUEL GAS FOE
PROPYlENE
BUTYl ENE S
400 EP GASOLINE
LCO FROM DESULFURIZED FEED
HCO FROM DESULFURIZED FEED
VOLUME GAIN
CAPACITY CONSTRAINT FACTOR
STEAM PROOUCTION, MLBS/B
COOLING WATER, MGALS/B
ELECTRICITY, KWH/B
FUEL CUNSUMEO, FOE/B
INCREASE FCCU CONVERSION
SCALEO BY 100 FOR CONVENIENCE
PRODUC T
STREAMS
0.159
0.156
0.2':'
0.547
0.5l8
b.2M
COKE
FUEL GAS, FOE
PROPYLENE
BUTYlENE
400 EP GASOLINE
LT CYCLE OIL
HEAVYCYCLE OIL
VOLUME GAIN
CAPACITY CONSTRAINT
STEAM PROOUCTION
COOLING WATER
ELECTRICITY, KWH/B
FUEL, FOE
FACTOR
(M-IS)
ALKYLATION & POLYMERIZATION
BUTYLENE ALKYLATION
FEED
CONS TR.
0.585
0.690
1.0
3.75
0.17
0.011
3.68
PRODUCT
STREAMS
1.00
0.215
FCCU BUTYLENES
ISO-BUTANE
MOTOR ALKYlA TE
VOLUME LOSS
CAPACITY CONSTRAINT FACTOR
ELECTRICITY, KWH/8
PROCESS FUEL, FOE/B
STE~M, MSCF/B
COOLING WATER, MGAL/S
pROpYLCNE ALKYLATION UNIT
FEEO
CONSTR.
0.706
0.723
1.0
3.75
0.17
0.011
3.68
pROOUCT
STREAMS
1.0
0.OB7
0.342
PROPYlENE
I SO--BUTANE
MOTOR ALKYlA TE
PROPYLENE TD FUEL FOE B/R
VOLUME LOSS
CAPACITY CONSTRAINT FACTOR
ELECTRICITY, KWH/B
PROCESS FUEL, FDE BIB
STHM, MLBS/B
COOLING WATER, MOAL/B
CATALYTIC POLYMERIZATION
FEEO
CONSTR.
1.2
1.0
1.7
0.200
0.630
PROOUC T
STREAMS
1.00
0.20
PROP YlENE
POLY GASOL! NE
VOLUME LOSS
CAPACITY CONSTRAINT FACTOR
ELECTRICITY, KWH/B
STEAM, MLBS/B
COOLING WATER, MGAL/B
-------
HYDROGEN & SULFUR PLANT
FEED
CONSTR.
0.07870
1.0
0.028
3.34
HYDROGEN PLANT
PRODUCT
STREAMS
1.0
0.0187
0.00432
FUEL GAS CONSUMED, FOE
HYDROGEN, MSCF
VOLUME LOSS
CAPACITY CONSTRAINT FACTOR
STEAM, ML8S/MSCF
CODLING WATER,MGAL/MSCF
ELECTRICITY,KWHI MSCF
AMINE PLANT
H2S REMOVAL BY MEA A8S0RPTION
FEED
CONSTR.
1.0.
0.005
0.015
0.016
FEED
CONS TR.
1.0
1.0
0.050
PROOUCT
STREAMS
1.0
SULFUR PLANT
PRODUCT
STREAMS
1.0
0.00336
H2S, SULFUR LBS
REGENERATED H2S, LBS S/LB
STEAM MLBS/LB
ELECTRICITY KWH/LB
COOLING WATER MGAL/LB
H2S,REGEN. AS SULFUR,LBS
SULFUR, LBS
CAPACITY CONSTRAINT FACTOR
STEAM PRDOUCED, MLBS/LB
ELECTRICITY KWH/B
UTILITIES
COOLING WATER UTILITY
MGAL
FEED
CONSTR.
0.820
PRODUCT
STREAMS
1.0
COOLING WATER, MGALS
ELECTRICITY, KWH/MGAL
ELECTRIC POWER GENERATION
FEED
CONSTR.
0.00145
PROfJUC T
STREAMS
10.0
ELECTRICITY, KWH
FUEL OIL, I"OE
GENERATE STEAM FROM SALEABLE FUEL
FEED
CONS TR.
1.00
(M-16)
PROOUCT
STR~AMS
4.4
FUEL, FOE
PROCESS STEA~, M LBS
-------
MISCELLANEOUS REFINERY TRANSFERS
BURN RFFINERY FUEL GAS
FEED
CONSTR.
1.0
PRODUCT
STREAMS
1.0
1.0
FUEL GAS
PROCESS FUEL
VOLUME LOSS
feED
CONSTR.
1.0
BURN PRODUCT FUEL IN REFINERY
FEED
CONSTR.
PROUUCT
STREAMS
1.0
1.0
1.0
HEAVY FUEL OIL
PROCESS FUEL.
VOLUME LOSS
FEED
CONSTR.
1.0
PRODUCT
FLARE F.XCESS REFINERY FUEL
FEED.
CONSTR.
1.0
PRODUCT
STREAMS
1.0
REFINERY FUEL FOE
EXCESS REFINERY FUEL FLARED
FEED
CONSTR.
1.0
BURN EXCESS HYDROGEN
FEED
CONSTR.
1.0
0.053
PRODUCT
STREAMS
0.053
.HYDROGEN
FUEL GAS. FOE
VOLUME GAIN
FEED
COr-.S TR.
ECUIVALENT
1.0
BURN PROPYLENE
. PRODUC T
STREAMS
0.615
1.0
BURN
BUTANES
PRODUCT
STREAMS
0.102
1.(;
BURN I~O-BUTANE
PRODUCT
STREAMS
0.102
1.0
BUY I SO-BUTANE
PRODUCT
STREAMS
1.0
PROPYLENE
REFINERY FUEL
VOLUME LOSS
NORMAL BUTANE
REFINERY FUEL
VOLUME LOSS
BURN ISO-BUTANE
I SO-BUTANE FOE
VOLUME LOSS
BUY ISO-BUTANE
VOLUME GAIN
(M-17)
-------
SUPPLEMENTARY DATA FOR KUWAIT OPERATIONS
HYCROCESULFuRIZATION OF KUwAIT
RESIDUU~ - SINGLE PASS OPE~ATION
FEED
CONSHI..
1.0
OelH6
".0
"'..~SltO
n...0312
0..9650
8.~OOO
C..0216
PROCUCT
STREAMS
0.028b
O~0050
0.0050
0.05CO
0.3850
D.3100
0.2710
KUWAIT VACUUM RESIO
fUEL GAS$ FOEIS
I SOeUTANE
NQR",Al BUTANE
REFQR"'ER FEED IKUWAITI
380-650 DES. KU" DISTILLATE
.650-1000 GAS OIL, LOW SULFa
DESULF. RESID.
VOlU~E GAIN
CAPACITY CONSTRAINT FACTOR
HYDROGEN, SCF/B
STEA,.., MLBS/B
COOLING WATER, MGAl/B
ELECTRICITY, KWH/S
fUEL OIL, FOE BIB
CAT CRACK KUWAIT GAS OIL
FEEC
CCNSTR.
1.0000
0.0260
D.CbeO
0.9150
O.8tHIO
0..2840
o.con
0.5040
PROCUCT
STREAMS
0.0580
0.0500
0.0530
0.02CO
0..0190
0..5040
0.2540
0.0500
4~OOOO
0.(1980
KUWAIT GAS OIL
FUEL GAS. FOE/a
PRCPYlEr!E
I SOIWT ANE
N-BUTANE
eUTYlENE
FCCU GASOLI NE
LIGhT CYCLE OIL, KU~AIT
HEAVY CYCLE OIL, KURAIT
SUl~UR TN H2S, LaS/a
CCKE YIEle
VCLL',,".E GAIN
CAPACITY CONSTRAINT FACTOR
STEAM PROCUCED, MLBS/B
COOLING WATER. MGAl/B
ELECTRICITY. KWH/B
PROCESS fUEL. FOE BIB
OCTANE ~DJUSTMENT, KU~AII
( M-J8)
HYCRGDESUlFURIlATION OF KUWAIT
RESIDUUM - RECYCLE TO EXTINCTION
FEEC
CCNSTR.
1.0
0.1560
1.3702
1.98ltO
0.0lt28
1.3220
11.5COO
0.0299
PIH'lCUCT
STKEAMS
0.0390
a.Ce70
0.0070
0.5270
0.0(,90
0.5e7a
KUWAIT VACUUM RESIO
FuEl GAS. FOF/B
ISOeUTANE
NOR"'AL 8UTANE
380-650 CESULF. KUWAIT D.
REFOR"'ER FEED IK~WAITI
-650-1000 GAS Oil, LOW SULFa
VCLUME GAIN
CAPACITy CONSTRAINT FACTOR
HYCROGEN, SCF/B
STEAM,MLBS/B
CGOlING WATER, "'GAL SIB
ElECTaICITv. KWH/8
FUEl Oil, FOE/a
KUWAIT CRUCE UNIT OPERATIO~
FEED
CCtoiSTR.
I.COCO
I.COOO
0.60CO
0.0250
0.0130
O.400G
PROCUCT
STREM~S
0.00lt8
(1.0040
0.0160
(1.0370
0.2050
0.165('1
0.0610
0.3120
0~1860
0.0032
O.q940
KUWAIT CRUOE
GAS, FOE
ISOeUTANE
NCRMAL BUTANE
KU~~IT C5 -IC6 lT GASOLINE
~UWAIT C6- 3BO wIDE NAPHTHA
180-560 LT. 0151. KU~AIT
560-050 hVY CIST. KU~AIT
~10-JIOO KUWAIT FCCU FEED
VACUUM RESICUE, KU~AJT
VClt..u,E lOSS
CAPACITY CONSTRAINT FACTOR
.ATER, P>\GAlIB
STEAM, /'Il8S/B
Fl.Jl::l. FOI:
ElECTRICTY, KwH/B
SU~fUR ADJ. NAPHTHA PRTRT
-------
FEED
CCNSTR.
1.0000
0.0700
1.0000
0.0130
0.7500
FEED
CDNSTR.
1.0000
1.0
0.0380
1.4400
.O.032Q
1.3000
0.'0060
SUPPLEMENTARY DATA FDR KUWAIT OPERATIONS
PRETREAT 400 EP NAPHTHA KUWAIT
HZS 1~ CRUDE + COKER P.QDElS
PRDOUCT
, STREAMS
0.9950
0.0050
FEED
CCNSTR.
1.0000
KUWAIT 275-380 EP NAPHTHA
HYDROGEN
TREATED KUWAIT REF. FEED
VOLUME LOSS
CAPACITY CONSTRAINT FACTOR
REFINERY FUEL, FOEIS
ELECTRICITY, KWH/B
1.0000
0.0647
p.OOIO
3.8000
0.7450
COkE KUWAIT REStD
PRODUCT
STREAMS
0.0108
o;oz~o
0.0050
0.0100
0'0130
0'0600
0.1870
0.2560
0.2150
4.6500
OJll56
0.'1592
0..10600
KUWAIT VACUUM RESIDUUM
FUEL GAS
PROPYLENE
ISO-BUTANE
N-BUTANE
BlJTYlENES
COKER IT GASOLINE
REFORMER FEED, KuwAIT COKER
LIGHT GAS OIL/FURNACE OIL
COKER GAS OIL 10 FCCU
SULFUR IN H2S, l~S/B
BYPRODUCT COKE, ~lBS/B
VOlU,..e lOSS
CAPACITY CONSTRAINT FACTOR
STEA~ CONSUMED, MlBS/B
ELECTRICITY, KWH/B
REFINERY FUEL, FOE/B
COLLING WATER, MGAl/B
SULFUR ADJUSTMENT, REF FEED
HYDROGEN ADJ. REF PP.TRT.
FEEO
CONSTR.
1.0000
0.0333
0.0710
1.1600
0.3460
0."-340
0.0034
(M-19)
REFORM ~UWAIT NAPHTHA, 90 OCTANE
PRODUCT
STREA~S
OHIOO
0'0080
0.0120
0.8560
0.5160
0.0140
TREATED KUWAIT NAPHTHA
FUEL GAS, FOE IS
I SO-BUT ANE
N-BUTANE
KUW~IT REFORMATE, 90 RON
HYOROGEfI:, SCF/B
VOLU"',E LOSS
CAP~CITY CONSTRAINT FACTOR
REFINERY FUEL, FOE/B
STEAM, MLBS/B
ELECTRICITY, KWH/B
WATER, MGAl/B
INCRE~ENTAl FCCU YIELDS, COKER G.O.
FROM KUWAIT CRUDE
PROOUCT
SIREA,...S
0.1:>40
O.lO~O
0.0860
0.3570
0;3450
6~OOOO
O~0640
0'0130
0.1120
COKER HEAVY G.Q. KUWAIT
COKE ON CATALVST, MlBS/6
FUEL GAS, FOE/B
PROPYLENE
BUTnENES
FUEL GASOLINE
KUWAIT CVClE OIL
H2S ~S SULFUR, lBS/S
I SD.-BUT ANE
"'-BUTANE
VOLUME GAIN
CAPACITY CONSTRAINT FACTOR
-------
SUPPLEMENTARY DATA FOR KUWAIT OPERATIONS
CATALYTICALLY CRACK GAS OIL FROM
RESIO HYOROOESULFURIZATION
OF KUWAIT VACUUM PITCH
FEED
CONSTR.
1.0
0.02136
0.0854
0..14373
0.648
0.225
0.00219
PROOUCT
STREAMS
0.476
0.0779
O~092S
0.0590
0;0130
0.0800
0.3830
O:.lQ30
Od610
0.0200
0.0426
LOW SULFUR KUWAIT GAS OIL
FCCU REGEN COKEy MlB$/8
HZS lBS SUlFUR/B
FUEL GAS FOE BIB
PROPYLE~E
ISO-BUTANE
N-BUT ANE
8UTYLENES
'tce EP GASOLINE
IT tYClE 0Il-QESULFURIIEQ
HVY CYCLE Oll-DESUlFURIZEO
DECANT OIL - OESUlFURIlED
VOLUM£; GAIN
CAPACITY CONSTRAINT FACTOR
STEAM PRODUCTION, MLBS/B
CCOLItlG ~ATER7 ~GAL/B
ELECTRICITY, KWH/b
FUEL FOE BIB
FCCU 800-950 GAS OIL
FEEO
CCNi JR.
1.0
0.03121
0.0699
1.13610
0.96000
0.340
0.00318
DESULF. OF IMPORTEO RESIOUAL FUEL
FEED
CONSTR.
1.0
0.0855
1.1240
1.0
0.0266
0.1519
6.26
0..0207
PROCUCT
STREAMS
0.0236
O;CC34
0.;0047
0.5,148
O~0411
O~2493
0.1826
1.0
PRODUCT
ST.REAMS
3..lt7
0;0822
O..oa07
o..OltOO
O~OlOO
0.0810
0..3960
0.277
0.083
0;020
0.09420
I'PORTED RESIOUAL FUEL OIL
FUEL GAS, FOEIB
1- BUTANE
N- BUT ANE
380-b50 OESUlF KUWAIT DIST.
REFORMER FEED (KUWAIT)
-b50-1000 GAS OIL, lOw FUlF.
DESUlF RESID
VOLUME GAIN
~YDROGEN, SCF/B
CAPACITY CONSTRAINT FACTOR
STEAM, MlBS/BR
COOlI~G kATER, MlBS/B
tlECTRICITY, KWH/8
fUEL Gll, fOE BIB
eyy FUEL GAS FOR HZ PlT
FCCU 8QO-QSO GAS all
FCCU REGEN COKE ~LBS/BBl
H2S, LBS SULFUR/BBL
FUEL GAS,FOE
PROPYLENE
ISO-BUTANE
N --BUTANE
BUTYL ENE
ItCD EP GASOLINE
lC()
tiCiJ.
DEI:ANT OlL
VClUME GAIN
CA?ACITY CONSTRAINT FACTOR
STEA~ PRODUCTION, MLBS/B
COOLING WATER, MGAL/B
ElI::"CTRICITV, KWH/B
FUEL FOE B/B
(M-20)
-------
GLOSSARY
Alkylation Unit
A type of refining unit which produces very high quality gasoline
from light olefins and isobutane. It is generally associated with FCCU's
or delayed cokers which tend to produce excess light olefins.
Barrel: 42 U.S. gallons
BPSD: Barrels per stream day
Capital Charges: The portion of the operating cost which is directly
proportional to the capital cost. Depreciation, insurance, total taxes,
and cost of maintenance are typical examples.
Crude Unit: A refining unit comprising an atmospheric or low pressure
first stage distillation unit, usually followed by a second stage distilla-
tion unit operating at vacuum. The unit comprises the distillation towers,
heat exchangers and heaters.
Coker; A refining unit which is used to convert vacuum residuum to
coke and distillates. A delayed coker is one type of coking unit.
Distillate: A refined or semi-refined material obtained by condensing
the portion of a mixture which is vaporized when heated.
pistillate, Middle: The portion of petroleum boiling between 330 *F
and about 700°F. This material usually includes the stocks blended
into No. 1 and No. 2 Fuel Oil.
FCCU: Fluid Catalytic Cracking Unit. This is one of the types of
units used for converting high-boiling hydrocarbons into lower boiling
hydrocarbons. These units use a finely divided catalyst which is con-
veyed from vessel to,vessel in an aerated state. The aerated solid
behaves like a fluid..
FCCU Feed Pretreater: A unit for pre-refining the feed stock charged
to an FCCU. These units usually involve hydrogenation. After hydro-
genation, sulfur and metallic compounds present in FCCU feed are
practically eliminated and nitrogen is reduced. ~Such purification
improves the operation of an FCCU.
Fuel, Distillate: No. 1 or No. 2 Fuel Oil
B-l
-------
Fuel, Residual: Any liquid fuel containing the residuum from crude
distillation or thermal cracking. In the context of this report, No. 6
Fuel Oil.
Gas Oil: While this term includes middle distillate, its more specific
use is to designate the heaviest vaporizable portion of petroleum, or-
dinarily boiling from about 550 F to 1000 F or more. It is FCCU Feed.
Hydrocracking^Unit: A recently developed process for cracking heavy
hydrocarbons to light products in the presence of rather high partial
pressures of hydrogen, and of a special catalyst. These processes
can convert gas oil completely into gasoline and lighter fractions, or
they can convert gas oils into high grade middle distillates. All hydro-
cracked products may be assumed to be essentially sulfur-free.
Hydrogen Plant: A unit for manufacturing hydrogen. In this study,
the steam reformer type of hydrogen plant has been assumed throughout,
Hydrogen or HZ Treater; A hydrogenating plant, often used to desul-
furize or otherwise purify hydrocarbons.
Incremental Cost: The difference in cost of operation between some
case, usually designated as a base case, and a related case involving
one or more variants.
Isobutane: A specific hydrocarbon compound containing four carbon
and ten hydrogen atoms. Isobutane reacts with light olefins in the
presence of sulfuric or hydrofluoric acids to make high octane alky-
late gasoline.
Iteration: A calculation which is repeatedly made in the course of
solving the matrix representing an L.P problem.
LP: Linear Programming. A mathematical operation, used for ar-
riving at an optimal solution in a system where a number of operable
solutions are possible. As used in the context of this study, the
technique is used to seek out the most profitable method of either
building a new refinery or operating or modernizing an existing refinery.
Matrix: An array of numbers, usually the coefficients of the variables
used in the mathematical model of the refinery. The LP computer code
recognizes the significance of these numbers, and manipulates them
to arrive at the optimal solution for a given problem.
B-2
-------
Qct'ahe: An abbreviation for octane rating. The octane rating is a
means of ranking gasolines according to their resistance to detonating
explosions when used as fuels in internal combustion engines. The
scale is based on 0 for normal heptane and 100 for 2-2-3 tri-methyl
pentane, often called iso-octane. Gasolines can also be rated above
100 octane by means of an extrapolation formula.
Olefin: A hydrocarbon which contains less atoms of hydrogen than its
full complement; olefins are particularly characterized by one double
bond between two carbon atoms. The double bond imparts character-
istic chemical behavior.
Propane De-Asphalter: A process which extracts FCCU feed from
residuum by its solubility in liquid propane.
Realization: A gross profit calculated by subtracting variable charges
that are affected by processing rate from value of products produced.
It is a convenient measure of profitability.
Reformer: A catalytic refining unit, usually employing a platinum —
containing catalyst. The reformer upgrades low octane gasolines into
high octane gasolines by rearrangement of the molecular structure.
Reformers generally yield by-product hydrogen.
Reformer Pretreater: A hydrogenation unit used for purifying the
feed to reformers. The pretreater in effect protects the investment
in platinum catalyst by insuring removal of catalyst poisons.
Residuum: The most non-volatile portion of petroleum, residuums
are sometimes called long-or short-residuums. Along residuum is
the non-volatile residue from an atmospheric pressure distillation;
whereas a short residuum is obtained from vacuum distillation.
Resid HDS Unit: A residuum hydro desulfurizer. In actual fact,
these units commonly also hydrocrack residuum. In any case, they
operate in the presence of a special catalyst and a fairly high hydro-
gen partial pressure. Hydrogen is consumed in the process and
sulfur is released as H2S.
Shadow Price: This is the mathematical quantity used by the LP as
a measure of profitability. It may be described as the cost of
making the last barrel of a particular product. If this cost is below
the value for that product, the profitability could be increased by
making more.
B-3
-------
Stream Day: A day on which a unit is operated at full capacity for
24 hours. Depending on the amount of maintenance work required, a
unit may accumulate 300 to 330 or even more stream days in a year.
Sulfur Plant: As used in this report, a plant for converting H£S to
sulfur, using the Glaus process.
B-4
-------
BIBLIOGRAPHY
Much of the information used in this study was obtained from private
unpublished sources. Included in this category are the assay data from
which crude unit yields were adopted as well as other yield'and cost
data used in the formulation of the LP matrix. Even where published
information was used, it was adapted to fit the requirements of this
study. For these reasons, complete documentation of all sources of
data is not practical.
The following is a list of selected references which may be of interest
to the reader. Those references specifically referred to in the text
of this report are noted under the section heading indicated.
THE NATURE OF THE PROBLEM
Bureau of Mines Data on Marketed Products
Blade, O. C. , "Burner Fuel Oils, 1963," U. S. Dept. of Interior,
Bureau of Mines, Petroleum Products Survey No. 31, Sept. 1963.
Blade, O. C. , "Diesel Fuel Oils, 1962," U. S. Dept. of Interior,
Bureau of Mines, Petroleum Products Survey No. 28, March 1963.
Kirby, J. G. , Messner, Walter G. , and Moore, Betty M. , "Crude
Petroleum and Petroleum Products," Preprint from Bureau of
Mines Minerals Yearbook 1962.
Bureau of Mines Literature Survey on Residual Oil Desulfuriz.ation
Carpenter, H. C. and P. L. Cottingham, "A Survey of Methods for De-
sulfurizing Residual Fuel Oil/'U. S. Dept. of the Interior,
Bureau of Mines, IC-8156 (1963).
The H-Oil Process
Galbreath, R.B. and A. R. Johnson, "H-Oil Process is Proven By
First Commercial Unit," Petroleum Refinery, Vol. 42, n. 9,
pp. 121-123 (1963).
The Gulf HDS Process
Beuther, Harold, and Schmid, B.K., "HDS Process Upgrade
Re sidues," Oil and Gas Journal, Vol. 61, No. 26, pp. 155-158
(July 1, 1963).
C-l
-------
Beuther, Harold, and Schrmd, B.K. , "Reaction Mechanisms and Rates
in Residual Hydr ode sulfur ization, " Proceedings of Sixth World
Petroleum Congress, Section III - Paper 20 - PD7. Frankfort,
Germany, June 19 - 26, 1963.
Propane De-Asphalting Process
Atteridg, P. T. "A Fresh Look at Solvent Decarbonizing, " The Oil &;
Gas Journal, Dec. 9, 1963, pp. 72-77.
Delayed Coking Procesj5
Mekler, V. and M. E. Brooks, "New Developments in Techniques in
Delayed Coking," Proceedings, American Petroleum Institute,
1959, Section III, Refining, American Petroleum Institute,
New York, New York, pp. 229 - 245.
Comparison of Fuel Oil Reduction Processes
The following paper by Slyngstad and Feigelman was used as a basis
for initiating this study. The paper was published in abridged form by
the Oil and Gas Journal.
Slyngstad, C. E., & Feigelman, S. ,"Fuel Oil R eduction, " Paper
presented at 38th Annual Meeting, California Natural Gasoline
Association, Anaheim, California, October 10 - 11, 1963.
Oil & Gas Journal, "Four Roads Offered to Less Fuel Oil," Jan. 6,
1964, pp. 65-68.
t
General Refinery Processing Methods
Refining Process Handbook Issue 1962, Hydrocarbon Processing and
Petroleum Refinery, Volume 41, No. 9, (September 1962).
Nelson, W. L. , Petroleum Refinery Engineering, 4th ed. McGraw
Hill Book Company, Inc., New York, New York (1958).
CRUDE SELECTION AND PRODUCT SLATE
MarmeJ s_ Je_t jTuel Study
Manne's LP study for jet fuel illustrates how a large scale LP simula-
tion is set up. This reference also includes other economic simulation
studies and provides useful background material.
C-2
-------
Manne, Alan S. , and Markowitz, Harry M. , "Studies in Process
Analysis - Economy Wide Production Capabilities," Cowles
Foundation Monograph 18, John Wiley and Sons, New York,
New York (1963). See Chapter 4 particularly.
California Crude fc Products Data
Slyngstad, C. E. , and Feigelman, S. (Reference listed above).
Brown, C. T. , "California Crudes Show Great Variation in
Properties," The Petroleum Engineering, Vol. 28, n. 1,
pp. C 9-C 14, Jan. 1956.
FORMULATING THE MATHEMATICAL MODEL
There appears to be no completely satisfactory reference dealing with
the technique of LP model formulation as related to the petroleum
industry. The book by Manne and Markowitz referred to above is
helpful as well as the following article.
Cheveny, John E. , three part article appearing in Oil and Gas Journal:
Part 1: "What is Linear Programming?" Vol. 589 No. 10, pp.
113-116, 119-120 (March 7, I960).
Part 2; "Putting Linear Programming to Work," Vol. 58, No.
12, pp. 108-110 (March 21, I960).
Part 3: "Putting Linear Programming to Work," Vol. 58, No.
14, pp. 114-115, 117-118 (April 4, I960).
Formulation of the matrix is to some extent related to the particular
algorithm and computer program used. This study employed a pro-
prietary program available from:
Bonner and Moore Associates, Inc.
500 Jefferson
Houston 3, Texas
Low cost natural gas fuel was considered in one LP case. Informa-
tion on the long term availability of gas and predicted use patterns for
it and other fuels is covered in two Federal Power Commission reports;
Federal Power Commission, National Power Survey, Advistory Com-
Committee Report No. 21 on Fuels for Electric Generation
(December 1963).
, Advisory Committee Report No. 18 on Fuels for Electric
Generation in Western United States. (July 1963)
C-3
-------
HYDROGEN BALANCE
Utilization of byproduct hydrogen in a refinery is described in detail
by Gwin.
Gwin, G. T. t "Optimal Utilization of Byproduct Hydrogen in an in-
tegrated Oil Refinery, " Proc. A.P.I., 1959, Section III,
Refining, pp. 193-201.
C -4
------- |