EPA-AA-SDSB-82-13
Technical Report
Distribution of Methanol as a
Transportation Fuel
by
R. Dwight Atkinson
June 1982
NOTICE
Technical Reports do not necessarily represent final EPA decisions
or positions. They are intended to present technical analysis of
issues using data which are currently available. The purpose in
the release of such reports is to facilitate the exchange of tech-
nical information and to inform the public of technical develop-
ments which may form the basis for a final EPA decision, position
or regulatory action.
Standards Development and Support Branch
Emission Control Technology Division
Office of Mobile Source Air Pollution Control
Office of Air, Noise and Radiation
U.S. Environmental Protection Agency
-------
Table of Contents
Page
I. Introduction 1
II. Overview of Existing Motor Fuel 1
Distribution System
A. Pipelines 1
B. Waterborne Transportation 6
C. Truck Tankers 8
D. Rail Tank Cars 10
£. Storage 12
1. Primary Distribution System 12
2. Secondary Distribution System 14
III. Technical Factors in Methanol Distribution 14
A. Technical Problems in Transportation 16
B. Technical Problems in Storage 17
C. Potential Dispensing Problems at the Pump 18
D. Summary of Technical Distribution Problems 19
IV. Economics of Synfuel Distribution 20
A. Overview of Illinois and Wyoming 21
Methanol Plants
B. Specific Costs in Transporting Methanol ..... .21
1. Pipelines .21
2. Rail 23
3. Storage Costs 23
C. Distribution Costs from Production Sites 23
to Bulk Terminals
1. Southern Illinois Plant 25
2. Wyoming Plant 25
3. Average Cost of Fuel Delivery to 27
Bulk Terminal (Long Range)
D. Distribution Costs from Bulk Terminal to 29
Retailer (Local Distribution)
E. Retailer Costs 29
F. Total Distribution Costs 32
V. Distribution Capacity Needed to Support a 32
Synfuel Industry
A. Future Fuel Needs 32
B. Long-Range Capacity 35
C. Bulk Terminal Capacity 36
D. Tank Truck Capacity 36
E. Retail Capacity 37
F. Cost Comparisons Between Synfuel Options 37
VI. Conclusions 39
References 41
-------
I. Introduction
This report examines the issues surrounding the distribution
of synthetic motor fuels, especially methanol. It is divided into
four basic sections. The first is an overview of the existing
transportation fuel distribution network. Capacities and histor-
ical trends are discussed in this section. The second major divi-
sion addresses the unique technical factors associated with meth-
anol distribution. Following this, the economics of synfuel
distribution are examined. Costs for long-range, local, and
retail distribution are presented. The last section draws upon
the previous three sections to determine the total distribution
capacity needed to support a viable synfuel (either methanol or
synthetic gasoline) industry. The total capital costs for such
distribution networks are given and compared to the capital cost
requirements for synfuel production facilities.
II. Overview of Existing Motor Fuel Distribution System
There are five major means whereby liquid fuels are trans-
ported today: pipeline, ocean tanker, inland barge, tank trucks,
and rail tank car. These transportation modes have blended into a
distribution network according to an evolution which has been
underway since Henry Ford popularized the automobile.
The petroleum industry transported about 7.5 million barrels
of gasoline per day by these methods in 1977. [1] Table 1 shows
the amount of petroleum products delivered by each mode in 1950,
1960, 1970, and 1977. 'As is apparent, the role played by pipe-
lines has been an .expanding one during this time frame as they
have not only carried increasing amounts but also the relative
percentage transported by pipeline has grown. While the percent-
age of petroleum products transported by water carriers and rail
tankers has significantly declined, it. should be noted that with
regard to total tonnage, only the rail mode has shown a decline.
The amount transported by trucks has increased substantially since
1950 but the percentage has remained fairly constant. Each of
these links in the motor fuel distribution system is discussed in
more detail below together with a discussion of storage facilities.
A. Pipelines
Pipelines are, in their simplest form, a series of steel
pipes welded together, with pumping stations placed at various
points along the route to maintain proper flow. They range in
diameter from 4 to 58 inches and vary in length from a few miles
to a few thousand miles.[1]
Pipelines currently transport over 72 percent of the crude
petroleum and roughly 37 percent of refined petroleum products in
the United States. Most are common carriers, that is they must
carry the product of any company meeting the pipeline's tariffs
-------
-2-
Table 1
Petroleum Product Transportation Methods[2]
Pipelines
Year
1950
1960
1970
1977**
Tons*
52.7
140.0
333.1
526.0
Percent
of Total
12.75
21.31
31.12
36.56
Water
Tons*
185.2
244.2
286.4
361.7
Carriers
Percent
of Total
44.85
37.17
26.75
25.14
* Tons rounded to nearest million.
** Preliminary.
Trucks
Tons*
130.8
242.5
425.2
524.6
Percent
of Total
31.66
36.93
39.72
36.46
Railroad
Tons*
44.4
30.2
25.8
26.4
Percent
of Total
10.74
4.59
2.41
1.84
Total
Tons*
413.1
656.9
1,070.5
1,438.7
-------
-3-
and regulations as governed by the Interstate Commerce Commission
and the Federal Energy Regulatory Commission.[2]
The petroleum products and crude oil pipeline networks and
capacities in the U.S. are given in Figures 1 and 2. These
figures also show the location of major terminals and refining
areas. As can be seen, there is a great concentration of pipe-
lines near highly populated areas and refining centers, especially
in the east and midwest. Few, however, are near the western coal
fields. thus, it is more likely that methanol produced in the
eastern coal fields will to some degree be able to use existing
pipelines than methanol originating in the west. The costs of
constructing new pipelines will be addressed in a subsequent
section.
Even though the eastern states have a relatively high density
of pipelines, there are technical factors which must be considered
before concluding that methanol can be transported in them. In
addition to material compatibility, discussed later, one such
factor of pipeline distribution is batching. This refers to the
separation of different products or grades of crude in a single
pipeline.[3] Since initial volumes of methanol will be small
compared to that of gasoline it is unlikely that entire pipelines
would be dedicated solely to methanol transfer. Instead, batching
would be used.
Shipments are batched in a continuous sequence with products
of similar quality being next to each other. Typically, they are
shipped in groups that move from lighter to heavier gravities and
then back to lighter in a sequence such as: gasoline-kerosene-
fuel oil-kerosene-gasoline. Cycles may be up to 10 days in
length, depending on such factors as pipeline capacity, refinery
scheduling, and market demand.[3]
If some of the methanol is to be used in blends with gaso-
line, then it would be desireable to batch it next to gasoline
since some mixing of adjacent products occurs.[4] To minimize
this interface, or the mixing of different products, batches are
sometimes physically separated by devices such as rubber spheres.
If the methanol being transported is to be used in a pure form,
then this technique would be desireable. Although it has not been
attempted, it is believed that pure methanol can be successfully
batched through petroleum pipelines.[1] Conoco is currently
planning a test program for the summer of .1982 to determine the
feasibility of batching methanol in a pipeline.[5]
Ethanol, a similar alcohol, has been successfully batched in
a pipeline, however.[4] This was done in January 1981 by William
Pipeline Co. of Tulsa, Oaklahoma.[4] They transported 5,000
barrels of ethanol approximately 200 miles through an 8 inch- pipe-
line. Unleaded gasoline was transported on either end of the
ethanol in a straight run configuration, i.e., the interface was
-------
Figure 1
CRUDE OIL PIPELINE CAPACITIES
( THOUSANDS OF BARRELS DAILY )
AS OF DECEMBER 31,1978
NATIONAL PETROLEUM COUNCIL
UGIUO
-------
Figure 2
PETROLEUM PRODUCTS PIPELINE CAPACITIES
THOUSANDS OF BARRELS DAILY *
AS OF DECEMBER 31,1978
A N A
NATIONAL PETROLEUM COUNCIL
IICINO
MAfOt 111*4111 IOCARON
MAJOI niMINAl ItOIAOl IOCAI1ON
nriuM CONMCIION
-------
-6-
unprotected by a physical apparatus. As a result, about 200 of
the 5,000 barrels mixed with the gasoline during transit. Since
the ethanol was ultimately used as a blending agent with gasoline,
this mixing was not undesirable.
Special care was exercised in this project to prevent the
ingress of water which could lead to phase separation when ethanol
(or methanol) is blended with gasoline. These precautions
consisted of using only sealed tanks with floating inner roofs
(tank designs are discussed in detail in a subsequent section).
Similar precautions would also be necessary with methanol if it
were to be used in a blending capacity. It should be noted that
over the past 10 years, essentially all newly constructed tanks
have been the floating roof variety, due to concerns over evapor-
ative hydrocarbon emissions.[4] Even tanks which pre-date this
time period are being retrofitted with floating roof tanks. Most
of these new and retrofitted tanks are not sealed, however, and
would therefore require some modification, as discussed later, to
further prevent the ingress of water.
This experience with ethanol indicates that methanol batching
should also be feasible. If true, then much of the existing east
coast pipeline network could be utilized to distribute methanol,
avoiding the necessity of constructing special lines solely to
handle methanol.
B. Waterborne Transportation
The nation's inland waterways system is illustrated in Figure
3. It consists of approximately 25,000 miles of waterways and
includes rivers, intracoastal waterways, canals, channels, and
other waterways.[6] In order to be considered navigable, a water-
way must facilitate the movement of a sufficient quantity of
products to be commercially economic. The width of the waterway,
its depth, and the navigability of its bends, locks, and channels
are key factors in determining this quantity. Nearly 25 percent
of the total inland waterways system is less than 6 feet deep and
almost 80 percent is less than 14 feet deep.[6]
As can be seen in Figure 3, navigable waterways, particularly
the Ohio and Mississippi Rivers and the Great Lakes, could be used
to link methanol production facilities in the eastern U.S. with
major population centers in the same region. Western production
facilities, in Wyoming and Montana for example, will not be able
to rely upon this transportation method. Instead, pipelines or
rail tankers (discusssed later in this section) will be needed for
long range distribution from the western sites.
The inland waterway industry consists of approximately 1,800
towing companies which serve 87 percent of the nation's major
cities.[6] These companies operate over 4300 towboats and tugs
with a combined horsepower of approximately 6.1 million. Tank
-------
-7-
Figurr-3. ConiintTcinily Navigable V.'utervvays of the UniU-ci States. $/
Atlantic
Iniracoastal
Waterway
CONTROLLING DEPTHS
•M 9 FEET on MORE
... UNDER 9 FEET
-------
-8-
barges number 3,971 with a total capacity of 71.3 million barrels
(refer to Table 2).[6]
Tank barges are either pushed by towboats or pulled by
tugboats. Towboats are flat bottomed vessels with multiple
rudders for maximum control. They are typically used where large
waves are not encountered. In deep water areas and along the
coast, the more streamlined tugboats are used to move barges. A
towboat may push as many as 20 barges at one time while a tug may
pull up to 4 in a simple operation. The larger towboat flotillas
have a capacity of up to 200,000 barrels, though in most cases
smaller flotillas are dictated by local restrictions.
Another aspect of waterborne transportation .is the self-
propelled tanker used in coastal and oceanic commerce. These
vessels range in capacity from 10,000 to over 1,000,000
barrels.[1] .Table 2 shows that as of July 1979, there were 352 of
these maritime carriers in the U.S. fleet with a total capacity of
over 97 million barrels.
Pipelines generally do not carry heavy products such as
residuals due to the relatively high pour points and viscosity of
these materials. Thus, domestically, barges and tankers often
complement pipelines in moving the heavy petroleum products.
Still, gasoline accounts for approximately 50 percent of domestic
volume petroleum product movements on inland waterways.[2]
Distillate fuel oil accounts for 23 percent and residual fuel oil
12 percent. When foreign tanker imports are included in volume
calculations, residual fuel oil is transported in the largest
quantities followed by distillate fuel oil, crude oil, and other
products.[2] This shift in rank is primarily due to the fact that
domestic refineries are optimized to produce gasoline and other
high-priced products. Consequently, the U.S. imports large quan-
tities of residual fuel oil.[2]
The large role played by tankers and barges in transporting
petroleum products and crude oil has important implications for
the future of methanol shipment for several reasons. Of primary
importance is simply their availability. Second, these carriers
are flexible with regards to the number of major markets they can
serve. The Ohio River Valley and Great Lakes for example, are
accessible without extensive construction projects which could be
needed for pipeline transport. Third, it is significant to note
that chemical grade methanol is currently moved by both ship and
barge.[1] Thus, the technical expertise gained in this practice
should prove beneficial if methanol were used to fuel the nation's
vehicles.
C. Truck Tankers
The role of the trucking industry in transporting the
nation's petroleum products has long been a vital one. This is
-------
Table 2
Inland Waterways and Maritime Carriers — Summary, July 1979[6]
Self-Propelled
Tank Vessels
(Tankships)
Non-
Self-Propelled
Vessels
(Tank Barges)
Greater Than 5,000 Barrels
Capacity
Waterways
System
1. East Coast
2. West Coast
3. Great Lakes
4. Alaska
5. Hawaii
6. Gulf &
Mississippi
TOTAL
1. East Coast
2. West Coast
3. Great Lakes
4 . Alaska
5. Hawaii
6. Gulf &
Mississippi
TOTAL
Number of
Units
152
55
8
0
12
. 74
301
335
91
91
9
6
2,731
3,263
Total Capacity
(barrels)
44,722,553
22,172,689
373,270
0
11,566,547
18,145,074
96,980,133
10,865,387
2,787,437
1,425,053
131,433
100,567
54,581,191
69,891,068
Less Than 5,000 Barrels
Capacity
Number of
Units
27
10
6
3
0
5
51
82
36
9
30
1
550
708
Total Capacity
(barrels)
24,190
6,391
7,988
1,244
0
5,169
44,982
140,591
67,363
13,614
44,523
818
1,180,519
1,447,428
TOTAL
Total
Capacity
Number of
Units
179
65
14
3
12
79
352
417
127
100
39
7
3,281
3,971
4,323
Total Capacity
(barrels)
44,746,743
22,17-9,080
381,258
1,244
11,566,547
18,150,243
97,025,115
11,005,978 T
2,854,800
1,438,667
175,956
101,385
55,761,710
71,338,496
168,363,611
-------
-10-
verified by the fact that nearly all oil products and LPG are, at
some point, transported by tank truck.[2] Table 1 indicates that
the total tons of petroleum products transported by trucks has
grown rapidly in recent years.
Tank trucks are primarily used for local distribution and are
optimum in the 25 to 50 mile range. [1] Truck tankers are also
used in longer hauls to supplement pipeline deliveries in seasonal
peak demand periods.[2] New tank trucks are getting lighter and
stronger with the increased use of aluminium alloys, stainless
steel, and reinforced fiberglass. This lighter weight permits
larger payloads, with capacities approaching 11,000 gallons (262
barrels).[2] Because of the many companies that own and operate
tank trucks, the short distances involved in most hauls, and the
extent of the nation's highway system, truck transport and hence
its capacity is difficult to document.
The National Petroleum Council (NPC) has attempted to deter-
mine this capacity by conducting a survey of the major trade asso-
ciations. [7] They estimate that as of December 31, 1978 there
were over 50,000 tank vehicles of over 3,500 gallons in capacity
in the U.S. The total fleet capacity of these tankers is about
364.4 million gallons (8.7 million barrels), 'it should be noted
that only approximately 80 percent of these tankers were designed
to haul liquid products, the remainder being used to transport
compressed gases. The NPC reports, however, that all of these
vehicles could be used to haul petroleum in an emergency. The
degree to which these tankers are compatible with methanol has not
as yet been determined. The general subject of materials compati-
bility will be addressed in a later section.
D. Rail Tank Cars
Several types of tank cars exist with some equipped with
heating coils or insulation to facilitate the transportation of
heavy fuels and asphalt. Pressurized cars carry LPG while unpres-
surized, unheated cars are usually relied upon to carry gasoline,
aviation fuel, and distillate fuel oils. Capacities of tank cars
range up to 50,000 gallons (1190 barrels) or more.[2] As many as
110 tank cars can be connected in a series to form a unit train,
making large volume transport possible by rail.fl]
As of June 15, 1979 there were 202,811 tank cars, with a
combined capacity of 3.6 billion gallons (85.7 million barrels),
in the U.S. fleet. [7] Table 3 gives a breakdown of these tank
cars according to their capacity and use. From this table it can
be seen that 107,552 tank cars, representing a capacity of 2.2
billion gallons (52.3 million barrels) are suitable for carrying
crude oil or petroleum products. Those not suitable are either
owned—by the railroad to transport diesel fuel for its locomotives
or have special design purposes such as acid or caustic soda
transport. Also evident from this table is that 77,166 cars
-------
Table 3
Summary of Demographic Breakdown of U.S. Tank Car Fleet [7]
(June 15, 1979)
Private Non-Pressure
Uninsulated
Coiled
Insulated
Coiled
Total
Private Pressure
Uninsulated
Insulated
Total
Unsuitable Total
RR Owned Total
Grand Total
Under
8,500
Gallons
996
3,875
387
1,578
6,836
0
599
559
12,755
233
20,423
8,500
to
13,499
Gallons
2,837
2,674
3,994
11,689
21,144
627
6,897
7,524
30,076
642
59,386
Age
All Years
13,500
to
20,499
Gallons
434
2,835
130
1,336
4,735
273
92
365
32,002
1,411
38,513
20,500
to
30,499
Gallons
10,390
14,429
712
18,067
43,598
2,060
2,457
4,517
11,060
2,518
61,693
Over
30,500
Gallons
561
106
10
176
853
15,350
2,031
17,381
4,549
13
22,796
Total
Number
15,218
23,919
5,183
32,846
77,166
18,310
12,076
30,386
90,442
4,817
202,811
Total
Capacity
306,131,695
429,313,277
64,048,976 '
583,110,649 P
1,382,604,597
583,595,930
209,331,725
792,927,655
1,327,217,268
88,283,117
3,591,032,637
-------
-12-
having a capacity of 1.4 billion gallons (33.3 million barrels)
are used for nonpressurized transport purposes. It is these cars
that would typically be relied upon to carry methanol.
Rail transportation currently constitutes the smallest market
share of petroleum product distribution with just under 2 percent
(see Table 1). Liquefied petroleum gases and coal gases are moved
in the greatest volumes followed by residual fuel oil, asphalts,
and lubricating oils.[2]
Despite the diminishing role of railroads in transporting
petroleum products, they may be more prominent in methanol trans-
portation. This can be expected since rail lines already connect
the coal mines with major markets. Since methanol plants are
likely to be sited near the mines, few new tracks would need to be
laid. This will be especially important in the west where there
is a paucity of navigable waterways, necessary for barge trans-
portation.
E. Storage
1. Primary Distribution System
The system of pipelines, tankers and barges that moves crude
oil from producing areas to refining centers, and the similar
facilities that move refined petroleum products in bulk to market-
ing areas, are generally referred to as the primary distribution
system.[8] In order to understand the storage, aspects of the
primary petroleum distribution system, one must be familiar with
the concepts of minimum and maximum operating inventories. These
are discussed below along with the national storage capacity.
The minimum operating inventory is defined as the inventory
required to fill such components as pipelines, tank bottoms, and
refinery process equipment in order to maintain normal operations;
this volume is considered unavailable for consumption.[8] Short-
ages and runouts would occur if inventory were to fall below this
level. Table 4 indicates that the minimum operating inventory for
the U.S. primary distribution system is approximately 720 million
barrels with just over 40 percent of that being crude oil.
The maximum operating inventory must also be considered when
examining storage capacities. This concept refers to the maximum
quantity that could be stored in assigned tankage (plus inven-
tories maintained outside of storage facilities) while still main-
taining a workable operating system.[8] In determining the maxi-
mum operating inventory a company must allow empty space in tank-
age to 1) provide room for thermal expansion, 2) receive inven-
tory, and 3) have a margin of error for emergency situations and
schedule changes. Exceeding the maximum operating inventory could
lead to slowdowns and interruptions in the system.
-------
-13-
Table 4
Overview of Primary Distribution System[8]
U.S. Primary Distribution System
Minimum Operating Inventory. — 1978
(Millions of Barrels)
Crude Oil 290
Gasoline 210
Kerosene 35
Distillate Fuel Oil 125
Residual Fuel Oil 60
720
U.S. Primary Distribution System
Total Shell Capacity of Tankage Under
September 30, 1978 Construction
(Millions of Barrels)
Crude Oil 462 12
Gasoline 438 5
Kerosene 90 less than 1
Distillate Fuel Oil 365 3
Residual Fuel Oil 162 _1
1517 21
-------
-14-
The tank capacities for crude oil and each of the major
refined products are shown in Table 4. The actual measure of
stocks lies at any one time somewhere between the minimum oper-
ating inventory and the total storage capacity. The National
Petroleum Council states that inventory has averaged about 50
percent of tank capacity for the past 30 years (see Figure 4).[8]
Since individual tanks alternate between empty and full, the NPC
concludes that no significant storage capacity exists for holding
emergency supplies.
It is likely that many new primary storage facilities will
therefore have to be constructed in order to adequately serve a
neat methanol fleet. This is to be expected since methanol's
lower energy content compared to gasoline or diesel fuel means
that nearly twice the capacity is needed to store the same amount
of energy.
2. Secondary Distribution System
The secondary distribution system is comprised of small
.resellers of petroleum products, such as gasoline service stations
or fuel oil dealers.[8] This system along with the consuming
sector contains substantial holding capacity. In an ..analysis of
the storage capacity for gasoline and distillate fuel oil in the
secondary/consumer network, the National Petroleum Council esti-
mates that capacity exists for at 'least 500 million barrels, or 60
percent of the primary storage- capacity for these products.
Since this network is by definition near marketing areas,
there should be no problems with regard to location in using its
capacity to distribute methanol.
III. Technical Factors in Methanol Distribution
This section identifies the unique technical problems
expected to be encountered when transporting, storing, and
dispensing methanol or methanol-containing fuels. Included are
discussions on the types of materials found in the distribution
system which are known to be incompatible with methanol, the
problem of preventing the ingress of water into storage tanks, and
ways to decrease the amount of time spent at filling stations (the
lower energy content of methanol vs. gasoline or diesel fuel could
necessitate more frequent or longer fueling stops).
It should be noted that the estimates below of the extent of
the various technical problems of methanol distribution are first
order estimates. There is a paucity of data along these lines and
more research will be necessary before the net costs associated
with a conversion to methanol can be better determined.
-------
TOTAL
SYSTEM
CAPACITY
SHELL
CAPACITY
OF
TANKAGE
MAXIMUM
OPERATING
INVENTORY
MINIMUM
OPERATING
INVENTORY
"COMPLETELY
UNAVAILABLE"
INVENTORY
TANK TOPS
SPACE IN THE SYSTEM WHICH IS NEEDED TO
MAINTAIN A WORKABLE OPERATING SYSTEM
AMOUNT OF PETROLEUM WHICH CAN BE
STORED, IN ADDITION TO THE MINIMUM
OPERATING INVENTORY, WHILE STILL
MAINTAINING A WORKABLE SYSTEM.
THIS INCLUDES SEASONAL INVENTORY.
ACTUAL INVENTORY ON REPORTING DATE
ADDITIONAL PETROLEUM REQUIRED TO
HAVE AN OPERABLE SYSTEM AND
AVOID PROBLEMS AND RUN-OUTS
TANK BOTTOMS
;£-$fi
;:»;:; PLUS: *
PIPELINE FILL, EQUIPMENT FILL
P IN-TRANSIT
Ln
I
Figure1/-. Slmpliricd Diagram of Terms Describing Petroleum Inventories and Storage Capacities. ]}/
-------
-16-
A. Technical Problems in Transportation
Methanol is highly corrosive to certain materials found in
the petroleum product distribution network and has little effect
on others. When in contact with ordinary carbon steel such as
that found in pipelines, fittings, and valves, it tends to
dissolve rust and to slowly corrode the steel. [1] The rate of
rust dissolution is fairly rapid but that of steel corrosion is so
slow as to warrant no concern. Carbon steel, in fact, is the
common construction material for methanol tanks and pipelines.[1]
Because methanol leaches out ordinary pipe dope, it is recom-
mended that either screwed flanges sealed with a teflon thread
sealant or welded flanges be used.[1] The normal practice of
welding underground piping and coating with asphalt or plastic
adhesive tape to prevent corrosion should be followed in the case
of newly constructed methanol pipelines.
The pumps used in methanol transfer are made of either carbon
steel or bronze lined steel. Either mechanical seals or square
asbestos packing is used to seal the pumps and prevent leaks.
Since these pumps are essentially the same as those used in handl-
ing petroleum products, no change would be necessary to facilitate
methanol transfer.[1]
Some plastics are also susceptible to deterioration by meth-
anol. Plastics are often used in the construction of gaskets,
0-rings, and packings.[1] Since the chemical industry uses meth-
anol in various processes, they have already faced this type of
problem. Neoprene and neoprene-asbestos mixtures were found to be
resistant to the debilitating action of methanol.[1] Plastics to
be avoided include polyurethane, Buna-N and Viton (common compo-
nents in some pipelines as well as automobile engines and
pumps.)[4,9,10,11]
One final issue relating to material compatibility concerns
the batching of methanol in pipelines with other petroleum
products.[19] Since methanol is such a strong solvent, it could
dissolve the rust inhibiting agents which typically line the
insides of pipelines. This potential problem will be addressed in
the Conoco research program mentioned earlier.[5]
In addition to the concerns about material compatibility
there is another potential problem associated with ordinary valves
in such places as transfer lines. This is their susceptibility to
blown gaskets and valve packing due to the expansion of methanol
when its container is heated by the sun's rays. Relief valves can
remedy this situation. Although gasoline also expands when
heated, its thermal expansion characteristics are not as
pronounced as those for methanol. Thus there are fewer concerns
over blown gaskets when shipping gasoline.[12]
-------
-17-
B. Technical Problems in Storage
Considerable industrial experience has been accrued in the
storage of pure methanol in mild steel tanks with satisfactory
results.[1] It has been demonstrated that old steel tanks can be
cleaned by removing rust and water with a solvent such as methanol
or isobutanol at a low cost. [9,10] These tanks can then be used
to store methanol on a permanent basis.
Two related problems in storing methanol are 1) preventing
water from entering the tank, which could cause phase separation
in methanol-gasoline blends, and 2) preventing breathing losses
from the tank as the ambient temperature changes and the level of
methanol fluctuates.[1,9,10] The approach taken to resolve these
problems depends on which of the two general types of bulk storage
tank is being used, floating roof or fixed roof.
Floating roof tanks are commonly used to store gasoline at
refineries and large bulk terminals. The floating roof tank has a
roof that actually floats on the surface of the liquid. Plastic
or rubber collars seal the border between roof and tank wall.
This design virtually eliminates breathing losses since the amount
of vapor space above the liquid is minimized. As mentioned
earlier, this is the reason why floating roof tanks have been in
such widespread use over the past 10 years. [4] Tests in Germany
and New Zealand indicate that a second seal is needed to prevent
water from entering the tank.[9,10] Another, cheaper, solution to
the water ingress problem is to simply construct a fixed roof over
the tank to keep out rainfall.[1,10]
Fixed roof tanks are typically used to store intermediate and
smaller quantities of petroleum products in bulk storage. This
tank design is also used for storage of chemical grade methanol.
Breathing losses are unavoidable with this type of tank. However,
water ingress as vapor or liquid can be prevented by either
putting desicators on the air vents or covering the surface of the
methanol with a dry gas such as nitrogen or carbon dioxide. Both
methods have been used in industrial applications.
Potential storage problems also exist at the retail level.
Typically, retail outlets have from one to six storage tanks, each
having a 10,000 gallon (238 barrel) capacity. Currently, there
are approximately 173,000 service stations nationwide.[1] If each
has an average of 4 tanks,[13] then there .would be approximately
700,000 tanks in use. These vessels have historically been made
of carbon steel which, as pointed out earlier, is a satisfactory
material for storing methanol. Since 1965 however, these carbon
steel tanks have been systematically replaced by tanks made of a
polyester-fiberglass laminate. This material offers superior
corrosion resistance to gasoline and its surroundings but is not
recommended for methanol storage since it can cause a noticeable
increase in the gum content of fuels containing methanol.[10,
-------
-18-
13,14] Roughly 10 percent of all retail storage tanks are now
made of the polyester-fiberglass laminate and by 2000, 25 percent
of all tanks are expected to be of this variety. [13] Replacement
of these tanks with carbon steel tanks would be necessary if
carbon steel tanks were not available at a station desiring to
carry methanol.
C. Potential Dispensing Problems at the Pump
The hoses, nozzles, etc. that are currently being used to
dispense gasoline are compatible with methanol.[1] Thus no modi-
fications should be necessary to this type of equipment to prevent
corrosion or deterioration of pumping components. However, some
modifications may be necessary to increase the pumping rate of
methanol because of its lower energy content.
Methanol contains about half the energy as gasoline on an
equal volume basis. Since methanol engines are anticipated to
have an approximately 20 percent higher fuel efficiency as
compared to gasoline engines, roughly 60 percent more methanol on
a volumetric basis would be needed to propel a car the same number
of miles it would otherwise travel on gasoline.[15] This implies
that motorists will not be able to drive as far in methanol vehi-
cles for a given size fuel tank.
A simple solution would be to increase the fuel tank volume
by 60 percent. Although this could give a methanol vehicle the
same driving range as a gasoline powered vehicle, there are._two
associated detriments: 1) increasing the tank volume ±-s incon-
sistent with current trends in downsizing, i.e., it would increase
vehicle weight and present possible spacing problems, and 2) since
more volume would be pumped, there would be more time spent per
fill-up. This could increase crowding at service stations. Even
without larger tanks there could be crowding since less energy
would be pumped per unit time than is customary. Whether or not
this increased crowding would be important is unknown, since the
current operating capacity of service stations at peak hours is
unknown.
Because of the space and weight constraints it is unlikely
that the full 60 increase in tank size would be realized. Some
increase, however, is likely so that motorists could have compar-
able driving ranges to those customary today. Thus the problem of
potentially crowded filling stations remains. A possible solution
would be to increase the fuel flow rate of the dispenser. This
too, however, has potential drawbacks.
These problems are related to a phenomenon known as "spit-
back" which results when fuel is fed too fast into the vehicle's
tank. Currently, car fill pipes and service station pumps are
designed for an approximate 12 gallon per minute (gpm) rate.[16]
Some self service stations, however, have nozzles calibrated to 10
-------
-19-
gpm to further reduce the likelihood of spitback. This practice
is also followed in areas using vapor recovery (such as in
California) where flow rates of 8 to 10 gpm are common.
If it were necessary to alleviate this spitback problem the
tanks of future vehicles could be constructed to accommodate a
higher fueling rate. One possible way to substantially increase
this rate would be to use a dual-nozzle delivery system. In this
design, the dispenser would have two nozzles instead of the
customary single approach. They would be inserted together into
separate fill pipes which would follow separate routes to the
tank. Thus as much as 24 gallons could be dispensed per minute to
the tank, but each fill pipe would only be delivering the usual 12
gpm. This could significantly reduce the amount of time motorists
spend actually refueling their vehicles and partially alleviate
the crowding problem at service stations.
An additional benefit of the dual-nozzle approach is that it
could prevent accidental misfueling with gasoline. The dual meth-
anol nozzle could not fit into the single-holed fill pipes of
gasoline tanks. Similarly, if the individual nozzle diameters of
the methanol system were slightly smaller than those of. the
unleaded gasoline dispenser, then gasoline could not be fed into a
methanol vehicle's tank. (The nozzles for unleaded gasoline are
smaller than those for leaded gasoline for this same reason, i.e.,
to prevent misfueling.) While this system would require addi-
tional equipment on both vehicles and dispensing pumps, the addi-
tional costs should be minimal.
D. S'lTmnary of Technical Distribution Problems
In conclusion, there are a number of technical factors influ-
encing the distribution of methanol. Of these, material compati-
bility is the greatest concern. When transporting methanol by
pipeline, it may leach out ordinary pipe dope, for instance.
Screwed flanges sealed with teflon or welded flanges could alle-
viate. this problem. There appears to be no compatibility problem
with the carbon steel of the pipeline itself or the pumps used to
facilitate transfer. Similarly, the steel construction of barges
and rail and truck tankers should present no problem for methanol
transfer. Some plastics, however, such as polyurethane , Buna-N,
and Viton found in gaskets, 0-rings and packings in some pipelines
are not compatible with methanol and would have to be replaced.
Noeprene and neoprene-asbestos mixtures are adequate replacement
materials.
In bulk storage, floating roof tanks are needed to minimize
both water ingress (especially when methanol is used as a blending
agent) and evaporative emissions (whether or not it is a blending
agent). As mentioned earlier, this type tank has become more and
more popular over the last 10 years due to concerns over reducing
-------
-20-
evaporative emissions. Tests indicate, however, that a second
sealing may need to be added to provide adequate water protection.
Storage at the retail level could present the most problems
if the current trend of relacing carbon steel tanks with ones
constructed of polyester-fiberglass laminate continues. This
latter material causes an increase in gum content of methanol
containing fuels. By 2000, at the current rate of replacement,
approximately 25 percent of all tanks would be the polyester-
fiberglass type.
There appear to be few compatibility problems at the dispens-
ing pumps. However, because of methanol's lower volumetric energy
content, there could be increased crowding at filling stations.
This is expected since 1) more frequent stops would be made by
motorists if their fuel tanks are kept at the current size for
gasoline vehicles; and 2) more volume per fill-up would be needed
if on board tank size is increased to make driving range more
compatible with that of gasoline vehicles. It is unknown a^t^this
time whether or not this increase in pumping time would adversely
affect service stations since their current operating conditions
are unknown. However, an increase in the fuel delivery rate,
possibly via a dual-nozzle pump, could alleviate this,_ problem if
it arose.
IV. Economics of Synfuel Distribution
In a draft study performed by DHR, Inc. for the Department of
Energy to investigate methanol use options, the cost of both meth-
anol and gasoline distribution from two hypothetical synfuel
production facilities to regional distribution centers was
considered.[1] One plant was located in the coal fields of
southern Illinois and the other in northeastern Wyoming. Each
plant has a 100,000 barrel per day capacity (methanol). The
logistics and costs of fuel delivery from these plants to speci-
fied sites is discussed below. Following this discussion, typical
local distribution and retailing costs are given and conclusions
reached about the total cost of distributing methanol and synthe-
tic gasoline.
It should be noted that a conservative approach in terms of
methanol distribution costs was followed in this report. That is,
in each aspect of distribution, long-range, short-range, and
retail, assumptions were made which would tend to increase the
cost of methanol distribution relative to that for synthetic gaso-
line. In most instances, for example, future methanol vehicles
were assumed to have the same energy efficiency as their gasoline
counterparts, although it is generally believed they will be
roughly 20 percent more efficient. This approach was followed in
order to help assure that the costs of distributing methanol were
not underestimated, since some costs of converting to methanol are
inevitably overlooked at this early stage.
-------
-21-
A. Overview of Illinois and Wyoming Methanol Plants
The southern Illinois plant would be located near several
major markets including St. Louis, Chicago, Toledo, Indianapolis,
Kansas City, Cincinnati, and Louisville. To simplify the study,
it was assumed that the entire product from this plant would be
distributed solely to St. Louis and Chicago. Both of these
markets are more than large enough to consume such quantities of
methanol.
Both St. Louis and Chicago are major oil refining areas
located by the Mississippi River and Lake Michigan, respectively.
Each area has major storage terminal facilities and is connected
by a network of common carrier product pipelines (refer to Figure
1) which could be used to distribute methanol. Systems for trans-
porting the methanol from the production plants to these areas,
however, would need to be devised. A pipeline from the Illinois
methanol plant to St. Louis would be approximately 100 miles long,
while one to Chicago would be roughly 300 miles in length. A
pipeline construction company contacted as part of the above
mentioned study reports that for a marketing split of roughly
one-third to St. Louis and two-thirds to Chicago, the necessary
pipeline diameters would be 10 inches and 12 inches, respec-
tively.
The Wyoming plant near Gillette is located further than the
Illinois plant from major markets, the closest ones being Denver,
Minneapolis, Omaha and Kansas City. As with St. Louis and
Chicago, these markets also have major terminal storage facilities
for petroleum products and are connected to other markets by pipe-
line. Minneapolis and Kansas City have access to inland water-
ways. Only Denver, however, has a pipeline which comes near
Gillette. A pipeline connecting the Wyoming plant with the
primary markets would be approximately 1,000 miles long and 10 to
24 inches in diameter.
B. Specific Costs in Transporting Methanol
As discussed earlier, the major means of transporting fuels
long distances within the U.S. are pipeline, barge, and rail tank
car. Of these, all but barge transportation are likely components
in the distribution network surrounding the two design plants;
each is too far from inland waterways to make barges feasible.[1]
Table 5 shows the average transportation cost for methanol for the
two remaining modes, pipeline and rail tank car. These are
discussed below. Costs for distributing synthetic gasoline will
be addressed in a subsequent section.
1. Pipelines
Of the two means of transporting methanol listed in Table 5,
pipelines are the cheaper. As can be seen from this table, the
-------
-22-
Table 5
Average Transportation Costs for Methanol
(1981 dollars per million Btu)[l,17]
Railroad Tank Car Transportation Costs (Unit Trains)
Distance, Miles
Distance,
Miles
50
100
200
500
800
1000
1200
100
200
400
600
800
1000
Pipeline Transportation
Bbl/Day 50,000
Pipe Dia. , Inch 10
.07
.12
.19
.41
.58
.72
.85
Cost
.28
.35
.58
.78
1.01
1.28
Costs
100,000
14
.049
.082
.12
.28
.41
.48
.59
200,000
20
.025
.033
.07
.19
.26
.33
.39
-------
-23-
operating cost decreases significantly as the volume transported
increases. These costs are similar on a volumetric basis to those
incurred in transporting gasoline via pipeline since the physical
properties of methanol and gasoline are similar.
The overall installed capital costs for pipelines in an aver-
age terrain outside of metropolitan areas is $11,000 per diameter
inch-mile (1981 dollars).[1] For metropolitan areas, the capital
costs rise to $33,000 per diameter inch-mile.[1] These costs
consist of right-of-way, pipe costs and installation and pumping
station equipment. Typically, capital charges of depreciation,
interest, taxes, and insurance account for 75 percent of the pipe-
line transportation cost with operating and maintenance costs
accounting for the remainder.
2. Rail
As mentioned earlier, there are two modes of railroad tank
car transportation: small volume transport with a few cars in a
mixed freight train or large volume transport as a unit train with
roughly 100 cars. Unit trains usually only apply to a fixed
route, but are less costly than individual tank cars in a mixed
train. The average cost for railroad transportation by unit train
is given in Table 5. From this table it can be seen that railroad
tank car transportation costs are 2.7-3.4 times higher than those
for a 14 inch pipeline carrying 100,000 barrels per day of meth-
anol. Operating and maintenance costs account for approximately
85 percent of the rail costs while capital charges account for the
remaining 15 percent.
3. Storage Costs
The bulk terminal storage and blending costs would be approx-
imately $0.13-0.17 per million Btu (mBtu) of methanol.[1] This
includes operating and maintenance and capital costs. The capital
costs for bulk storage facilities constructed of standard mild
steel are given in Table 6.
C. Distribution Costs from Production Sites to Bulk
Terminals
Distribution costs for transporting pure methanol and gaso-
line from both the Southern Illinois and Wyoming sites discussed
earlier to markets (bulk terminals) are presented in the following
sections. In both cases, the assumption was made that methanol
would be carried by pipeline to markets and then stored in bulk
terminals and blended or marketed as needed. In addition, distri-
bution costs from another study will be discussed and compared to
that of the DHR report.
-------
-24-
Table 6
Bulk Storage Capital Costs
(1981 Dollars)[1,17]
Storage Tank Size,
Bbl Cost
125,000 $ 313,000
250,000 474,000
500,000 719,000
1,000,000 1,090,000
-------
-25-
1. Southern Illinois Plant
Table 7 shows the distribution costs to a bulk terminal from
this plant for the two fuels as determined by DHR. The distri-
bution costs are $0.31 per mBtu for synthetic gasoline and $0.56
per mBtu for methanol. Note that these costs include terminaling
costs. Terminaling costs are the same for both methanol and gaso-
line since they include such factors as labor, grounds keeping and
to a smaller degree electricity which are essentially the same for
both products.[18]
The gasoline costs reported by DHR, however, appear to be
based on the assumption that the capital cost of a gasoline pipe-
line would be roughly one-half that of a methanol pipeline trans-
porting the same amount of energy. (Recall that the energy
density of methanol is about half that of gasoline.) This is
overly simplistic since the volumetric flow of a pipeline varies
with the cross-sectional area and, as indicated by DHR themselves,
pipeline capital cost is proportional to the radius and not the
area or volumetric capacity.
To estimate a revised cost of transporting gasoline by pipe-
line, it will be assumed that the DHR statement that 75 percent of
such costs are due to capital charges, and the remainder operating
and maintenance, is correct. As determined by DHR, pipeline
capital costs are assumed to be proportional to pipeline
diameter. Operating and maintenance costs for gasoline pipelines
are estimated to be 25 percent less than a comparable methanol
pipeline transporting the same amount of energy.[18] This last
estimation is based on the assumption that two-thirds of operation
and maintenance costs in a gasoline, pipeline are due to such
factors as labor and maintenance and one-third due to pumping
costs. For a methanol pipeline the labor and maintenance costs
would be very nearly the same but the pumping costs would be
roughly twice as high since twice the volume is being trans-
ported. [18] Following this procedure, synthetic gasoline would
cost approximately $0.37 per mBtu to distribute to a bulk terminal
from the southern Illinois plant.
The capital costs of the methanol pipeline (as determined by
DHR) and gasoline pipelines (as determined by the above procedure)
are also given in Table 7. Together, the methanol pipelines cost
roughly $65 million. To transport the energy equivalent in gaso-
line, the 100 mile pipeline would be approximately 7 inches in
diameter and the 300 mile pipeline would have a diameter of 8.5
inches. Together, the gasoline pipelines would cost $46 million.
2. Wyoming Plant
In this example, DHR assumed that one 14 inch diameter pipe-
line 1000 miles long would carry all the methanol to either
Minneapolis, Chicago, or Kansas City. The capital costs for this
-------
-26-
Table 7
Distribution Costs to and Including a Bulk Terminal
For a Southern Illinois Methanol Plant, $/mBtu
(1981 Dollars)[1,17]
Pipeline, 100 miles to
St. Louis, 33%; 300
miles to Chicago, 67%
Bulk terminal operating cost*
Added storage in bulk terminal
Pure
Methanol
.31
.15
.10
.56
DHR
.16
.15
.31
Gasoline
Revised
.22
.15
.37
Pipeline
Capital Cost; (millions of dollars)
100 mile, 10 inch diameter 15.3
100 mile, 7 inch diameter 10.8
300 mile, 12 inch diameter 49.7
300 mile, 85 inch diameter 35.2
* No interest charges are made for the stored fuel.
-------
-27-
pipeline would be $165 million (see Table 8).[1] Following the
procedures outlined in the previous section, a pipeline to carry
the same amount of energy in the form of synthetic gasoline would
be 10 inches in diameter and cost roughly $118 million.
The costs associated with distributing methanol and synthetic
gasoline from the Wyoming plant to a bulk terminal are given in
Table 8. DHR estimated these distribution costs to a bulk termi-
nal to be $0.40 per mBtu for gasoline and $0.73 per mBtu for meth-
anol-. As was the case with the southern Illinois scenario,
however, DHR apparently assumed the cost to transport gasoline by
pipeline would be one-half that of methanol. Using the aforemen-
tioned capital cost for a 10 inch pipeline for gasoline and the
same 75/25 split of capital costs and operating and maintenance
costs discussed earlier, synthetic gasoline would cost $0.50 per
mBtu to distrbute.
3. Average Cost of Fuel Delivery to Bulk Terminal (Long
Range)
If one assumes that half of the nation's synfuel plants are
located in the west and half in the east, then an average nation-
wide distribution cost from production facility to bulk terminal
can be obtained from the previous two sections. By averaging the
values given in Tables 7 and 8, one obtains a typical methanol
distribution cost of $0.65 per mBtu and a revised cost of $0.44
per mBtu for synthetic gasoline.
In a less detailed report than the DHR study described above,
ICF, Inc. also examined the costs associated with methanol/
synthetic gasoline distribution.[19] Both short- (268 miles) and
long- (1,039 miles) range pipeline scenarios were examined in that
study also. There, the average cost of distribution from produc-
tion facility to bulk terminal was estimated to be $0.44 per mBtu
for methanol and $0.22 per mBtu for synthetic gasoline. These
costs, however, did not consider terminaling costs; the DHR study
did. When these costs are subtracted from the DHR estimates given
above, average distribution costs from plant to bulk terminal are
estimated to be $0.40 per mBtu for methanol and $0.29 per mBtu for
synthetic gasoline (revised). As can be seen, the two reports are
in close agreement, especially for methanol where the DHR estimate
is $0.04 per mBtu less. The gasoline estimates are further apart
and the ICF projection is the lower of the two. This is due to
the fact that ICF made the same simplistic assumptions concerning
the relative costs of methanol and synthetic gasoline transport
discussed earlier. . That is, gasoline would cost one-half as much
to distribute via pipeline as methanol. Using the original DHR
estimate based on this assumption for comparisons, an average
distribution cost for synthetic gasoline would be $0.21 per mBtu.
This is very close to the ICF value of $0.22 per mBtu.
-------
-28-
Table 8
Distribution Costs to and Including a Bulk Terminal
For a Wyoming Methanol Plant, fc/mBtu
(1981 dollars)[1,17]
Pure Gasoline
Methanol DHR Revised
Pipeline, 1000 miles to .48 .26 .35
Minneapolis, Chicago, or Kansas
City
Bulk terminal operating cost* .15 .15 .15
Added Storage in bulk terminal .10
.73 .40 .50
Pipeline
Capital Cost; (millions of dollars)
1000 mile, 14 inch diameter 165
1000 mile, 10 inch diameter 118
* No interest charges are made for the stored fuel.
-------
-29-
D. Distribution Costs from Bulk Terminal to Retailers
(Local Distribution)
Local distribution is primarily done by tanker truck. Since
trucks would be making periodic trips, as opposed to the fixed
pipeline carrying fuel, it is likely that the costs per unit
volume would remain fairly constant. Some economies of scale
could probably be realized from a switch to methanol, but overall
more trips will have to be made since trucks cannot increase in
size by the necessary amounts due to state weight limitations.
.This approach was followed in Table 9, which gives the costs
of transporting synthetic gasoline and methanol by tank truck for
distances from 5 to 75 miles. These costs are based on trucks
with 9,000 gallon capacities with one stop delivery and an empty
backhaul. Capital costs account for roughly 40 percent of total
costs while operating and maintenance accounts for the remaining
60 percent. Tank trucks cost $76,000-82,000 each with $49,000-
55,000 for the tractor and $27,000 for a trailer having vapor
recovery equipment.[1] For the purposes of this study, a typical
distance of 50 miles will be used. Thus, local methanol distri-
bution would cost $0.28 per mBtu while synthetic gasoline would
cost $0.14 per mBtu to transport this-distance.
E. Retailer Costs
The costs of retailing fuel are more like those of long-range
distribution than local distribution. That is, the costs of
retailing are primarily fixed costs, such as land or rent. Unlike
the case with long-range distribution, however, large capital
investments should not be required. Instead of installing more
storage tanks to facilitate the sale of larger volumes of meth-
anol, it would be more likely that the service station would mere-
ly receive more frequent deliveries from the bulk terminals.
Instead of filling their methanol tanks once a month, for example,
the retailer would fill these tanks every two weeks. Such a
procedure would avoid the capital costs associated with installing
new storage tanks, unless of course his existing tanks are not
compatible with methanol'. Retailing also differs from both long-
range and local distribution in that the critical marketing factor
is fuel energy used and not volume. The following example should
clarify this point.
First, let it be assumed that gasoline and methanol cost the
same per unit energy (e.g., gasoline is $2.00 a gallon and meth-
anol is $1.00 per gallon). Also, let it be assumed that both
gasoline and me.thanol engines have the same fuel efficiencies, so
that a person with a methanol-fueled auto buys twice the volume of
fuel as a person with a gasoline-fueled car. With methanol, each
retailer would sell twice the volume of fuel as compared to gaso-
line, but the same amount of energy, and would obtain the same
amount of revenue. His operating costs would change very little,
-------
-30-
Table 9
Tank Truck Distribution Costs[1,17]
(1981 Dollars)
Distance Cost, $ per mBtu
(Miles) Methanol Gasoline
5 .13 .07
25 .21 .11
50 .28 .14
75 .37 .19
-------
-31-
since his fixed costs dominate and actual pumping costs are negli-
gible.
The above scenario is very reasonable, except that it is
likely that methanol engines will be as much as 20 percent more
fuel efficient than gasoline engines and less total energy will be
required. If this were the case, then each vehicle would require
less fuel (in terms of energy) per fill-up or fill-up less often.
In either case, retailers would see a reduction in sales, but
costs would remain at or very near prior levels. If all stations
remained in business, then the total cost of fuel retailing would
remain constant under this scenario, as it did under the prior
scenario. Since the total amount of energy distributed will be 20
percent less, the cost for retailing methanol per mBtu will be 25
percent higher. However, experience with the current gasoline
retailing situation might say that the current drop in demand for
gasoline is forcing some retailers out of business, and reducing
the overall cost of retailing. In the extreme, then, one might
expect that a 20 percent drop in fuel (energy) usage would result
in the need for 20 percent less retailers and reduce overall
retailing costs by 20 percent. In this case, the cost for retail-
ing methanol per mBtu would be the same as gasoline since both the
amount of energy and the cost of retailing would decrease by 20
percent. Since it is impossible to determine which of these situ-
ations would actually occur, both will be used to bracket the
possibilities.
Typical retailer mark-ups are estimated to be in the range of
$0.05-0.18 per gallon of gasoline.[20] However, since the lower
mark-ups are usually associated with the high-volume stations, the
average mark-up per gallon of gasoline sold in the U.S. should be
nearer to the lower limit, about $0.09 per gallon, or $0.76 per
mBtu. For methanol, the cost would lie between this value and 25
percent more since the total amount of energy distributed would be
20 percent less due to the expected higher efficiency of methanol
engines. Thus, the cost of retailing methanol would be $0.76-0.95
per mBtu.
In deriving these retail costs, no attempt was made to
account for any additional costs the retailer would bear when
methanol is first introduced. For example, he will have to make
some monetary allowance for the initial small volume of custom-
ers. The retailers in some instances will also incur costs asso-
ciated with installing new tanks if the existing ones are incom-
patible or unavailable due to large demands for the specific fuels
they contain. (This topic is addressed in Section V.) The afore-
mentioned retailing costs should therefore be considered as
long-term costs, after the methanol market stabilizes.
As mentioned earlier, in the cases of long-range and local
methanol distribution costs, it was assumed that methanol vehicles
would have the same fuel efficiency as their gasoline counter-
-------
-32-
parts. Thus, the volume of methanol needed was twice that for
synthetic gasoline. Although this approach appears inconsistent
with that followed to determine retail costs, each procedure is
conservative with respect to the estimation of methanol distribu-
tion costs. That is, in all instances the assumptions were made
which would tend to increase the cost of distributing methanol
relative to gasoline. This was done to help assure that the costs
of distributing methanol were not underestimated, since some of
the costs of conversion are inevitably overlooked.
F. Total Distribution Costs
The total cost of distributing methanol and synthetic gaso-
line can now be determined by combining the costs presented in the
last three sections. Methanol would cost $1.69-1.88 per mBtu to
distribute; gasoline would cost $1.34 per mBtu. Gasoline has a
significant advantage over methanol in terms of percentage (21 to
29 percent), but the absolute difference is only $0.35-0.54 per
mBtu.
In order to understand the implications of these higher meth-
anol distribution costs, it is useful to know what fraction of the
total fuel cost is due to distribution. In Table 10! the plant-
gate, [21] distribution, and at-the-pump costs for methanol and
synthetic gasoline are given. As can be seen, distribution
accounts for 13-22 percent of methanol's price at the pump and
8-15 percent for gasoline. Thus, it is clear that the dominating
factor in determining the retail price of each fuel is production
costs.
V. Distribution Capacity Needed to Support a Synfuel Industry
In the first two sections of this report the existing petro-
leum fuel distribution system and technical factors affecting
methanol distribution were discussed. A key question to be
answered when considering the viability of a given synfuel is "how
much distribution capacity is needed to support it?" This section
expands upon the discussions in the first two sections in an
attempt to answer this question. Also, this section will discuss
the total capital cost of implementing this capacity based on the
findings of the previous section.
A. Future Fuel Needs
First, it will be necessary to know approximately how much
methanol or synthetic gasoline must be distributed to fuel the
nation's automotive fleet. Table 11 shows the expected consump-
tion of gasoline and diesel fuel for selected future years as
projected by Data Resources, Inc.[22] To simplify this discus-
sion, it will be assumed that synfuels will be used to meet 20
percent of the nation's transportation fuel needs by the year 2000.
-------
-33-
Table 10
Synthetic Fuel Costs*[21]
( $ per mBtu)
Methanol Synthetic Gasoline
Plantgate Costs 5.90-12.42 7.35-15.29
Distribution Costs 1.69-1.88 1.34
Cost at Pump 7.59-14.30 8.69-16.63
* The range of plantgate costs for each fuel reflect
differences due to capital charge rates of 11.5 and 30 percent.
Bituminous coal was used as a feedstock to each. The methanol
price range is from the Texaco process (low) and the Koppers
process (high). That for synthetic gasoline was based on the
Mobil MTG process.
-------
-34-
Table 11
Future Motor Gasoline Demand
(Billions of Gallons)[22]
1985 1990 1995 2000
Motor Gasoline
Regular
Unleaded
Diesel
89.6
23.2
66.4
20.5
81.1
9.7
71.4
29,8
78.2
9.2
69.0
36.7
78.4
8.7
69.7
4 3 .'8
110.1 110.8 114.9 122.2
-------
-35-
The 122.2 billion gallon total for gasoline and diesel fuel
in 2000 has the same energy content as 267.4 billion gallons of
methanol.[14] Thus, if methanol were chosen to meet 20 percent of
these needs, approximately 53.5 billion gallons of that fuel would
suffice. Similarly, if synthetic gasoline is relied upon for 20
percent of the nation's transportation needs in 2000, then roughly
27 billion gallons of it would be necessary. The total volume of
transportation fuel in the year 2000 under the 20 percent methanol
scenario is roughly 151.3 billion gallons, while that for the
synthetic gasoline option is 122.2 billion gallons.
B. Long-Range Capacity
In order to produce 53.5 billion gallons of methanol annual-
ly, 35 methanol plants would be needed, each producing 100,000
barrels per day. Without knowing where each of these plants would
be located it is impossible to know the exact nature of the asso-
ciated distribution network. To simplify the discussion here and
to represent an upper limit of needed capacity it will be assumed
that new pipelines will be constructed to facilitate all long-
range distribution.
In the previous section it was determined that for western
markets, a 14-inch pipeline would be sufficient to transport
100,000 barrels of methanol per day 1,000 miles to major markets.
Similarly for eastern markets, two pipelines, 10 and 12 inches in
diameter, stretching 100 and 300 miles, respectively, could also
facilitate 100,000 barrels per day to major markets. Assuming
that 17 of the necessary 35 methanol plants are located in the
west and 18 in the east, then there would be a need for 17,000
miles of 14 inch pipelines, 5,400 miles having a 12 inch diameter
and 1,800 miles with a 10 inch diameter.
The capital costs for these individual new pipelines were
given earlier and can be found in Tables 7 and 8. Using these
values, the aggregate cost to the nation for new methanol pipe-
lines for the above methanol scenario would be roughly $4.0
billion.
Following a similar approach for the synthetic gasoline
option, 17,000 miles of 10 inch diameter pipeline would be needed
to connect western production facilities with major markets. For
the 18 production facilities in the east, 5,400 miles of 8.5 inch
diameter pipeline and 1,800 miles of 7 inch pipeline would be
needed. Using the individual capital costs given in Table 7 and
8, the aggregate cost to the nation for the new synthetic gasoline
pipelines would be approximately $2.8 billion. Thus, pipelines to
facilitate the methanol option would cost roughly $1.2 billion
more nationwide than that needed for the synthetic gasoline route.
-------
-36-
C. Bulk Terminal Capacity
Typical gasoline bulk terminals are large enough to facili-
tate an approximate 10 day supply from the refinery.[13] If this
rule of thumb is also applied to methanol, then the average bulk
terminal would need to hold 1 million gallons for each methanol
plant that sends its total production to it. Since, however, a
given volume of methanol is in effect replacing half that volume
of gasoline, the storage capacity that previously went to the
gasoline should be available for methanol. Thus only 500,000
gallons of capacity would have to be added to each terminal for
each 1 million gallons of methanol it would store. From Table 6
it can be seen that a storage tank of this size would cost $719
thousand. The total costs to the nation to construct enough meth-
anol storage tanks at bulk terminals to store the production of
3.5 billion barrels per day would be roughly $25 million. Note
that this cost does not include the purchase of land on which to
locate the additional tanks. This was not included since the
degree such land would be needed and the associated costs are
unknown. These costs, however, should be minimal compared to that
for tank purchases.
For the synthetic gasoline option, it is likely that existing
gasoline terminals would also be accessible to synthetic gaso-
line. Thus no additional tanks at bulk terminals would be needed.
D. Tank Truck Capacity
As noted earlier, the National Petroleum Council estimates
that as of December 1978 there were approximately 40,000 tank
vehicles hauling liquid products.[7] The total capacity of these
trucks was approximately 291.5 million gallons.[7] The average
size truck in such a fleet would have a capacity of nearly 7,300
gallons. For that same year, Data Resources reports a total of
122.8 billion gallons of motor fuel were consumed.[22] Thus, each
tank truck has a yearly haul of approximately 3 million gallons,
or approximately A20 hauls per truck per year.
Since a total of 151.3 billion gallons of fuel (methanol,
gasoline and diesel) are needed under the 20 percent methanol
scenario, there would be 28.5 billion gallons more fuel consumed
in 2000 than in 1978. If the number of hauls per truck per year
remains constant, and assuming the same ratio of tank truck to
nationwide capacity, the same individual tanker capacity as that
sited above, and no spare capacity in either the 1978 or 2000
systems, then an additional 9,300 tank trucks would be needed.
Tank trucks this size cost roughly $78,000 each. [1] Thus the
total cost of 9,300 tankers would be approximately $725 million.
Since nearly the same volume of fuel was consumed in 1978 as is
projected for 2000 under the synthetic gasoline option, no addi-
tional capacity would be needed in 2000 for gasoline over the 1978
level.
-------
-37-
If, however, the trucks dedicated to methanol transfer make
twice as many hauls per year to retail outlets as their gasoline
counterparts then no additional trucks would be needed over that
necessary for gasoline transportation. This is a likely scenario
given the significant cost of tanker trucks.[23] Although this
approach would not require an initial capital investment for meth-
anol delivery trucks, the fact that they would be operating twice
as many miles per year as the gasoline trucks indicates they will
need replacing twice as often. Thus, in the long term, the $725
million additional dollars for methanol tankers will have to be
spent, but the impact would be spread out over a number of years.
E. Retail Capacity
If, as depicted above, approximately the same volume of fuel
was consumed in 1978 as is projected for 2000 under the synthetic
gasoline option, then no additional retail capacity would be
needed for gasoline in 2000 over 1978 levels. Given the likely
scenario whereby retail methanol tanks would be filled twice as
often as gasoline tanks, there would similarly be no need for
additional storage tanks for methanol. Those constructed of
fiberglass laminate will, however, need to be replaced by carbon
steel tanks due to their incompatibility with methanol. Since, as
mentioned earlier, fiberglass tanks will only comprise roughly 25
percent of all retail tanks in 2000, and since methanol will only
comprise a fraction of service station storage needs, the degree
of this substitution is likely to be minimal.
F. Cost Comparisons Between Synfuel Options
The total costs of implementing distribution networks for the
two synfuel options can now be approximated by combining the
individual costs for each distribution component. This has been
compiled in Table 12. As can be seen the distribution network for
a scenario which yields 20 percent of the nation's transportation
fuel needs in 2000 by. methanol would cost roughly $4.75 billion.
A similar network whereby the same fraction of fuel was supplied
by synthetic gasoline would cost $2.8 billion, or $1.95 billion
less than the methanol option.
In order to put these findings into perspective, one must
also consider the differences in the production facility capital
costs associated with each scenario. It has been estimated that a
methanol plant of the size depicted in this study would cost at
least $680 million less than a similar facility which yields gaso-
line. [21] The relative capital costs here are approximately $1.99
billion for a methanol plant and $2.67 billion for the synthetic
gasoline facility.
Thus, to produce enough methanol to meet 20 percent of the
fuel demand in 2000 would require an investment of approximately
$69.7 billion. If the synthetic gasoline route were chosen, the
-------
-38-
Table 12
Capital Costs of Synfuel Distribution in 2000
(Billions of 1981 Dollars)
Pipelines
Bulk Terminal
Tank Trucks
Retail C
Total
Methanol
4.0
lal 0.025
0.725
icity 0.0
4.75
Synthetic Gasoline
2.8
0.0
0.0
0.0
2.8
-------
-39-
total capital costs for the requisite production facilities would
be $93.5 billion. As can be seen, the capital costs for distri-
bution are dwarfed by production capital costs which are 15 to 33
times as great.
VI. Conclusions
It has been noted that the petroleum product distribution
network is very complex and extensive. Since there are few pipe-
lines and navigable waterways near the western coal fields, it is
likely that new pipelines will be needed to support synfuel
production in this region. Railroads are typically located near
these fields and thus could be relied upon in the initial stages
of synfuel implementation. Their high costs relative to pipe-
lines, however, precludes their long-term use. In the eastern
half of the U.S., there are many pipelines and navigable waterways
which could support synfuel distribution to some degree. Here
also, railroads which already connect coal fields with major
markets could be relied upon in the early stages.
The extent to which the existing distribution system can be
used to support a synthetic fuels industry is impossible to deter-
mine accurately due to the lack of information concerning the
exact location of future synfuel plants and unanswered questions
regarding the compatibility of these new fuels, especially meth-
anol, with materials in the existing network.
Material compatibility problems of methanol which have been
identified include: 1) its leaching out of ordinary pipe dope in
pipelines, 2) the dissolution of rust inhibitors from pipeline
inner walls during batching, 3) the deterioration of certain plas-
tics and rubber compounds, and 4) the tendency of methanol fuels
to gum when stored in polyester-fiberglass laminate tanks. None
of these problems appear to be significant at this time. Research
has been initiated by major oil companies to examine these and
related issues in more detail.
The cost of distributing methanol from plantgate to customers
at a retail outlet is approximately $1.69-1.88 per million Btu.
Similarly, synthetic gasoline would cost roughly $1.34 per million
Btu to distribute. These costs are responsible for 13 to 22
percent of methanol's price at the pump and 8 to 15 percent for
synthetic gasoline using production cost estimates developed else-
where .
In order for methanol to meet 20 percent of the nation's
transportation .fuel needs in the year 2000, approximately $4.75
billion would be required to increase current capacity in the
motor fuel distribution network. If synthetic gasoline were
relied upon to meet this same requirement, then $2.8 billion would
be needed. Thus, choosing methanol would necessitate the spending
-------
-40-
of $1.95 billion over that required for synthetic gasoline to
facilitate distribution.
These capital costs are quite small, however, compared to
those necessary to construct the requisite production facilities.
A single methanol production facility yielding 100,000 barrels per
day costs, for example, approximately $680 million less to
construct than a comparable synthetic gasoline plant ($1.99
billion versus $2.67 billion). Thus, the construction costs asso-
ciated with building enough methanol plants to meet 20 percent of
domestic needs in 2000 would be approximately $69.7 billion.
Synthetic gasoline plants necessary for this amount of fuel would
cost a total of roughly $93.5 billion. The methanol route would
therefore be $23.8 billion less expensive in terms of plant costs.
-------
-41-
References
1. "Methanol Use Options Study," DHR Incorporated for DOE,
Draft, Phase I, Vol. Ill, Appendices D-F, Contract No. DE-AC01-
79PE-70027, December 1980.
2. "Environmental Aspects of Synfuel Utilization,"
EPA-600/7-81-025, March 1981.
3. "Petroleum Storage and Transportation Capacities, Vol.
Ill, Petroleum Pipelines," National Petroleum Council, December
1979.
4. Personal communications with William Samp, Willaims
Pipeline Co., Tulsa, OK, December 1981.
5. Personal communication with Richard Tinneal, Conoco,
Houston, TX, December 1981.
6. "Petroleum Storage and Transportation Capacities, Vol.
V. Waterborne Transportation," National Petroleum Council, Decem-
ber 1979.
7. "Petroleum Storage and Transportation Capacities," Vol.
IV, Tank Cars arid Trucks, National Petroleum Council, December
1979.
8. "Petroleum Storage and Transportation Capacities," Vol.
II, Inventory and Storage, National Petroleum Council, December
1979.
9. "BP New Zealand Experience with Methanol-Gasoline
Blends," G.R. Cassels, et al., BP New Zealand Ltd., Proc. 3rd Int.
Symposium on Alcohol Fuels Tech., Asilomar, CA, May 29-31, 1979.
10. "Gasoline-Methanol Fuel Distribution and Handling
Trial, "R.W. Hooks and R. Sagaive, Deutsche Shell AG, PAE-Labor,
Hamburg, West Germany, Proc. 3rd Int. Symposium on Alcohol Fuels
Tech. Asilomar, CA, May 29-31, 1979.
11. "New Zealand's Methanol-Gasoline Transport Fuel
Programme," Graham, E., et al., University of Canterbury, Proc.
3rd Int. Symposium on Alcohol Fuels Tech., Asilomar, CA, May
29-31, 1979.
12. Personal communications with Jerry Hurley, American
Petroleum Institute, Washington, D.C., December 1981.
13. Personal communications with F.B. Killmar, Owens Corn-
ing, December 1981.
-------
-42-
14. Personal communications with Bryce Cecil American
Petroleum Institute, Washington, D.C., December 1981.
15. "Methanol As An Alternative Transportation Fuel," R.
Rykowski, et al, Presented at Sixth Symposium of Env. Aspects of
Synfuel Conversion Technologies, Denver Co, October 26-30, 1981.
16. Personal communications with Robert Maxwell, U.S. EPA,
Ann Arbor, MI, April 1981.
17. "Economic Indication," Chemical Engineering, p. 7,
April 6, 1981.
18. Personal Communication with Bill Taylor, Marathon Oil
Co., Findlay, Ohio, Apri 1982.
19. "Methanol From Coal: Prospects and Performance As A
Fuel and As A Feedstock," ICF, Inc., December 1980.
20. Personal communications with John Ayling, Lundberg
Survey, Inc., North Hollywood, CA, February 1981.
21. "Indirect Liquefaction Processes," John McGuckin, U.S.
EPA, Ann Arbor, MI, December 1981.
22. Energy Review, Data Resources, Inc., Winter 1980.
23. Personal Communications with Lana Butts, American
Trucking Association, Washington, D.C., April 1982.
------- |