Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
ENERGY APPLICATIONS
OF BIOMASS
Edited by
MICHAEL Z.LOWENSTEIN
Solar Energy Research Institute, Golden,
Colorado, USA
FOREWORD
The National Meeting on Biomass R&D for Energy Applications was supported
by the Council of Biomass Energy Technology Sponsors (CBETS) and was
organized and hosted by the Solar Energy Research Institute (SERI). The
Biomass Energy Research Association (BERA) provided technical assistance.
CBETS was founded on July 14, 1983, as a forum for communication and
cooperation among managers of the major biomass energy programs in the
United States, including various federal and state government organizations,
industry institutes, and associations. Council membership includes the American
Public Power Association; the Electric Power Research Institute; the Gas
Research Institute; the Hawaii Natural Energy Institute; the Legislative
Commission on Minnesota Resources; the National Rural Electric Cooperative
Association; the New Mexico Energy Research and Development Institute; the
New York State Energy Research and Development Authority; the North
Carolina Alternative Energy Corporation; the Tennessee Valley Authority; the
U.S. Department of Agriculture; and the U.S. Department of Energy. The
Councils affiliate members include the American Gas Association; the BioEnergy Council; the Biomass Energy Research Association; the Fiber Fuels
Institute; the National Wood Energy Association; the Renewable Fuels
Association; and the Wood Heating Alliance.
CBETS has two primary objectives: programmatic coordination to maximize
the value and usefulness of biomass energy research and development activities
sponsored by Council members; and timely and effective transfer of the results
of biomass energy technology advances to industry and other interested parties.
The national meeting held in Arlington, Virginia, on 13 October 1984, was one
of CBETSs continuing efforts aimed at achieving the goal of timely technology
transfer. Attendees represented industries, colleges and universities, federal and
state governmental agencies, and foreign countries.
Each of the three sections in the following Proceedings has a specific focus.
Section 1 contains discussions of issues important to the various sectors of the
biomass energy community; Section 2 discusses in detail the research interests of
biomass energy sponsors; and Section 3 provides highlights of significant
biomass energy research efforts.
vii
The organizers thank all the authors for having contributed their reports and
the CBETS members for initiating and assisting with the planning that led to the
success of the meeting.
Golden, Colorado
M.Z.Lowenstein
August 1985
CONTENTS
Foreword
2.
10
3.
17
4.
30
46
7.
55
8.
63
9.
66
75
10.
ix
11.
89
12.
100
13.
114
14.
120
15.
130
16.
135
140
18.
147
19.
158
20.
169
21.
185
22.
197
23.
219
24.
229
J L KUESTER
25.
240
26.
255
27.
266
List of Attendees
275
SECTION I
Issues of Importance to Biomass Energy
Research
SYNOPSIS
As a result of the depletion of our supplies of oil and gas, which were created by
the process of photosynthesis, it is now necessary to develop renewable fuels for
the future because of the environmental problems associated with the expanding
development of coal and oil shale, particularly the carbon dioxide problem. The
most immediate source of renewable fuels is annually growing green plants,
some of which produce hydrocarbons directly. We can select new plant sources
that have high potential for production of liquid fuels and chemicals. Suggestions
are made for the modification of both the product character and the productivity
of the plants. Ultimately a totally synthetic device will be developed for the
conversion of solar quanta into useful chemical form, completely independent of
the need for arable land.
1
INTRODUCTION
We are here largely because of the awakening that occurred in 197475 as a
result of the action of the OPEC nations, and we are now more conscious of
energy problems than we once were. American energy use for 1983 (Fig. 1) still
emphasized fossil fuel energy (natural gas, coal, and oil) as the major sources.
These are basically the products of ancient photosynthesis when plants were laid
down and converted from primarily carbohydrate into carbon, hydrocarbon, and
gas. The other energy sources (nuclear, geothermal, and hydro) are relatively
minor.
How long can we expect fossil fuel energy to continue as a major source? One
of the simplest ways of answering that question is to examine the energy costs of
finding and extracting a barrel of oil. One way of expressing that cost is to
determine the number of barrels of oil per foot of well drilled, which was 35 in
1945 ; this quantity in 1975 was less than 18. Another way of expressing these
same data is to say that the energy cost today for finding and extracting that
barrel of oil has risen, and we are very close to the edge of energy economy. A
barrel of oil contains 6 million Btu, and when drilling and exploration uses more
energy than the energy content of the barrel itself, it becomes questionable
economically to sell that barrel of oil (Ref. 1).
2
ECOLOGICAL CONSTRAINTS
Major coal producers in the three major coal-producing countries of the world
(the Soviet Union, China, and the United States) consider that coal is a viable
energy source to replace petroleum and natural gas, and it is likely that in the
planned economies, at least, coal will become the major energy source. A reason
for concern is that even today, when coal represents less than 25% of the total
energy usage, our biosphere cannot absorb the carbon dioxide at the rate of
injection that burning fossil carbon in all its forms produces (Refs. 2, 3). Today
the C02 is removed from the atmosphere at roughly half the rate at which it is
injected. If coal production increases markedly, the removal of C02 from the
atmosphere will be a still smaller fraction of what we inject, meaning that the
rate of rise of the CO2 concentration in the atmosphere will double. The C02
concentration has risen from 295 ppm in 1880 to 330 ppm in 1980, and if the use
of coal greatly increases, the C02 concentration will rise at an even more
substantial rate. The amount of C02 produced from the combustion of coal is
roughly twice that from petroleum or natural gas because combustion of coal
burns carbon only, and not carbon and hydrogen.
The rising C02 level produces the greenhouse effect as a result of the peculiar
properties of the C02 itself. Concomitantly, the temperature measured at various
places on the earths surface has also been rising in the last 100 years, increasing
about 0.4C from 1860 to 1980, which represents a very large rise (Ref. 4). One
of the best ways to show the result of this increase in temperature is to examine
the satellite photographs of the South Polar ice cap. One can estimate the amount
of ice that has disappeared in the last 20 years as about 1.2 million km2 from an
approximate total of 12 million km2 (Ref. 5). Thus the global mean sea level has
been rising about 2 to 3 mm per year during the last 50 years, compared to about
1 mm per year during the previous half-century (Ref. 6). A recent report by the
Environmental Protection Agency (Ref. 7) projects a global rise of sea level
between 144 cm (4.8 ft) and 217 cm (7 ft) by the year 2100, with the low
estimate of 56 cm (1.9 ft) and high of 345 cm (11 ft). Also, along most of the
Atlantic and Gulf Coasts of the United States, the predicted rise will be more
than the global average; i.e., 18 to 24 cm (0.6 to 0.8 ft).
It is therefore obvious that to alleviate the trend of rising C02 concentration
and its subsequent problems, an alternate energy source must be found (Refs. 8, 9).
The best annually renewable source of energy is the green plant, and we are
looking for green plants that can produce hydrocarbons from carbohydrates.
There are many such plants, in several different families, and they are found all
over the world.
3
BIOMASS FOR RENEWABLE ENERGY
As you know, the primary productivity of biomass is carbohydrate (sugar, starch,
and wood) which must be converted into a much more concentrated form, such
as hydrocarbon. One of the first efforts has been made in Brazil where sugar
cane has been used directly as an energy source as well as a source of
carbohydrate. In 1983, 4.3 x 109 L of fermentation alcohol were produced from
sugar cane on the autonomous sugar cane plantations. This fermentation alcohol
is used directly in automobiles and is a chemical feedstock for Brazilian
industry.
In Puerto Rico an energy cane has been developed that can be used not only for
its sugar content but also for its total energy content to fire the boilers in heating
plants on the south coast of Puerto Rico (Ref. 10).
Some plants, such as Euphorbia lathyris, a member of the Euphorbiaceae
family, produce hydrocarbons directly from C02. Plantations for E. lathyris have
been developed not only in the United States (Ref. 11) but in Spain as well (Ref. 12).
Euphorbia produce 8% oil and 20$ sugar upon extraction of the whole plant.
Another species, Asclepias speciosa (milk-weeds) , produces approximately the
same combination. That species has been studied extensively in Utah (Ref. 13).
The seed oils are also being developed (Ref. 14). For example, sunflower seeds
produce an oil that can be used directly in a mixture as a diesel fuel or easily
converted by transmethylation to a diesel fuel without any additives by replacing
the glycerine of the triglyceride with methanol. The best commercial seed
oil producer is the palm, which is being grown on a large scale in Brazil and
Malaysia as a source of oil for fuel and materials.
The processing sequence to recover oil (terpenoids) and fermentable sugars
from E. lathyris (Fig. 2) was worked out in the laboratory and is calculated here
for 1000 dry ton/day of material, which would yield 80 ton of crude oil and 200
ton of fermentable sugars that could produce 100 ton of alcohol (Refs.15, 16).
About 500 ton of bagasse are used to run the process, with a resulting 200 ton of
bagasse that could be used to distill the alcohol. The fermentation alcohol that is
a by-product is, of course, a starting point for an entire petrochemical industry.
The whole process is self-contained.
Studies of E. lathyris have shown that while it is possible to crack the material
from this plant, it might be more economical to determine whether this species
and its products contain more useful materials for chemical feedstock production
(Ref. 16). The crude oil from the E. lathyris has been converted, using special
zeolite catalysts, to the usual products such as olefins, paraffin, aromatics, and
nonaromatics. This confirms the desirability of the products of E. lathyris as
possible raw materials to substitute for crude oil. It now appears that a price of
$100/barrel for the oil from E. lathyris, only 2.5 times more than the 1982 OPEC
price per barrel of crude, might be a realistic projection. The price will almost
certainly turn out to be less when larger-scale energy agriculture operations are
commenced.
The plant chemicals that constitute the black oil are mostly triterpenes (C30),
sterols, and sterol esters, which can be cracked to make desirable products. The
biosynthetic route to the terpenes in plants such as E. lathyris has not been
completely determined, but it is probably similar to that for rubber biosynthesis
except that the end products from the E. lathyris are lower molecular weight
compounds. The route is from sugar via the glycolytic cycle to pyruvate, which
then builds up to mevalonic acid and goes on to give isopentylpyrophosphate
(IPP). The IPP polymerizes into a variety of isoprenoids, and in E.lathyris the
material goes through the isoprenoid biosynthetic pathway to squalene (C30),
which is then folded up to make the C30 terpenoid (steroidal) alcohols that are the
greater percentage of the oil.
We also looked for trees that might produce oil directly when they were
tapped or had fruits extracted. Such an agronomy would conserve soil and water.
One tree in Brazil, the Copaifera multijuga of the family Leguminosae, does
exactly this. It grows in the Amazon area of Brazil, and the material from the tree
can be used directly in a diesel engine. It is tapped twice each year and produces
20 L of cyclic sesquiterpene oil per tap. The Brazilians are now beginning
plantation experiments with these trees to improve yields of oil (Ref. 17).
Another family of trees, the Pittosporaceae, has representatives in various parts of
the world. In the Philippines the fruit of the Pittosporum resiniferum is used by
the natives for illumination and for its oil content. The fruit, about the size of a
prune, contains about 50% terpenes of three different componentsmyrcene,
alpha-pinene, and limonene (Ref. 18). These three components can be used
almost directly as cracking stock or fuel. A species that grows in California,
Pittosporum undulatum, has smaller fruits with slightly different oil content.
Another tree, Jatropha curcas, which is a member of the Euphorbiaceae family, is
being cultivated on a large scale in Thailand; this tree is closely related to the castor
bean.
Another type of plant that does not involve agricultural land directly and
which produces hydrocarbon is the microalga Botryocoocus braunii, which can
grow in either fresh or salt water. The algae secrete an oil that is roughly 50%
70% of the dry weight and is mostly C30, which can be cracked in the same way
as crude oil (Ref. 19).
The energy yields for different plants are shown in Table 1. It is clear that the
most important component in the table is the energy in liquid fuels as
hydrocarbons in millions of Btu per acre per year per inch of water. There is no
question in my mind that it would be possible to introduce into the United States
a substantial energy agriculture, if there were the right type of economic
incentives for such a biomass production (Ref. 20).
The projections for energy use in the United States for the year 2000 (Fig. 3)
for the first time include biomass as a significant component of the total,
representing about 6% of the total energy requirement. This is an important
TABLE 1
XBL 8493892
ACKNOWLEDGEMENTS
The work described in this paper was supported, in part, by the Assistant
Secretary for Conservation and Renewable Energy, Office of Renewable Energy,
Biomass Energy Technologies Division of the U.S. Department of Energy under
Contract No. DE-AC0376SF00098.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
Fig. 2. Conceptual processing sequence to recover oil and fermentable sugars from
Euphorbia lathyris
Fig. 3. Projections for energy flow in the United States for the year 2000
After more than ten years of study, by numerous organizations, that have led to
reports that must weigh tons, the size of the U.S. biomass resource base still is
largely unknown and estimates are highly uncertain. The reason for this
ambiguity is not difficult to trace. Unlike fossil fuel resources that are discovered,
biomass resources are manufactured by photosynthesis. The principal resource
base is land of appropriate quality. But such land can be used for other purposes.
Furthermore, the quantity of biomass resource that can be produced on a given
land (or aquatic) area varies with the technology employed to grow the biomass.
For these reasons, the size of the resource base cannot be known with a great degree
of precision or accuracy.
Despite the uncertainties, the size of this resource base is important for
decisions to be made by governments and industrial organizations. If the biomass
resource base were barely sufficient for food and fiber production, it would be
improper to divert some of this precious resource to fuel and chemical use. On the
other hand, if the resource base is likely to be underutilized and low prices are
likely to prevail because of farm surpluses, an additional set of major uses could
strengthen the entire biomass system.
1
BIOMASS RESOURCE ABUSE
Certain syndromes are caused by abuses of biomass; they need to be confronted
and banished so that the potential of biomass can be measured. The abuses are
more serious than they sound.
1.1
Quadromania
Biomass energy systems tend to be small compared with nuclear and coal-based
energy systems. This is one of the advantages of biomass. Quadromania is
insistence on multiple quad output from all biomass energy systems.
Quadromania is unrealistic and impedes progress because it encourages rampant
extrapolamania.
11
1.2
Rampant Extrapolamania
Not only government officials and private sector executives have biomass abuse
problems. Those engaged in both technological and economic research and
development activities must share the blame for biomass abuse. Rampant
extrapolamania takes on many manifestations. Here are a few examples.
Hyperyieldemia. The researcher obtains a high yield with a small plot that is
tended regularly by graduate students who water it, weed it, and remove every
insect that approaches it. This same high yield is extrapolated to all farmland that
is considered to be appropriate for growing the crop at all. The yield estimate
should be reduced as the quantity of land to be devoted to the crop is increased.
Hyperareomania. Hyperareomaniacs find a crop with good yield and project a
large number of quads of energy from the crop by using every plot of land that
they can in the United States to grow the crop, even if the land area is needed for
food, paper products, highways, and cities.
Liebnitzs Syndrome.Liebnitz stated that the world was getting better and
better. The comparable biomass abuser takes relatively low yields and/or high
costs and extrapolates to unproven improvements to obtain a rosy future.
Bunyans Syndrome.Many of those who extrapolate the use of forest residues
for biomass energy production forget that substantial transportation obstacles
impede delivery of these products to central plants for efficient production of
fuels or chemicals. These biomass abusers assume that Paul Bunyans free time
and energy is going to carry the forest residues to the central location.
Chronoverleukemia. New biomass technologies may take 5 to 15 years of
steady development before the first commercial facility is built. Then, companies
tend to observe the success or failure of the first one or two plants for years
before rapid construction of an entire industry ensues. The time required for
initiating such an industry frequently is overlooked.
2
PRICE/VOLUME RELATIONS
Realistic estimates of biomass resource size need to begin and end with estimates
of realistic prices. Recent Gas Research Institute studies, (Refs. 1, 2) indicate
that:
The gas industry is seeking to have gas available from biomass for
approximately $6.00/106 Btu in the year 2000 (1984 $).
Approximately half of the gas cost is likely to be consumed in processing the
biomass to methane, leaving about $3.00/10 Btu for biomass cost.
These price considerations are not to be considered permanent, but they provide
broad guidelines as to what constitutes biomass resources that are likely to be
exploited versus those that are likely to be ignored.
Given the energy content of a crop (10 Btu per ton), the crops likely yield in a
given geographical area (tons/year), and the guidelines for what gas companies
might pay ($/ton), an estimate of revenues for farmers or forest managers can be
derived ($/acre year). This revenue can be compared with agronomic or
silvicultural costs of managing the land for this purpose; in agricultural circles
this comparison is known as a crop production budget, which is the sum of the
cost of the seed, fertilizer, and other inputs plus the cost of land and capital plus a
reasonable charge for labor and profit. If the revenue to be obtained by selling
biomass is likely to exceed the crop production costs, biomass production for fuels
or chemical applications becomes interesting. It still may not win out over
alternatives such as growing crops for food or fiber applications. However, any
positive value for revenue over crop production budget costs is likely to receive
attention at this time when agricultural profits are not outstanding.
3
AN EXAMPLE: CORN/SOYBEAN LAND AS A BIOMASS
RESOURCE BASE
More than 150 million acres now are devoted to production of corn and soybeans
in the U.S. grain belt. Most of the corn/soybean farmers are making small profits
at best, and enough have been losing money consistently to cause widespread
alarm. Therefore, corn/soybean farmers could be quite interested in production
of alternative crops for energy production. Applying the simple relationships
suggested in the previous section of this paper, the following conclusions can be
drawn.
Crop production budgets indicate that corn/soybean farmers need about $300
per acre of revenue, preferably more.
If the biomass energy buyer plans to convert the biomass into methane that
has value of $3.00/10 Btu of methane allocated for biomass, and if each ton of
biomass provides 10106 Btu of methane, then one ton of biomass is worth
$30 to a natural gas utility.
Given that the farmer needs $300 of revenue and will get $30 per dry ton, then
the required biomass yield is 10 dry tons per acre per year.
A yield goal of 10 dry tons per acre in the grain belt is higher than yields
obtained with conventional corn grain or silage corn; however, sweet sorghum
can produce such yields in at least part of the grain belt. Therefore, this back of
the envelope calculation helps to clarify the type of crop that might be useful on
current corn/soybean land for biomass fuel purposes.
13
One of the reasons that such a high yield is required for natural gas production
is that the lignin contained in the biomass makes no contribution to natural gas
production. This constraint does not apply to biomass grown for combustion to
generate steam and electricity. Because these buyers might purchase low-sulfur
coal instead of biomass for approximately $2/10 Btu, they might not be willing
to pay $3 for biomass. However, many forms of biomass produce approximately
15 x 106 Btu/ton. At $2.50/106 Btu, the allowable biomass cost would be $37.50/
ton. With the needed revenues set at $300/acre, a yield of 8 tons/acre year could
be tolerated. A much greater percentage of grain belt land could yield 8 tons/acre
year than could yield 10 tons/acre year. The preferred crop might be a variety of
sorghum, but silage corn or other crops also may be appropriate.
The size of the biomass resource base that would be created by diverting part
of the corn/soybean land to energy crops is difficult to estimate for several
reasons. As land is diverted from corn and soybeans to energy crops, their price
will rise and thus reduce the attractiveness of producing energy crops. On the
other hand, gas and electric utilities could offer long-term contracts to farmers
that could free them from the vicissitudes of the corn/soybean business.
Considering the volatility of the corn/soybean prices, 10%20% seems to be the
maximum that might be diverted to energy crops and 5%10% appears to be
practical in a non-energy-crisis situation. The quantity of biomass thus made
available would be on the order of 120 million tons per year, which is equivalent
to about 1.8 quads.
4
OVERALL BIOMASS POTENTIAL
Using methods that are analogous to those described for corn/ soybeans, the
overall biomass potentials shown in Table 1 were derived. The entries other than
corn/soybean land need to be explained briefly.
Table 1:
Overall biomass potential
Corn/soybean land
Diverted cropland
Latent farmland
Agricultural residues
Forest biomass
Total
106 tons/yr
input quads
120
100
100
50
200
570
1.8
1.5
1.5
0.8
3.0
8.6
Diverted cropland is defined as the acreage that farmers are paid to leave
fallow so that prices of grains and other major commodities will not be
reduced too much. This land is good cropland but generally not quite as good
as the land in production. In recent years the diverted cropland has ranged
from about 16 to 80 million acres per year, and the program costs taxpayers
billions of dollars per year. Production of 100 million tons per year of
biomass on this land could save taxpayers money and reduce the federal
budget deficit.
Latent farmland is land that currently is in pasture, rangeland, or forests. This
land is reasonably level and is of sufficiently high quality to be reclaimed as
farmland. However, it is less likely to be used because it may not be
controlled by those interested in energy crops.
Agricultural residues total approximately 450 million tons/year, but only a
small fraction is believed to be concentrated enough in its availability to be
worth collecting, especially considering the potential for soil erosion when it
is removed. The degree of geographical concentration of this residue is
illustrated in Fig.1.
The forest biomass resource potential appears hugh because the inventory is
growing every year in all geographical areas in the United States except for the
Pacific Coast (Table 2). Furthermore, substantial percentages of the trees that
currently are harvested are not managed or put to good use. In addition,
considerable progress has been made in recent years in developing silvicultural
plantations that have potential for high yields of biomass on a sustainable basis.
When relatively conservative discounts
Table 2:
Growing stock inventories of trees (million dry tons)
region
1976
North
2,850
South
3,230
Rocky Mountain
1,497
Pacific Coast
3,552
United States
11,130
Source: USDAForest Service 1982.
1990
2000
3,603
4,003
1,615
3,124
12,346
4,030
4,435
1,693
3,043
13,201
are taken for the numerous problems that are involved in growing, collecting,
and transporting forest biomass, there still appears to be an opportunity for 200
million tons per year, or approximately 3 quads.
A major issue arises as to whether it is legitimate to add together the items
shown in Table 1 to obtain a total energy potential of almost 9 quads. In the
opinion of this author, using the total of these possibilities would be a form of
biomass abuse. Only an extreme energy crisis could or should mobilize this
overall biomass potential. It is more likely that one, two, or three of the
alternatives will prove to be viable and will gain the attention and interest of
15
those developing biomass energy. The others will remain latent until the
economics or technology changes.
Currently, there is considerable pressure to reduce the federal budget deficit
and to help corn/soybean farmers. Therefore, the corn/soybean land, diverted
land, or possibly both (which are interrelated in any event) could be mobilized.
The forest products industry is making considerable progress in making its own
energy by managing its biomass resource steam and electricity for captive use
(Ref. 3). The momentum thus gained may be extrapolated into sale of more
electricity to the grid on a commercial basis in areas that have high avoided
electricity costs and large biomass resources.
5
CONCLUSIONS
This author believes that biomass has considerable potential as a source of fuels
and chemicals. The overall potential is the sum of many financially small and
local decisionsa situation that may be frustrating for government planners but
is actually an assurance of ultimate success. We all need to recognize that
government agencies strive for large-scale and widespread use of results,
biomass researchers strive for high yields, industry strives for profits, and
bankers strive for safety of capital. Pursuit of these goals is not biomass abuse but
being blinded by these goals is the source of biomass abuse.
6
ACKNOWLEDGEMENTS
The author thanks the Gas Research Institute and Electric Power Research
Institute for partial support of the research described in this paper.
REFERENCES
1.
2.
3.
Mishoe, J.W., Boggess, W.G. and Kirmse, D.W. Biomet: A simulation model for
study of biomass to methane systems, Preprints of the 1984 International Gas
Research Conference, 459463, Aug. 1984.
Lipinsky, E.S., Young, B.A., Sheppard, W.J. and Jenkins, D.M. Review of the
potential for biomass resources and conversion technology status, Preprints of the
1984 International Gas Research Conference, 440 450, Aug. 1984.
Lipinsky, E.S. and Anson, D. Research and development opportunities, in
Proceedings: EPRI/TVA Workshop on the Use for the Generation of Electric Power,
EPRI AP-3678, Sept. 1984.
involved with peasant agriculture and therefore retains very low status, meagre
funds, and inadequate research (Refs. 2, 4, 5).
3
FOOD PRODUCTION
The amount of energy the world obtains from food and fuel in a year represents
only about 10% of the annual photosynthetic energy storage. In other words,
photosynthesis already stores ten times as much energy as the world requires
(90% of it in trees)an amount equivalent to our proven fossil fuel reserves
(Refs. 6,8).
Human croplands represent only 0.53% of all standing phytomass and
currently occupy 11% of all land area. The worlds flora has over 1 million
species, but only 10 annual grains (rice, wheat, corn, sorghum, millet, rye, barley,
common bean, soybean, and peanut) now provide 80% of the plant nutrients we
consume, and 75% (dry matter) of annual food production (Ref. 5).
Between 1950 and 1980 world food production doubled. The rate of increase
in grain production per capita slowed since 1970, reaching a peak in 1978, then
actually declining for two years. For the most part this was due to governments
in Canada and North America paying farmers not to grow grain in an effort to
maintain world prices, to bad weather, and to poor harvests in Russia. Since
1980, when world stocks of grains stood at 186.7 million t, the trend has actually
been a steady increase. World grain stocks will be around 206.9 million t in
198485. The grain harvest for 1984 is expected to be 1600.9 million t with
record and near record crops in the United States and the European Economic
Community (EEC) (Refs. 14, 7).
Despite the increases in agricultural production in the developing world
120% since the early 1950sthe average per capita increase in food has been
modest, while in Africa (except white South Africa) there has actually been a
continued decline in food supply of around 15% (Refs. 7, 8).
Because of its more modest population increases, the developed world has
seen a substantial upward trend in per capita food production. Crop yields
expanded at three times the rate of those in the developing world. Figures for
both corn and wheat yields in the United States in the past 30 years have seen
threefold increases.
Despite increased production, maize farming took up 6.7 million fewer
hectares in 1980 than it did in 1940. UK wheat yields increased by 98% and
barley by 62% between 1954 and 1981; and they are still climbing.
In India, the 1982 rice yields, although variable throughout the country, were
3.3 t/ha, a 99% increase since 1960. And the 1984 crop was 58.5 million t27%
higher than in 198283. Even so, potential rice yields in the tropics are far
higher, around 1315 t/ha. Wheat yields have risen from 0.65 t/ha to 1.85 t/ha in
the past 30 years (Refs. 9, 10, 11, 12, 13, 14).
A recent study by the FAO (Ref. 17) estimated the size of population that
most third world countries could support using three levels of analysis: low-level
inputs, approximately equivalent to subsistence farming; high-level inputs,
approximately equal to Western farming technology; and intermediate inputs,
halfway between these two farming systems.
The result was surprising. Even with low levels of input, the less developed
countries could support about one and one-half times their projected population
(3600 million people) at the end of the century. At the highest level of input,
Africa could theoretically support 16 times the population estimated for the year
2000, and South America 3 times. However, these figures relate to an egalitarian
world, where rich and poor eat the same, mostly vegetarian, minimum diet.
There was no provision for nonfood crops, woodfuel, or fruit; about 66% of the
tropical rainforest would have to be felled, and every available acre of
cultivatable land would be utilized. Taking these factors into account it is
reasonable to reduce the estimate of potential global food production by at least
one-third. And when the area most likely to be under cultivation is taken into
account, by 2025 only Asia will be able to support its population with low
inputs, while Africa will only be self-sufficient for 40% of its people. Central
America would have to go beyond intermediate levels to feed its people
unlikely following present trendswhile even at highest levels the Middle East
would only be able to feed about 65% of its population (Refs. 17, 40).
Some countries, mostly in central Africa and South America, have brighter
prospects. The point to be emphasized is that this FAO study indicates that the
planet could support the predicted 10 billion world population of the future. It
suggests some practical ideas on how to cope with the problem: for example,
conservation measures alone could reduce the number of critical African
countries by seven; and growing crops more suited to the soil and climate of a
particular area could increase Africas food production by 58% without any extra
inputs.
About 400 million peoplea tenth of the worlds populationare malnourished; and this figure is probably still increasing. Another 600 million live
constantly on the brink of undernourishment. Fifteen million children die each
year from starvation. Forty million human beings die from hunger and its related
diseases every year (Refs. 6, 4).
This misery is avoidable even now. The world grows sufficient food that, if
equally distributed, would provide a nutritious diet for everyone. For the
developed countries this would mean the consumption of more basic staples and
considerably less animal protein. Effective distribution of food within a country
is essential, as is dramatically illustrated by Sri Lankas long life expectancy (66
years in 1975, 39% higher than would be predicted from world norms) and
infant mortality (67% lower than would be expected) as results of such a
programme (Ref. 15). One of the most expensive food distribution projects ever
undertaken, its beneficial effect is emphasized when compared to the situation in
India. In 197374, although enough food was produced to provide an adequate
diet for all Indians, 38% of the population had a deficit in daily calorie intake;
the poorest 5% had a deficit of 1100 calories per day (Ref. 16).
On average, a person in the developing world gets about 188 calories per day
from animals; the figure is 1073 calories in the developed countries. About 40%
of all cereals producedand 90% of the soya bean productiongoes into
feeding livestock; and in the richer countries it is nearer 75% (90% in the United
States) . Diverting only one quarter of the world soya bean harvest from feed
use to direct human consumption would provide 5 kg per year of high protein
food (750 kcal per day) for everyone in the world. But the situation is not so
simple. Most feed use of cereals occurs in the developed countries. Even if many
people are desperately short of food, not using cereals as livestock feed would not
mean they would reach the needy. The actual situation is thus very complex, but
the worlds food problem is really a political one, and essentially one of
irradicating poverty. The money required to provide adequate food, water,
education, health and housing for everyone in the world has been estimated at
$17 billion a year. It is a huge sum of moneyabout as much as the world spends
on arms every two weeks (Refs. 18, 5).
4
BIOMASS
The realisation that the production of power alcohol from sugar and starch crops
is the quickest means of introducing a new liquid fuel source is receiving
worldwide recognition. Individual countries such as Brazil, the United States, the
Philippines, Malawi, Kenya, and Zimbabwe have made substantial investments
in fuel alcohol production. Worldwide, yearly government expenditures on power
alcohol schemes are over $2 billion. The largest programme is in Brazil, where
the government currently spends around $1.3 billion on subsidizing the
production of fuel alcohol, mostly from sugarcane. The aim is to increase
production from the 2.4 billion L produced in 1978 to 10.7 billion L by 1985,
and reaching 14 billion Lwhich would replace 75% of the 197879
consumption of gasolineby 1987. If, as seems likely, the 1985 target is met,
the alcohol produced will mean a 60% saving on petrol (Ref. 19).
Over 1 million cars now run on hydrated alcohol in Brazil, with the remaining
7 million or so using a 20% alcohol blend. In 1983 the number of pure-alcoholfueled cars produced accounted for 75% of vehicle production and 90% of all
sales. Recently a new policy has been initiated to make alcohol production more
self-sufficient.
Between 1976 and 1982 the land used to grow sugarcane increased by around
1.2 million hactares and if the 1985 target of 10.7 billion L of ethanol is to be met,
a further 1.6 million hactares will have to be planted in relation to the land area
planted in 1982. But in 1980 sugarcane accounted for only 5.4% of the
approximately 48 million ha of cultivated land, and only 28% of the total sugar
crop went directly to produce alcohol. A more serious source of competition for
agricultural land came from the everincreasing area devoted to cash crops,
which from 1976 and 1982 was between 2.2 and 2.6 million ha. Soybeans alone
took up 50% more land than did sugarcane. Industrial crops have increased by
some 46%, while food crops only accounted for 18% of this new agricultural
land and have actually decreased from 64% to 59% of the total area cultivated
(Ref. 19).
The potentially cultivatable land (between 226 and 268 million ha) is more
than enough to accommodate the extra 4.8 million ha required to meet the 1989
alcohol production target. And by increasing both the yield and sugar content of
sugarcane, and making more efficient use of the distilleries, the yield of alcohol
from a tonne of sugarcane could be improved by about 45%from 65 L to 90 L/
t and productivity from 3600 L to 7200 L/haan improvement of 100% (Ref. 20).
Production costs for alcohol are about $50$55/bbl of petrol replaced. Petrol
made from imported oil costs $41/bbl at the refinery. Taking the considerable
foreign debt and the interest charges into account, alcohol production in Brazil
can be justified on purely economic grounds (Ref. 19).
Zimbabwes alcohol production began in 1980 in a plant designed to produce
120,000 L/day, or 10 million L/year, about 12% of the countrys petrol
requirements. In 198384, there were approximately 34 ,000 ha of cane planted,
producing some 410,406 t of sugar. The area of land needed to grow cane to
provide alcohol for all liquid fuel requirements plus all domestic sugar needs
(but not for export) would be about 52,000 haless than double the present area.
Land is relatively plentiful, so there would be little competition with food (Ref. 21).
Production of alcohol reduces foreign exchange earnings. However, because
the price for sugar on the international market is notoriously variable, and a
considerable strategic advantage is gained from greater self-sufficiency, plus the
fact that alcohol is a renewable resource, the balance is more than in favour of
home alcohol production for Zimbabwe (Ref. 21).
Since U.S. alcohol production comes largely from surplus grains, its future
will depend on technological developments, economics, and considerations
concerning the provision of food for the hungry Third World. A present U.S.
food aid stands at around 5 million t, about 5% of exports, and about the limit
that the government or the public seems willing to accept. In theory, the United
States could convert its export grain (around 123 million t) to alcohol, with no
effect on the domestic market. This would produce 11.5 billion gal of ethanol,
around 10% of the annual petrol consumption, replacing 164 million bbl of oil
and saving 7.1% of the annual oil imports. However, in 1980 prices, the ethanol
produced would be worth $6.6 billion, while the value of the exported corn alone
was worth $7.7 billion (Ref. 22).
The U.S. alcohol industry now produces about 500 million gal from a total of
65 plants (80% from only 8 of these plants). Assuming a 10% blend, the
potential demand is 6 billion gal of ethanol, or 12 times the present capacity.
However, it is expensive to produce, with corn accounting for 50% of the price.
Ethanol currently costs $1.70 to make, so that a subsidy of 80 cents/gal to bring
the price down to the production cost of gasoline is essential to the survival of
the alcohol industry. But the production of ethanol has to be seen against a
backdrop of economic, strategic, and political factors. In 1983, the cost of
policing the sea routes to ensure uninterrupted supplies of oil amounted to double
the market price of the oil actually exported from those areas. And in the same
year, the cost of maintaining a strategic oil reserve was five times the amount of
tax relief given to the alcohol industry (Ref. 41).
Consumption of power alcohol has risen from almost nothing in 1979 to 455
million gal in 1983. But what would be the impact of alcohol production on U.S.
agriculture? The production of a small amount, 2.5 billion gal requiring 1 billion
bushels of corn, would have little effect on food prices. The main problem would
seem to be the moral one of turning so much food into (relatively) so little fuel.
If 10 billion gal of ethanol were produced, the required 4 billion bushels of corn
would need an additional 41 million acres, while the soybean crop would be
reduced by 25 million acres. Corn prices would go up by between 20% and 40%
(Refs. 22, 23, 24).
The United States, and more especially Brazil, have developed alcohol
programmes on the strength of agricultural surpluses. Is Europe in a position to
do the same with the substantial surpluses produced by the CAP (Table 1)?
Table 1:
Self-sufficiency in certain agricultural products (%).
product
EEC 10 (1)
USA (2)
Wheat
118
315
Rice
83
250
Sugar
124
64
Grain maize
62
160
Soy beans
185
Skimmed milk powder
126
199
Butter
118
121
Cheese
106
99
Beef/veal
102
93
Poultry
108
106
Eggs
101
104
Pork
101
98
Cotton
11
265
EEC self-sufficiency for concentrated milk was 154, whole milk powder 337, barley
112, and rye 107 .
(1) crops, average for 197879, 197980, 198081; animal products, average for 1978,
1979, 1980.
(2) sugar and animal products, average 198182; crops other than sugar: average 1981
82, 198283. Source: Ref. 25.
Although these agricultural surpluses are substantial, they are very small
compared to the volume of the oil market. Between 197881 the potential
quantity of alcohol that could be (expensively) produced from unmarketed fruit
and wine averaged at 223 million L, some 204,000 tons of oil equivalent (toe), or
only 3% of total oil demand. And converting 35% of the export sugar would
really only amount to a nonpayment of subsidies. Before production of
agricultural alcohol becomes viable, new lower cost feedstocks will have to be
found. At present there is no concerted programme for producing fuel alcohol, or
even blending with petrol (Ref. 24).
Unproductive agricultural land in the United States could produce between 19.
7 and 32.9 billion L of alcohol. And if 30.1 million acres that farmers were paid
not to grow crops on in 1984 were used to produce ethanol from corn, between
20.3 and 40.6 billion L of alcohol could have been produced. More intensive
cultivation could increase the quantity of alcohol produced to replace around 10%
of the current petrol consumption. An EEC study suggests that by shifting 8% of
the agricultural land to grow sugar beets for conversion to alcohol, between 6%
and 8% of the projected petrol consumption for 1990 could be met (Ref. 24).
Elimination of dairy surpluses in the EEC would release no less than 1.4 million
ha that could be used for energy crops and also reduce the import bill for animal
feeds and grains. In 1980 the United States over-produced an equivalent in milk
of around 5.8 million t. Removing this surplus would release 510,400 ha from
grainlands, and a further 406,000 ha from land cultivated to grow feed crops
(Ref. 24).
Table 2:
Recoverable potential energy production of the main agricultural residues and wastes
country
Canada
2.3
1.8
4.1
1.9
United States
20.2
5.0
25.2 1.4
Japan
0.4
1.2
1.6
0.4
Finland
0.4
0.2
0.6
2.4
France
5.2
2.4
7.6
4.0
Germany
2.2
1.6
3.8
1.4
Italy
2.5
0.9
3.4
2.5
Sweden
0.8
0.2
1.0
2.0
Switzerland
0.1
0.2
0.3
1.2
United Kingdom 1.3
1.3
2.6
1.2
Note: Figures are an average of 19781980 and are in million toe.
Source: Ref. 24.
In the United States by the year 2000, agricultural residues and wastes could
provide up to 30.64 million toe annually, and animal waste could contribute 7.1
million toe annually. In the EEC, the total net energy potential for animal wastes
is 11.5 million toe, and for crop residues is 12.5 million toe (Refs. 2427).
The U.S. timber industry now gets 50% of its energy requirements from wood
residues. By 2000, wood could provide 235.7 million toe per year (OTA). In
1979, Germany had the potential to provide 6.2 million toe from forest waste and
industrial residues, Finland 8.6, France 6.1, Sweden 10.4, and Canada 32.1.
France has reserves of coppice that could provide between 1.5 and 2 million toe
per year. Finland has stands of unproductive hardwoods that could give some 1.
01 million toe per year, and forest thinnings would be worth 0.8 million toe/year
(Refs. 26, 27).
In the EEC, by 2000, forestry and short rotation plantations could provide 28
million toe net. And in the United States, herbaceous plants and short-rotation
forestry combined could provide 117.9 million toe in 15 years with a value of
$26,390 million in 1984 prices. The United Kingdom produces enough wastes to
produce, with the current technology, about 5.7 million toe, currently worth $1,
215.4 million. The United States could, by 2000, obtain between 5% and 15% of
its total energy requirements from biomass, some 141.4391.3 million toe/year.
In the same period the EEC is estimated to be able to produce 85.8 million toe
from biomass (Refs. 24, 39).
5
SUBSIDIES
In the United States in 1983 agricultural subsidies came to around $19 billion in
payment-in-kind (PIK) and cash payments, to which must be added indirect
subsidies, tax concessions, and support grants, etc. PIK will not operate in 1984
85. Direct government payments are forecast to be $810 billion in 1983 and $6
10 billion in 1984. Price support and direct payments to farmers in the United
States went up by 792% between 1970 and 1982. It is projected to rise by a
further 115% to $68 billion by 2000. In the EEC in 1983, agricultural subsidies
amounted to some $15 billion. The projected total cost of EEC subsidies in 1984
is a colossal $23.83 billion. The cost of EEC farm policies to the UK consumer
in 197980 came to an astounding 3,350 million or nearly 13,000 for every
farmer, while the gross product of British agriculture was then only 5,420
million (Refs. 13, 14, 28, 25).
6
RESEARCH
The green revolution has enabled crop yields to be increased dramatically, but
the high yield varieties used respond to expensive inputs, limiting their use to
richer farmers in more favoured areas. With dwindling reserves of fertile soils,
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
Fig. 1. Purchasing power index of petroleum and cash crops exported by developing
countries in terms of imported manufactures, 197182.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
1
INTRODUCTION
When the Reagan Administration took office, it brought with it the view that
governments role in research and development (R&D) should be quite limited.
Specifically, with the exception of defense and space, government should
support only the most fundamental research. The application of research results
should be left to the private sector.
While not yet born again, the Administration has, in practice, been much
more pragmatic than its ideology would suggest. In some areas it has gone
beyond its strictly defined limitation. But when it has, it has done so as an ad hoc
activity without any clearly defined overall program for attempting to ensure that
federal R&D expenditures produce new industries, new products, greater
efficiencies, or other potential benefits. The thesis of this brief paper is that this
problem must be addressed in a more pragmatic, nonideological way. It is
maintained that a federal-state energy R&D partnership should be established to
help ensure that national energy R&D programs produce meaningful, tangible
benefits at the state and local levels.
Given the U.S. Department of Energys (DOE) proposed $25 million
expenditure for technology transfer, it is essential that this partnership be
established as soon as possible.
2
BACKGROUND
Effective linkages between those who acquire knowledge and those who apply it
have always been essential for knowledge and new technologies to be translated
into widespread, productive applications. No era illustrates this better than the
1950s and 1960s, a period of extraordinary success in realizing the tangible
benefits of R&D in the United States.
During those decades, the research contract and research grant mechanisms
developed during World War II were used to establish linkages among
point out that in many of these new areas, including energy, government is
neither the customer nor the consumer. It is also true that other countries have
certain advantages, such as newer, more modern plants and lower labor costs.
Some of these changes are probably significant. But whether taken separately
or together, they do not provide an altogether satisfying explanation of the
limited success of our national R&D programs in these new areas. I believe a
major part of the problem in energy R&D is that in most publicly supported
energy R&D programs, government has meant only the federal government.
But the federal government is not the level of government best equipped to
bridge the gap between national-level programs and actual applications. The
latter always takes place at specific locations at the state and local levels.
Effective promotion of applications based on generic, national-level R&D
requires a detailed knowledge of specific local conditions and needs. And it is
regional and state level rather than national organizations that have this
knowledge and know best how to use it effectively to produce tangible results.
4
THE PROPOSAL
Thus a federal-state energy R&D partnership is needed. In particular, DOE and
regional and state energy R&D organizations, such as those in the Northeast and
in California, North Carolina, New Mexico, and New York, should jointly plan,
fund, and implement energy R&D programs in selected areas. Such a partnership
would draw on the strengths of both DOE and regional and state energy R&D
organizations. Briefly stated, the successful application of knowledge gained in
large-scale national research programs is much more likely to occur when
regional and state-level organizations sensitive to local needs, opportunities, and
conditions are involved. These organizations have closer, more direct
relationships with private sector energy and energy technology consumers.
The proposed partnership would incorporate an approach emphasizing:
An overall program rather than an individual project orientation
Joint planning and joint funding in specific program areas of mutual interest
to DOE and individual regional/state energy R&D organizations
Effective utilization of the full range of regional/state-level resources in
technology transfer
Results that lead to new industries, new products, greater efficiencies, and
other potential benefits.
With few exceptions, cooperative research involving DOE and regional/ state
energy R&D organizations is undertaken on a project-by-project basis. Such
efforts tend to be ad hoc and identified for the most part on a target-ofopportunity basis. A much more rational approach would be to identify specific
program areas of mutual interest to DOE and a regional/ state organization, have
DOE and the regional/state organization jointly identify R&D needs in these
areas, and then jointly develop and fund an R&D program designed to meet these
needs. Such an approach would provide a mechanism for making more efficient
and effective use of scarce energy R&D sources than is likely to occur using the
current approach.
An important measure of the effectiveness and efficiency of energy R&D is
whether results are translated into beneficial applications. At the regional and
state levels, effective applications mean that new industries are established, new
products are produced, greater efficiencies are achieved, and other potential
benefits are realized. For this to happen, technology transfer cannot stop with
published reports, conferences and workshops, videotapes, and traveling
roadshows. Slick public relations wont do it. Someone has to be an effective
communicator/translator/ promoter at the regional, state, and local levels. A
broad range of economic development and technical assistance capabilities
already exists at these levels, and regional and state energy R&D organizations
know how to mobilize the delivery of these services. They know how to identify
the individuals and firms likely to benefit from energy R&D results, and they
know how to help these potential users apply these results at specific places to
meet specific needs and to take advantage of specific opportunities.
5
THE BIOMASS EXAMPLE
Biomass is an area ripe for the proposed federal-state partnership. In this case,
the resource base itself varies regionally. The overall national research program
would benefit from the direct participation of regional and state energy R&D
organizations. But the more compelling point is that there will not be much of a
payoff from biomass R&D unless regional and state-level R&D organizations are
a part of the program. Comprehensive knowledge about supply and technologies
for either converting or utilizing biomass resources will produce few practical
results unless local needs, opportunities, and conditions are taken into account.
Quite apart from the standard questions about technical and economic feasibility
usually addressed by R&D projects, a myriad of other factors stand between
knowledge and application; for example, the effects of state and local laws and
regulations, attitudes, and business conditions.
Under the leadership of the Coalition of Northeastern Governors (CONEG)
and the New York State Energy Research and Development Authority
(NYSERDA), a biomass partnership is developing in the Northeast. The CONEG
program, funded by DOE, is devoted largely to the promotion of wood energy
through studies and various information dissemination activities. Specifically,
the program supports wood energy expertise in state governments, analyses of
environmental and economic impacts of expanded utilization of wood for
energy, and development of workshops and printed material designed to bring
the advantages of wood fuel to the attention of potential users. This attempt at
technology transfer is useful but not sufficient to bring new technology to the
marketplace.
NYSERDAs regional biomass effort, on the other hand, is an example of the
kind of program being proposed here. It transcends the traditional boundaries of
government contract R&D, and it directly involves users at an early stage of
technology development to address in advance the broad spectrum of practical
problems connected with implementation.
Several examples illustrate how the program works. Two research projects are
being conducted in conjunction with private developers to establish the basis for
investment decisions to construct innovative facilities that would produce fuels,
energy-intensive chemicals, and possibly food by-products from wood.
NYSERDAs direct wood-to-energy conversion efforts include not only research
but also coordinated technical assistance and financial aid to innovators. To
ensure long-term biomass supplies, a formal program involving NYSERDA, the
Gas Research Institute, and a consortium of New York based utilities sponsors
research in fast-growing hardwoods. The knowledge and technical capability
created by this type of research has enabled NYSERDA to organize separately a
commercial tree cultivation demonstration at a manufacturing facility in the
northern part of New York State.
The program is regional both in its outlook and its institutional support. Much
of the work we support addresses opportunities applicable to the Northeast as a
whole and takes place outside New York State with support from other states or
private organizations. For example, NYSERDA, the Commonwealth of
Massachusetts, and several New England gas utilities are sponsoring a methanefrom-sewage sludge project in Salem, Mass. A number of other states are also
involved with NYSERDA in a project to take place in Pittsfield, Mass., designed
to contribute to better control of dioxin emissions from municipal waste
combustion.
What is lacking in this Northeastern effort is program support from DOE.
Such support would help DOE be more effective in its technology transfer
efforts.
6
A CALL TO ACTION
While we are all interested in and support research undertaken to produce
knowledge, we also want to reap the benefits of actual applications of this
knowledge. The reality is that knowledge and new technologies do not get
applied automatically. We all know that potential users of knowledge and new
technologies need to be involved at an early stage so they can help shape R&D
products to meet their needs. But even then, they often need encouragement, if
not a hefty push or pull to actually apply these research products. Providing
encouragement and a boost through technical assistance and risk-sharing
programs is something that seems to be done best by regional and state
organizations. This is why we propose that DOE and regional and state energy
R&D organizations combine forces. A partnership is needed to ensure that our
federal and state R&D expenditures produce the maximum possible meaningful,
tangible benefits. And it is needed now!
REFERENCES
1.
2.
SYNOPSIS
The forest products industry utilizes residual material from trees to processes.
The trend is toward replacement of natural gas, oil, and coal by produce steam,
hot air, electricity, and combustible gas for its various wood-derived biomass as
well as self-generation of the electrical requirement. This trend suggests a
doubling of wood biomass use from 2 to 4 quads from 1980 to 2000. It will be
shown that in a managed conifer plantation system at steady state there is a slight
excess of energy over production requirements. Technology improvements in
harvesting and energy production make residual biomass the vehicle of economic
choice in most instances. In the United States, short-rotation hardwood
plantations may eventually be used by the industry as a source of pulp chips, but
not primarily for energy.
1
INTRODUCTION
My assignment is to put in perspective the utilization of biomass for energy in
the forest products industry in the United States. I will not address biomass for
other potential energy uses such as municipal electric generation, ethanol
production, or residential heating.
In the forest products industry, the biomass of interest is really phytomass, i.e.,
tree tissue, and normally consists of bark, foliage, limbs, sawdust, and other
residues. This material is called energy fiber to distinguish it from pulp fiber. A
less sophisticated term is wood waste. This energy fiber is combusted to produce
hot air, low energy content gas, steam, and, by cogeneration, electricity for mill
processes.
Table 1 compares the use of wood (trees) with other biomass material and the
total U.S. energy use in 1980 with that projected for 2000. The doubling of the
use of wood is a conservative estimate, for it does not include the probable
increase in the number of pulp mills but only the replacement of fossil fuels by
energy fiber in current mills.
Table 1:
Biomass energy in the United States
source
Wood
Agricultural wastes
Municipal solid waste
U.S. total energy use
2000
2.2
0.1
0.2
75
4
0.5
1
95
2
TRENDS IN THE INDUSTRY
For the past decade, the amount of fossil fuel and purchased electricity per ton of
product has been steadily decreasing as a result of conservation technology and
replacement of these fuels by biomass. As a result, the energy self-sufficiency of
pulp, paper, and lumber mills has been increasing (Fig. 1). In terms of biomass
(energy fiber) use, it has increased from 15 to 31 million tons per year (Table 2).
This quantity of wood is equivalent to nearly 210 barrels of oil per year or about
40 days of current U.S. imports. The reduction in the use of fossil fuel is even
more impressive when it is realized that total production of pulp and paper
increased by 24% during the same period. For the 197282 period, the energy
use per unit of output (Fig. 2) is seen to decline due to conservation. Further, the
biomass contribution increases in absolute terms and the fossil fuel use declines
in both absolute and relative terms. By 1990, it is estimated that the total will
drop to 26 million Btu/ton and the biomass component use to 18 million Btu/ton.
By 2000, the ultimate conservation effort may result in about 20 million Btu/ton,
with 90% or more being supplied by tree biomass.
Table 2:
U.S. pulp and paper industry biomass use
year
106 ton
1972
1976
1980
1984
15
19
25
31(est.)
and, most important, replacement of fossil fuels in the boilers and lime kilns by
energy fiber.
Table 3:
Weyerhaeuser mill (2400 t/day) boiler and lime kiln areas
103 bbl oil equivalent
1981 fossil fuel consumption
Planned production increases
Planned fossil fuel reductions conservation
more steam from black liquor
replacement by biomass
1990 fossil fuel consumption
1166
427
242
174
1137
0
3
STEADY-STATE PLANTATIONS
In the lands owned by the forest products industry, the relatively slow-growth
native forest or marginal farmland is being converted to scientifically managed
high-yield tree plantations. These plantations will be thinned, harvested, and
replanted on the most economic schedule. At steady state with present
technology, an average pine plantation in the coastal plain of the southeastern
United States will yield about 18 m3/ha yr (or 8 dry ton/ha yr) over a 30-yr
rotation. Each tree will be broken down to yield the highest value; i.e., the
maximum lumber will be first priority followed by the maximum pulp chips. The
residue is assigned to energy fiber.
Table 4:
Loblolly pine plantation system
Assumptions
Average S.E. U.S. plantation site (18 m3/ha yr, 8 dry ton/ha yr)
30-yr rotationtwo thinningsall above-ground biomass
Each part of biomass to highest value
1,000 ton/day pulp mill, lumber mill to match
Result
1.45105 ha plantation
21-km radius (15 km < average), 100% ownership Products
8.4105 dry ton/yr chips
1.5105 dry ton/yr lumber
2.9105 dry ton/yr energy fiber
Energy self-sufficiency (steam and electricity)
produce the resultant products listed in Table 4. The energy fiber produced in
this flow is more than sufficient to meet all of the steam, gas, and electrical
requirements of the mill processes. This is displayed quantitatively in Table 5.
The total available energy exceeds the demand by 1.6106 GJ.
Table 5:
Energy balance
GJ106
kWh106
6.4
0.7
7.1
2.0
9.1
316
35
351
4.1
6.6
10.7
1.6
1
ECONOMICS
The availability of enough energy fiber from a steady-state plantation system
does not, however, ensure its utilization. This is determined by the economics of
production and delivery of the material. The value of energy fiber is compared
with coal and oil at representative 1984 costs and also with higher valued wood
products in Table 6. From these data, it is clear that wood that will produce clean
pulp chips will not be used for energy. It also follows that for a delivered cost to
a boiler of coal at $60/ton (average for Weyerhaeuser mills), energy fiber must
be delivered at $33/dry metric ton or less to be competitive. Since coal is readily
available anywhere in the United States, it, not oil, sets the standard for
allowable wood cost at any given mill site.
Table 6:
Relative value of wood1984
$/dry metric ton (delivered to mill)
Energy fiber
Aspen flakesOSB
Pulp chips
Loblolly pine
33 at $60/ton coal
75 at 30/bbl ($l80/m3) oil
40
6010
705
705
9010
The cost of the energy fiber delivered to a mill using best available technology
from loblolly pine plantations on flat or rolling land is analyzed in Table 7. It has
three components: stumpage, harvesting, and transportation. The stumpage value
of a plantation is the total cost of establishment and silvicultural management
over the life of a stand treated as a capital investment that must earn 6%8%
compounded net of inflation.
Harvesting is the cost of cutting and moving the whole tree (above-ground
portion) to the landing, and breaking it down into its components: sawlogs, pulp
chips, and energy fiber. Transport includes loading and unloading and movement
to the pile at a mill site.
Depending on various accounting assumptions and the distance to the mill, a
reasonable upper and lower bound on the delivered cost of energy fiber can be
given for 1984 technology.
Table 7:
Cost of energy fiber from loblolly pine plantations
factors
$/dry ton
high
Low
Stumpage
(Cost weighted by product value)
(No cost for waste-energy fiber)
Harvesting
(Stump-to-truck equal for all products)
(No cost for waste-energy fiber)
Transportation
50 km
15 km
Total
14
13
33
7
7
4.1
Stumpage
First the total stumpage value per hectare of the entire plantation is calculated.
Then the part attributable to energy fiber is determined either by calculating its
weighted relative value (high case) or by assuming it to be zero on the basis that
the plantation was grown for the sawlogs and pulp chips and the energy fiber is
waste (low case).
4.2
Harvesting
The high case is calculated by dividing the total cost per hectare for cutting,
moving, and separating the three products at the landing by the total biomass
processed. The low case again treats the energy fiber as waste and hence assigns
no cost.
4.3
Transportation
A well-planned plantation wood basket for a scalp pulp mill (1,000 ton/day)
would normally have an average distance from the landing to the mill that did
not exceed 50 km (high case). If the wood basket consisted of the minimum
required land (100% ownership) completely surrounding the mill, the average
transportation distance would be about 15 km (low case).
If these numbers are added, it is seen that with current technology the cost of
energy fiber would be between $7 and $33/dry ton and therefore would be
competitive with coal at a delivered cost of $60/ton. Further technology
advances in silviculture and harvesting, coupled with ultimate pressure on coal
suppliers and higher oil prices, should continue to widen the gap and make tree
biomass for energy the fuel of economic choice in most instances.
In the case of a native forest that is being clearcut, the ratio of energy fiber to
lumber and chips is higher than in plantations due to the extra defective, dead,
and small trees and unutilized species. Therefore, in all forests in the temperate
zone that are being cut for lumber or pulpwood, the mills can be self-sufficient
using biomass for their energy requirement. The costs of harvesting the energy
fiber as a come-along residue from natural forests on steep ground will be
higher than for level plantations and in some cases may then exceed the cost of
coal.
4.4
Short-Rotation Hardwood Plantation
In recent years, a great deal of effort has been directed toward the development of
short-rotation hardwood plantations as a possible source of biomass energy.
Yields as high as 2030 dry ton/ha yr have been claimed for small experimental
plots. The combined cost of stumpage, harvesting, and transportation for this
energy fiber does not appear to be competitive with coal in the near term.
However, as technology advances, such plantations may have high value and
utility as raw material for pulp chips to produce fine paper.
5
SUMMARY
The forest products industry is, and probably will remain, the prime user of
biomass for energy in the United States. Its use of wood residuals for energy fiber
will probably double by 2000 in replacement of fossil fuel and purchased
electricity. Beyond 2000, use of biomass will continue to grow with the total
production of pulp and paper. It is unlikely that the forest products industry will
install short-rotation hardwood plantations for energy, but it may well adopt this
tree technology to provide raw material for fine paper and composite boards as well
as simultaneously providing the energy fiber to manufacture these products.
Fig. 2. Pulp and paper industry energy use per unit of output
SECTION II
Research Interests of Biomass Sponsors
SYNOPSIS
The purpose of this paper is to describe the U.S. Department of Energy (DOE)
Bio-Energy Coordinating Committee (BECC), an organization for information
exchange and communication among the DOE offices involved in bio-energy,
and to provide a brief overview of bio-energy programs and activities conducted
by DOE.
1
THE DEPARTMENT OF ENERGYS BIO-ENERGY
COORDINATING COMMITTEE
The Bio-Energy Coordinating Committee (BECC) is an internal DOE committee
comprising representatives of all organizations involved in bioenergy. The
purposes of BECC are:
to achieve effective coordination of DOEs bio-energy research and
development (R&D) programs;
to assure optimum use of DOEs existing expertise in bio-energy R&D;
to provide a way for industrial and other users to access information rapidly
on DOE bio-energy programs/technology; and
to achieve rapid communication within DOE of new developments,
opportunities, and problems in bio-energy research and technology
development.
The committee was originally established in 1980 as the Biomass Energy
Coordinating Committee. However, in July 1984, bio-energy was substituted
for biomass energy to expand the scope of the committee to encompass all
areas of biotechnology that play a role in the production, conversion, and use of
energy resources. The impetus behind this decision was the realization that a
common technology base in biochemistry and genetics was being applied
independently in a number of DOE programs involving both biomass and
nonbiomass energy resources. The committee agreed that BECC could provide a
much needed forum for communication among and coordination of these diverse
programs.
Currently, the following DOE organizations are represented on BECC:
Office of Renewable Energy
Biomass Energy Technology Division
Energy from Municipal Waste Division
Office of Conservation
Renewable Energy
Conservation
Energy Research
Fossil Energy
Bonneville Power Administration
The Bonneville Power Administration conducts R&D in bio-energy using ratepayer funds and the support it receives from DOEs Biomass Energy Technology
Division for managing the Alaska and Pacific Northwest Regional Biomass
Energy Program. These R&D activities are aimed at the development of biomass
as an integral part of the regions electric energy resources. BPAs major
technical thrusts include biomass conversion technology for the electric utility
industry and the identification and mitigation of adverse environmental impacts
associated with using biomass for energy.
2.5
Summary of R&D Programs
Figure 1 summarizes the R&D activities conducted by each DOE office in terms
of the major categories, feedstock production, and conversion. Note that of all
the technologies shown, biochemical conversion research is being performed
universally by all of the DOE offices active in bio-energy R&D.
The technology areas addressed by DOEs bio-energy R&D effort are
summarized in Fig. 2. In addition to heat, electricity, and fuels derived from
biomass
feedstocks,
DOEs
bio-energy
research
interests
also
include applications in the manufacture of chemicals and in environmental
assessment and protection.
APPENDIX
DOE BIO-ENERGY COORDINATING COMMITTEE MEMBERSHIP AND
REPRESENTATIVES
CE-321
CE-323
CE-112
CE-121
CE-122
CE-13
CE-112
(202)2526741
(202)2528021
(202)2529130
(202)2522898
(202)2522455
(202)2528055
(202)2521484
CE-20
(202)2529104
CE-80
(202)2521277
ER-17
ER-32
ER-75
(301)3532873
(301)3534355
(301)3535778
FE-13
(301)3532773
PE-15
PE-26
(202)2526296
(202)2524760
EIZ-53
(202)2529775
SYNOPSIS
Biomass offers an alternative to fossil fuels and hence an opportunity for the
United States and the Tennessee Valley region to become less dependent on
imported petroleum. While biomass can meet only a portion of the fuel
requirements, it has potential to make valuable contributions. Biomass also
offers potential for less centralized energy production relative to electrical
generation, petroleum refining, and pipeline distribution systems as a strategy in
the unfortunate event of war. Moreover, biomass relies on renewable energy
supplies rather than on mining a depletable (stock) resource.
The southeastern United States, of which the Tennessee Valley region is an
integral part, has substantial biomass potential because of the relatively long
growing season, abundant rainfall, and underutilized land and forest resources.
However, research and development activities are needed to transform these
potentials into realities.
1
PROGRAM
The Tennessee Valley Authoritys (TVA) Biomass Fuels Program is designed to
develop information to assist industry in commercializing the use of renewable
resourcesprimarily for energy purposes. This would reduce dependence on
foreign oil and hence provide for improved national defense and balance of
payments; it would improve resource use; and it would provide economic
development, especially for rural areas. The program emphasizes the use of
underutilized hardwood resource of the region with a focus on development of an
efficient and economical process for producing liquid fuel from wood.
Significant amounts of underutilized (marginal) land also exist. Avenues are
being pursued to develop this resource into possible renewable energy
production, thereby providing potential to increase farm income without adverse
impacts on the environment and food production.
While the program has several integrated facets designed primarily for Valley
conditions, the technologies and concepts have national and international
applications and implications. Activities in the program include: (1) forestry
resource assessments, (2) forestry management studies, (3) forestry harvesting
techniques, (1) development of technologies for conversion of biomass (both
woody and nonwoody) primarily into energy products, (5) technical monitoring
of the U.S. Department of Energys (DOE) loan guarantee program, (6)
management of the Southeastern Regional Biomass Energy Program (SERBEP)
for DOE, (7) technical assistance to the Agency for International Development
(AID) of the U.S. Department of State in developing and implementing a
bioenergy program for less-developed countries, and (8) transfer of technology.
2
RESOURCES POTENTIAL
The Tennessee Valley has an abundant and currently underutilized hardwood
resource. Over 50% of the 58.5 million acres of land in the valley is forested, and
80% of the forests is hardwoods. Even before the recent recession, which was
particularly deep in the forest industry, only about one-third of the annual growth
of hardwoods was harvested each year. Preliminary estimates indicate that with
current forest management practices the wood supply available for biomass fuel
uses in the Tennessee Valley area is sufficient to (1) displace nearly one-third of
the oil and natural gas currently used for space heat and process steam in
commerce and industry, and (2) replace one-fifth of the current automobile fuel
needs of the valley region. TVA foresters carry out activities for the Biomass
Fuels Program in the areas of wood availability, cost of harvesting and
transporting wood for energy, management of forests for energy, and technical
assistance to industry.
In addition to the forestry resource, nearly 2 million acres of underutilized
open lands exist in the Valley, some of which may be suitable for producing
energy crops. The Tennessee Valley is one of the best suited regions of the
nation for biomass production because the relatively warm climate allows a long
growing season, which is complemented by high annual rainfall (4065 in.); the
heavy rainfall, however, does pose concerns in terms of soil erosion.
Agriculturists have addressed some of the questions related to land availability
for fuel energy crop production.
3
EXPERIENCE
TVA has a history of work in biomass activities dating back to the late 1940s;
and, of course, work in forestry and agriculture was organized soon after the
formation of the agency. The U.S. government constructed a wood hydrolysis
plant at Springfield, Ore., in the 1940s based on research improvements to the
Scholler acid hydrolysis process. This plant was not as successful in its technical
performance as had been anticipated, and combining this with changing political
and economic situations led to its abandonment. The Forest Products Laboratory
(FPL-Madison, Wis.) of the U.S. Department of Agriculture (USDA) then asked
TVA to further develop the research by scaling up a proposed modified process
(commonly referred to as the Madison process). Subsequently, TVA
constructed and operated a pilot plant at Muscle Shoals. In this pilot plant, wood
was hydrolyzed to sugars via a percolation acid hydrolysis route. The sugars
(primarily C6 since the process tended to degrade C5 to furfural) were
successfully concentrated to a blackstrap molasses substitute. When molasses
prices declined substantially shortly thereafter, interest vanished and further
development was discontinued. Numerous other joint research projects were
conducted with the FPL over the next three decades, including development of
fire retardant materials, which are now widely recommended and used
throughout the United States.
The 50 successful years of experience in research, development, and
technology transfer of TVAs National Fertilizer Development Center provide a
solid foundation for TVAs biomass technology development and transfer
program. About three-fourths of the commercial fertilizers used in the United
States have been developed or improved at TVAs National Fertilizer
Development Center. The mechanism for similar results in the biomass area is
already in place, with a comprehensive program ranging from basic research
through bench-scale, pilot-plant, and technology-transfer activities. TVA also is
charged with aspects of national defense, valley programs for reforestation, land
use, and economic development, which complement biomass program objectives.
4
PROJECTS SUPPORTED BY CONGRESSIONAL
APPROPRIATIONS
Growing interest in and concern for fuel energy and resource development led
TVA to organize a formal biomass effort in 1980. The emphasis is on the
underutilized hardwood resource, which has potential to provide liquid fuel, our
nations real energy need. With FPLs and TVAs cooperative efforts
reestablished, a two-stage, short retention time, acid hydrolysis concept was
adopted for research and development. A Swedish scientist, Karl Cederquist, had
envisioned this process following his association with FPL representatives at the
Springfield, Ore., wood hydrolysis plant. Research to develop an efficient and
economical process for converting hardwoods to liquid fuels became the core of
the program, with work on developing the Cederquist concept receiving the
primary emphasis.
The improved concept for production of alcohol from hardwood, which includes
greater utilization of all cellulosic components, is being studied in TVA
laboratories. This concept involves two-stage hydrolysis of wood with short
hydrolysis retention times, explosive release to physically disrupt the wood, and
use of dilute sulfuric acid to form solutions of predominantly five-carbon and sixcarbon sugars, respectively, from the two stages (Fig. 1). These sugars are then
available for fermentation to ethanol. Design work is in progress, and equipment
for a 2-ton-per-day, pilot-plant facility is being purchased at this time.
To improve the economics for commercial use of woody biomass, TVAs
forestry programs talent and experience are employed to develop practical
techniques to reduce the cost of woody biomass, transfer technology to industry
and private ownership, and assure wood availability, as well as other
management and marketing aspects. Improved harvesting systems specifically
designed for hardwood fuels use are being tested for technical, economic, and
environmental acceptability. Study is under way to determine methods to most
economically and efficiently remove trees from the forest under various
conditions; demonstrations of the techniques will be used to transfer this
information to industry and landowners. These are all important aspects to
successful commercialization of an ethanol-from-wood or other energy-fromwood process and must be developed in parallel to the conversion technology.
This phase of the Biomass Fuels Program is supported by direct appropriations
from Congress. Activities supported by TVA power program funds that directly
replace electricity and hence peak loads, such as the residential wood heater
project, are not part of the Biomass Fuels Program per se; hence no funds from
TVA power revenues are used in the Biomass Fuels Program.
5
PROJECTS SUPPORTED BY CONTRACTUAL
ARRANGEMENTS
The appropriated funds have been supplemented by contract work. Contracts
exist with DOE and AID.
5.1
U.S. Department of Energy
TVAs Office of Agricultural and Chemical Development has been conducting
contract work with DOE since 1980 on small-scale technology for converting
alternative agricultural crops to ethanol. A contract has also been executed to
provide technical monitoring assistance for the DOE loan guarantee program.
TVA manages SERBEP for DOE and is just beginning work on improved
harvesting equipment for wood for energy.
5.1.1 Office of Alcohol Fuels: In the loan guarantee program, DOE will
guarantee loans for construction of privately owned plants to produce ethanol
from corn or molasses. Plant sizes range from 15 to 20 million gallons per year.
DOE incurs a liability only if the firms default on the loans; hence, technical
assistance is necessary to ensure that the plants are properly constructed and
effectively operated. TVA engineers provide this service for DOE. The National
Fertilizer Development Center has provided assistance to the fertilizer industry in
plant construction and startup and therefore has demonstrated experience in this
general area as well as knowledge of alcohol production technology.
The New Energy plant at South Bend, Ind., is nearing completion; it is a 50million-gallon-per-year facility. Tennol, Inc. has completed final preconstruction
requirements with DOE and is beginning construction of a 25-million-gallon-peryear plant. Consideration is being given also to construction of plants in
Louisiana, Maine, Minnesota, and Nebraska.
5.1.2 Ethanol from Agricultural Materials: DOE has, through the Solar Energy
Research Institute (SERI), funded research at Purdue University on the
conversion of agricultural residues (such as corn stover and wheat straw) to
ethanol. A concentrated sulfuric acid hydrolysis process employing low
temperatures and pressures is used. DOE then asked TVA to modify an existing
small-scale pilot plant (previously built with DOE funds) to test this process. The
modification consists of addition of the following process steps: grinding the
agricultural residues, converting the hemicell-ulose and alpha cellulose fractions
to fermentable sugar solutions (principally C5; and C6, respectively), and
separating the lignin fraction. Communications among SERI, Purdue University,
and the staff of TVA have resulted in the design of an experimental facility as a
front-end modification of the existing 10-gallon-per-hour ethanol unit built by
TVA to obtain benchmark data on grains and numerous alternative starch and
sugar crops. The equipment is installed and shakedown tests are being conducted.
Figure 2 is a flowsheet for the process.
5.1.3 Southeastern Regional Biomass Energy Program (SERBEP):
Regional programs have been established by DOE in four sections of the
country, and TVA was assigned responsibility to manage the program in the
southeastern part of the United Statesan area covering Alabama, Arkansas,
Florida, Georgia, Kentucky, Louisiana, Missouri, Mississippi, North Carolina,
South Carolina, Tennessee, Virginia, and West Virginia. A purpose of the
program is to promote effective use of regional biomass resources to meet
regional energy needs. The focus is on information development/ transfer and
technology transfer for a broad range of biomass resources, conversion
technologies, and end uses. Much of the effort is carried out under competitive
contracts selected with the assistance of key representatives of industry,
academia, and government.
Specific objectives include (1) establishing the availability of biomass
resources within the defined regions through resource assessment studies, (2)
enabling industry to match local resources with conversion technologies that will
permit private sector investments in biomass energy technologies, (3)
transferring results of research and development to the private sector, and (4)
establishing a partnership with industry through cost-shared projects that will
build private sector confidence in adopting biomass energy technologies.
Our farmers are unable to market all of their crops; therefore, government
programs to idle portions of cropland have been maintained. If we can develop
energy crops that are economically competitive with fossil fuels, we could use
those excess acres for productive gain to the farmers and reduce government
program costs to the nation.
Two weeks ago, at the Washington Conference on Alcohol Fuels, Secretary
Block restated his continued support for biomass energy. I am proud that
alcohol production has gone from 40 million gallons of alcohol production in
1980 to over 400 million gallons projected in 1984 using close to 200 million
bushels of grain. Im proud that our alcohol fuel production capacity has
increased fourfold since 1980.
I am proud that the production of other biomass including wood has greatly
increased over the last four years.
Without question, continued growth in the domestic ethanol industry will
have a major impact upon U.S. agriculture. Increased demand for ethanol as an
octane enhancer will increase the demand for corn we are using. In turn, this will
increase the price, profitability, and utilization of corn. That means jobs and
economic activity for America.
Our policy wasand still isto allow market forces to determine the types
and quantities of biomass energy produced and consumed. We also feel that the
tax exemption has been very helpful in helping develop an alcohol fuels
industry.
We shall continue our research and development activities for biomass
energy. The goal shall be to improve conversion technologies and lower
production costs that will allow biomass energy to someday be substituted for
liquid, gaseous, or solid fuels.
We believe this research is vital to reducing the costs of biomass energy so it
can compete head to head in the marketplace with conventional fuels.
In line with that effort, we will be conducting a forum next month to look
into new uses for farm products. This is the second in a series of what I call
challenge forumsto address some of the major issues facing agriculture.
This two-day session will focus on bringing together the public and private
sectors for an in-depth look at new products, new uses, and new markets for
agricultural products.
We have, at least temporarily, excess production capacity in our agricultural
plant. Farmers want to produce. We can use the jobs. Lets put these resources to
work.
The transformation of crops into energy for fuel offers an exciting
opportunity for agriculture to serve this country in a nontraditional, but new and
very vital, way.
For those engaged in forestry, fuel production is not a new venture but it has
renewed emphasis. We now use in excess of 2.6 quads of wood energy. The
forest products industry consumed over 1.6 quads of wood residue as fuel in
1983 About 1 quad was used as household fuel in stoves for space heating.
title
location
INT-1603
INT-1753
Logan, UT
Reno, NV
70
50
Provo, UT
10
INT-1703
number
title
INT-1752
location
Provo, UT
10
Grand Rapids, MN
20
Columbia, MO
Rhinelander, WI
10
50
Parsons, WV
20
Durham, NH
15
Orono, ME
Warren, PA
15
15
Delaware , OH
15
Durham, NH
Berea, KY
Corvallis, OR
20
100
20
Table 1:
Silviculture energy-related research projects in the Forest Service, September 1984
(concluded)
number
PNW-1207
title
location
Olympia, WA
30
Wenakhee, WA
Redding, CA
10
20
Honolulu, HI
25
Honolulu, HI
Redding, CA
25
20
Fort Collins, CO
Laramie, WY
20
25
Albuquerque, NM 50
Rapid City, SD
50
Flagstaff, AZ
10
number
title
location
RM-2251
SE-1102
Lincoln, NE
Asheville, NC
10
20
Clemson, SC
25
Gainesville, FL
25
Charleston, SC
15
Alexandria, LA
15
Sewanee, TN
10
Stoneville, MS
20
Auburn , AL
Monticello, AR
10
25
SE-1118
SE-1153
SE-R&D
SO-1102
SO-1105
SO-1110
SO-1116
SO-1117
Table 2 lists tree genetics energy-related projects in the Forest Service. Genetics
can play a major role in the selection of hybrids and superior trees to obtain
maximum biomass production and other desired traits, such as disease
resistance. Clones of hybrid poplar trees developed at our northeastern station are
commonly used in plantations to produce biomass for energy in the northern
United States.
Table 2:
Tree genetics energy-related projects in the Forest Service, September 1984
number
title
location
NC-1401
NE-1401
PNW-1401
PSW-1401
SE-1499
SO-1401
Rhinelander, WI
Durham, NH
Corvallis, OR
Berkeley, CA
Raleigh, NC
Gulfport, MS
15
15
15
15
15
15
title
location
energy interest %
Missoula, MT
30
Missoula, MT
30
Carbondale, IL
Delaware, OH
45
10
Princeton, WV
Portland, OR
Portland , OR
Athens, GA
Alexandria, LA
Madison, WI
Madison, WI
40
25
40
25
20
30
20
Madison, WI
20
Madison, WI
Madison, WI
Madison, WI
30
20
15
Madison, WI
30
Madison, WI
30
number
title
location
energy interest %
FPL-34 11
Madison, WI
100
Madison, WI
20
FPL-3506
We are particularly optimistic about research that may lead to fermenting fivecarbon sugars to ethanol. This could bring about a doubling of the ethanol yield
by acid hydrolysis of hardwoods that are currently in large surplus in the eastern
United States. Another prospect for increasing the ethanol yield from wood is
through research on enzyme hydrolysis.
We have some strong research and development programs to improve the
processing of wood to various products. Improvements in processing go hand in
hand with energy conservation. Based on past experience, we can signifi cantly
increase future processing efficiencies. For example, recent Forest Products
Laboratory research on the press drying of paper has resulted in a new process
that can reduce the total energy consumed in making paper from hardwoods by
19%.
At the Forest Products Laboratory, renewed emphasis is being given to
research on combustion and gasification of wood. Researchers have obtained
information on the kinetics of wood combusted in suspension that may be used to
design more efficient wood-burning equipment.
Fundamental work on wood combustion is an essential step toward increasing
efficiency above the best levels now attainable. We hope that some new research
efforts on pyrolysis and gasification of wood will also result in a more favorable
position for wood fuel in comparison with gas, oil, and coal.
Because of changed building construction practices as a result of rising energy
costs, more building insulation is being used and air leakage in buildings is
greatly reduced. These factors can combine to create moisture problems within
walls, floors, and roofs of wood-frame homes. Researchers at the Forest Products
Laboratory are working to ensure that moisture accumulation does not become a
major problem. A theoretical model, which considers the effect of air leakage,
has been developed to analyze moisture movement through walls. Data on
seasonal moisture changes in walls have been obtained with field experiments.
Manufacturers of siding and insulation materials are using results of these studies
in their recommendations to builders.
At other research stations researchers are working on more intensive
utilization of local species and are evaluating the effects of more intensive
removal of biomass from forests on soil nutrients and soil erosion. Utilization
researchers are also working on the assessment of existing biomass.
Traditionally, timber inventories have provided information only on the
availability of wood for high-value uses such as lumber, veneer, poles, and
pulpwood, but equations are now being refined to accurately determine total tree
biomass from timber inventory data.
title
location
NC-3701
NE-3701
PNW-3701
SO-3701
Houghton, MI
Morgantown, WV
Seattle, WA
Auburn, AL
30
20
25
25
title
location
INT-4101
Ogden, UT
20
number
title
location
NC-4101
NE-4101
PNW-4101
PNW-4103
RM-4101
RM-1102
SE-4101
SO-4101
Resource Evaluation
Resource Evaluation
Resource Evaluation of PNW
Resource Evaluation of Alaska
Analysis Techniques
Inventory Techniques
Resource Evaluation
Resource Evaluation
St. Paul, MN
Broomall, PA
Portland, OR
Anchorage, AK
Fort Collins, CO
Fort Collins, CO
Asheville, NC
Starkville, MS
10
20
10
10
15
20
20
20
Table 6:
Economies, marketing, and taxation energy-related research projects in the Forest
Service, September 1984
number
title
location
NC-4203
NC-4252
Duluth, MN
St. Paul, MN
20
10
Broomall, PA
Princeton, WV
Princeton, WV
Burlington, VT
Durham, NC
10
20
20
75
20
New Orleans, LA
Madison, WI
10
10
NE-4201
NE-4204
NE-4206
NE-4207
SE-4203
SO-4202
FPL-4151
title
location
INT-2103
RM-2110
Fire Behavior
Air Quality
Missoula, MT
Fort Collins, CO
10
70
Insect and disease research projects (Table 8) provide technology for controlling
infestations in tree plantations. We hope to find ways to conserve trees that
would otherwise be rendered unavailable for energy and other uses.
Table 8:
Insects and disease energy-related research projects in the Forest Service, September 1984
number
title
location
NC-2203
NC-2205
NE-2210
15
10
10
SYNOPSIS
The Canadian Government program on Biomass R&D is a reflection of a policy
decision to increase Canadian energy independence by substituting a renewable
resource for imported petroleum-based fuels and feedstocks. The challenge was
to mount an integrated and coordinated R&D program on problems as diverse as
the genetic engineering of bacteria for the biological conversion of
lignocellulosics, environmental impact assessment, and thermochemical
liquefaction of wooda diversity matched by the number of government
departments and agencies that have related jurisdictional mandates.
Since the program began, notable progress has been made in programs such as
resource assessment, setting of standards for combustion equipment, gathering of
data for the improvement of efficiencies and emissions from large industrial
boilers, development and commercial demonstration of a high rate anaerobic
reactor, development of fluidized-bed gasification technology, a successful
bench-scale flash pyrolysis process for wood liquefaction, and a plantation tree
harvester.
Currently the major elements of the Canadian program are changing
considerably. Most of the biomass inventory questions have been answered; we
have completed a major techno-economic assessment of forest biomass
conversion technologies, and based on this assessment it is clear that the nearterm Canadian biomass opportunities are with combustion applications in the
forestry and related industries. There are, however, continuing R&D
requirements in emissions and regulatory aspects of combustion as well as in
forest residue recovery and new harvesting concepts.
A major portion of the bioenergy R&D effort has been toward the supply of
liquid fuels. In thermochemical conversion, one conceptthe pressurized O2,
fluidized-bed reactoris being demonstrated while the prospects for direct
liquefaction have been recognized as being longer term in the light of an IEA
study.
In the area of the application of biotechnology for lignocellulosic conversion
to liquid fuels, it has become apparent to us and to other countries that
economics will continue to be a barrier for some time, particularly if the present
forecasts for oil prices hold true.
At the National Research Council (NRC), the biotechnology strategy for
bioenergy is heavily dependent on the development of multiple uses for the
product streams produced from our major lignocellulosic resource base. Central
to this concept are the fractionation concepts that are being extensively evaluated
in our program along with downstream processes such as fermentation,
separation, and waste treatment.
1
INTRODUCTION
In Canada the majority of renewable energy other than conventional
hydroelectricity is derived from forest biomass. Climate and geography dictate
that the major biomass production is in the form of forests. Agricultural biomass
production takes place on less than 1% of the land mass and is primarily aimed
toward a large export of wheat on the world markets. The zone of productive
forest biomass in Canada is spread along its southern latitudes. Above the
productive zones there is a gradation of marginal boreal forest, tundra, and, in the
extreme north, ice and rock. Agricultural zones occupy a relatively small share
of the total land mass. The area of highest forest productivity is found along the
western coast ranging from a timber density of about 175 m3/ha to over 420 m3/
ha, equivalent to a biomass mean annual increment of approximately 24 ovendried metric ton per hectare (ODt/ha). The major portion of the productive forest
zone has a timber density of 70175 m3/ha equivalent to a mean annual
increment of about 12 ODt/ha of standing biomass. These values are for natural
forest stands and do not take into account the potential gain that could be realized
through silviculture, intensive breeding programs, and clonal forestry. An
example of what could possibly be achieved in Canadian forestry is illustrated by
the clonal plantations of hybrid poplar in Eastern Ontario. Yields of 710 ODt/ha
have been reported, approximately five times the yield of natural forest poplars.
Intensive management would easily double the yield of Canadian forests (Ref. 1)
without improvements in the genetics of planting stock. However, a recent study
by Forintek Canada (Ref. 2) has shown that the fibre of fast-grown trees through
intensive management is inferior to mature trees in a natural stand. But since the
quality of a tree is genetically determined (Refs. 3, 4), it is clear that intensive
management must be integrated with intensive breeding for varieties that will
make optimum use of the silvicultural inputs and at the same time possess other
desirable characteristics such as fibre quality, yield, and disease resistance.
Clonal forestry through vegetative propagation (using cuttings and, in particular,
tissue culture) is the only means for a rapid exploitation of the full genetic gain
achieved in a breeding program (Refs. 3, 4, 5, 6). With the state of forestry in
Canada deteriorating rapidly, and projections of impending shortages of suitable
fibre at the right price (Refs. 7, 8), it is imperative that tree improvement and
biomass with almost equal proportions of 25% each, while stand conversion for
regeneration will contribute about 10%. The price for this biomass (Fig. 2) has a
broad range, from almost zero to above $100/ton (Canadian), moisture- and ashfree (maf). Most of the mill residues (about 12 million tons) range from $0 to
$15/ton. Another 8 million tons consisting of mill residues and a mix of material
from salvage, surplus, stand conversion, and logging residues range from $15 to
$30/ton. The bulk of material (almost 40 million ton) consisting of the above mix
will cost between $30 and $15/ton, while another 20 million ton will cost as
much as $60/ton. The present demand for biomass from forest operations is about
16 million ton (maf), not counting domestic space heating. This suggests that
even with a doubling of biomass use through combustion and gasification,
almost 30 million ton would still be available for chemicals or liquid fuels
production, for a theoretical yield of about 3 billion gal (U.S.) of ethanol.
Note also that since substantial amounts of material from salvage operations
(fire and disease killed) and collection of logging residues and stand conversion
can be extracted at a rather favourable cost, this should give added incentive and
generate greater profitability in Canada for a more substantial forest regeneration
effort.
3
AGRICULTURAL BIOMASS
The existence of a major grain export industry based on the prairie, a region of
low population density having relatively low-cost energy available as natural
gas, has resulted in only minor use of grains for motor fuel in contrast to the U.S.
gasohol program. In consequence, the research emphasis has been based on crop
residues such as straw, chaff, stover, and potato culls or on animal residues and
manures, especially intensive husbandry of beef and pork. Preliminary
investigations of straw availability by Agriculture Canada showed that, in the main
cereal grain areas of the prairie provinces, soil tilth and moisture concerns
demand the replacement of all the straw into the soil. This is neither practiced
nor physically possible with conventional machinery, and thus some controversy
continues. However, with a possible increase in no-till methods in the future, a
substantial quantity of straw could be available for local energy use, probably
only for combustion purposes.
Considerable work has been undertaken to hydrolyze the inulin and ferment the
fructose in Jerusalem artichokes. Near-term predictions would suggest that liquid
fuels from Jerusalem artichokes will not be economical but that it may be
profitable to produce a high-fructose syrup instead. In fact, due to the suitability
of the crop to land used for growing tobacco, it is possible that farmers may grow
it as a replacement crop due to a diminishing tobacco market.
4
HISTORY OF BIOMASS R&D IN CANADA
A number of federal government departments and agencies are involved in R&D
funding for bioenergy (Fig. 3). Since the beginning of the federal program, the
Canadian Forestry Service of Environment Canada, through the ENFOR program,
received the majority of contract R&D in silviculture, mechanization,
pretreatment, materials handling, resource and environmental assessment, and
conversion (including thermochemical and biological processes). On April 1,
1984, the management of the conversion projects was handed over to the
Department of Energy, Mines, and Resources. EMRs primary function is setting
energy policy for the country. They have some inhouse projects in combustion of
biomass, but most of their biomass R&D activity is subcontracted. Some of these
projects are large in terms of dollars because the activities are close to
demonstration and commercial application. One is the boiler test program which
has turned out to be very successful. It points out major inadequacies in boiler
design from the point of view of efficiencies and emissions. In most cases minor
modifications have caused considerable savings and improvements. The
information being generated is transferred to the boiler manufacturer and is
helped by direct involvement of the Canadian Boiler Manufacturers Association
in the project. Three other major projects are managed by EMR. The Iotech/
Monenco 1 ton of wood/day plant that uses steam explosion, enzyme hydrolysis,
and fermentation to ethanol will come on line early in 1985.
The gaseous HF solvolysis process, originally started by Canertech, is now
being managed by EMR. Issues that still need to be addressed are reprocessing
and recycling of HF and toxicity to the microbes. The lignin is unfortunately
highly modified and is expected to be used as fuel. The other is the Biosyn
projecta $21 million 10 ton/hour oxygen-blown, fluidized-bed gasifier for
MJV gas. Researchers will optimize parameters such as moisture content of
feedstocks, operating pressure, and oxygen. It is expected that after cold tests this
winter, full testing will begin in 1985.
The Natural Science and Engineering Research Council (NSERC) is a federal
government agency that grants university R&D funding. They have identified the
strategic importance of biotechnology and energy production. In these two
categories there are a number of projects of interest to bioenergy development.
Through NSERC and a number of in-house projects at NRC, some badly needed
basic research is ongoing. It is our feeling that all basic research is in real
jeopardy in Canada for very much the same reason as the U.S. Quadromania
syndrome in biomass R&D as mentioned by Dr. Lip insky.
NRC is the prime research agency of the government. The NRC Bioenergy
Program does not have its own laboratory facilities but funds projects in our other
NRC laboratories such as Chemistry and Biological Sciences. The major R&D
effort is, however, subcontracted to universities, provincial research agencies,
and the private sector including Forest Engineering Research Institute of Canada
(FERIC) and Forintek. The R&D interests of the NRC Bioenergy Program span
a wide spectrum from the use of tissue culture for producing superior trees to
downstream processing by reverse osmosis.
Agriculture Canada has parallel R&D interests to forest biomass on
agricultural residues, energy crops, and biogas. Overall, coordination among all
of these interest groups has worked reasonably well. Part of the reason has been
that the funding has come from a special envelope that is distinct from the A
base funds of any department. This has allowed each department to be an equal
partner in the relationship.
Figure 4 shows the total Bioenergy R&D expenditure trend, starting in 1977
with less than $1 million to about $14 million at the end of FY 1983 1984. The
second oil shock occurred three years after the beginning of the program. At this
point, preliminary results of the ENFOR program justified the expansion of
Bioenergy R&D in 1980, leading to an increase in the budget for conversion,
energy crops, and resource assessment and development. This expansion is also
illustrated in Figure 5, which shows the diversification of the program in a
comprehensive Bioenergy R&D effort with the greater involvement of other
departments along with Environment Canada. This diversification is further
illustrated in Figure 6, which reflects the development of the major areas of
bioenergy R&D over the last 7 years. In this analysis, the projects related to
environmental assessment have been grouped under silviculture.
5
RESOURCE ASSESSMENT
As shown in Figs, 4 and 6, resource assessment has been a significant activity for
locating the quantity of forest biomass from coast to coast. The work includes
conversion of existing volume data on timber to actual mass data on the whole
tree and the identification of biomass in relation to specific geographic location
and by groupings of type and species. It will also be possible to determine how
the quantity of biomass available for energy varies with harvesting methods and
wood utilization and to determine the accessibility of the resource. This ENFOR
activity, in cooperation with in-house work on new inventories of unsensored
areas, is close to completion, and trials have already started in some regions of
the country to determine the reliability of data.
The last two years have seen an increased effort by Agriculture Canada. Much
of this is associated with the assessment of residue availability and crops
specifically suited for energy production, in conjunction with both existing
agriculture and the cultivation of underutilized land. The purpose of this longterm program is to identify energy crops through planned cropping trials,
observation of existing plant stands, and development of agronomic practices.
6
SILVICULTURE
Projects under this title could suitably be subtitled Forest Management for
Biotnass Production in an Environmentally Sound Manner. Two areas are
achieving prominence: the growth of energy plantations, and the use of computer
simulation to evaluate the consequence of different management and
silvioultural practices. The former is the subject to considerable federal/
provincial collaboration, and participation in the IEA has resulted in significant
attention being paid to this new area of silviculture. The computer model
FORCYTE (FORest nutrient Cycling and Yield Trend Evaluator) developed
under ENFOR is an ecologically based forest management simulation designed
to estimate the long-term consequences of intensified biomass production and
harvesting. Several versions of the model have been developed, calibrated, and
tested; although originally designed and calibrated for Douglas fir ecotypes, the
model is now being calibrated for and applied to other Canadian forest ecotypes
through a series of ENFOR projects. The model has attracted considerable
scientific interest worldwide and is now being tested in several countries.
It is likely that in Bioenergy R&D, silviculture will maintain its relative share
of both budget and priority as the program evolves.
7
HARVESTING AND PLANTATION TECHNOLOGY
Considerable effort has been concentrated in this area. Under ENFOR and NCR
there have been several prototype systems constructed and evaluated in
conjunction with Forest Energy Research Institute of Canada (FERIC) and the
private sector. Highlights include:
RECUFOR. For collecting and preprocessing logging residues that remain
after short-wood harvests. A prototype was constructed and field-tested; the
concept and general design features are considered proven. Its large size and
high cost may restrict its application.
Logging Residue Processor (LRP; previously Recufor S). For collecting and
processing (fully) logging residues accumulated along roads and at landings
during full-tree harvests. Prototype was field-tested in summer 1984.
Crabe Combine. This machine, a much modified Canadian version of the
Finnish Palari Brush Harvester, is for swath harvesting and chipping brush
and small trees. The high cost and very heavy carrier may limit the potential of
this machine to special applications such as clearing rights-of-way.
Separator Shear. For chunking large pieces and separating biomass and debris
that accumulates in West Coast log sortyards. Prototype has been constructed
and successfully tested.
10
PRETREATMENT
This area covers research from physical size reduction to lignocellulosic
fractionation prior to cellulose hydrolysis. In terms of technology development,
it is a curious hybrid of commercial processes, such as the Bioshell process,
demonstration activity such as the Iotech/Monenco 1 ton/day steam explosion
process for a wood-to-ethanol plant, Westons Biohol extrusion process, and
R&D such as the University of Sherbrooke thermomechanical procedure for
lignocellulosic fractionation. Increased effort in fractionation of lignocellulosics
into lignin, cellulose, and hemicellulose is justified in terms of potential value
added to the separated fractions.
11
BIOTECHNOLOGY AND BIOGAS
These two areas have shown consistent growth within the funding envelope for
Bioenergy R&D. The scope of effort ranges from genetic engineering of yeasts
to enable them to simultaneously saccharify cellulose and utilize the sugars, to the
operation and monitoring of 400-m3 fixed-film anaerobic digesters.
The anaerobic digestion program is an example of the time scale that is
required to develop a laboratory system for a commercially viable process. It is
also a useful model for future bioenergy development since it illustrates the
economic environment in which a bioenergy process is likely to be viable.
Figure 7 shows the pattern of program support since 1976. Prior to 1976, NRC
had maintained a program looking at the treatment of food processing wastes. This
served as the basis for the development of the Down Flow Stationary Fixed Film
(DSFF) concept. Up to 1980, the majority of the work was in the laboratory,
though a number of early demonstration and pilot projects were undertaken by
Agriculture Canada. The early demonstrations, with the exception of the Roslyn
Park Farm, were technical or economic failures (or both) . The early failures of
bulk anaerobic digesters brought into focus the developments called for in
creating a viable technology. NRC, Agriculture Canada, and the private sector
have worked together to bring the DSFF and other advanced concepts to full
scale. From one successful operating STR (Roslyn Park) in 1980 there are now
11 advanced systems under construction or operation. These use both animal
manures and industrial wastes such as cheese whey.
Economically, the animal systems are successful if both energy is produced
and protein is recovered. The increased emphasis on energy/protein recovery and
environmental improvement makes such advanced digesters economic for farms
with 100150 animals.
The NRC DSFF technology has been scaled up from 50 cm3 working volume
to 400 m3, while a mobile trailer with a 1-m3 unit is being used to test the
different industrial wastes in eastern Canada. The current demonstration units are
Fig. 1. Forest biomass souree and distribution as a percentage of the total available in
1990 for energy and chemieals
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
Fig. 3. Percentage of year total expenditures 19771983 (in dollars of the year)
structure of the program and new targets were approved, extending the use of
neat alcohol in cars.
1
ORGANIZATION
It was established at the beginning that alcohol production would be undertaken
by the private sector. The government role has been to promote PROALCOOL
through long-term loans, to coordinate its implementation to achieve proper
production in the areas of high demand, to assure equal opportunities for ample
participation of the society to define regulatory policies, to enforce adequate land
use, to prevent environmental impacts by the effluents, and to guarantee social
benefits for the rural areas.
The policy-making body is the Alcohol National Council (CNAL), which is
made up of the Minister of Industry and Commerce (Chairman), the SecretariesGeneral of the Ministries of Mines and Energy, Agriculture, Interior, Labor,
Planning, and Finance, the Director for Technology of the Joint Staff of the
Armed Forces, and representatives of the private sector.
The Executive Secretariat of the National Alcohol Commission (CENAL)
implements the policies defined by CNAL. CENAL is chaired by the SecretaryGeneral of the Ministry of Industry and Commerce, and composed of the
President of the National Petroleum Council, the President of the Alcohol and
Sugar Institute, the Executive Secretary of the Industrial Development Council,
and the Secretary for Industrial Technology.
2
OPERATIONAL ASPECTS
Great effort was made to avoid establishing new institutions. Existing
government agencies performed the appraisal and follow-up of the projects. To
shorten the time between the application and the contract of the loan, the project
proposal submitted to CENAL is sent for technical and economic appraisal
simultaneously. The technical aspects of alcohol-producing units based on sugar
cane are evaluated by the Sugar and Alcohol Institute (IAA). For other raw
materials, the technical evaluation is carried out by the Secretariat of Industrial
Technology. The economic analysis is conducted by the financing agencies.
Official investment and commercial banks or private investment institutions
approved by the Central Bank of Brazil participate in the lending operation.
The financing conditions stimulate the participation of the private sector in
PROALCOOL and prevent undesirable impacts on land use. Up to December
1980, any sugar-cane-based distilleries would receive up to 80% of the total
industrial investment, and the units using other raw materials or cooperative or
farm association would be entitled to 90% of the total investment. Recently,
these values were reduced to 70% and 80%, respectively. The loans are
repayable in 12 years with a 3-year grace period. The interest rate was a fraction
corresponding to 10% of the government bonds indexed to the inflation rate and
a fixed interest rate, which varies from 3% to 5% depending both on the new raw
materials that the government decided to promote and on regional necessities. For
instance, in the depressed northeastern region, the interest rate is 3%.
Recently, however, these requisites were altered, particularly regarding
indexing, which has reached 85% plus interest. The financing of the agricultural
sector presently uses the same rate applied to other major crops.
3
PRODUCTION CAPACITY OF PROALCOOL
The operational system described above has contributed significantly to reaching
the goals established by the Brazilian government. By last October, 86 new
projects had been approved, corresponding to a producing capacity of 10.8
million m3 which is slightly above the target established in 1979. Additional
projects are being evaluated in order to meet the recently established target of 14.
3 million m3 . Individual distillery capacity ranges from 30,000 to 1,000,000 L
of alcohol per day; however, most units have a capacity of about 120,000 L/day.
Small units with a capacity of 1000 to 5000 L/day are under evaluation.
4
STORAGE AND COMMERCIALIZATION OF ETHANOL
Since liquid fuel production is under government administration, the alcohol
production program was established according to regional demand. The National
Petroleum Council defines the program for alcohol delivery and pays for the
products according to an ex-factory price. The retail price includes additional
costs for transportation and commercial return.
The storage facilities are concentrated mainly in the producing units. The
present commercialization procedures establish that 1/12 of the total production
has to be removed monthly from the distilleries. For autonomous distilleries, 1/9
of the total production is delivered monthly to the fuel distributors. In addition,
regional storage facilities are controlled by oil companies, which subsequently
distribute the product.
Anhydrous alcohol is received by Petrobras, which is the only company
responsible for its blending with gasoline. A well-planned distribution network
minimizes transportation costs.
5
ALCOHOL QUALITY
The conditions in which alcohol is produced vary from region to region and
among distilleries. Consequently adequate quality control is required to
two new programs: vegetable oils to extend diesel oil and forest energy products
to reduce the fuel oil demand.
In conclusion, the success of the alcohol program has to be viewed within the
framework of the Brazilian social and economical conditions. It may, however,
be possible that our experience in the biomass utilization might serve the
planning and launching of other alternative energy programs, particularly in
developing countries, notwithstanding the site specificity of any energy from
biooass project.
SYNOPSIS
This paper presents the status and progress of the Biomass Demonstration
Program created by the State Agricultural and Forestry Residue Utilization Act
(SAFRUA) of 1979. The paper highlights the 17 projects that have received
SAFRUA funding to date and discusses various aspects of the program,
including economics, financing, risk sharing with industry conversion
technologies, and resource potential. The paper also explains how to participate
in the program.
To accelerate the development of biomass resources, the state legislature
established the Biomass Demonstration Programto promote the immediate
development and implementation of residue conversion as an energy generating
technology, and to provide funds to encourage the development and
demonstration of residue conversion.
The state has recognized that converting biomass residues to energy is
instrumental in reducing industrys dependence on costly, nonrenewable
resources. Biomass is a low-cost energy source compared to natural gas and oil.
For the industrial sector, abundant supplies of biomass fuels were available at
costs ranging from $0 to $2.50 per million Btu (1983 dollars) while costs of
distillate fuel oil and natural gas prices were $5.80 and $5.75 per million Btu,
respectively (1983 dollars).
Biomass energy projects generate jobs and revenues and have created a
fledgling industry in California. The $8.8 million provided by the SAFRUA
account is supported by private industry commitments of $45.6 million. The 17
SAFRUA projects are estimated to generate $54 million in capital invest ment,
$244 million in gross sales, $73 million in gross income, and over 3600 jobs.
SAFRUA projects also enable seasonal operations in the agriculture and forest
industries to extend productivity, produce fuel, and preserve jobs.
1
DEMONSTRATION PROGRAM: PROJECT STATUS
The California Energy Commission (CEC) originally focused on funding
projects that would demonstrate collection equipment and three major energy
conversion technologies: direct combustion, methane fermentation, and
gasification. Emphasis was placed on projects that would be technically
successful and easily duplicated at other sites. As the program progressed, projects
that proposed designing and building new prototype biomass equipment were
included.
Of the 17 projects in the SAFRUA program (see Table 1), three are completed
and have repaid $270,000 in SAFRUA loans. Five projects are in the final
shakedown stage and operating successfully. Four projects are experiencing
shakedown problems, and special consultants have been brought in through the
CECs technical support contract to identify and resolve the problems. Five
projects are in the engineering, design, or construction phase.
As a result of the states effort over the past four years, a biomass energy
industry is slowly emerging in California. In addition to the 17 SAFRUA
projects, many other biomass systems are under construction or in operation
throughout the state. But without further demonstration of some technologies,
many privately funded projects will not be constructed. To ensure the growth of
a healthy biomass industry in California, the state must continue to assist in the
demonstration of new and risky biomass energy projects.
2
BIOMASS ENERGY DEVELOPMENT IN CALIFORNIA
Many of the 17 SAFRUA projects are the first in California to convert
agricultural and forestry residues to energy. Today California leads the nation in
the development of biomass energy as a result of technical and economic data
gained from the demonstration of these projects. Collecting, processing, and
converting biomass residues to energy will keep dollars in the states economy,
creating jobs and revenues. Figure 1 shows the typical annual generation and
current energy use of biomass residues in California.
There are major costs, uncertainties, and unique technical problems associated
with converting biomass residues to energy. Orchard and vineyard prunings were
the first agricultural crop residues to be used for fuel due to the high costs of
current disposal practices. Californias agricultural crop residues with the greatest
energy potentialcotton and corn stalks and rice, wheat, and barley straws
have not been demonstrated due to collection and energy conversion difficulties.
Some agricultural industry residues, such as shells and fruit pits, are currently
being used as fuel in biomass energy systems throughout California. Others, such
as rice hulls and gin trash, have substantial problems that must be overcome.
Each year in California, over 2 million tons of animal manures are collectible for
Table 2:
Number of constructed facilities by type as of December 1983
Technology Agricultura Agricultur Manure Forestry
l Crop
al Industry
Residual
Forestry
Industry
Direct
4 (2)
10* (5)
1 (1)
0
97* (3)
Combustio
n
Methane
N/A
1
5 (3)
N/A
N/A
Fermentati
on
Gasificatio 0
1
N/A
0
0
n
Collection 10 (5)
0
N/A
0
N/A
**
Note: N/A Indicates not applicable.
* Indicates number of boilers.
** Includes combustion projects using agricultural crop residues.
( ) Number of projects in SAFRUA program.
Source: CEC staff calculations
Chaparral
and Urban
Residues
2
N/A
1
4
3
ECONOMIC BENEFITS
Californias agriculture and forest industries offer the greatest opportunities for
biomass energy investments. Energy costs and availability are dominant factors
affecting a firms ability to produce products competitively. As electricity, oil,
and natural gas prices increase, finding new methods for controlling energy costs
and selecting more reliable, cost-effective energy sources become increasingly
important. Several firms have found it economical to modify existing boilers or
install new boilers to use biomass as a primary energy source. A cost comparison
of various biomass and conventional fuels is illustrated in Fig. 2. Residues such
as manures can be generated and used on-site at little or no cost.
The cost of energy from biomass is influenced by the type of conversion
technology, feedstock cost, project location, and ownership structure. The
economic benefits are twofold in projects where biomass that was previously
hauled away at a cost for transportation and dumping is now being used to
generate energy and revenues.
The economics of biomass energy projects are improved considerably when
electricity is generated through cogeneration because overall system efficiency is
greater when compared to generating electricity and process heat separately.
Firms can either use the electricity in-house, thereby achieving significant cost
savings by deferring the cost of purchasing electricity from the local utility, or
sell all or a portion of the generated power to a utility at published avoided
cost prices.
The industrial sector can produce electricity from biomass at a lower cost than
purchasing electricity from the utility. Figure 3 shows the projected prices of
electricity and natural gas to the industrial users and the approximate cost of
producing electricity in a biomass-fueled power plant. As the cost of purchased
electricity increases, more industrial energy users will develop biomass energy
projects. Biomass energy facilities could also locate in industrial parks to meet
several end-users steam and electricity demands. A mutually beneficial
arrangement allowing the energy producer to receive a reasonable payment for
electricity and still charge the industrial user less than the cost of electricity
purchased from the utility has the added benefits of increasing energy selfsufficiency and stimulating the local economy through economic growth and
jobs. Figure 4 displays the economic benefits resulting from the SAFRUA
programs $55 million investment in biomass energy projects.
4
A STEP TOWARD SELF-SUFFICIENCY
Heavily dependent on conventional energy sources, forest and agricultural
operations can be jeopardized by disruptions in energy supplies. This risk,
coupled with rising electricity costs, makes using biomass resources for energy
conversion an attractive alternative in California. California produces 50 million
dry tons of biomass residues each year. However, only 17 million dry tons are
available for energy purposes, and only 10% of this amount is being used.
While the energy equivalent of nearly 17 million barrels of oil is available
annually from biomass, the forest and agricultural industries of California used a
total of 53 million barrels of oil equivalent in 1980. These industries generate a
variety of biomass residues annually, including straws, stalks, and prunings; fruit
pits, nut shells and hulls, gin trash, and other processing wastes; animal manures;
thinnings, slash, and culled logs; edgings, shavings, bark, trim, and sawdust; and
chapparal and other woody residues. Industry residues are more economical to
convert to energy than crop residues or forest slash because they are centrally
located. The agricultural and forest industries will benefit greatly from biomass
energy development because of the increasing uncertainty of conventional
energy costs and supplies. These industries are also burdened with high residue
disposal costs and environmental disposal problems. Figure 5 illustrates that both
the agriculture and forest industries can be energy self-sufficient.
The 17 SAFRUA projects displace many forms of conventional energy used
by the forest and agricultural industries. Five of the projects cogenerate; five
produce only electricity; three produce only process heat or steam; and the
remaining four involve fuel collection or fuel cleaning and handling.
The annual energy that the SAFRUA projects produce is equivalent to 622,000
barrels of crude oilenough energy for 60,000 homes. All five cogeneration
projectsFarmers Cooperative Gin, Superior Farming Company, Big Valley
CEC projects
Air pollution
TSP
RHC
NOx
CO
Solid waste
1,500
1,700
550
18,000
37,400
14 ,000
16,000
5,900
183,000
321,000
6
RISK SHARING WITH INDUSTRY
The advantages of biomass energy are weighed against the costs and availability
of conventional fuels, the competing uses of biomass resources, and the capital
investment for conversion facilities. Because biomass energy is relatively new,
potential users and investors are uncertain of the advantages and risks involved
with constructing and operating a biomass facility. To accelerate the near-term
applications of energy-from-residue technologies, the state legislature established
the SAFRUA program to pro vide technical support through staff expertise and
outside consultants to resolve problems that may occur.
The industrys commitment of over five dollars to every state dollar invested
demonstrates the success of the program. The $8.8 million encumbered by the
SAFRUA program has been matched by $45.6 million in industry investments.
Not only is the money providing short-term financing for projects that will result
in net benefits to investors of almost $180 million during the estimated 20-year
life of the projects, but most of the $8.8 million will return to the SAFRUA
program for reinvestment in additional biomass projects.
Money is not the only support that the SAFRUA program provides. The CEC
offers technical assistance through staff engineering, troubleshooting, and
financial and marketing expertise. Over the longer term, SAFRUAs success will
be measured by the extent of further applications of biomass energy technologies
by commercial enterprisesusing their own capital.
7
TECHNOLOGY TRANSFER
The states ability to develop a biomass energy industry depends on more than
providing money for demonstrations, and developing a successful biomass
project takes more than just seeing a similar project in operation.
Since the state funds only a few projects compared to the total potential for
biomass energy applications in California, the CEC must market successful
demonstrations to achieve widespread technology development. Marketing is the
states way of communicating with the private sector, which finances and
constructs spinoff biomass energy projects. These spinoff projects represent
the force behind the creation and sustenance of a biomass energy industry in
California.
The CEC markets biomass energy in several ways. For instance, SAFRUA
demonstrations are developed into showcases that encourage others to invest in
similar projects. Staff members prepare brochures on the technical, economic,
and environmental aspects of each project. Workshops and press conferences,
held near the completion of a projects shakedown, enable potential users,
manufacturers, consultants, utility representatives, government officials,
legislators, the media, and others to discuss what has been learned and how
future projects can be improved.
The CEC also sponsors innovative equipment demonstrations. In 1982 and
1983 collection equipment demonstrations for forest slash, precommercial
thinnings, and chaparral attracted over 500 people.
The CEC prepares reports, handbooks, and other written information on a
variety of subjects. The CEC prepared an investment prospectus, which was sent
to banks, lending institutions, and potential investors outlining the financing
options, tax incentives, and the overall economics of biomass energy projects.
Project developers need information on biomass fuel suppliers and users,
financing, feasibility studies, economic and technical data, and resource
availability. Advertising the success of SAFRUA-funded demonstration projects
will stimulate private industry investment in other biomass energy projects.
8
BIOMASS ENERGY TECHNOLOGY
The three major biomass energy conversion technologies are direct combustion,
methane fermentation, and gasification. Technologies to collect and process
biomass residues are also important, since most of Californias biomass is in
fields or forests.
Direct combustion converts biomasscomposed primarily of carbon and
hydrogento useful heat. Examples of direct combustors include incinerators,
kilns, and fluidized-bed or suspension-fired burners. Direct combustion of woody
residues is well demonstrated. Agricultural residues that contain high
concentrations of ash, nitrogen, and moisture, such as rice straw, make
combustion difficult and have only recently received developmental and
commercialization attention.
In methane fermentation, organic matter in the absence of oxygen is
biologically broken down into organic acids and carbon dioxide, which are then
bacterially converted to biogasprimarily methane and carbon dioxide. The
biogas is burned in systems that typically use natural gas to produce space heat,
process steam, or electricity. Methane fermentation is a mature technology in
municipal wastewater treatment but is not fully developed using biomass as a
feedstock.
Like direct combustion systems, gasification systems involve combustion of
biomass, but in an oxygen-deficient environment. They are designed to produce
a combustible gas rather than to release heat immediately. The gasification
process produces a product gas that can be converted into process heat, steam,
electricity, or liquid fuels. Gasification has potential for small-scale electric
generating systems less than 1 megawatt (MW). In California, gasification
systems that will be used with boilers are currently in shakedown operation.
Gasification systems to be used with engines have not yet been demonstrated as
commercially feasible. Because gasification is cost-effective for special
applications such as irrigation pumps and can utilize large amounts of residues,
the CEC has emphasized gasification in its current project solicitation.
Technologies to collect and process forest slash, agricultural crop residues,
and orchard prunings have not been developed extensively. This, coupled with
the dispersion of these residues, is an impediment to more widespread use of
biomass for energy. Funding collection systems will help identify and resolve
uncertainties such as hardware, operational management, and economics. In the
future the CEC also will focus on demonstrating projects that integrate biomass
collection with energy conversion technologies.
9
HOW THE BIOMASS DEMONSTRATION PROGRAM
WORKS
The CEC seeks biomass demonstration project proponents through the Request
for Proposal (RFP) process, including bidders conference and public
workshops.
Before selecting a proposal, the CEC analyzes project-specific information on
the biomass resource base, system design and engineering, environmental
impacts, and technical and economic feasibility. The CEC staff works with the
project proponent to obtain permits and financing and resolve other barriers as
they occur. During the construction, shakedown, and operation phases of the
projects, the SAFRUA program is supported by technical experts who are called
upon when problems occur.
Funding from the SAFRUA account is accomplished by purchase/buyback
agreements between the state and the project proponents. The state provides up
to 50% of project funding for the purchase of biomass energy equipment. Once a
project meets performance criteria established during contract negotiations (see
Table 4), the project proponent is obligated to buy back the equipment within
90 days. If a project does not meet the performance criteria, the state may secure
the equipment for resale; otherwise a reduced repayment, in proportion to the
degree to which the project meets specifications, can be negotiated. Once
repayment is accomplished, the monies are returned to the SAFRUA account to
be used for other projects.
SAFRUA provides the project proponent with short-term capital
corresponding to the cash flow requirements of a project during its construction
and shakedown phases. The funding also provides a demonstrated system for a
financial institution to evaluate prior to financing its further commercialization.
TABLE 4
CRITERIA FOR SELECTING SAFRUA PROJECTS
IMPACT ON TECHNOLOGY COMMERCIALIZATION
Fig. 1. Residues generated, available, and used for energy (million dry tons)
Fig. 4. Economic impacts and benefits of Safrua program (Investment of $55 million)
The biomass program of the New York State Energy Research and Development
Authority (NYSERDA) encompasses a wide range of topics in resource
development, direct conversion, production of fuels and chemicals, and resource
recovery. In the following paper, I will outline the underlying philosophy and
rationale that tie together what would seem to be a disparate collection of
activities, and then I will report on some recent progress in five major program
elements.
1
PHILOSOPHY AND RATIONALE
Biomass research and development sponsored by New York State fills a gap left
between the activities of the U.S. Department of Energy (DOE) and those of
private institutions. This work bridges Congressman Fuquas river between the
federal promotion of long-term, high-risk technologies for the benefit of the
nation as a whole and the pursuit by private investors of shorter-term, limitedrisk opportunities that promise a high return. NYSERDA accepts greater risks at
earlier stages of technology development than private corporations can on their
own, but we are not restricted to high-risk technologies that must be justified in
terms of potentially very large contributions to national energy supplies. States
are also in a position to be more sensitive to local needs and opportunities than a
federal agency is likely to be; they are also motivated by a broader range of
public interests than business profits alone.
From a national perspective, our program represents a medium for technology
transfer, while at a local level, potential adopters of new energy technology view
us as a resource for innovation. Our role is some what analogous to that of
industry consortia like the Electric Power and Gas Research Institutes, but it is
defined in terms of geography rather than industry sector.
Some of the apparent technical diversity of our program is the result of
following a market pull as distinct from a technology push paradigm of R&D
management. While federal programs sometimes begin by developing a
technology that appears promising to the research community and then attempt to
commercialize it afterward, our approach seeks first to identify a need and then
to support the innovation required to meet it. For this reason, the unifying
concepts in our program revolve around classes of users and applications rather
than categories of technologies to be applied.
Based on NYSERDAs energy experience, states are uniquely suited to
facilitate traffic in technology between the public and private sectors. Aware of
local needs, state officials can better pursue solutions to identified problems in
contrast to seeking applications for identified technologies. Also, smaller, less
minutely compartmentalized agencies are more capable of devising approaches
that would involve many separate jurisdictions and missions at the federal level.
Unfortunately most states, especially smaller ones, are without the means to
support individual R&D programs. To overcome this limitation, NYSERDA has
attempted to enlist other states in the Northeast to pool resources in a joint
biomass endeavor based on several common interests: heavy oil dependence,
similar forest and agricultural resource characteristics, like climatic conditions,
interdependent economies, incidences of rural poverty throughout the region, and
a shared natural environment and common coastline. Benefits would include
elimination of gaps and duplications through joint planning and coordination,
improved capacity to sponsor worthwhile projects too large for individual states
to support separately, reduction of technical risk through diversification, and
more efficient utilization of the specialized capabilities of research institutions
within the region.
2
PROGRAM OVERVIEW
The NYSERDA biomass and waste programs are composed of five elements,
funded at annual levels of approximately $0.5 million each. This amount is
leveraged two to three times through cofunding arrangements with other
institutions. The elements, organized by market application, are resource
development, direct conversion, production of fuels and chemicals,
resource recovery from municipal solid waste, and energy from sewage
treatment processes.
Research in resource development is designed to increase the availability of
suitable biomass raw materials over the long term and to minimize the
environmental and economic costs of exploiting them as an energy source.
Direct conversion technologies permit generation of useful heat and mechanical
power, primarily from wood in small- to medium-sized installations located
where the biomass grows. Production of high-valued fuels and chemicals
provides transportable and marketable substitutes for products now refined from
petroleum and natural gas. Finally, the municipal solid waste and sewage
treatment process R&D is designed to reduce the cost of environmentally
acceptable disposal through generation of energy as a revenue-producing byproduct. The following discussion details progress made within the past year and
planned future actions in each of these areas.
2.1
Resource Development
We agree that surplus wood is sufficient, at least in New York and throughout
most of the Northeast, to support foreseeable biomass energy applications for the
next several years. We are also concerned that the resource may eventually
become limited either through expansion of the traditional forest products
industry or through new uses of trees. For this reason, our resource development
research addresses potentially large sources of supply for the next decade and
beyond. Recent accomplishments include the expansion of the State University of
New York College of Environmental Science and Forestrys research into the
suitability of different hybrid poplar varieties for intensive energy farming under
soil, climate, and pest conditions prevalent in New York. The data and expertise
resulting from NYSERDA-supported work will provide a unique resource for
organizations considering raising fast-growing trees for energy. The Reynolds
Metals Company in Massena, which now has approximately 200 acres in
cultivation, is an example of the type of audience that receives these research
results through technical bulletins and workshops, as well as formal reports.
In addition to working to develop fast-growing trees, we are cooperating with
several other institutions to evaluate the mechanisms of possible forest decline
and the effects of atmospheric deposition on forest nutrient cycles and tree
metabolic processes. The knowledge generated from this activity may help to
preserve the existing forest resource for all uses, including energy.
After several years of research supported by NYSERDA and the gas industry,
the State University of New York at Stony Brook has established a successful
test farm for promising seaweed species in Long Island Sound. Data from the
farm will form part of the basis for an evaluation of the economic prospects for
maricultural energy, and future work, if warranted, could be expanded to include
processes for converting seaweeds to fuels and chemicals. The growing and
processing of seaweeds represent a potential new industry for coastal
communities as well as a source of energy.
NYSERDA also has completed a survey of the prospects for energy from
herbaceous (nonwoody) plants in several regions of New York. The study
evaluated the advantages and disadvantages of growing 49 different species, and
although this resource is generally costly compared to surplus wood today, the
report will be useful for future developers of energy farming concepts.
Finally, within the past year the Authority began to build upon an earlier
statewide survey of peat resources by assessing the technology, economics, and
environmental impacts of exploiting peat as a fuel in New York. This multiyear
project, managed with the advice of a formally constituted panel of
representatives from relevant state agencies and interested institutions to assure
that the analysis is objective and thorough, will enable the state to formulate
policies that balance energy benefits with environmental costs and will provide
$20 million facility in the state to process 630 tons of wood per day. This type of
plant will make possible the production of fuels and energy-intensive chemicals
as other downstream conversion technologies emerge.
Ethanol (ethyl alcohol) has the advantage that it can be derived from biomass
and blended readily with gasoline for motor vehicles. Although ethanol for fuel
blending is now generally made from grain, it also can be produced from wood,
which is abundant in the Northeast. NYSERDA completed an investigation last
year into the most promising near-term technology for conversion of wood to
ethanol and identified the Brink acid hydrolysis approach as closest to
commercial application. As a result, the Authority is sharing the cost of a testing
and evaluation project designed to lead to construction of a $25 million
installation to produce 2.75 million gallons of ethanol per year and a variety of
coproducts at a site in Franklin County.
A longer-term wood-to-ethanol prospect is being evaluated in a site-specific
design and feasibility study for a facility employing enzyme hydro lysis at a
location in Jefferson County. The project, jointly sponsored by NYSERDA and
DOE, will explore the potential for commercial adoption of a technology
advanced in part through federal programs. The next stage of development, if
warranted by the studys fundings, is likely to be supported by NYSERDA and a
private industrial concern, with diminished dependence on DOE.
Another milestone in our petrochemical substitutes program was the
completion of the first phase of a laboratory effort at Rensselaer Poly-technic
Institute, cost shared by two industrial concerns, to perfect a process for
converting sugars derived from wood to useful chemicals that could substitute
for petroleum- or natural-gas-derived materials. The initial results indicate that
fermenting 2, 3-butanediol from glucose is feasible and economically promising.
Planned subsequent work will examine xylose as a raw material and will attempt
to produce glycerol and succinic acid as additional products.
2.4
Municipal Solid Waste(MSW)
One of the principal barriers to further construction of MSW combustion
installations is the public concern over dioxin emissions. In addition to
cosponsoring a national workshop to standardize dioxin sampling and testing
protocols, NYSERDA is supporting one project in Albany and organizing
another in Pittsfield, Mass, to ascertain the relation of emissions of dioxins and
similar substances to the operating characteristics of MSW-fired boilers. The
results will be useful for improving system design, introducing more effective
operating procedures to control emissions, and informing the public on resource
recovery facility siting.
Other projects within our program involve pretreatment processes for MSW.
At one facility we are testing a slow speed rotary shear shredder in a side-by-side
comparison with an existing hammermill. At another site, a novel rotary drum air
classifier design is being tested with both shredded and unshredded municipal
waste.
After completing several projects on landfill gas recovery and utilization,
NYSERDA and the Gas Research Institute are conducting trials of leachate and
condensate recycling as means to accelerate gas production and to mitigate
environmental impacts. The project, located near Binghamton, should result in
better ways to design and operate landfills from the perspective of economical
gas production as well as environmentally acceptable waste disposal.
2.5
Sewage Treatment Processes
NYSERDA supports a wide range of activities that contribute to improved
energy recovery from sewage. Two research projects are designed to enable
improved management of anaerobic digestion processes. One involves the use of
pure starter cultures to establish optimal microbial populations. The second is
exploring parameters other than temperature and acidity that could provide early
indications of process upsets and eventually form the basis for improved
feedback control.
In part to addressing the unique problems associated with disposal of sewage
sludge contaminated with chromium from tannery wastes, NYSERDA, the
Commonwealth of Massachusetts, and gas companies from New York and New
England are supporting construction of a pilot plant that employs the Institute of
Gas Technologys two-stage anaerobic digestion process. The plant will provide
the design basis for a possible full-scale installation at Salem, Mass., and at other
communities, in New York and elsewhere, faced with similar sludge disposal
problems.
Our new program of sewage treatment plants as total energy systems
contributes to solving immediate problems at New York municipal installations
on a cost-shared basis. Projects initiated at various locations this year include
oxygen compressor modifications, new process air blowers, fume incinerator heat
recovery, improved sludge dewatering and incineration, and wastewater coupled
heat pumps.
SYNOPSIS
The Gas Research Institutes (GRI) Program on Methane from Biomass and
Wastes was originally an initial high-risk, long-term project to achieve a large
quad impact from a marine biomass feedstock. The program evolved to a more
comprehensive yet integrated effort to produce low-cost gas from wastes and
biomass/waste blends in the near- to midterm, and cost-competitive gas from the
much larger biomass resource base in the longer term. A systems approach is
utilized to develop and integrate the multidisciplinary technologies to provide
supplemental supplies of natural gas in the near term and to meet long-term gas
demands in the future.
Emphasis has been placed on the development of regional biomass and waste
to methane programs to provide the gas industry with the flexibility to scale
methane production systems up or down in response to local energy demand.
Participation and cofunding by regional entities are measures of success of these
programs and have been encouraged to further leverage GRIs R&D dollar and
promote earlier market penetration by alternative energy systems.
1
INTRODUCTION
Baseline projections of U.S. energy supply and demand indicate that
approximately 8 quadrillion Btu (quads) of gas per year may be required to
supplement diminishing supplies of natural gas after the turn of the century.
Researchers have estimated that roughly 25% and 40% of the U.S. gas supply
will be from supplemental sources by the years 2000 and 2010, respectively
(Ref. 1). Supply technologies under development are expected to provide lowcost gas within this timeframe to the mutual benefit of the gas consumer and the
gas industry.
One of the three supply options being investigated by GRI is the production of
methane from biomass and wastes. GRIs Methane from Biomass and Wastes
Program has evolved from an initial high-risk, long-term project using a marine
4
METHANE FROM BIOMASS
The objective of this component is to develop the technology base to produce
low-cost methane from biomass through the application of engineering and
agricultural sciences. To meet this objective, we are addressing research and
development sensitivities in biomass-to-methane systems to yield low-cost gas
by the year 2000, and applying advanced biotechnologies to biomass production
and conversion systems to reduce the cost even further.
Presently applied R&D is focused on conversion of sorghum, napier grass, and
water hyacinth, and on high- and low-solids anaerobic conversion systems. By
1990, one crop and the appropriate conversion system will be selected to
demonstrate the component parts of a biomass-to-methane system. The technical
and economic feasibility of this system will be established and if warranted a
subsequent development effort will be supported to bring an integrated system to
the point of commercialization by the year 2000. In a second parallel effort,
longer term R&D will be conducted to apply advances in biotechnology to
further reduce the cost of methane by improving plant yields while reducing
energy inputs such as fertilizer and water addition, and by increasing
bioconversion rates and yields.
5
REGIONAL METHANE FROM BIOMASS AND WASTE
PROGRAMS
GRI has adapted a regional strategy to develop the technologies to produce
methane from biomass and wastes and to foster the necessary infrastructure to
eventually commercialize such systems. This approach, involving local
institutions, researchers, agricultural organizations, and gas companies through
cofunding and participation in the R&D, benefits both the gas industry and the
regional participants. A local supply of energy provides socio-economic benefits,
can be utilized at a relatively small scale, requires less risk capital, and has a
relatively short time frame for facility implementation. Estimated capital costs of
biomass and wastes systems are in a range affordable to municipalities and other
local entities as well as individual industries. The R&D dollars required for
demonstration and scale-up are also less than those required for most other
supply options.
5.1
The GRI/IFAS Southeastern Regional Program
This pioneering regional program, cofunded by GRI and the State of Florida
through the Institute of Food and Agricultural Sciences (IFAS) at the University
of Florida, was initiated in July 1981. Florida consumes 2.5 quads (2.610 GJ)
of primary energy, of which 0.1 quads (4.110 GJ) is methane. Only 10% of this
amount is from in-state gas wells, which are being rapidly depleted.
Floridas number one industry is agriculture; there is a plentiful supply of
biomass and an environment that is conducive to year-round biomass production.
As a result, research on biomass energy systems was under way at many of IFASs
22 research centers and in 18 academic departments at the University of
Floridas Gainesville campus prior to the GRI/IFAS regional program. Much of
this existing expertise was brought into the GRI/IFAS regional program by
refocusing existing projects and initiating new projects with the common goal of
producing low-cost, pipeline-quality methane.
An initial objective was to examine a variety of energy crops to determine
their potential for methane production and to define energy crop management
techniques to maximize biomass production. Over 250 varieties were evaluated.
In many cases biomass yield could clearly be doubled simply by close-spaced
planting for energy, which is not permissible for food or fiber crops. Napier grass
and water hyacinth were selected as model feedstocks on which to focus R&D for
an integrated biomass-to-energy system. Other promising biomass feedstocks are
still being pursued as part of a longer-term research effort.
Growth and production models developed for napier grass and water hyacinth
exhibited an excellent fit to actual field data. In accordance with the solids
content of the two feedstocks, both high- and low-solids bioconversion systems
are being developed. Emphasis was placed on a high-solids type conversion
system in the IF AS program, and a series of large bench-scale units are being
used to test a two-phase leachate bed/packed bed conversion system with highsolids feedstocks such as napier grass. This work is supported by other highsolids conversion studies being conducted at Cornell University as part of the
Northeast Regional Program and low-solids work which is being conducted
primarily at the Institute of Gas Technology (IGT) in Chicago.
As part of the longer-term, more advanced portion of the program, work on
tissue culture propagation and genetics is under way. Somatic hybridization and
other genetic approaches are being applied to produce new plant strains, and
fluidized gel seeding techniques for planting tissue culture embryos are being
developed. New strains of methanogenic bacteria have been isolated that have both
cyst-like and disaggregated stages in their life cycles. The importance of these
bacteria to methane generation is being studied. Work on cellulases has isolated
and characterized cellulase mRNA and cDNA. This work is being focused on the
important cellulytics found in anaerobic digesters.
A site-specific study conducted in the Lake Apopka Natural Gas District
located near Orlando revealed that the preferred approach at this site was a
dedicated energy farm to grow either napier grass or water hyacinth (Ref. 4).
Economies of scale indicated that for pipeline quality gas, a facility size of at
least 10 Btu/yr would satisfy most of the gas needs of the local gas district. Land
or lake requirements for energy crop production to supply this need were
available. The study also examined the cost of methane production using state-of-
Biomass yields of hybrid sorghum grown in small test plots are twice those for
commercial grain sorghum, 16 versus 8 dry ton per acre-yr (35.8 versus 17.9 dry
metric ton/ha-yr) . Some of these high-energy sorghums are also low in lignin,
which significantly improves bioconversion to methane as illustrated in Fig. 3.
High methane yields, 5.0 to 6.4 ft3/lb (0.3 to 0.4 m3 /kg) volatile solids (VS)
have been obtained with some of these varieties. The theoretical maximum is 7.5
ft3/lb (0.46 m3/kg) VS. This indicates that plants such as Brown Mid Rib can be
tailored for maximum methane production. A reduction in volatile organic
matter of 94% is also possible with sorghum, the highest observed to date for any
biomass energy crop.
Economic and technical sensitivities are being determined and incorporated
into a system model to evaluate various sorghum production/ conversion system
configurations. Analyses indicate that feedstock costs and methane yields have
greater impact on total costs than does methane production rate. To obtain
maximum methane yields from the feedstock, a conversion process that provides
a long solids residence time is desired. Consequently a simple, low-cost
conversion system for methane production and upgrading to pipeline-quality gas
is being developed.
5.3
The GRI/NYSERDA/NYGAS Northeastern Regional Program
A third regional program has been recently established in the northeastern United
States, which depends on imported oil at the rate of about $20 billion/yr to meet
the bulk of its energy needs. About 0.5 quads/yr (5.210 GJ/yr) of energy is
derived from biomass in the Northeast, mostly from the direct combustion of
wood. Researchers estimate that this quad impact could be increased fourfold
simply by utilization of the areas 73 million acres (29.2 million hectares) in
forest land resources (Ref. 5).
The northeastern region of the United States also has one of the highest rates of
gas usage per capita. The development of a regional supply of gas could be
particularly important in winter months when localized gas shortages can occur.
While the goals of this regional program are similar to those of the other regional
programs, the structure and development of the program are quite different. This
program arose from the recognition that several projects being cofunded by GRI,
New York State Energy Research and Development Authority (NYSERDA) ,
and the New York Gas Group (NYGAS) would be best integrated into one R&D
program. The management structure of the program is based on the cooperative
efforts of these three funding organizations, none of which conducts in-house
R&D. Projects for inclusion into this program come from unsolicited proposals or
solicitations, and the three participants must agree on new projects.
The northeastern program began in 1983 as a collection of existing projects,
and is evolving into a more focused and structured program. The projects
initially addressed included landfill gas recovery, woody and marine biomass
2.
3.
4.
Holtberg, P.D., Woods, T.J., Ashby, A.B., Dryfus, D.A. and Hilt, R.H. GRI
baseline projection of U.S. energy supply and demand 1983 2010, Gas Research
Insights, 1984.
Bird, K.T. and Ashby, A.B. Recent economic results of converting biomass to
methane, Symposium Papers, Energy from Biomass and Wastes VIII, Institute of
Gas Technology, January 1984.
Wilkey, M.L. and Edwards, G.N. An economie analysis of the biological
gasification of municipal solid wastessludge blends by the RefCoM process,
Energy from Biomass and Wastes IX, January 1985.
Warren, C.S. et al. The methane from biomass and wastes program task 1:
Evaluation of the Lake Apopka Natural Gas District, Gas Research Institute
Topical Report, GRI 84/0015.1, 1984.
7.
1
CASE STUDIES
Case Study Number 1
Palo Alto, Calif., is adapting its water treatment plants incineration process to
decrease natural gas required for incineration. Methane from the citys landfill
adjacent to the plant will also be recovered and used for supplemental power
generation. The project reduces natural gas requirements by 50% (41,250
therms) and provides about 2.2 MW of capacity and 17,600,000 kWh of energy.
Total project cost is estimated to be $5 million.
Case Study Number 2
The municipal utility at Lakeland, Fla., has been harvesting cattails and water
hyacinth used for tertiary water treatment as a supplemental fuel supply at its coaland-garbage-fired, 364-MW McIntosh No. 3 Generating Unit. The utility is also
test burning peat and wood chips as supplementary fuels.
Case Study Number 4
scrutiny are systems involving a downtown area, providing process steam or hot
water to commercial users, or the option of cross-connecting a coal-fired,
cogeneration facility run by a paper company.
Hibbing has been selected by the Iron Range Resources and Rehabilitation
Boardan agency of the State of Minnesotaas its prime site for a wood
products park. Hibbing was not only the most suitable for immediate
construction, but the utility could use the wood residue to fire one of its boilers.
Hibbing will also extend its district heating system to provide both heating and
process heat to the new industrial park. Both Wood Industrial Park and Hibbing
Utility customers will benefit from lower fuel costs and expanded utility use. The
first phase of the wood products park would include a high-efficiency, small-log
sawmill to meet regional demand, a lumber concentration yard for upgrading, a
dry kiln, a planner mill, a hog mill, a chemical mixing plant to supply regional
processors with formulations for wood stabilization and preservation, and a wood
dip-treating facility. The first phase of construction will require approximately $5
to $6.5 million. Funding will be provided through a combination of grants, loans,
and bonds from the Iron Range Resources and Rehabilitation Board, industry,
and financial institutions; tax increment financing from the City of Hibbing; and
industrial revenue bonds. The 180-acre Hibbing Industrial Site, known as Agnew
#3, owned by the city and purchased for industrial growth, will be totally
dedicated to the Wood Industry Development.
Researchers working for the City of Virginia are cooperating with the
Minnesota Department of Energy and Economic Development to test use of
peat, a sizable state resource, as a fuel in conventional boilers for production of
electric and thermal energy for district heating systems. The researchers hope to
prove peats value and determine maximum sustainable capacity of the boiler
relative to its design capacity using peat and peatcoal mixtures, boiler efficiency,
and gas and particulate emissions from the boiler. Three test burns have been
used to determine the feasibility of using peat as a boiler fuel or fuel extender in
a power plant. The utility burned a mixture of 25$ peat and 75% coal during the
first test, 50% peat and 50% coal in the second test, and 75% peat and 25% coal
in the third. Also, the Virginia utility is cooperating with local suppliers to
examine the feasibility of using wood wastes as an alternative fuel extender in
their conventional boilers to produce electric and thermal energy.
Case Study Number 6
Endicott, N.Y., has committed to the construction of a peaking plant that will
burn methane gas produced by the villages sewage treatment plant. The plant
will be an adjunct of the sewage plant and will store methane gas for use during
the peak winter months. The 750-kW facility will also serve as an emergency
back-up power supply for the treatment plant. Gas produced during off-peak
months will be stored in stationary tanks adjacent to the plant. A feasibility study
completed under grants from the Appalachian Regional Fund and APPAs
DEED programindicated that the village could save $500,000 over a 20-year
period if the plant is built.
Case Study Number 7
The utilities in Sturgeon Bay, Wis., have adopted an innovative energy system
involving heat pumps on a newly constructed sewage treatment facility. The
facility won Wisconsins State Award for Engineering and Energy Conservation
and honorable mention in the same category at the national level. Methane gas is
produced in anaerobic digesters to heat the digester and stand-by generators. A
heat pump system using heat from the sewage effluent heats treatment plant
buildings. Two 350-kW generators that operate on natural gas and methane were
installed; each is large enough to operate the entire wastewater treatment plant.
Generators are also installed in parallel with the electric system for peak shaving.
2
CONCLUSIONS
Experience from these projects and recent developments points to future
initiatives that include controlled breeding programs to enhance biomass
production; recovery of plant oils as replacements for petroleum; and the use of
genetic engineering to enhance methane production by anaerobic bacteria, to
recover resources from wastes, to control pollution, and to create synthetic
systems for photosynthesis. With locally available bioenergy resources,
appropriate load characteristics, favorable institutional climates and a lower cost
of money, and community energy requirements wellscaled to smaller generating
facilities, the nations publicly owned electric utilities have proved that they can
economically convert biomass to energy.
In the late 1970s at the Alaska Village Electric Cooperative (serving some 40
villages all over Alaska), the cost of fueling their diesel generators was becoming
prohibitive. The manager of the cooperative began to search for an alternative.
Together with the State of Alaska, the cooperative financed the construction of a
wood-waste-fueled biogas production unit that could be hooked up to a diesel
generator and, with modifications, fuel that generator. The 150-kW gasifier was
produced by a company in California and moved to Anchorage for test
operations. Tests proved to be successful, and the cooperative, together with
Marenco (a local engineering firm) and the State of Alaska, is now working on a
250-kW unit.
At Walton Electric Membership Corporation in Monroe, Ga., the rural electric
system, together with the Georgia Institute of Technology and a local dairy farm,
has built a manure-to-methane, electric-generating operation. At the Mathls
Dairy, which has approximately 100 cows, the manure is collected and slurried
to an underground digester tank. The gas produced is burned in a modified 50kW Waukeshau gas-spark-ignition engine, and the electricity is used on the
premises to run milk chillers and other equipment.
Anoka Electric Cooperative in Anoka, Minn., together with an engineering
firm, Perennial Energy, Inc., of Oakbrook, Ill., is mining the local landfill for
methane, and the mined gas is burned to 2 Minneapolis Moline 160-HP gasspark-ignition engines with 105-kW generators attached to each. Since beginning
operation in January 1984, the two generator sets have produced nearly 600,000
kWh (a plant factor in excess of 70%). Problems with the occurrence of poisons
(chlorinated hydrocarbons), which can reduce the lifetime of the engine have
been reported, but the project is considered a success.
The Goodhue Electric Cooperative in Zambrota, Minn., and a local dairy
farmer with a 50-cow herd have also built an anaerobic digestion system and are
using the methane produced to generate electric energy. The digester is a
converted Butler silo, and the engine generator is a natural gas internal
combustion engine with 10-kW generator attached.
Along the same lines, at a chicken farm in Rushford, Minn., with the
encouragement of the Tri-County Electric Cooperative, the chicken manure from
more than 50,000 chickens is anaerobically digested to produce methane gas.
The gas is then burned to produce heat for the chicken house and an alcohol fuels
plant next door.
On a much larger scale, at Dixie Electric Power Association in Laurel, Miss., a
local paper pulp mill has installed 225 MW wood-waste-fired cogeneration
units that supply the local mill and the rural electric system with power at offpeak times. The mill, which has a 1000 ton/day pulp dryer, is now 95% energy
self-sufficient.
At Boone Electric Cooperative in Columbia, Mo., together with the University
of Missouri Agricultural Engineering Department, the U.S. Department of
Agriculture, and a local hog-feed-lot operator, an operation has been set up
where the hog manure is anaerobically digested to methane gas and the methane
137
gas is burned in a modified engine generator set to produce electricity for onfarm use and thermal energy for an alcohol still next door. The digester is a
converted grain bin that handles 510 kg of manure per day, producing 285 m3 of
55% methane, 408 kWh of electricity, and 3 GJ of thermal energy each day.
At North Carolina Electric Membership Corporation (NCEMC) in Raleigh,
N.C., there has been an interest in using local peat resources to fire a power plant
(much as is being done in Finland, Iceland, and Russia) for some time now.
Considerable experimentation with harvesting technologies has been done, and
NRECA and the Electric Power Research Institute of Palo Alto, Calif.,
performed a study of the suitability of this concept. The study, which has been
released, found that the proposed site was the most desirable (in terms of drying
climate and peat quality) in the United States for such a plant and that the
economics are competitive with a coal-fired plant of the same size.
Presently, NCEMC is awaiting developments on their potential access to the
peat resources in question before they proceed with additional
development. Basin Electric Power Cooperative (with its headquarters in
Bismarck, N.D.) also has an old, inefficient, small, coal-fired power plant. One
of the major crops in the area is sunflower seeds. The process of producing
sunflower oil and other products from these seeds produces a great deal of hulls,
which are waste materials.
Basin, in 1979, decided to test the feasibility of using these hulls as a
supplemental fuel in the generating station. The tests showed that the hulls had
good handling characteristics and almost as high a Btu content as the lignite coal
the plant was using as a primary fuel. Basin accepted a proposal from a sunflower
processor in 1980 to build a sunflower processing plant next to the power plant.
The power plant is providing the processor with waste heat from its boilers and,
in exchange, is receiving sunflower hulls as a supplemental fuel.
At Minnkota Power Cooperative in Grand Forks, N.D., bordering North
Dakota and Minnesota, there is a great deal of interest in the use of local
agricultural surpluses to produce useful liquid fuel as substitutes to expensive
imported oil. There is an effort under way, spearheaded by several private
developers and supported by Minnkota, to build a number of grain-to-alcohol
fuels plants in their service territory. This effort has picked up speed during the
last year with the pending ban by the Environmental Protection Agency of lead
additives as octane enhancers for gasoline and the increasing interest by the
major refiners in ethanol to fill that role. The first of these plants is already up
and operating. The plant at Walhalla, N.D., is a great local success. It uses 5
million bushels of barley per year to produce 11 million gallons of ethanol fuel
per year. The plant was designed by Ultrasystems, Inc. of Irvine, Calif.
At Verendrye Electric Cooperative in Velva, N.D. (where sunflowers are a
very large local crop) , a substantial amount of sunflower oil is available. The
cooperative, remembering the oil embargo of 1973 and the importance of the
continuity of agriculture and ensuring a supply of fuel for the local farm
machinery, has been involved with and is helping to finance a project at North
Dakota State University in Fargo to fuel tractors and other farm machinery
(which normally run on diesel) on sunflower oil.
Results so far are encouraging. The sunflower oil seems to be almost as high
an energy content fuel as diesel and can be readily produced on a small scale by
local farmers with a press and filter. However, it has a much higher viscosity
than diesel oil in cold weather and can produce problems on cold start in the
winter. Testing is continuing on what impact sunflower oil has on the long-term
life of a diesel engine. Preliminary results indicate no excessive buildup of deposits
on the cylinder walls, piston rings, or other critical ports.
The National Rural Electric Cooperative Association in Washington, D.C., and
East River Electric Power Cooperative in Madison, S.D., have for a number of
years been supporting a project to produce alcohol from waste materials at South
Dakota State University in Brookings, S.D. The process involves breeding
special bacteria that break down the cellulose in materials like old newspapers,
corncobs, grain chaff, etc., by converting the cellulose to starches and sugars.
These materials can, in turn, be fermented into alcohol.
These projects are actually just a sampling of bioenergy projects at rural
electric systems all over the United States. The trend is away from rural electric
system involvement moving toward third party development and interconnection
under the Public Utility Regulatory Policies Act (PURPA). Rural electric
systems will probably continue toward their trend of investing in large nuclear
and coal-fired power plants, and the abundant biomass resources in their service
areas will be developed by third party entrepreneurs who will sell the power
produced back to them under PURPA.
SECTION III
Biomass Energy Research Projects
SYNOPSIS
The worldwide energy shortage and Arab oil embargo of the early 1970s
encouraged many nations to look for new sources of oil, electricity, and gas.
Renewable resources such as biomass were often highlighted as a long-term
solution to the energy problem because of their nondepletable, renewable nature.
While the first biomass sources considered were the readily available ones such
as corn or wood, it was apparent that new biomass sources should also be
developed, among them aquatic species.
The purpose of the SERI/DOE Aquatic Species Program is to improve the
productivity, conversion to fuels, and cost efficiency of aquatic plant culture
technology. The emphasis of the program is on developing a mass culture
technology for cultivating lipid-yielding microalgae in the American Southwest
(Fig. 1). It was determined that fuels from microalgal lipids presented better
options than converting the microalgal biomass to either alcohols or methane.
All lipids can potentially be catalytically converted to gasoline, or the fatty acids
can be converted to substitute diesel fuels. The Southwest has the necessary lowcost resources available for this technology, including large expanses of flat,
underutilized land, and carbon dioxide available from either natural deposits or
flue gas from industrial plants. The amount of saline water available will
probably determine how much fuel can be produced from aquatic species. This
question should be answered during 1985.
1
THE AQUATIC ALGAE RESOURCE
Aquatic algae may be divided into two groups: macroalgae and microalgae.
Macroalgae range in size from small filamentous forms to very large complex
forms such as Macrocystis. Various concepts have been developed for culturing
several species of macroalgae for biomass. The giant kelp Macrocystis is being
cultivated in near-shore areas off California in less than 20 meters of water.
Floating or benthic species such as Sargassum or Gracilaria have been cultured
1980 Shifrin reported that microalgae under nutrient stress could accumulate up
to 72% of their weight as lipids. Even though large accumulations of lipids are
found only under stress conditions of nutrient limitation or salinity, certain
microalgae are obviously metabolically capable of producing these high-energy
compounds in large quantities. The heat of combustion for a typical algal mass is
in the range of 8,00010,000 Btu per pound dry weight, but it is strongly affected
by the low heat of combustion of nonlipid products. The specific heat of
combustion for algal oils and lipids is approximately 16,000 Btu per pound, and
for the purpose of energy or fuel production by microalgae, it is desirable to
increase lipid content of these organisms.
Microalgae lipids can be classified as polar, membrane lipids or non-polar,
storage lipids, such as hydrocarbons and triglycerides. Storage lipids, which
accumulate when microalgae cells are stressed, offer the most potential since
they more closely resemble petroleum-derived compounds. Imposing stresses,
such as nitrogen limitation, increases the percentage of storage lipids.
Further initiative for emphasizing lipids for fuels was provided by reports of
the direct synthesis of hydrocarbons by various microalgae. Specifically, large
quantities of C-30 hydrocarbons were identified in the freshwater species
Botryococcus, an organism that is postulated to be responsible for present
petroleum reserves. It was initially assumed that hydrocarbons extracted from
such organisms could be readily processed by the existing petrochemical
industry and used to produce gasoline, although no process evaluations were
performed to verify this hypothesis.
3
SITING CONSIDERATIONS
Most biomass production technologies require a large, inexpensive resource base
to be economically competitive with conventional (mined) fuel sources. The
advantages of microalgae are that they can grow in any climatic region of the
world, since they require only light and nutrients, and that they are renewable
and therefore not limited to fixed deposits like oil, coal, and natural gas.
Development of a biomass technology that can exploit the underutilized,
marginal resources of the Southwestflat land, saline water, and high incident
solar radiationoffers potential for the production of a high-value energy
feedstock from microalgae. This concept has unique characteristics, few
competitive impacts, and enormous potential for displacement of exhaustible
conventional energy resources.
Arizona, New Mexico, Southern California, West Texas, and parts of Nevada,
Utah, and Colorado have abundant lands that support relatively little
conventional biomass productivity. Resource assessment studies completed in
1983 estimated that up to 25 million acres may be highly suitable for microalgae
cultivation, but the definition of criteria was too gross to provide definitive
estimates of total resource availability (Fig. 4). The abundance of available saline
water has also become an issue. While scarce fresh water in this region limits
conventional agricultural possibilities, saline and brackish waters are known to
exist in large underground aquifers, and large quantities of agricultural drainage
waters are present in existing irrigation canals. These water resources are
typically underutilized because of their marginal quality and characteristics. In
some cases, such as irrigation runoff, they are an economic nuisance that is
costly to alleviate.
A wide variety of microalgal species can grow in highly saline desert waters.
Therefore, despite the fact that desert saline waters differ from marine waters
because of different ionic ratios of bicarbonate, calcium, magnesium, and
sulfate, these differences may not preclude adaptation of abundant marine
microalgae to desert waters. Thus, the natural variability of both marine and desert
species may prove to be useful in developing species for desert mass cultivation.
There is no doubt that many areas of this technology will need long-term
development before a barrel of lipid oil is economically produced and converted
into a liquid fuel. The current estimate of the state-of-the-art cost is between
$250 and $350 per barrel. To compete by the year 2000, it is estimated that the
cost will need to be reduced to $60-$85 per barrel. Although the development of
the technology is mid-term, it provides a potentially valuable source of
renewable high-energy liquid fuels.
4
RESEARCH ACCOMPLISHMENTS
During 1984, research was carried out under three tasks: biological, engineering,
and analysis. Biological research was aimed at improving photosynthetic
efficiencies and lipid yield of species that can be cultivated using mass culture
technologies operated in the American Southwest. Emphasis has been placed on
screening for productive species, developing culture and management techniques
for growing desirable species, and understanding photosynthetic and lipid
physiology as it applies to increasing yields. Engineering research focused on the
development and analysis of harvesting schemes applicable to species of
microalgae that grow in saline waters. Three system designs and analyses were
initiated in 1984, and these designs will be completed in 1985. The analysis task
is designed to support technology development through the determination of cost
goals, assessment of resources, and evaluation of emerging technologies. A
comprehensive technical and economic evaluation was completed during 1984,
and this analysis and assessment provided insights into where program emphasis
should be placed for the next ten years.
SYNOPSIS
The ocean is an ideal environment for the production of energy crops, because it
represents 70$ of the earths surface and is not heavily used for other purposes.
Its plant life, particularly the macroalgae (seaweeds), is highly productive and
readily converted to fuel at high efficiency. With few exceptions, such plants are
not in demand for other purposes such as food or fiber.
Because they are not widely used, seaweeds have not been grown
commercially in most parts of the world. The technology for their large-scale
cultivation, particularly in the open sea, is therefore lacking. Exceptions are
found in the Orient and Southeast Asia, where certain macroalgal species are
used locally for food or worldwide for their gelatinous polysaccharides (agar,
alginic acid, carrageenin). Such species are grown near shore, usually in
protected embayments or impoundments, in commercial culture, using
technologies that are, for the most part, both crude and labor intensive.
Examples that are discussed and illustrated include Porphyra (nori) culture in
Japan, Laminaria (kelp) culture in China, Gracilaria culture in Taiwan, and
Eucheuma culture in the Philippines. Culture methods are described and biomass
yields are summarized for each species.
1
INTRODUCTION
The use of terrestrial plant crops as biomass for fuel is complicated because such
products are generally worth more as food or fiber; in addition, the land on which
energy crops might be grown could be used to produce the more valuable food or
fiber crops (Ref. 1). The growing world demand for food and the current
existence of famine conditions that affect a significant fraction of the human race
make the concept of energy crops not only economically but also morally
questionable.
The oceans, however, are virtually unused by man for the production of crops
of any kind. Yet marine algae are highly productive, capable of yields as great as
the best terrestrial crops (Ref. 2) , and such plants are also among the most
efficient for conversion to fuel (Ref. 3). Marine algae that are unused for food or
other purposes by most human societies could be grown as energy crops without
competing for demand for their use or for the area in which they might be
grown.
There is no well-developed or widely used technology for the commercial
cultivation of seaweeds. However, in the Orient and Far East, a few species are
now grown for food or for their chemical products, principally the
polysaccharides, agar, alginic acid, or carrageenin. For the most part, the existing
seaweed cultivation practices may be characterized as low-technology, low-yield,
and labor-intensive. Cultivation is successful only because of the low cost of
labor and the high value of the product. Despite their primitive nature, however,
several sizable industries have developed. They provide a useful background of
information for any future consideration of seaweed production systems for
energy.
The present discussion will be restricted to four of the larger commercial
seaweed culture systems currently in operation in the world: Porphyra (nori)
culture in Japan, Laminaria (kelp) culture in the Peoples Republic of China,
Eucheuma farming in the Philippines, and Gracilaria culture in Taiwan.
Technical aspects of each culture system will be reviewed briefly, along with the
small amount of information that is available on production and economics.
2
Porphyra (NORI) CULTURE IN JAPAN
Several species of the red algae genus Porphyra have been grown in Japan,
where it is commonly known as nori, as a highly prized food since the
seventeenth century.
The large sporophyte Porphyra plant (the edible stage) is an irregularly shaped,
flat, deep red blade that may grow to a foot or more in length, depending on the
species and growing conditions. The Japanese crop comes primarily from P. tenera
and P. yezoensis but includes at least four other species. It matures in late fall
and may be harvested several times throughout the winter by cutting back the
thallus without destroying its attachment.
In 1978, 60,000 hectares (ha) of sea surface produced 21,150 dry tons of nori
that had a value of $540 million (U.S.), by far the most economically important
seaweed crop in the world (Ref. 4).
Originally, cultivation of Porphyra consisted of driving leafless tree branches
into the sea floor just above the mean water level along the open coastline in the
fall of the year. The nonmotile monospores of Porphyra settled on the branches,
which were moved to more nutrient-rich habitats at the mouths of or within
estuaries. The attached monospores developed into leafy sporophyte thalli, the
edible stage and portion of the plant, which was harvested by cutting throughout
the winter.
All of these stages in the life cycle, from the shedding of the zoospores to the
development of the young sporelings, take place in the nursery between June and
October. When the outside water temperature falls below 20C, in mid to late
October in Tsingtao but in late September to early October in the more northern
Dalien area, the spore curtains are taken off their wooden frames and moved to
the ocean, where they are suspended between parallel rows of large buoyed and
moored ropes. At this point, the sporelings are 2 to 4 cm long, and there are some
50,000 of them per 50-m spore curtain.
When they are small, the sporelings are tended daily. Each spore curtain is
lifted from the water, the sediment and attached plants and ani mals are
meticulously brushed from each plant, and the entire curtain is immersed into a
tub of concentrated liquid fertilizer. When they reach the size of about 10 cm,
after 25 to 30 days in the ocean, the entire young crop is harvested, manually
stripped off the strings to which they are attached. Bundles of four sporelings are
inserted into the weave of larger, 5-cm-diameter, coarse, loosely woven ropes
that are again tied across the parallel suspending lines. The bundles remain in
these ropes until they are harvested over a six-week period beginning in early
June.
The plants are no longer individually tended after they are transplanted, but
the crop is usually fertilized. Formerly, this was done by attaching to the ropes
ceramic containers of fertilizer through which the nutrients could diffuse slowly.
Now liquid fertilizer is sprayed daily over the kelp beds.
In Tsingtao, where the growing season is 230 days, the kelp reach a length of
about 3 m at the time of harvest. Yields average 12 dry ton/acre/yr. In the colder
Dalien region, the season is perhaps one month longer. The kelp plants there
reach a length that may exceed 5 m, and yields of 20 dry ton/acre/yr are reported.
The mature kelp plants are harvested in the late spring by manually hauling the
lines with plants attached into a fleet of rowboats each manned by three to four
workers.
4.
Eucheuma CULTURE IN THE PHILIPPINES
Eucheuma is a multiple-branched, fleshy red alga that is utilized for its
polysaccharide, carrageenin. Originally harvested from wild stocks in Southeast
Asia, such supplies were cut off by political upheavals in the 1950s, which led to
the development of cultivation methods for the species by the joint efforts of
Marine Colloids Inc. (Rockland, Maine), M.S.Doty (University of Hawaii), and
the Philippine Bureau of Fisheries and Aquatic Resources.
Unlike Porphyra and Laminaria culture, which require tending throughout all
stages of their complex life cycles, Eucheuma is grown purely vegetatively by
planting fragments of the fleshy thallus, allowing them to increase in size, and
simply breaking or cutting off the new growth. Initially, the Eucheuma was grown
on nets closely resembling those used for Porphyra farming in Japan. These nets
also are suspended off the bottom and parallel to the water surface by tying them
to bamboo stakes. Eucheuma fragments are tied to the nylon mesh intersections.
Typically a family Eucheuma farm consists of four modules of 200 nets
covering an area of about 2500 m2; the total farm occupies 1 ha (Ref. 7). More
recently the net system has been replaced by the use of monolines, 10-m lengths
of nylon monofilament staked 0.5 m apart. Eucheuma propagules are tied every
2.5 m along the line. The monoline method provides easier access to the plants
and accommodates as many as 100,000 plants per hectare (Ref. 8).
Like Porphyra culture, Eucheuma farming is a highly labor-intensive
operation, with constant tending of the plants, manual removal of epiphytes and
predators, mending the nets and repairing the moorings, and harvesting the plants
as they grow (Ref. 9). Siting of the farm is also critical because the alga requires
a vigorous exchange of seawater for rapid growth. Thus favored sites are highenergy areas such as behind fringing reefs, but the surf should not be so violent
as to break off the seaweed from the nets or otherwise damage the operations.
Two species of Eucheuma, E. cottonii and E. spinosum, are farmed in the
Philippines. These contain, respectively, the different chemical isomers kappa
and iota carrageenins, which have different gelling properties and hence different
commercial applications. Productions increased from a few hundred dry metric
ton during the 1960s to a peak of 6590 ton in 1974. At that time nearly half the
harvest remained unsold, the price dropped almost tenfold, and many farmers
dropped out of the market. Since then there has been gradual recovery of the
industry, with about 400 E. spinosum and 600 E. cottonii farms in operation in
1977 producing more than 4000 ton of dried seaweed.
5
Gracilaria CULTURE IN TAIWAN
Growing the red seaweed Gracilaria in southern Taiwan was first attempted in
1962, using ponds that were originally constructed for fish culture. The
Graeilaria is used locally or, after preliminary processing, is shipped to Japan for
the extraction of agar.
Although Gracilaria normally grows attached to rocks or other substrata and
undergoes the complex alternating life cycle between sexual and asexual
reproduction that is typical of red algae, certain species may grow unattached, in
a drifting mode, on the bottoms of shallow ponds and estuaries. Such plants are
usually sterile and grown entirely vegetatively, larger clumps breaking up into
smaller fragments by the action of waves, current, and other forms of natural
turbulence.
The milkfish ponds in which Gracilaria is now grown in Taiwan are usually
rectangular, 1 to 10 hectares in area, and about 1 m deep when filled to capacity.
The pond bottoms are hard, sandy loam, since a soft mud bottom is undesirable
for both growth and harvesting. The ponds are located adjacent to estuaries so
that they may be filled and drained by tidal exchange, assisted as needed by
The low cost of land, labor, and other factors make the above examples
economically irrelevant to potential U.S. practices, but the yields of 10 to 20 dry
ton/ha/yr for three of the four cases described, all simple, nonintensive farming
practices, are encouraging. More sophisticated cultivation methods, including
genetic improvement of stocks, better nutrition, and mechanization of seeding
and harvesting practices can be expected to increase yields several times. Smallscale, experimental yields of Gracilaria using highly intensive culture methods
have already exceeded the equivalent of 100 dry ton/ha/yr (Ref. 2).
Modern cultivation technology in agriculture has evolved slowly, over
centuries. Yet current yields from farming the land average no more than about
twice those now achieved from the seaweed culture methods that have been in
practice for no more than a few decades (Table 1).
Farming the sea, while in its early infancy, thus shows sufficient promise to
merit further investigation. By the time a world shortage in fossil fuel has
developed, marine biomass could, with its available resource base, make up
much or all of the deficiency with only a modest annual investment in research
and development effort in the intervening years.
Table 1: Annual yields of agricultural crops (total plant) (from Ref. 2) and of seaweeds
(this report)
crop
Temperate Zone
Rye grass
Kale
Sorghum
Maize
Potato
Sugar beet
Wheat (spring)
Subtropical Zone
Alfalfa
Sorghum
Bermuda grass
Sugar beet
Potato
country
U.K.
U.K.
U.S. (Illinois)
U.K.
Canada (Ontario)
Japan
U.S. (Iowa)
U.S. (Kentucky)
U.K.
Netherlands
U.K.
U.S. (Washington)
U.S. (Washington)
23
21
16
17
19
26
16
22
23
22
23
32
32
U.S. (California)
U.S. (California)
U.S. (Georgia)
U.S. (California)
U.S. (California)
33
47
27
42
22
crop
country
Wheat
Rice
Maize
Mexico
U.S. (California)
Egypt
U.S. (California)
18
22
29
26
Tropical Zone
Napier grass
El Salvador
Puerto Rico
Sugar cane
Hawaii
Oil palm
Malaysia
Sugar beet
Hawaii (2 crops)
Cassava
Tanzania
Malaysia
Maize
Thailand
Peru
Rice
Peru
Mean (all crops, all countries)
Median (all crops, all countries)
Seaweeds
Laminaria
China (Ref. 4)
Eucheuma
Philippines (Ref. 9)
Gracilaria
Taiwan (Ref. 10)
(Ryther, personal observation)
85
85
64
40
31
31
38
16
26
22
31
26
15
13
10
14
REFERENCES
1.
2.
3.
4.
5.
6.
7.
Greely, R.S. Land and freshwater farming, Proc. Conf. on Capturing the Sun
through Bioconversion, Wash. Center for Metropol. Stud., vol. 197, 1976.
Lapointe, B.E. and Ryther, J.H. Some aspects of the growth and yield of Gracilaria
tikvahiae in culture , Aquaculture, vol. 15, 1978.
Fannin, K.F., Srivastava, V.J. and Chynoweth, D.P. Unconventional anaerobic
digester designs for improving methane yields from sea kelp, Symp. Papers
Energy from Biomass and Wastes IV, Lake Buena Vista, Fla., January 2529,
1982.
Tseng, C.K. Commercial cultivation, Chap. 20 in The Biology of Seaweeds, C.S.
Lobban and M.J. Wynne, eds., Bot. Monogr. 17., U. Cal. Press, 1981.
Bardach, J.E., Ryther, J.H. and McLarney, W.O. Aquaculture, Wiley-Interscience,
1972.
Tseng, C.K. Personal communication.
Doty, M.S. Farming the red seaweed, Eucheuma, for carrageenins, Micronesia,
vol. 9, 1973.
8.
9.
10.
11.
Hansen, J.E., Packard, J.E. and Doyle, W.T. Mariculture of red seaweeds, Cal.
Sea Grant College Prog. Publ. T-CSGCP-002, 1981.
Parker, H.S. The culture of the red algae genus Eucheuma in the Philippines,
Aquaculture, vol. 3, 1974.
Shang, Y.C. Economic aspects of Gracilaria culture in Taiwan, Aquaculture, vol.
8, 1976.
Hanisak, M.D. Recycling the residues from anaerobic digesters as a nutrient
source for seaweed growth, Bot. Mar., vol. 24, 1981.
1
INTRODUCTION
For more than two decades, aquaculture systems employing plants grown on
sewage have been proposed as alternatives to conventional wastewater treatment
processes. In addition to removing pollutants, some of these systems can provide
biomass in substantial amounts for conversion to methane. This concept requires
the integration of at least three technologies: (1) anaerobic digestion; (2) biomass
management for achieving maximum yields; and (3) aquaculture treatment of
wastewater to meet federal, state, and local standards.
Previous investigations of hyacinth sewage treatment have mainly emphasized
removal of BOD5 and suspended solids (secondary treatment) and reduction of
phosphorus and nitrogen (tertiary treatment) (Refs. 15). Hyacinth treatment of
sewage is usually conducted in shallow ponds or channels less than 1 m deep.
Pilot-scale hyacinth treatment systems have typically achieved removals of 75%
to 95% of BOD5 and TSS under sewage loadings of 20 to 150 kg/ha day (Refs. 4
, 5), but hyacinths in these studies, however, were not managed and harvested to
maximize biomass production.
This paper describes the status of research in progress to develop a system that
combines methane production with a wastewater treatment concept that produces
a high-yielding crop of water hyacinths from sewage. This integrated concept
has the potential to provide a stable source of low-cost methane while offering the
community a cost-effective process for wastewater treatment and water reuse.
2
APPROACH
The general approach used by this project has been to concentrate research on an
integrated field test facility at Walt Disney World, near Orlando, Fla., while
relying on systems analyses and engineering trade-offs to direct R&D toward
process performance goals that achieve cost-competitive methane. The system
concept, depicted in Fig. 1, consists mainly of water hyacinth channels for
4
WASTEWATER TREATMENT
Hyacinth wastewater treatment studies and hyacinth productivity research were
conducted in five 0.1-ha hyacinth test channels, each with dimensions of 8.8
m110 m0.35 m deep, constructed of reinforced concrete blocks and lined with
20-mil PVC sheet. Previous studies on these channels showed that secondary
effluent standards could be achieved under low sewage feed rates (BOD5
loadings) typically applied to aerobic ponds without hyacinths, amounting to
about 70 to 90 kg BOD5/ha day. In 1983 and 1984, four channels were fed with
primary sewage (obtained from the Walt Disney World wastewater treatment
settling basins) at loadings of 55, 110, 220, and 440 kg BOD5/ha day
corresponding to hydraulic retention times of 24, 12, 6, and 3 days, respectively.
The technical objective of this study was to measure channel BOD5 and TSS
removal efficiencies under high loadings that would stress the systems
treatment capabilities. Wastewater treatment data collected from the channels
included influent and effluent BOD5, suspended solids (SS), pH, temperature,
dissolved oxygen, and various forms of nitrogen and phosphorous.
Results from this study over a 9-month test period (from November 1983
through July 1984) indicate that a single hyacinth channel is capable of removing
72% to 90% of the BOD5 (81% average) and 70% to 90% of the SS (80%
average) in wastewater at loading rates as high as 440 kg BOD,-/ha day [(3 day
hydraulic retention time (HRT)] . Average steady-state data for BOD5 and SS
removals are presented in Table 1. Average effluent BOD5 and SS
concentrations from the hyacinth channels met federal standards for secondary
treatment at loadings up to 220 kg BOD5/ha day during all but two of the coldest
months of the test period; during that time, secondary effluent standards were met
at the 110 kg BOD5/ha day loading rate. A statistical analysis of influent and
effluent data as well as channel profile measurements taken over the past four
years is in progress to allow more accurate correlations between treatment
performance and hyacinth channel operating conditions (e.g., loadings,
temperature, retention time, hyacinth density). Preliminary analysis of the
performance data for the four channels suggests that treatment efficiencies may
be only marginally improved by extending retention time and that it may be costeffective to use staging of the unit processes to achieve a high compounded
removal efficiency at an equivalent retention time as opposed to increasing the
hydraulic retention time of a single-stage channel. Experiments with a highthroughput, two-stage hyacinth system are under way.
Table 1:
Hyacinth wastewater treatment performance summary
sewage loading (kg/ HRT (days)
ha day)
BOD5
SS
55
110
220
440
24
89
71
12
87
72
6
84
83
3
81
80
*Based on a 9-month operation from November 1983 through July 1984 and an influent
BOD5 concentration of 200 to 260 mg/L.
5
WATER HYACINTH PRODUCTION
Maximum growth yields of water hyacinth are desirable in the sewage channels
because rapid growth is associated with efficient wastewater treatment and
results in larger quantities of biomass available for conversion to methane.
Numerous factors can influence hyacinth yields, and the most important are
temperature concentration of CO2, sunlight capture, nutrient availability, and
planting density (or current rate). Of these, the most controllable parameters
studied thus far include, nutrient availability and planting density.
Hyacinth productivity experiments have been conducted in the hyacinth
channels (0.1 ha) and in small field test units (1.7 m2). Measurement of biomass
yield in each channel was performed using 10 m2 Vexar mesh baskets placed
about 18.3 m apart. The methods and results of this work are described in more
detail in Ref. 7. The large channels were used to observe the effects of sewage
loading, channel retention time, seasonal temperatures, and sewage treatment
efficiency on hyacinth yield. The small experimental systems were used to
optimize hyacinth production by controlling parameters such as nutrient
availability, plant density, and aeration.
Hyacinth productivity under unoptimized conditions ranged from 45 to 58 kg/
ha yr. The use of a harvesting schedule that provides an optimum planting
density of 36 kg/m2, however, has increased hyacinth yields to 60 to 70 dry
metric ton/ha yr. Preliminary tests in small field units suggest that still further
yield increases of 30% to 50% are possible through the discretionary use of
aeration.
6
ANAEROBIC DIGESTION PROCESS DEVELOPMENT
Anaerobic digestion was selected for the processing of mixtures of water
hyacinth and primary sludge since it produces methane as the principal product
(60% to 65%) and since the process is compatible with the conversion of
feedstocks with a high water content. The technical objective of this R&D effort
is to develop a data base for the design and operation of an optimized system for
biogasificatlon of the hyacinth and sludge, and to integrate this process with the
hyacinth wastewater treatment facility.
6.1
Laboratory Studies
A number of experiments were conducted to evaluate the performance of a
continuous stirred tank reactor (CSTR) and an upflow solids reactor (USR)
designed to promote long solids retention times under high hydraulic load-ings.
A schematic of the USR is shown in the diagram of the digester of Fig. 1. This
reactor employs little or no mixing. Influent is fed at the bottom of the tank, and
effluent is removed near the top of the liquid contents. The laboratory USR and
CSTR units were fed with a 3:1 blend of hyacinth and primary sludge at loadings
between 1.6 and 6 .4 kg VS/m3 day (corresponding to HRTs between 31 and 8
days). Steady-state data plotted in Fig. 4 show that in side-by-side tests, the USR
consistently achieved 10% to 25% greater methane yields over a wide range of
HRTs. The superior performance of the USR was attributed to the reactors
ability to increase the solids and microorganism residence time significantly
above the HRT through sedimentation of particulate solids. Thus, the USR
reactor achieved greater methane production with substantially less mixing.
These results provided the basis for the selection of the USR design for testing at
the experimental test unit (ETU) scale.
6.2
Biogasification Experimental Test Unit
In 1983, a 4.5-m3 (160 ft3) upflow solids ETU was designed and constructed,
located beside the five existing hyacinth channels at Walt Disney World. The
technical objective of the first phase of the ETU study was to evaluate reactor
performance, scale-up, and materials handling parameters at several different
loadings of hyacinth/sludge blends. The ETU facility is capable of processing up
to 910 kg or 1 ton of a wet hyacinth/sludge feed blend (5% total solids) each
day. The ETU is sized to ensure that the demand for biomass feedstock does not
exceed the availability of hyacinth from the channels during the coldest months
of the winter when hyacinth pro-ductivity is at its lowest. Major components of
the facility include two feed tanks for short-term storage of sludge and chopped
hyacinth, a feed blend tank, an upflow solids reactor (4.5 m3), an effluent storage
tank, and gas compression and storage. The ETU was initially tested as an upflow
solids reactor test facility, though it is designed to allow flexibility in simulating
and testing other reactor designs as well. The current testing phase of the ETU is
expected to provide:
engineering parameters
Methane yield
Methane production rate
Gas composition
Volatile fatty acids
Temperature
pH
Alkalinity
Organic matter reduction
Materials balance
Solids
Carbon
Nitrogen
Phosphorus
Energy balance
Materials handling
Scale-up of laboratory performance
Results from the second ETU steady-state condition illustrate the effect of feed
blend composition on methane yield. When feed conditions were shifted from a
2:1 to a 1:1 blend of hyacinth and sludge at a constant HRT (16 days) and
loading (3.2 kg/m3 day), the methane yield was increased from 0.29 to 0.39 m3/
kg VS added. These results are consistent with previous batch reactor tests which
indicated that the ultimate methane yield of hyacinth (0.30 to 0.37 m3/kg VS
added) is lower than that of sewage sludge (0.40 to 0.45 m3/kg VS added). A
higher sludge content in the ETU feedstock mix should therefore result in higher
2.
3.
4.
5.
6.
7.
8.
9.
Tchobanoglous, G., Stowell, R., Ludwig, R., Colt, J. and Knight, A. The use of
aquatic plants and animals for the treatment of waste-water: an overview, in
Aquaculture Systems for Wastewater Treatment/MCD 67, available from General
Services Administration, Denver, CO, Document No. EPA430/980006, p. 35,
1979.
Reed, S., Bastian, S. and Jewell, W. Engineering assessment of aquaculture
systems for wastewater treatment: an overview, in Aquaculture Systems for
Wastewater Treatment/MCD-68, available from General Services Administration,
Denver, CO, Document No. EPA 430/980007, p. 1, 1980.
Wolverton, B.C. and McDonald, R. Upgrading facultative wastewater lagoons
with vascular aquatic plants, Jour. Water Poll. Control Fed., vol. 61, 305, 1979.
Kruzic, A.P. Water hyacinth wastewater treatment system at Walt Disney World,
in Aquaculture Systems for Wastewater Treatment/MCD 67, available from
General Services Administration, Denver, CO, Document No. EPA 430/980006,
p. 35, 1979.
Dinges, R. Upgrading stabilization pond effluents by water hyacinth culture,
Jour. Water Poll. Control Fed., vol. 50, 833, 1978.
Smith, M.D., Filard, R.E., Curran, G.M., and Miller, G.R. Economics of methane
from hyacinth wastewater treatment systems, Proceedings of the International Gas
Research Conference, Government Institutes, Inc., Rockville, MD, in press, 1984.
Reddy, K.R. Water hyacinth biomass production in Florida, Biomass, vol. 6, 167,
1984.
Chynoweth, D.P. et al. Biogasification of water hyacinth and primary sludge,
Proceedings of the International Gas Research Conference, Government Institutes,
Inc., Rockville, MD, in press, 1984.
Wastewater Engineering, Metcalf & Eddy, Inc., McGraw-Hill, 1972.
Fig. 4. Methane yield performance vs. hydraulic retention time for bench-scale USR and
CSTR units operated on a 3:1 hyacinth/sludge feed blend
primary
conversion
process
Land-based
biomass
Trees
Separation
Combustion
Thermal
Energy
Steam
feed stock
primary
conversion
process
Plants
Grasses
Pyrolysis
Hydrogenation
Anaerobic
Solid fuels
fermentation
Aerobic
fermentation
Biophotolysis
Partial oxidation
Steam
Gaseous
reforming
fuels
Chemical
hydrolysis
Enzyme
hydrolysis
Other chemical
conversions
Natural
processes
Water-based
biomass
Single-cell algae
Multicell algae
Water plants
Organic wastes
Municipal
Industrial
Agricultural
Forestry
Electric
Char
Combustibles
Methane (SNG)
Hydrogen
Low-Btu gas
Medium-Btu gas
Light
hydrocarbons
Methanol
Liquid
fuels
Ethanol
Higher
hydrocarbons
Oils
Chemicals
1.1
Wood Fuels
Residential firewood usage is greatest in the urbanized areas of the Northeast and
North Central States; according to a recent study, 9% to 11% of U.S. space
heating input is from firewood. For example, during the 1978 1979 heating
season, about 34.7106 cords of wood were burned in residences. Wood now
accounts for 33.8% of all residential heating in Vermont, and more than 3.310
cords were burned in New York State during the 1980 1981 heating season. At
an energy equivalency of 3.9 bbl oil/cord of wood, this consumption level
corresponds to the heating value of 12.9 million BOE. From the New York State
study we concluded that wood is a major heat source in 707,000, or 18%, of the
states households, excluding metropolitan New York City and Westchester
County, and that the direct and indirect economic impacts of solid fuel use are of
tremendous importance to the New York citizens. Estimates indicate the sales of
wood heating units are about 11.5 mlllion/yr in the United States; this correlates
with increased fuel-wood consumption. Greater residential fuelwood
consumption has increased the accompanying pollution, resulting in legislation
to control fuelwood pollution. A number of the resulting ordinances and laws are
quite severe. For example, new stoves sold in Oregon after 1986 will have
to purge particulates and carbon monoxide from the smoke with devices such as
dual-combustion chambers or catalytic converters. Aspen, Colo., has an
ordinance that allows only one traditional fireplace per new structure, even for
apartment buildings.
To facilitate future large-scale increases in fuelwood usage for residential
space heat, the following parameters must be addressed in detail: fuel
availability, form, and cost; certified combustion systems and installation
methods; fuel storage; health and safety; and the development of low-cost
automated feeding and combustion control systems. Some research work is
under way to examine these issues on an integrated basis, but much more must
be done to open residential fuelwood markets and attract new users. The use of
fuelwood in residences is now too labor-intensive for most areas of the United
States.
Direct wood combustion is used by only a few utilities for supplemental
electric power production. Table 2 lists power plants currently on standby or in
operation that use mill residues (bark, sawdust, shavings, slabs, hogged bark, and
other mill residues) as fuel. The largest system, the Washington Power Company
plant, is used for peaking purposes. The largest wood-chip-fueled power plant is
the 50MW plant in Burlington, Vt., which was placed on-line in June 1984. The
10to 50MW, wood-chip-fueled plant to be operated by California Power and
Light Corporation in Madera, Calif., is still in the permitting stage, but
construction was expected to start in 1984.
About 85% to 90% of the total biomass contribution to U.S. primary energy
consumption, about 2.6 quads, is derived from wood and wood wastes. The major
consumer is the pulp and paper industry, while the electric utility industry
currently uses less than 0.1 quad of wood to generate power. Although the
technology for power generation in wood-fueled steam-electric plants is well
developed, the only economic wood-fueled power plants in the United States are
the larger plants. A catch-22 situation results because if a plant is larger than 50
MW, the wood must be transported for distances often greater than 50 to 75
miles, thereby increasing fuel cost. Thus, 50 MW seems to be the maximum size
at the present time.
Table 2:
operator
period
size (MW)
comments
Cowlitz, WA
Cowlitz
County Public
Utility District
192360,
1970s-present
27.5
Dixville, NH
New
Hampshire
Electric
Cooperative,
Inc.
1.9
Eugene, OR
Eugene Water
and Electric
Board
194180
33.8
Kettle Falls,
WA
The
Washington
Power Co.
Dec. 1983
42.5
1979-present
2530 from
wood
Steam
produced now
for
Weyerhauser
Company
One of 3
generators (totalling 1.9
MW) on line,
but down
because of fire
Steam-electric
plant now
down because
of reduced
demand
$82.5 million
plant uses 450,
000 ton/yr of
wood residues
Cofiring of
wood and coal
1981-present
Lake Superior
District, WI
Lake Superior
District Power
Co.
Red Wing, MN North States
Power Co.
Cofiring with
20% wood
residues, 80$%
coal
Note: Mill residues defined as bark, sawdust, shavings, slabs, hogged bark, and other
mill residues. This tabulation may not include all U.S. utilities that convert mill
residues to power, and does not include the plants operating on wood chips.
The Republic of the Philippines is developing a nationwide program of woodfueled, 3-MW power plants, each of which is supplied by a tree plantation of
about 2,500 acres of Leucaena. The goal is to have 210-MW total capacity (6%
of the total electrical requirement) by 1987; 27 plantations were established by
the end of 1981 and 20 additional plants are under construction.
A direct combustion technology that may make smaller plants more
economical uses pressurized combustion of wood, clean-up of the combustion
gas with cyclones to remove particulates, and passage of the gas through a
gas turbine (Aerospace Research Corporation). Initial studies with a 375-kW
system indicated satisfactory performance and no turbine blade erosion. A 3-MW
plant being constructed at Red Boiling Springs, Tenn., to test the system on a
larger scale will consume 100 ton/day of green wood, and the power will be sold
to TVA.
number
status
design capacity
range (ton/day)
total capacity
(ton/day)
Mass burning,
waterwall
Mass burning,
waterwall
Mass burning,
waterwall
Mass burning
Operating
2401600
6,250
15
Negotiation/
Construction
Planning
6003000
24,110
>500
17,060
200500
NA
15
Numerous Planning
plant type
number
status
design capacity
range (ton/day)
total capacity
(ton/day)
Mass burning,
starved air
Mass burning,
excess air
RDF
RDF
28
Operating
20240
2,912
26
Operating
50500
6,301
11
9
2003000
10003000
19,500
14,300
RDF
Operating
Negotiation/
Construction
Closed
5002000
7,400
97,833
1.3
Dioxin
For energy-from-biomass-and-waste combustion processes, the dioxin issue
continues to attract extensive media coverage, although human exposure to
dioxin has been scientifically linked to only a few health problems. The
symmetrical isomer, 2, 3, 7, 8tetrachlorodibenzo-p-dioxin, referred to by most
scientists as dioxin, is apparently one of the most toxic of 75 possible isomers,
and has one of the lowest LD50s of those determined to date. Extensive research
is in progress to develop data and information on the toxicity of dioxins as
related to human health.
The importance of the subject for energy-from-biomass-and-waste technology
is that evidence now being accumulated indicates combustion processes fueled
with biomass and wastes can form dioxins. In addition to formation as a trace
contaminant in the manufacture of the insecticide 2, 4, 5trichlorophenoxyacetic
acid (2, 4, 5T), which has been banned from most agricultural uses, dioxin is
formed on combustion of materials containing chlorinated phenols and its
precursors. For example, wood-fueled stoves produce dioxins in the soot because
the natural chlorine content of wood, about 1484 ppm, reacts with the lignins.
Dioxins also form in trace amounts in the fly ash and soot from industrial and
municipal incinerators supplied with municipal solid wastes.
The hazards associated with these emissions are not known with certainty. But
presuming some toxicological conditions exist because of the highly toxic nature
of dioxins, the problem is clearly not severe because of the broad distribution of
biomass- and waste-fed combustion systems. This apparent lack of hazard may
be caused by the strong affinity of dioxins for solid adsorbents such as fly ash
and soils, and the fact that some biomass combustion equipment is operated
above the thermal decomposition temperature of the dioxins, or 750C. The
hazards of dioxins appear to be minimal in biomass and waste combustion
heating of the reactor (electric power was priced at $0.04/kWh for all processes).
The Omniful fluidized-bed gasification process, which has not yet been operated
at elevated pressure as assumed in the economic analysis, afforded methanol at
the lowest cost, $0.88/gal. This system also turned out to have the highest overall
thermal efficiency. In all cases, however, the total production cost of methanol,
including capital charges, was not competitive with natural gas derived product.
Certain sites with special attributes that help to reduce the cost might therefore
be needed to make biomass-derived methanol competitive using current
technology. These site-specific characteristics might include lower feedstock and
electric power costs, as well as good markets for by-products such as C02. The
authors of this analysis concluded that the profitability of the various plants is
clearly related to the $0.40/gal federal tax subsidy, and that inclusion of the
subsidy would make several of these systems show a net after-tax income.
Without the subsidy, the per-gallon operating costs of most plants would
somewhat exceed the initial market prices of methanol. It should be noted,
however, that the federal tax subsidy if applicable is actually a tax forgiveness in
that it is not collected by anyone in the marketing sequence. Therefore, it does
not reduce the manufacturing cost.
2.2
Anaerobic Digestion Research
Laboratories throughout the world are continuing research on anaerobic digestion
(or methane fermentation) to increase knowledge of microbial processes and
biochemical mechanisms, to evaluate different types of waste streams and
biomass feedstocks as substrates for various reactors, and to develop processes
with improved reaction kinetics and methane yields. New information has been
reported on the mixed obligately thermophilic cultures, originally cultured from
350C waters along the East Pacific Rise, that were reported last year to be
capable of growth at 100C and 1 atm to produce CH4, CO, and H2 in mineral
media. These complex communities of thermophiles are capable of
chemolithotrophic growth at pressures of 265 atm and temperatures of at least
250C. A gas sample from a culture at 300C contained 9.6% CH4 (presumably
mol %), 2.2% H2, and 0.1% CO. The balance of the gas was not stated, but it was
probably water or CO2. Doubling times were 8, 1.5, and 0.67 h at 150, 200,
and 250C, respectively. These findings open up the possibility that bacteria may
exist and grow within the earths crust at temperatures exceeding 250C, and that
microbial growth is limited not by temperature, but by the existence of liquid
water, assuming all other conditions necessary for life are provided. The
investigators also concluded this work greatly increases the number of
environments on earth where life can exist. However, it should also be noted that
this work supports the possibility of microbial methane formation from minerals
within the earth at depths where the pressure-temperature relationships are such
as to maintain water in the liquid state. It will be very interesting to determine the
converting fatty acids to methane into one digester. The combined two-phase
system outperforms the other processes to a considerable degree.
Several research projects are in progress to evaluate various biomass species
as sources of methane. Application of a long-term, batch anaerobic
biogasification potential assay to several herbaceous species afforded methane
yields in the range 4 to 6 ft3/lb VS added. Sorghum exhibited a methane yield up
to 6.4 ft3/1b VS added, corresponding to a reduction in organic matter of 92%.
Methane yields as high as 3.5 to 5.1 ft3/lb VS added were observed for several
hardwood species without pretreatment, while sodium hydroxide pretreatment
resulted in improved methane yields or production rates with several species.
Based on this work and many other studies of the digestability of biomass, there
is no shortage of degradable feedstocks. Almost all land-based, aquatic, and
marine species examined to date either have good digestion characteristics or can
be pretreated to promote digestion. The cost, of course, is reduced if
pretreatment is not necessary.
One of the biomass species that does not require pretreatment for conversion
to methane by anaerobic digestion is giant brown kelp. A recent detailed economic
analysis was made of the growth, harvesting, and biological conversion of kelp
to methane for an integrated near-shore kelp farm and a shoreline digestion
plant. Depending on the assumptions made in the analysis, baseline methane
costs were $13.47/106 Btu. Kelp and gas yield increases, which are believed to
be achievable, reduced the methane price to $6.14 to $8.66/106 Btu. Thus, the
economic feasibility of kelp as a feedstock for methane production appears to be
within reach. Additional research on kelp growth and conversion will help
determine whether kelp-to-methane systems are competitive.
Although the economic feasibility of simultaneous waste disposal-methane
production by anaerobic digestion has been established in many cases, the
economic feasibility of producing methane from urban wastes is yet to be
demonstrated, according to some analysts. The 100-ton/day (design) RefCom
plant in Pompano Beach, Fla., which produces methane from MSW by anaerobic
digestion is the largest research project in the world that has the goal of
determining this feasibility. The RDF-preparation section of this plant was
recently modified to overcome operating problems. The entire system will now
be operated to evaluate the performance of newly installed disc screens in the
RDF section; dewatering of the digested solids by belt press, cone press, and
vacuum filter; the effects of digestion temperatures less than 60C; and the
effects of various municipal sludge: RDF ratios.
3
LIQUEFACTION *
Liquefaction of biomass and wastes is accomplished by natural, direct and
indirect thermal, and fermentation methods. Natural liquefaction systems were
referred to in Sec. 2.2 of this paper in connection with certain arid-land plants
and microalgae growth and the resultant formation of lipids and hydrocarbons.
Other natural processes that produce liquids suitable as fuels are performed by
certain tree species (e.g., the Brazilian Copaifera langsdorfii tree that yields
sesquiterpenes that can be used as diesel fuels without modification, and plants
that bear oil seeds; e.g., sunflowers). Research is continuing in all of these areas.
One of the most interesting approaches to the natural production of liquid
fuels by biomass is under investigation by Nobel prize winner Melvin Calvin
using a combination of natural photosynthesis and genetic manipulation. The
overall process consists of three steps: hybridization of Euphorbia lathyris with E.
esula, which produces fewer hydrocarbons than E. lathyris but grows as a
perennial rather than an annual; modification of the photosynthetic pathway of
the hybrid to cyclize C15 intermediates so that sesquiterpenes are formed; and
transfer of the gene that codes for sesquiterpene production from C. langsdorfii
to the plant. Conceptually, this sequence would optimize for sesquiterpene
production by a herbaceous plant that can be grown in the United States at high
annual yields without replanting each year. This process would provide a
significant advance over present techniques of liquid hydrocarbon production
from biomass.
Currently, more research is being done on direct and indirect thermal
liquefaction methods for biomass and wastes than on the other methods. Direct
liquefaction is either reaction of biomass components with smaller molecules
such as H2 and CO (e.g., PERC and LBL processes) or short-term pyrolytic
treatment, sometimes in the presence of gases such as H2. Indirect liquefaction
involves successive production of an intermediate, such as synthesis gas or
ethylene, and its chemical conversion to liquid fuels. In 1983, after several years
of laboratory and pilot-plant work on the PERC and LBL processes, which
involve reaction of product oil or water slurries of wood particles with H2 and
CO at temperatures up to about 370C and pressures up to 4000 psig in the
presence of sodium carbonate catalyst, researchers concluded that neither
process can be commercialized for liquid fuel production without substantial
improvement. The most attractive approach to such improvement is believed to
be a combination of solvolysis with a pyrolysis or reduction step. However, the
oxygen content of the resulting complex liquid mixture is still high (~6 to 10 wt
%), and considerable processing would appear to be necessary to upgrade this
material.
A convenient classification of biomass pyrolytic processes is shown in
Table 4. Maximum liquids yields are usually obtained in the intermediate
temperature range if the residence time is short. Among the short residence-time
processes (0.5 to 5 s) under development are vacuum pyrolysis at about 300 to
400C and 0.3 atm (U. of Sherbrooke, Canada), flash pyrolysis at about 500 to
650C and 1 atm (U. of Waterloo, Canada), hydropyrolysis in an atmosphere of
*Note: Ethanol and methanol fuels are discussed in the next two sections.
hydrogen at about 500 to 600C andS to 6atm (HYFLEX , IGT), and flash
pyrolysis in atmospheres of hydrogen or methane at 600-1000C and 1 to 70 atm
(Brookhaven National Laboratory). An interesting report of a relatively long
residence time (10 to 15 min heat-up, several hours at temperature) pyrolysis
study at reduced pressures of 0.0004 to 0.004 atm and temperatures of 250 to
320C of wild cherry wood seems to contrast with the results of several reports
on flash pyrolysis. In this study, about 70 wt % of the sample was volatilized at
290 to 315C over the pressure range studied; the major products were
methanol, acetone, acetic acid, cresols, and substituted phenols. These results
suggest that the combination of lower temperature, reduced pressure, and long
residence time may provide a technique for minimizing char and heavy tar
formation. In any case, the liquid products from all direct pyrolysis processes are
highly oxygenated and acidic. Chemical rather than fuel applications would
appear to be more feasible with these wood oils at this time.
Fundamental studies of the mechanisms of biomass pyrolysis continue to shed
more light on the complex chemistry of direct thermal conversion. One of the
most interesting techniques developed by the Solar Energy Research Institute
(SERI) for this work uses direct mass spectrometrio sampling of pyrolysis
products from wood. The goal of these studies is to determine, in molecular
detail, the chemistry and kinetics of the primary and secondary pyrolysis
processes for biomass and its constituents. Fingerprints characteristic of the
particular biomass used and identification of the broad range of compounds
formed including specific polynuclear aromatics will undoubtedly make this
technique very useful.
Table 4:
Biomass pyrolysis classifications*
pyrolysis type heating rate (C/ temperature (C) residence time
s)
Conventional
<2
primary products
<500
1983 is the competition caused by the reduction in gasoline prices. Obviously the
profit margins of the ethanol fuel producers who use corn feedstocks have
significantly diminished.
The largest ethanol fuel effort in the world, Brazils Proalcool program, is
making excellent progress and should achieve its production goal of 2.83109
gal/yr (10.7109 L/yr) by 1985, or the equivalent of 101,000 BOE/day.
Production in 1982 and 1983 was estimated at 1.1109 gal and 1.4109 gal; 1984
production is projected to be 2.3109 gal. Although discrepancies among the
various market figures are reported, about 136 ,000 neat-ethanol-fueled cars were
sold in the first quarter of 1983, and about 500,000 to 600,000 all-alcohol vehicle
sales were projected for 1983. Alcohol shortages are expected, so the Brazilian
government has stated that it will allow the production of neat-ethanol cars only
as long as fuel supplies are guaranteed.
In accordance with the requirements of the U.S. Energy Emergency Preparedness Act of 1982, the USDA and DOE published a study in 1983 on the
potential of a Strategic Alcohol Fuel Reserve (SAFURE) similar to the existing
Strategic Petroleum Reserve (SPR). In the event of a petroleum disruption,
ethanol (methanol was considered also) would be withdrawn from the SAFURE
and used as a gasoline extender. Two of the major conclusions from this study
were: the highest value use for ethanol during a disruption is as an octaneenhancing blending component for unleaded gasoline, and none of the SAFURE
alternatives compared favorably with equivalent crude oil storage in the SPR for
the base case set of assumptions. The overall, conclusions of the study were that
the costs of acquiring, storing, and managing an alcohol fuel reserve are
substantially higher than the costs of the current SPR program, and that the value
of a barrel of SPR crude oil stored in the Gulf Coast salt domes is substantially
greater than a barrel of ethanol stored under any of the SAFURE options
considered. The apparent built-in difficulty with this study is that it focused on
the use of alcohol as a high-octane gasoline additive; few alternatives were
evaluated. Consequently, this defined limitation became part of the conclusions,
and the intrinsic value of using ethanol (or methanol) for neat-alcohol fueled
vehicles, for example, to permit complete independence from petroleum
disruptions was not considered.
There is much current research on new ethanol fermentation methods and
improvements to increase productivities, permit the use of low-grade biomass or
biomass not readily converted, and conserve energy through more efficient
separation processes, thereby increasing net energy production. Many of these
projects should make important contributions to the development of superior
fermentation technology. Some are expected to include improved alcohol
fermentation processes for complete use of both the hexoses and pentoses;
simultaneous hydrolysis and fermentation of cellulosics; low-cost, multistage
acid hydrolysis processes for cellulosics; nondistillation separation of ethanolwater mixtures via extraction, adsorption-desorption on solids, and selective
Motor Company has supplied California with 506 methanol-fueled Escorts for
use in fleet tests, and the California Energy Commission has sponsored the
construction of 33 methanol fueling stations.
The role that biomass will have as a feedstock for methanol production is not
clear at this point. Numerous studies have been carried out to develop economic
estimates, and a few projects have been announced; but to date none of these
activities has resulted in an operating biomass-fueled plant in North America.
The most recent project, now in the detailed engineering phase, will process 300
ton/day of sawmill residues using the Omnifuel pressurized, fluid-bed gasifier
and a methanol train to yield 165 ton/day of methanol. But this scale of operation
is far removed from the projected demands for methanol fuel, since about 14 or
15 50,000-bbl/day methanol plants based on coal would be able to supply enough
methanol to displace about 10% of the countrys daily gasoline consumption.
The amount of dry biomass equivalent needed as feed for such plants would be
about 16,000 ton/day. This is about 25 times the feed rate of the largest woodfueled power plant believed to be feasible at this time because of the limitations
on fuel transport distance. Thus, present technology will limit methanol-frombiomass facilities to smaller dispersed plants, with maximum methanol
capacities of 2000 to 3000 bbl/day (~50% thermal efficiencies to methanol, ~1.5
bbl methanol/green ton feed).
The total production cost of methanol including capital charges for a 1070-bbl/
day plant supplied with green wood feedstock (wood chips at $15.50/ton, 50 wt
% moisture) ranges from $0.88 to $1.16/gal (1983 dollars) for a broad range of
gasification processes. As of December 21, 1983, methanol was selling for $0.43
to $0.45/gal, so clearly methanol produced by conventional wood gasificationsynthesis gas reduction technology could not compete in the open market.
Sensitivity analyses for the Omnifuel process showed that increases in plant sizes
to 500 and 1000 ton/day of methanol reduced the methanol price to $0.67 and $0.
63/gal, which is still not competitive. A similar analysis conducted with landfill
gas at $1.50/10 Btu for a 568-bb1/day plant afforded methanol at $0.77/gal.
CONVERSION OF LIGNOCELLULOSIC
BIOMASS TO ETHANOL
L.J.DOUGLAS*
*Solar Energy Research Institute, Golden, Colorado
1
INTRODUCTION
The Solar Energy Research Institute (SERI) has been assigned field
responsibility by the Department of Energy, Biomass Energy Technology
Division, for the direction of the National Alcohol Fuels Program. The principal
objectives of the program are to develop the conversion technology for
transforming lignocellulosic biomass into fuels alcohols, and to provide a
mechanism for the early transfer of mature processes to the private sector.
Premium liquid fuels, particularly in the transportation sector, represent an area
of potential shortfall in the event of another disruption in imported oil, and in the
long term liquid fuels are expected to be in short supply. This paper discusses
why ethanol is a useful fuel, and why lignocellulose is a suitable feedstock for
ethanol.
2
ETHANOL AS A FUEL
Ethanol is a high-value liquid fuel that has excellent octane-enhancing properties
and readily blends with the existing hydrocarbon-based fuel supplies and
distribution system. Ethanol has very clean combustion properties and offers an
environmentally safe material for octane replacement for tetraethyllead, which is
being phased out of gasoline.
3
LIGNOCELLULOSIC BIOMASS AS A FEEDSTOCK FOR
ETHANOL
Ethanol is principally made from grain and ethylene. Ethylene-based ethanol
plants are being closed down as the price of petroleum increases because
ethylene-based ethanol cannot compete with grain-based ethanol in the
marketplace. However, substantial increases in the amount of grain based
ethanol would cause disruptions in the food chain and raise the often heard food
versus fuel arguments.
Lignocellulosic feedstocks provide an inexpensive, renewable resource for the
production of ethanol. Use of these feedstocks would minimize competition with
the food chain, since dedicated wood crops could be grown on marginal,
nonagricultural lands. The conversion processes are environmentally benign and
can be adapted to such lignocellulosic materials as crop and forest residues,
hard- and softwoods, and MSW.
4
FEEDSTOCK CHARACTERISTICS
Lignocellulosic materials are in the form of a matrix that consists of three
polymeric components: cellulose, hemicellulose, and lignin. These three
components are situated in the material in a manner that provides mechanical
strength and forms a barrier to chemical and biological attack. Cellulose is the
principal component of the complex and is a linear polymer of hydroglucose
units. Cellulose is always found in nature in microfibrils, which are 40 cellulose
chains or 35 angstroms in cross-section. These very rigid structures are
embedded in a matrix of amorphous hemicellulose and lignin, which forms the
resistant cell walls.
The feedstocks chosen for this program are hardwoods and crop residues.
Figure 1 shows the relative composition of a hardwood, in which the cellulose
fraction constitutes about 50% of the material, hemicellulose about 25%, and the
lignin about 25%.
5
RESEARCH OBJECTIVES
The objectives of the program are to
Develop the technology base for the economic conversion of lignocellulosic
biomass to ethanol and other liquid fuels
Improve the overall utilization of lignocellulosics through research on
Acid hydrolysis processes
Enzymatic hydrolysis processes
Supporting research activities.
Figure 2 shows the current conversion efficiency of developed processes and
indicates that only cellulose is used to produce fermentation sugars. Thus only
about 30% of the feedstock is used to produce an end-product fuel. Figure 3
shows ethanol yield goal of the program in which about 70% of the feedstock is
used to produce sugars for fermentation to ethanol and other liquid fuels. In
addition, the lignin fraction can be converted to liquid fuels via thermal
processes that will boost the total yield to about 90% of the feedstock. Clearly,
the research objectives are to produce the maximum amount of liquid fuels from
the feedstock; however, relating the changes in conversion yield to specific
improvements in the cost of production of the ethanol is not an easy process.
Figure 4 shows the year 2000 cost goals for renewable ethanol technology,
which is to produce ethanol for $1.21/gal without any federal subsidy. This goal
will be accomplished by meeting several research objectives that include
reduction in the feedstock cost via the short rotation woody crops program and
utilization of higher percentages of the native feedstock through higher
conversion yields. Note that the feedstock contribution to the selling price of
ethanol becomes less significant as the conversion yields increase. If the process
efficiency can also be improved, then the cost of the feedstock will no longer
dominate the process economics.
6
ROUTES TO ETHANOL
The several approaches to production of ethanol are shown in Fig. 5. The
fermentation of sugars, conversion of starch to sugars, and fermentation of
sugars to ethanol are mature technologies and will not be described in this paper.
The conversion of lignocellulosic biomass to ethanol via hydrolysis of cellulose
and hemicellulose followed by fermentation is a more difficult approach because
of the highly resistant properties of this material. The two basic types of
hydrolysis of cellulose are acid and enzymatic hydrolysis. The various system
configurations for each of these processes can accommodate variations in several
process parameters, including type and concentration of acid, operating
temperature, reactor configuration, microbial system, and pretreatment.
Acid hydrolysis approaches included for study in the program are dilute
sulfuric acid and concentrated sulfuric acid processes. Two dilute acid processes
are being investigated. One uses Dartmouths plug-flow reactor design, which
features a high-temperature, short residence time system and will be discussed in
more detail later. The other is a moderate-temperature (180C) system that is a
variation of the Madison percolation system developed during the 1940s.
Three enzymatic hydrolysis systems are being developed under the program:
separate hydrolysis and fermentation (SHF), simultaneous saccharifi cation and
fermentation (SSF), and direct microbial conversion (DMC) processes. The SHF
system uses separate hydrolysis and fermentation reactors for processing the
feedstock and is the standard (base case) approach for microbial conversion,
conceptually the most simple. This system will also be discussed in more detail
later. The SSF system combines the hydrolysis and fermentation steps in one
reactor. While this variation provides several technical improvements, it also
requires additional process control. The SSF process appears at present to be the
most feasible system in the mid-term. The DMC approach uses thermophilic
bacteria that feed directly on the cellulose and hemicellulose portions of the
feedstock to produce sugars and use the sugar to produce ethanol directly, in one
step. Conceptually, this should be the preferred system in the long-term;
however, this approach has the highest risk and the most technical barriers to
success.
7
BASE CASE STUDIES
In this paper only one variation for each of the general process types will be
discussed: (1) a high-temperature, dilute-acid hydrolysis process, and (2) a basecase enzymatic hydrolysis process (SHF) that uses steam explosion as the
pretreatment.
7.1
Acid Hydrolysis Process
The base-case acid hydrolysis process employs a plug-flow reactor in the
hydrolysis unit that operates at 240C, 1% acid, and a 7-s residence time. The
substrate is ground aspen wood, which is loaded into the reactor as a 15% solids
slurry with water. The unreacted solids are pressed and burned to provide heat
for the process. The liquid stream containing the soluble sugars is neutralized
using calcium hydroxide and treated to remove any materials that would be toxic
to the fermentation organisms. Standard fermentation and distillation convert the
glucose sugars to ethanol and purify the ethanol to fuel quality. The base-case
design was sized to produce 50 million gal/yr of anhydrous ethanol from aspen
wood; the cost estimates were based on a Gulf Coast location and constant 1983
dollars. The 50 million gal/yr size reflects the dispersed nature of biomass and
assumes a maximum collection radius of 25 miles. The estimate for the selling
price of ethanol from the base-case system as configured was $2/gal.
Parametric analyses were performed for several operating conditions and
system configurations. The items determined to be the dominant cost-controlling
factors are feedstock cost, solids loading in the reactor, and xylose sugar
utilization.
The cost of the feedstock is the dominant factor in determining the costeffectiveness of production of ethanol in the acid hydrolysis process. Figure 6
shows the relationship of the selling price of ethanol in $/gal versus the cost of
the aspen wood. This relationship dictates the necessity of reducing the cost of
the feedstock or increasing the yield for the process to reduce the production cost
per unit of ethanol.
The amount of water in the process streams has a large effect on the process
economics. The presence of large amounts of water increases the steam
requirements as well as the equipment size and capital cost of the plant. In
addition, low solids loading yields a dilute sugar stream to the fermentation
section, resulting in a low ethanol concentration in the beer and increased energy
requirements for the distillation section. Figure 7 shows the effect of solids
loading on the selling price of ethanol. The optimum solids loading appears to be
about 30%. This is the target value in our current process design; however, reliable
equipment for handling solids at this level may be difficult to obtain.
Process changes that favor the use of the xylose sugars in the feedstock can
greatly reduce the selling price of ethanol. One approach being studied is to
develop, through genetic engineering techniques, a yeast that can ferment both
xylose and glucose in the same fermentation vessel. This research issue is being
pursued with vigor because meeting this goal has the potential to reduce the
selling price of ethanol to about $1.10/gal. A more imminent alternative is to
convert the xylose to furfural and use the by-product credit to subsidize the
selling price of ethanol. Figure 8 shows the effect of credit for furfural on the
selling price of ethanol. The current market price for furfural is about $0.60/lb,
and a net selling price for furfural of $0.10/lb could reduce the selling price of
ethanol to less than $1/gal. With the current demand for furfural, one or two 50million-gal/yr plants would saturate that market. However, there is a large
potential market for derivatives of furfural that are based on well-known
chemical processes if the price of furfural were low enough to encourage the
chemical industry to change feedstocks.
7.2
Enzymatic Hydrolysis Process
The base-case enzymatic hydrolysis process was a separate hydrolysis and
fermentation (SHF) system that was studied at SERI using the Chem Systems
model, which was developed under a SERI subcontract. The data include recent
research results from Lawrence Berkeley Laboratories.
The base case consisted of steam-explosion pretreatment, enzyme production
by a fed-batch fermentation of the RUT-C30 strain of the fungus Triehoderma
reesei, hydrolysis of the cellulose to glucose by enzymes followed by
fermentation of the glucose in a separate process step, and vapor reuse
distillation. The operating conditions for the base case were:
Pretreatment for 5 seconds with 400 Ib of steam per ton of wood at 560 psig
and 247 C
Substrate loading of 20 wt %
Enzyme loading of 25 FPU per gram of substrate (to yield 80% conversion of
cellulose to glucose).
The base case for the enzymatic process was also sized to produce 50 million gal/
yr and was subjected to the same constraints as the acid hydrolysis case. The
estimated selling price of fuel-grade ethanol from the process as configured was
$2.30/gal.
presentation are accurate to only 35% at best. The real value in this exercise
comes from comparing various process options and plant configurations, and
using the results of these studies to identify the research and process activities
that have the greatest potential for cost reduction.
Fig. 4. Cost goals for renewable ethanol technology (based on feedstock cost of $40/dry
ton)
Fig. 8. Selling price of ethanol as a function of the net furfural by-product credit
Enzyme Productivity
(IU/liter/hr)
Enzyme Costs
($/gal of ethanol)
Fig. 11. Research improvement
1980
1985
1990
2000
50 IU
100 IU
200 IU
400 IU
1.20
.60
.30
.15
SYNPOSIS
Hemicelluloses are major components of plant biomass. In hardwoods and
agricultural residues, xylose is the principal hemicellulosic sugar. Xylose and
other hemicellulosic sugars are recovered from lignocellulose more readily but
are fermented with greater difficulty than is glucose. Xylose metabolism
employs pathways distinctly different from those involved in the utilization of
glucose. With most yeasts, xylose metabolism requires air. Aeration results in
cellular respiration (as opposed to fermentation) and low ethanol yields. It is
possible, however, to suppress respiration by feeding small amounts of glucose
during the xylose fermentation. Some yeasts, such as Pachysolen tannophilus,
will metabolize xylose anaerobically. Alternately, other yeasts will anaerobically
ferment the keto isomer of xylose, xylulose, after it is formed from xylose by the
action of xylose isomerase. In both instances, the fermentation rates are low.
Improved strains of P. tannophilus have been obtained by UV mutagenesis
followed by enrichment for faster growth in nitrate-xylitol broth or by selecting
for yeast strains incapable of using ethanol as a carbon source. Several yeasts
have been described as superior xylose fermenters, including (in approximate
ascending order) : Candida troplcalis. Kluyveromyces marxianus. P. tannophilus,
the mutant Candida sp. XF 217, and Candida shehatae (and its sexually perfect
form, Pichia stipitis) . The xylose fermentation rate of C. shehatae is 3 to 5 times
higher than that obtained with P. tannophilus, but the yields of ethanol from
xylose are similar with the two organisms. The glucose fermentation rate and
ethanol yield are lower with C. shehatae) than with P. tannophilus. Unstable
petite and grande strains of C. shehatae have been obtained on urea+xylitol agar,
and some show markedly different fermentation rates and products. Further
strain improvement and process development should soon provide commercially
practicable technology for the fermentation of xylose.
1
INTRODUCTION
Within the realm of liquid fuel production from biomass, utilization of
lignocellulose has focused largely on the problem of cellulose saccharification.
Various approaches have been tried including hydrolysis by extra-cellular
streptomycete and fungal cellulases, simultaneous saccharification and
fermentation by cellulolytic bacteria, and acid hydrolysis followed by
fermentation. To a certain extent, the attention given to cellulose is justifiable.
Cellulose comprises about half of the total weight of lignocellulose, and its
fundamental constituent, glucose, is an excellent fermentation substrate. In a larger
context, however, consideration of cellulose to the exclusion of the two other
major constituents, hemicellulose and lignin, is futile. One reason is that it is
uneconomical to throw away almost half of the feedstock. Another is that
cellulose has appreciable commercial value as fiber. Converted to pulp, a ton of
cellulose is worth $400 to $700; converted to ethanol, it is worth less than $300.
In the kraft pulping process, lignin and hemicellulose are extracted under
alkaline conditions and then burned to recover chemicals and energy. In some
instances, the lignin is recovered for other applications. The hemicellulose is
largely degraded to organic acids prior to combustion and has no current
commercial value. Other technologies are being developed that will enable the
efficient fractionation of lignocellulose into pulp-grade cellulose, useful lignin
derivatives, and useful hemicellulosic sugars including xylose.
The objective of the research described in this paper is to improve our
knowledge of pentose metabolism in yeasts and to thereby provide the means for
more efficient utilization of xylose.
1.1
Sources and Recovery
Hemicellulosic sugars are major constituents of wood and agricultural residues.
Table 1 shows the average proximate composition of seven commonly occurring
hardwood (angiosperm) and softwood (gymnosperm) species (Ref. 13) along
with a few major U.S. agricultural residues (Refs. 46).
Table 1:
Proximate composition of some major lignocellulosic materials*
lignooellulose
hemicellulosic
sugars
cellulose
lignin
Hardwoods**
Softwoods
Wheat straw
Corn stalks
24
19
29
28
45
43
31
30
21
29
14
lignooellulose
hemicellulosic
sugars
cellulose
lignin
Soybean residue
19
37
Most yeasts use a xylose reductase with a specific requirement for NADPH as
a cofactor to reduce xylose to xylitol. Next, xylitol dehydrogenase specific for
NAD oxidizes xylitol to xylulose. Consequently, assimilation of xylose converts
NADPH into NADH. In Candida utilis, the organism best studied in this regard
(Ref. 8), NADPH is supplied by the oxidative phase of the pentose phosphate
pathway (PPP) in a closed cycle (Fig. 1). Under oxidative conditions, the only
mode of fungal xylose assimilation known until 1981, NADH is recycled to NAD
by respiration. Under anoxic conditions, NAD cannot be regenerated, and xylose
assimilation ceases (Ref. 9). NADPH is used primarily in metabolic syntheses,
and is generated mainly by the oxidative PPP (Ref. 10). Thus, production of
NADPH by the oxidative PPP is thought to provide the means to assimilate
xylose for aerobic production of ethanol by Candida tropicalis and other yeasts
(Ref. 12). More recently, certain yeasts capable of fermenting xylose to ethanol
in the absence of oxygen (anoxically) (Ref. 13) have been shown to possess
xylose reductase(s) capable of using either NADH or NADPH as a cofactor
(Ref. 14). If NAD(H) can be used for both the reductive and oxidative steps of
xylose metabolism, the balance between NAD and NADH can be maintained
under anoxic conditions (Fig. 2) and xylose utilization is not dependent on
aeration. The observation that P. tannophilus will ferment but not grow
anaerobically on D-xylose could be attributed to the insufficient production of
metabolic reductant or energy for growth.
2
PROCESS ALTERNATIVES
2.1
Coupled Isomerization and Fermentation
In 1980, Wang, Shopsis, and Sohneider (Ref. 15) showed that yeasts are able to
ferment xylulose to ethanol under anoxio conditions. This finding had immediate
implications because the conversion could be carried out readily by using
commercial xylose (glucose) isomerase. The discovery was immediately seized
upon and became the basis for considerable research and development in this
field. As proposed for commercial practice, the technology would employ
exogenous, immobilized xylose isomerase (already commercially derived from
bacteria) to convert xylose to an equilibrium mixture of xylose and xylulose. The
xylulose would then be fermented to ethanol and the residual xylose recycled
over the xylose isomerase. The process would be continued until all xylose was
consumed. Several variations on the basic process are possible and most have
been attempted, but the principal remains the same. Xylose isomerase could be
incorporated directly into the fermentation vessel or the xylulose could be
produced exogenously and separated from the xylose prior to fermentation.
g/Lha
g/ghb
Xylulose
S. cerevisiae
0.058
c
S. cerevisiae
0.25
0.025
C. tropicalis
0.7
C. tropicalisc
1.2
S. pombe
0.086
S. pombe
0.10
Xylose
K. marxianus
0.040.1
P. tannophilus
0.12
0.05
c
P. tannophilus
0.023
2.0
P. tannophilus
0.12
C. shehatae
0.28
Glucose
P. tannophilus
1.0
0.22
C. shehatae
0.5
S. cerevisiae
11.8
1.78
Z. mobilis
8.0
2.5
aGrams ethanol/liter of reactor per hour.
reference
Chiang et al. (24)
Suikho and Poutanen (25)
Jeffries (21)
Jeffries (21)
Roman et al. (26)
Chiang et al. (21)
Margaritis and Bajpai (27)
Jeffries et al. (28)
Slininger et al. (29)
Slininger et al. (30)
du Preez and van der Walt (31)
Jeffries et al. (28)
this report
Maiorella et al. (32)
Rogers et al. (33)
fermentation rate
cells
g/Lha
g/ghb
bGrams ethanol/gram dry wt cells per hour.
clmmobilized cells.
reference
3
PROCESS VARIABLES
3.1
Effects of Aeration
Aeration stimulates cell growth and occasionally stimulates fermentation as well.
Only a few yeasts are capable of (limited) anaerobic growth. This inability stems
in part from a biochemical requirement for molecular oxygen in the synthesis of
membrane steroids. However, Schneider and co-workers (Refs. 7, 8) have shown
that the inability of P. tannophilus to grow anaerobically cannot be overcome by
the addition of ergosterol or unsaturated fatty acids. Maleszka and Schneider
have also shown that oxygen and mitochondrial function are also required for S.
cerevisiae to grow on xylulose (Ref. 39). These observations suggest that the
anaerobic metabolism of pentoses supplies metabolic energy (ATP) fast enough
to satisfy only the basal metabolic demand but does not provide enough ATP to
allow cell growth.
Aside from affecting growth, aeration strongly affects the specific
fermentation of glucose by P. tannophilus. This was first shown by Scheffers and
Wiken (Ref. 40). Unexpectedly, the stimulation does not extend to xylose
(Table 3). Aeration does increase the volumetric fermentation rate, but this
stimulation can be attributed to increase in cell mass.
Table 3:
Effect of aeration on the specific and volumetric rates of glucose and xylose fermentations
by P. tannophilus (Ref. 25)
aerobic
g/Lh*
anaerobic
g/gh**
Glucose
0.98
0.22
Xylose
0.12
0.05
*Volumetric rate=gethanol/(liter-h)
**Specific rate=g ethanol/(g dry wt cellsh)
g/Lh*
g/gh**
0.11
0.06
0.10
0.05
amount of carbon going into xylitol is deducted from the calculation, because it
accumulates early in the fermentation pathway and essentially represents sugar
that is not metabolized. The yield of xylitol decreases under aerobic conditions
and increases under anaerobic conditions (Ref. 28).
3.2
Effect of Glucose on Ethanol Yields from Xylose
The aerobic ethanol yield from xylose can be improved by adding small amounts
of glucose during the fermentation (Table 5). By using this approach, a high rate
of ethanol production can be achieved with relatively little ethanol loss. The
improvement in yield is not observed under anaerobic conditions, and control
experiments show that adding glucose at the low concentrations employed does
not affect the rate of xylose assimilation. So the observed improvement in yield
is attributed to a decrease in the rate of ethanol respiration (Ref. 28).
Table 4:
Product yields of P. tannophilus from glucose and xylose under anaerobic conditions
(Ref. 28)
ethanol*
co2*
xylitol*
acetic*
Ethanol
(xo-xi)**
Xylose
0.33
0.32
0.30
0.03
0.47
Glucose
0.48
0.44
0
0.02
0.48
*Yield expressed as g product g sugar consumed.
**Yield expressed as g ethanol (g xylose consumedg xylitol produced).
Table 5:
Effect of glucose additions on the yield of ethanol from xylose under aerobic conditions
(Ref. 28)
ethanol yield
g/g*
4.5% xylose alone
0.28
4.5% glucose alone
0.43
3% xylose+0.5% glucose
0.41
at T0, 24 and 48 h
0.43
*g ethanol/g sugar consumed
**actual yield divided by theoretical (0.51 g/g)100
% of theoretical**
55
84
81 (xylose)
84 (glucose)
3.3
Effects of Nitrate on Ethanol Production
Nitrate increases the levels of PPP enzymes in yeasts, fungi, and plant cells
(Refs. 10, 4557). The enhancement occurs because the PPP is the primary
3.5
Candida shehatae as a Rapid Xylose Fermenter
Although mutation and selection methods have been successful in obtaining
incremental improvements in the xylose fermentation rates of laboratory strains,
enrichment and screening of yeast strains from natural sources has led to the
identification of C. shehatae as a species capable of fermenting xylose at four to
five times the specific rate of P. tannophilus (Ref. 31). Candida shehatae
produces up to 3.8% (w/w) ethanol from 16% D-xylose (Fig. 8) and about 5%
ethanol from 16% D-glucose (Fig. 9). In comparison to P. tannophilus, which
forms much more ethanol on D-glucose than on D-xylose, with C. shehatae the
final ethanol concentrations on these two sugars and the ethanol yields (after
deducting xylitol production) are about the same (Table 6).
Table 6:
Product yields from xylose and glucose (g/g) (Ref. 25)
sugar conc. (%)
xylose
ethanol
xylitol
glucose
EtOH/(Xo-Xi)*
4
0.307
0.052
8
0.267
0.151
12
0.264
0.236
16
0.242
0.274
20
0.255
0.230
*ethanol/(xylose consumedxylitol formed) .
cells
ethanol
cells
0.322
0.313
0.343
0.334
0.328
0.096
0.069
0.047
0.040
0.039
0.322
0.345
0.350
0.312
0.253
0.100
0.072
0.044
0.035
0.026
Grande
0.86
0.14
0.92
0.03
0.05
0.84
0.16
0.94
0.06
0.92
0.08
0.93
0.07
Petite
1.440.20
1.120.22
0.970.11
0.410.12
1.440.10
1.230.16
0.710.13
1.460.26
0.670.28
1.380.20
0.640.23
1.760.41
0.440.31
The xylose fermentation characteristics of petite and grande strains are related to
respiratory activities. When a tetrazolium agar overlay is applied to colonies
growing on urea+xylitol agar, five different colony types can be distinguished
(Table 8). Some of these strains, occurring in low frequency, exhibit a small
colony diameter and a strong tetrazolium reaction on xylitol. These strains are
poor ethanol producers, but they form other products. The petite colonies
showing a weak tetrazolium reaction on xylitol tend to produce less xylitol and
glycerol, but the higher overall yield of all products formed by grande
tetrazolium-positive strains tends to suggest that these strains have lower
endogenous respiratory activity when growing fermentatively. Neither grande nor
petite strains show significant tetrazolium reactions on xylose. Preliminary
studies in my laboratory have shown that a similar petite-like variation occurs in
strains of P. stipitls. An improved understanding of this petite-like variation should
eventually contribute to the isolation of yeast strains capable of fermenting
xylose economically under practical conditions.
Table 8:
Average colony diameters, tetrazolium reactions, and fermentation patterns of C. shehatae
ATCC 22984 strains exhibiting different characteristics on xylitol agar (Ref. 54)
freq.
dia. (mm)
tetrazolium reaction
ethanol
g/g*
g/g h**
g/g/*
g/g/*
0.08
2.440.30
3+
0.03
1.610.14
4+
0.84
1.150.01
1+
0.01
0.92 0.05
4+
0.04
0.650.19
1+
*g product/g xylose consumed.
**g product/g dry wt cells h.
other products formed.
0.23
0.20
0.18
0.11
0.17
0.09
0.05
0.07
0.02
0.08
xylitol
glycerol
0.26
0.40
0.14
0
0.10
0.11
0.16
0.04
0
0.04
3.6
Comparison of Various Xylose-Fermenting Yeast Strains
Various researchers have used many different media and cultural conditions in
studying different yeast strains for their abilities to ferment xylose. While these
strains doubtless possess different optima for ethanol production, it is useful to
compare them under a single set of fermentation conditions. My lab has recently
done such a comparison. Results show that although the mutant strains of P.
tanophilus and Canida sp. performed better than their parent strains, all strains of
C. shehatae were better than any other strain tested. These results show that further
strain development with C. shehatae as well as enrichment and selection of new
isolates should continue.
4
CONCLUSIONS
1. Xylose is widely available in angiosperm residues and more readily
recoverable than glucose from lignocellulosic materials.
2. Although a two-stage isomerization and fermentation of xylose is feasible,
direct fermentation of xylose to ethanol can proceed at a higher specific rate and
is more likely to have a lower overall cost.
3. Xylose fermentation rates and ethanol yields are still much lower than
commercial glucose fermentations, but they are improving. For specialized
situations where waste xylose streams constitute a disposal problem,
fermentation to ethanol may be economical.
4. Biochemical, genetic, and strain selection studies have only recently been
undertaken, and it is expected that they should result in better strains and
fermentation conditions.
5
ACKNOWLEDGEMENT
The author wishes to thank Henry Schneider and his coworkers at the National
Research Council, Ottawa, Ontario for useful discussions and for sharing
references and preprints of unpublished data.
6
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
Jeffries, T.W. Utilization of xylose by bacteria, yeasts and fungi, Adv. Biochem.
Eng./Biotechnol., vol. 27, 132, 1983.
Chen, W.-P. Glucose isomerase (a review), part 1, Proc. Biochem., vol. 15,
3035, 1980.
Chen, W.-P. Glucose isomerase (a review), part 2, Proc. Biochem., vol. 15,
3641 1980.
Hsiao, H.-Y., Chiang, L.-C., Chen, L.-F. and Tsao, G.-T. Effects of borate on
isomerization and yeast fermentation of high xylulose solution and acid hydrolysate
of hemicellulose, Enzyme Microb. Technol., vol. 4, 2531, 1982.
Venkatasubramanian, K. High fructose corn syrup: plant design and process
economies, in Enzymes, The Interface Between Technology and Economics,
J.P.Danehy and B.Wolnak, eds., Marcel Dekker, 1978.
Jeffries, T.W. Fermentation of xylulose to ethanol using xylose isomerase and
yeasts, Biotechnol. Bioeng. Symp., vol. 11, 315324, 1981.
Maleszka, R., and Schneider, H. Fermentation of D-xylose, xylitol and D-xylulose
by yeasts, Can. J.Microbiol., vol. 28, 360363, 1982.
Suihko, M.-L. and Dravic. Pentose fermentation by yeasts, Biotechnol. Lett., vol.
5, 107112, 1983.
Chiang, L.-C, Gong, C.-S., Chen, L.-F. and Tsao, G.T. D-xylulose fermentation to
ethanol by Saccharomyces cerevisiae, Appl. Environ. Microbiol., vol. 42,
284289, 1981.
Suihko, M.-L. and Poutanen, K. D-xylulose fermentation by free and immobilized
Saccharomyces cerevisiae cells, Biotechnol. Lett., vol. 6, 189194, 1984.
Roman, G.N., Jansen, N.B. and Tsao, G.T. Ethanol inhibition of Dxylulose
fermentation by Schizosaccharomyces pombe, Biotechnol. Lett., vol. 6, 712,
1984.
Margaritis, A. and Bajpai, P. Direct fermentation of D-xylose to ethanol by
Kluyveromyces marxianus strains, Appl. Environ. Microbiol., vol. 44, 10391041,
1982.
Jeffries, T.W., Fady, J.H. and Lightfoot, E.N. Effect of glucose supplements on the
fermentation of xylose by Pachysolen tannophilus, Biotechnol. Bioeng., 1984.
Slininger, P.J., Bothast, R.J., Black, L.T. and McGhee, J.E. Continuous conversion
of D-xylose to ethanol by immobilized Pachysolen tannophilus. Biotechnol.
Bioeng., vol. 24, 22412251, 1982.
Slininger, P.J., Bothast, R.J., Van Cawenberge, J.E. and Kurtzman, C.P.
Conversion of D-xylose to ethanol by the yeast Pachysolen tannophilus,
Biotechnol. Bioeng., vol. 24, 371384, 1982.
du Preez, J.C. and van der Walt, J.P. Fermentation of D-xylose to ethanol by a strain
of Candida shehatae, Biotechnol. Lett., vol. 5, 357362, 1983.
Maiorella, B.L., Blanch, H.W. and Wilke, C.R. Economic evaluation of alternative
ethanol fermentation processes, Biotechnol. Bioeng., vol. 26, 10031025, 1984.
Rogers, P.L., Phil, D., Lee, K.J. and Tribe, D.E. High productivity ethanol
fermentations with Zymomonas mobilis. Proc. Biochem., vol. 15, 711, 1980.
Kreger-van Rij, N.J. W., ed. The yeasts, a taxonomic study (3rd edition), Elsevier
Science Publishers, 1984.
Barnette, J.A., Payne and Yarrow, D. The yeasts, Cambridge University Press,
1983.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
Toivola, A., Yarrow, D., van den Bosch, E., van Dijken, J.P. and Scheffers, W.A.
Alcoholic fermentation of D-xylose by yeasts, Appl. Environ. Microbiol., vol. 47,
12211223, 1984.
Maleszka, R., Wang, P.Y. and Schneider, H. Ethanol production from Dgalactose
and glycerol by Pachysolen tannophilus. Enzyme Microb. Technol., vol. 4,
349352, 1982.
Neirinck, L.G., Maleszka, R. and Schneider, H. The requirement of oxygen for
incorporation of carbon from D-xylose and D-glucose by Pachysolen tannophilus,
Arch. Biochem. Biophys., vol. 228, 1321, 1984.
Maleszka, R. and Schneider, H. Involvement of oxygen and mitochondrial
function in the metabolism of D-xylulose by Saccharomyces cerevisiae, Arch.
Biochem. Biophys., vol. 228, 2230, 1984.
Scheffers, W.A. and Wiken, T.O. The Custers effect (negative Pasteur effect) as a
diagnostic criterion for the genus Brettanomyces, Antonie van Leeuwenhoek, vol.
35, A31-A32, 1969.
Tyagi, R.D. Participation of oxygen in ethanol fermentation, Proc. Biochem., vol.
19, 136141, 1981.
Hoppe, G.K. and Hansford, G.S. The effect of microaerobic conditions on
continuous ethanol production by Saccharomyees cerevisiae , Biotechnol. Lett.,
vol. 6, 681686, 1984.
Maleszka, R. and Schneider, H. Concurrent production and consumption of ethanol
by cultures of Pachysolen tannophilus growing on D-xylose, Appl. Environ.
Microbiol., vol. 44, 909912, 1982.
Bruinenberg, P.M., de Bot, P.H. M., van Dijken, J.P. and Scheffers, W.A. The role
of redox balances in the anaerobic fermentation of xylose by yeasts, Eur.
J.Biotechnol., vol. 18, 287292, 1983.
Hankinson, O. and Cove, D.J. Regulation of the pentose phosphate pathway in the
fungus Aspergillus nidulans, J. Biol Chem., vol. 249, 23442353, 1974.
Jessup, W. and Fowler, M.W. Interrelationships between carbohydrate and
nitrogen metabolism in cultured plant cells, Part 3, Effect of the nitrogen source on
the pattern of carbohydrate oxidation in cells of Acer pseudoplatanus grown in
culture, Planta, vol. 137, 7176, 1977.
Osmond, C.B. and Rees, T.A. Control of the pentose phosphate pathway in yeast,
Biochim. Biophys. Acta., vol. 184, 3512, 1969.
Jeffries, T.W. Effect of nitrate on the fermentation of xylose and glucose by
Pachysolen tannophilus, 1983 Biotechnology, vol. 1, 503 506, 1983.
Jeffries, T.W. Mutants of Pachysolen tannophilua showing enhanced rates of
growth and ethanol formation from D-xylose, Enzyme Microb. Technol., vol. 6,
254258, 1984.
McCracken, L.D. and Gong, C.-S. D-xylose metabolism by mutant strains of
Candida sp., Adv. Biochem. Eng./Bioteohnol., vol. 27, 3455, 1983.
Schneider, H. Personal communication, 1984.
du Preez, J.C., Prior, B.A. and Monteiro, A.M.T., The effect of aeration on xylose
fermentation by Candida shehatae and Pachysolen tannophilus, Appl. Microbiol.
Technol., vol. 19, 261266.
Dellweg, H., Rizzi, M., Methner, H. and Debus, D. Xylose fermentation by yeasts.
3. Comparison of Pachysolen tannophilus and Pichia stiptis, Biotechnol. Lett., vol.
6, 395400, 1984.
Fig. 7. Ethanol production from 4.5% D-xylose by mutants of P. tannophilus. Symbols are
the same as in Fig. 6.
NOVEL DEVELOPMENTS IN
BIOREACTOR DESIGN AND
SEPARATIONS TECHNOLOGY
C.D.SCOTT*
*Chemical Technology Division, Oak Ridge National Laboratory,
Oak Ridge, Tennessee
1
INTRODUCTION
Much of the recent activity in the new biotechnology is associated with the
production of high-value products at a relatively small scale. However, the more
conventional fermentation industry has a long history of large-scale production
of commodity-type products. This type of technology will gain increasing
importance as the production of fuels and chemicals from renewable resources
again becomes economically viable. Although many advanced research and
development concepts will have applications in either type of biotechnology, the
primary interest here will be to consider the large-scale systems, with particular
emphasis on bioconversion systems.
In general, an integrated bioconversion system using advanced concepts may
have several processing steps, including preparation of the feed material,
production of the biocatalyst, the bioconversion step, and product concentration
and purification (Fig. 1). The heart of the system is the bioreactor; however,
separation processes also play important roles in several of the possible
processing steps. Recent advances in more efficient separation processes and
advanced bioreactor concepts will be discussed in greater detail.
2
SEPARATIONS
Separation processes are important components in many of the possible
processing steps in advanced biotechnology (Fig. 1). They may include:
Fractionating biomass feed materials to supply necessary process substrates
Removing or recovering particulates in suspension such as microorganisms,
microbial fragments, raw material fractions, or nonbiological solids
Recovering and concentrating macromolecules such as proteins and enzymes
that will either be the end product or be used as biocatalysts
behind on the packing (Fig. 4). After the sorbent bed had been loaded with
product, it would be taken off line (at least two packed beds would be required
for continuous operation) and regenerated. Again, a thermal process would
probably be used; in this case, however, the solid sorbent would be extremely
nonvolatile and, in turn, thermal efficiencies should be quite high.
Current research is oriented toward developing a sorbent material that has high
capacity and high specificity. Both synthetic molecular sieves and neutral resins
show promise for this application (Fig. 5) (Refs. 4, 5), and there will undoubtedly
be other candidates in the future.
2.3
Integrated Processes
The most optimum separation systems may include a combination of two or
more concepts into a single processing step. For example, sorption or solvent
extraction in conjunction with distillation may have an advantage over either
single process. It is also beneficial to combine the primary separation step with
the fermentation step in a single processing vessel. This would result in a
decrease in the total number of processing steps, but it also might enhance the
fermentation step since product inhibition could be decreased.
3
ADVANCED BIOREACTOR SYSTEMS
Most conventional, large-scale bioreactor systems are batch-fed stirred tanks;
however, important research advances are being made on innovative, new
concepts that would be more efficient, controllable, and productive. It would be
desirable to have continuous bioreactors with high concentrations of the
biocatalyst (e.g., microorganisms) that also operate in a multistage mode. The
latter point is important in instances where product or substrate inhibition affects
the microbial kinetics.
3.1
Continuous Columnar Bioreactors
Columnar systems operating continuously with high concentrations of
immobilized biocatalysts appear to be very attractive as high-productivity
bioreactors. Such systems with large biocatalyst particles and low product flow
rates will operate as fixed-bed bioreactors, while systems with small biocatalyst
particles and higher flow rates will be operated as fluidlzedbed bioreactors in
which the particles are suspended by the upflow of the feed stream (Fig. 6).
Continuous columnar systems have been shown in the laboratory to be many
times more productive than batch-stirred tanks, (Refs. 6, 7, 8) and they can
apparently be scaled up in a straightforward manner (Fig. 7).
3.2
Immobilization of Biocatalyst
Advanced columnar bioreactor systems require that the biocatalyst be
immobilized onto or into a particulate that remains in the reactor, even at high
flow rates. In some fixed-bed systems, especially for wastewater treatment, a
biological film is formed on the external surface of the packing material (Ref. 9).
However, systems for producing commodity-type chemicals will require a much
better controlled immobilization technique.
Of particular interest is the incorporation of the biocatalyst into a solid support
matrix. For example, natural gels, such as carrageenan, can be used to entrap
microorganisms into biocatalyst beads (Fig. 8) (Refs. 7, 10). Such materials,
when used in a fluidized-bed bioreactor for producing ethanol from glucose,
have been shown to increase productivity by over an order of magnitude when
compared with conventional technology (Ref. 8).
4
CONCLUSIONS
Advanced technological approaches promise to significantly increase the
productivity and to decrease the cost and energy requirements of biological
processing systems. This seems to be especially true for largescale fermentation
for commodity-type chemicals. As feed materials for bioprocess become
competitive with fossil-based processes, a resurgence in the bioconversion
industry will undoubtedly follow.
REFERENCES
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Fig. 2. Simplified solvent extraction system for product recovery and concentration
Fig. 4. Simplified solid sorbent separation system for product recovery and concentration
Fig. 3. Potential ex tractants for ethanol from dilute aqueous solutions KD=(concentration
in organic phase/ concentration in aqueous phase), a=(concentration of ethanol/
concentration of water) in extradant
Fig. 5. Ethanol breakthrough curve for a 72.5-mL packed bed of molecular sieves
operating at 21C and 185 mL/h
Fig. 8. Carrageenan gel beads containing Z. mobllis cells at approximately the theoretical
concentration
1
INTRODUCTION
A thermochemical conversion process to convert various biomass materials to
diesel type fuels has been under development at Arizona State University (ASU)
since 1975. An indirect liquefaction approach is used, i.e., gasification to synthesis
gas followed by liquefaction of the synthesis gas. The primary virtue of an
indirect liquefaction approach for cellulosic feedstocks is that oxygen contained
in the materials is easily separated. Thus the hydrocarbon liquid product is free
of oxygenated compounds and can therefore be tailored to match transportation
fuel products currently derived from petroleum. Approximately 100 biomass
materials are being studied as received from private industry, government
laboratories, and other university laboratories. The feedstock candidates include
industrial wastes, agricultural and forest residues, and crops that could be
deliberately grown for energy conversion purposes. The product of the process is
a liquid hydrocarbon transportation grade fuel similar to diesel. This can be
upgraded to high-octane gasoline via catalytic reforming if desired. The products
should be compatible with existing engine designs and fuel distribution and
marketing systems. The major virtue of the process is that a renewable, often lowvalued material is used as the feedstock to produce a quality product.
Current efforts on the project are designed to maximize diesel fuel yields for a
variety of feedstocks of commercial interest and to establish corresponding
reliable, reproducible mass and energy balances. Auxiliary tasks include
alternative feedstock assessment, alternative product development, system
simplification and automation, gasification and liquefaction catalyst
development, environmental assessment, and scale-up to a 10 ton/day facility.
The primary purpose of the larger scale facility is to produce a sufficient amount
of product for applications testing and to minimize the risk to a commercial scale.
With the present status of the process, a profitable scale would require about 300
ton/day of feedstock. This number is expected to decrease with further
improvements in the process that are currently in progress in the research
reformer
490
400
11
3
PROCESS STUDIES
The basic objective of the project is to maximize yields of high-quality, oxygenfree liquid hydrocarbon fuel suitable for transportation use in existing engines.
The oxygen in the biomass is converted to carbon monoxide, carbon dioxide, and
water in the gasification step. In the liquefaction step, the carbon monoxide is
converted to paraffinic hydrocarbons, water, and normal propanol via the
following reactions:
With proper manipulation of the reactions, the oxygen in the biomass will end up
in water, carbon dioxide, and normal propanol. Carbon dioxide and water will be
vented from the gasification system regenerator, and an immiscible alcoholwater phase will be separated from an oxygen-free paraffinic hydrocarbon phase.
Past and present efforts on the project have been aimed at optimizing the
implementation of this scheme via process variable studies involving the
assessment of various feedstocks and factor studies on the conversion system.
The feedstock materials studied in our laboratory are received from private
industry, municipalities, government laboratories (from various nations), and
other university laboratories. A photograph of the feedstock storage area (milled
materials) is shown in Fig. 4. A range of characterization data for the various
ranges
7,40012,700
0.135.9
0.125.3
0.120.2
0.039.20
010.4
0.526.6
7.828.8
17.746.7
5.114.9
Table 2:
Factor studies
Gasification:
1.
Reactor system configuration
2.
Feedstock characterization
3.
Heat transfer media/catalyst
4.
Fluidization gas composition
37.760.9
4.78.8
28.954.4
0.31.7
< 0.01
5.
Residence time
6.
Temperature
7.
Pressure
8.
Recycle effects
Liquefaction:
1.
Catalyst composition
2.
Catalyst preparation method
3.
Catalyst calcination, reduction, pretreatment
1.
Reactor system configuration
5.
Conversion temperature
6.
Conversion pressure
7.
Conversion residence time
8.
Feedgas composition
9.
Recycle effects
Table 3:
Synthesis gas composition (mol %)
Hydrogen
Carbon monoxide
Olefins
Paraffins
Carbon dioxide
range
typical
1053
660
539
633
426
32
32
10
15
11
Table 4:
Properties of Fischer-Tropsoh product and commercial fuel oils
Specific
gravity
Gravity, API
Boiling point
range,
Fischer-Tropsch product
almond
prunings
feedstock
guayule
bagasse
0.8360
0.8108
0.7586
0.7902
0.7950
37.8
43
55
47.6
46.5
336
410
479
47.8
147
302
438
48.3
235
352
471
45.3
238
414
535
55.7
21676
22440
19354
21043
0F
10% 369
evaporated at 50% 458
90% 563
Calculated cetane
45.9
index
Heating value, Btu/lb 19383
Factor studies in the gasification system indicate that low pressure and
residence time, ~1500F, and a combination of steam (hydrogen source) and
liquefaction reactor off gas (paraffin source for cracking to olefins and
hydrogen) for fluidization are favorable. The fluldized solid candidates are still
under investigation to satisfy the criteria of operational reliability and selectivity
(catalysts). For the liquefaction system, the best catalyst is an impregnated cobalt
catalyst with conversion conditions of 500F, 140 psig, 15 to 3s single-pass
residence time, and 3/1 recycle (weight basis) . The H2/CO mole ratio in the
synthesis gas can be manipulated over a broad range for a given feedstock (say 0.
5 to 8.0), but the olefin composition is heavily feedstock dependent. Typical
synthesis gas compositions considered achievable for virtually any biomass
feedstock are indicated in Table 3.
Without any post reactor refining, the product quality is most similar to JP-4
jet fuel due to the presence of materials in the C7C10 range. A simple distillation
will produce a product in the No. 2 diesel fuel range. Further tuning of process
conditions is expected to establish the flexibility to manipulate the product
quality without the necessity of a separation step.
Details of the various feedstock and factor studies performed in the laboratory
are presented elsewhere (Refs. 15).
4
CONTINUING WORK
Tasks in progress in the laboratory include product yield improvement,
throughput optimization, alternative feedstocks and products assessment,
environmental compatibility, process simplicity, and automation. A contract has
been let to Ultrasystems, Inc. to design a larger scale version (10 ton/day of
feedstock) of the process with the primary purpose of minimizing the risk at a
commercial scale. The continuing objective is to minimize the break-even
commercial scale and thus increase the number of applications for the process
technology.
5
ACKNOWLEDGEMENTS
The research project described in this paper is currently sponsored by the U.S.
Department of Energy, Office of Industrial Programs and the U.S. Department of
Agriculture. International Business Machines Corporation (IBM) has provided
computational equipment via a joint study program with Arizona State
University.
REFERENCES
1.
2.
3.
4.
5.
Kuester, J.L. Conversion of cellulosic wastes to liquid fuels, DOE interim report,
COO-2982108, Contract No. DE-AC0276CS40202, May 1984.
Kuester, J.L. Diesel fuel from biomass, in Proceedings of the Energy from
Biomass and Wastes VIII Symposium, Institute of Gas Technology, 1984.
Kuester, J.L. Liquid hydrocarbon fuels from biomass, Chapter 8 in Biomass as a
Non Fossil Fuel Source, D.L. Klass, ed., ACS Symposium Series 144, American
Chemical Society, 1980.
Kuester, J.L. Conversion of cellulosic wastes to liquid fuels, Chapter 15 in
Energy from WasteVol. I, T.C. Franiewicz, ed., Ann Arbor Science Publishers,
1980.
Kuester, J.L. Diesel fuel from biomass via indirect liquefaction, presentation at
BioEnergy 84, Gotenborg, Sweden, June 1981.
1
INTRODUCTION
The Energy Research Advisory Board (ERAB) has estimated that biomass could
potentially supply the nation with about 10.51015 kJ (101015 Btu) by the year
2000 (Ref. 1). Similarily, the Office of Technology Assessment (OTA) has
estimated that, with proper resource management and the development of
efficient conversion processes, the potential contribution of biomass to U.S.
energy demand could range as high as 181015 kJ (171015 Btu) per year (Ref. 2).
To place these estimates in some perspective, they represent, respectively, 14%
and 23% of the nations total estimated energy consumption in 1982 (Ref. 3).
This potential contribution is greater than that of any other renewable energy
technology. In addition, biomass is the only renewable technology that can
contribute to the need for transportation fuels. Thermochemical conversion
processes are expected to contribute a majority of the total. In thermochemical
conversion elevated temperatures convert biomass materials to more useful
energy forms. Wood and crop residues are 96% of the biomass feedstocks
available for conversion to liquid and gaseous fuels. Thermochemical processes
are capable of converting 85% to 95% of the organic material in these feedstocks
with high efficiency and relatively little sensitivity to variations in the feed
material.
In addition to utilizing diverse biomass resources efficiently, thermal
conversion processes can produce a broad spectrum (Fig. 1) of energy products
that fit existing U.S. energy use patterns.
Biomass feedstocks have unique properties that offer great potential
advantages for thermochemical conversion processes. These advantages include:
High VolatilityBiomass feedstocks contain 70% to 90% volatile material for
wood versus 30% to 45% for typical coals. Thus a large fraction of most
biomass feedstocks can be pyrolyzed (devolatilized) rapidly at relatively low
temperatures. Figure 2 compares weight loss due to devolatilization versus
temperature for wood and a typical coal.
241
Gasification technology
Liquid fuels technology.
A discussion of the complete program can be obtained elsewhere (Refs. 4, 5).
Selected projects with nearer term potential for technology commercialization by
the private sector are discussed.
2.1
Direct Combustion Technology
The Biomass Thermochemical Conversion Program is sponsoring innovative
direct-combustion research projects focused on converting the heat released from
direct combustion directly into mechanical power. By directly producing
mechanical power without the use of an intermediate working fluid, such as
steam in a boiler-steam turbine system, operating economies can be realized in
industrial-scale applications.
Aerospace Research Corporation, Roanoke, Va., is conducting research to
determine whether combustion gases from wood can be used to directly power
gas turbines. Hot combustion gases from a pressurized, wood-fired suspension
burner are passed through a series of cyclones to remove particulate matter and
are injected directly into a gas turbine. A general flowsheet for the process is
shown in Fig. 5. Trials using a 375-kW combustor/ turbine system indicated no
significant signs of erosion or corrosion following 500 hours of operation. Tests
showed that 80% to 90% of the particulates entering the gas turbine were less
than 0.5m in diameter. Total particulate loading entering the turbine ranged
from 9 to 13 mg per dry normal cubic meter of combustion gas. These results
strongly suggest that turbine erosion will not be a problem.
A 3000-kW combustor/gas turbine electrical generating system was
constructed at Roanoke. Particulate sampling tests conducted during shakedown
tests indicated that the particulate loadings entering the turbine are similar to
those observed with the smaller system. The 3000-kW system has been moved to
Red Boiling Springs, Term., where it will undergo long-term testing while
generating electrical power, which will be sold to the Tennessee Valley
Authority. Over 80% of the funds for the 3000-kW unit are being provided by
the private sector. The Allison Division of General Motors is donating the gas
turbine for the project. Figures 6 and 7 show the combustor/cyclone systems and
the turbine/generator for the 3000-kW unit.
2.2
Gasification Technology
Biomass gasification technology can be divided into processes that produce a
low-energy gas with an energy content of 3.5 to 7.0 MJ/m3 (90 to 180 Btu/ft3)
and those that produce a medium-energy gas containing 12 to 21.5 MJ/m3 (300
243
to 550 Btu/ft3). Several systems for producing low-energy fuel gas are
commercially available. The versatility of the fuel gas produced by these lowenergy systems is restricted and subject to the following limitations:
The low heating value of the gas requires that it be consumed at or near the
site of production in a close coupled process
Retrofitting oil or natural gas-fired boilers to use low-energy fuel gas usually
requires boiler derating or extensive retrofit modifications
The high nitrogen content of low-energy gas precludes its use as a synthesis
gas for the production of liquid fuels.
The Biomass Thermochemical Conversion Program has recently initiated a small
research effort aimed at resolving problems associated with using low-energy
gas to fuel internal combustion engines.
However, the main thrust of the gasification research and development efforts
sponsored by the program is directed toward technology for the production of
medium-energy gases. Medium-energy gas is a versatile fuel with the following
advantages:
A higher heating value, which allows it to be used in nearly all retrofit
applications on a particular industrial site without major modifications or
derating problems
Higher flame temperatures, making it suitable for retrofitting processes, where
higher flame temperatures are critical
Two to five times the energy density of low-energy gas, allowing it to be
transported moderate distances by pipeline at a reasonable cost
Required for the synthesis of derived liquid fuels such as methanol and
Fischer-Tropsch liquids, and for the production of pure methane (SNG).
Gasification research sponsored by the Biomass Thermochemical Conversion
Program is directed toward optimizing medium-energy gasifiers to produce fuel
gas or synthesis gas for subsequent conversion to liquid fuels, and developing
innovative reactor designs that reduce or eliminate the requirement for oxygen by
exploiting the high reactivity of biomass.
Battelle-Columbus Laboratories (BCL) is conducting research on a process to
produce a medium-energy fuel gas without requiring pure oxygen. This process
involves indirectly heating an entrained bed gasifier by circulating low-density,
hot, incandescent sand to the gasifier. The entrained sand and any unreacted char
leaving the gasifier are separated from the product gas in a cyclone. The char is
burned in a fluidized-bed, air-fired combustor, and the hot sand is recirculated
back to the gasifier. A schematic flowsheet for the process is shown in Fig. 8.
In previous research using a 15.2-cm (6-in.) diameter gasifier, BCL
successfully produced a medium-Btu fuel gas with a heating value of about 18.6
MJ/m3 (475 Btu/ft3) using both wood chips and coarsely shredded bark. This
heating value did not vary significantly for feedstocks containing from 8% to 40/
6 moisture. Gasifier throughputs of up to 8660 kg/h m2 (1800 lb/h ft2) were
achieved.
The gasifier was subsequently modified to a 25-cm (10-in.) diameter to more
fully examine critical operating parameters. It has operated successfully,
obtaining even higher carbon conversions under similar operating conditions
than were obtained in the smaller unit. Preliminary estimates indicate that a 17.4
to 21.6 MJ/m3 fuel gas can be produced by this process for less than $4.27/103 MJ
($4.50/106 Btu).
The University of Missouri-Rolla is conducting a research program to
investigate the technical feasibility of using a metal, fire-tube heat exchanger to
provide heat to a fluidized bed gasifier. This concept, suggested by Davy
McKee, would allow the production of medium-energy gas or synthesis gases
without using expensive oxygen. The university uses a 0.5-m (20-in.) diameter
gasifier fitted with an internal heat exchanger consisting of thirty 2.5-cm (1-in.)
diameter U-tubes spaced on a 5-cm (2-in.) pitch. Hot combustion gases,
produced by burning a portion of the product gas or other fuels such as char, are
passed through the heat exchanger.
The system has been operated at feed rates up to 205 kg/h (450 lb/h) using
10% moisture wood and temperatures up to 740C (1365F). Current research
efforts are aimed at determining the optimal process conditions for producing
medium-Btu gas and critical design parameters for the heat exchanger. A view of
the reactor, with the top removed exposing the heat exchanger, is shown in
Fig. 9.
The Solar Energy Research Institute (SERI) is developing a pressurized,
oxygen-blown, fixed-bed, down-draft gasifier (Fig. 10) for the production of
methanol synthesis gas or medium-energy fuel gas. The 0.15-m I.D. gasifier has
a nominal capacity of 900 kg of wood per day and a design operating pressure of
1.03 MPa (150 psig). In down-draft operation, pyrolysis tars and oils produced in
the upper zone of the bed are forced through the hotter zone at the bottom of the
bed where they are cracked. Typical hydrogen to carbon monoxide ratios ranged
from 0.5 to 0.8 in the synthesis gas produced. Oxygen consumption averaged 0.
43 kg/kg of wood gasified.
The Institute of Gas Technology (IGT) is conducting research on an oxygenblown pressurized, fluidized-bed research gasifier (Fig. 11) designed to produce
either a synthesis or fuel gas. The 0.30-m (12-in.) diameter research gasifier has
a design capacity of 410 dry kg (900 lb) of wood per hour. The effects of
pressure up to 3.5 MPa (500 psia) oxygen consumption are being explored. Data
obtained to date indicate that oxygen consumption has ranged from about 0.20 to
0.35 kg/kg of wood (~10% moisture) gasified. Gas heating values have ranged
from 11.5 to 14.7 MJ/m3 (292 to 372 Btu/ft3). Hydrogen to carbon monoxide
ratios of the gas produced have ranged from 1.8 to 2.2. An important activity
associated with this project has been the design and construction of a reliable
pressurized wood lock-hopper feeding system.
245
2.3
Liquid Fuels Technology
The ability to convert biomass to liquid fuels has several advantages over the
biomass resource itself. Liquid fuels have a higher energy density and can be
transported and stored more economically. Liquid fuels also match existing
energy end-use patterns better, particularly in the transportation sector. As noted
previously, biomass is the only renewable energy technology that can contribute
to the supply of transportation fuels.
It must be noted, however, that the molecular composition of biomass can lead
to biomass-derived liquid fuels that are different chemically from petroleum
fuels. The challenge, then, is to generate high-value liquid fuels from biomass
that can both supplement existing liquid fuels and be economically competitive.
The liquid fuel research sponsored by the DOE Biomass Thermochemical
Conversion Program is directed at:
Identifying reaction pathways and methods to produce high-value
intermediates for liquid fuels
Improving yields and quality of liquid fuels
Producing economically competitive fuels in the post-2000 time frame.
The liquid fuels research is divided into two primary areas: pyrolysis and direct
liquefaction. The differences in these approaches make it possible to generate a
variety of fuels ranging from fuel oil substitutes to olefinic/aromatlc
hydrocarbons, helping to meet the wide range of demands from the liquid fuel
market.
Pyrolysis refers to the heating of biomass in the absence of air. Traditionally,
it has been used to produce charcoal. Conventional pyrolysis typically produces
products consisting of about one third each gases, pyrolysis oil, and solid char. In
recent years, the concept of rapid pyrolysis has emerged as a promising
alternative. Using rapid heating rates, yields of gases and liquids as high as 95%
can be produced.
Georgia Tech University, Atlanta, Ga., is conducting research with an
entrained flow pyrolysis unit with the goal of generating high yields of liquids at
low cost. During 1983, Georgia Tech finished construction of an entrained flow
pyrolysis reactor, shown schematically in Fig. 12. The system consists of an
upflow, entrained pyrolysis reactor and an oil recovery system that allows partial
on-stream fractionation of the product. Experimental operation of the unit is
currently under way. Reaction temperatures of 400 to 550C have been
examined and wood feed rates up to 10 kg/h (90 Ib/h) have been obtained.
Preliminary results indicate that greater than 50% oils by weight (moisture-free
basis) can be obtained, well over double the yields from conventional pyrolysis
systems. Results from related research indicate that yields as high as 60% by
weight may be obtained as operating parameters are optimized. While the
pyrolysis oils are quite different chemically than petroleum liquids, the low
projected costs of the biomass oils indicate that upgrading may be economically
feasible.
SERI is investigating the use of an ablative reactor for fast pyrolysis to
determine how to obtain the high heat fluxes needed for rapid pyrolysis and to
investigate fundamental reaction behavior. The unique reactor -supplies heat for
reaction by the ablation of biomass particles forced against a hot reactor wall.
Heat-up rates of up to 500,000C/sec can be obtained at the sample surface.
Contact of the biomass with the reactor surface converts the biomass into a liquid
layer, which subsequently is vaporized. A schematic of the reactor system is
shown in Fig. 13. The initial products formed in the reactor section are primary
pyrolysis vapors that can be condensed as pyrolysis oils at yields of about 60%.
The primary tars can also undergo secondary cracking in a vapor cracker to form
highvalue products such as benzene, toluene, xylene, and ethylene, plus carbon
monoxide and hydrogen.
Biomass can also be converted to liquid fuels via direct liquefaction
technology. Direct liquefaction research at this time is based primarily on a
concept proposed by the Pittsburgh Energy Technology Center (PETC). In this
concept, biomass is mixed with recycle wood oil and sodium carbonate catalyst
along with a mixed H2/CO reducing gas of mixture. The mixture is injected into
a high-pressure vessel at 21 MPa (3000 psi) and heated to about 350C.
During 1981, tests of the PETC process in a DOE research facility located at
Albany, Ore., confirmed the technical feasibility of this approach. Over 5000 kg
(11,000 Ib) of oil resembling No. 6 fuel oil were produced in one run alone.
However, the results indicated that the process is not now economically
competitive, partly because of the large recycle oil requirement. The Biomass
Thermochemical Conversion Program is attempting to improve the
competitiveness of direct liquefaction by:
Improving the economics of direct liquefaction through the use of increased
feedstock slurry concentrations
Improving product quality by reducing oxygen content and molecular weight.
The University of Arizona, Tucson, is conducting research on an advanced
concept for direct liquefaction that would use very concentrated biomass slurries.
The goal of this work has been to use a polmer extruder as a slurry feeding/pumping
device. The modified extruder/feeder system is capable of handling slurries as
concentrated as 60% wood solids in biomass oil. Conventional systems, by
comparison, typically cannot handle slurries containing over about 25% wood.
During 1984, the university designed and began constructing an integrated
extruder/static mixer liquefaction system. The static mixer is expected to allow
adequate mixing and agitation of the viscous slurries. Construction of the unit is
nearing completion, and shakedown operation is expected in early 1985. A
schematic flowsheet of the Arizona reactor system is shown in Fig. 14.
247
3
ADDITIONAL INFORMATION
Detailed descriptions of all the research and development projects funded by the
Biomass Thermochemical Conversion Program are given in Ref. 5.
REFERENCES
1.
2.
3.
4.
5.
Solar energy research and development: federal and private sector roles, draft report
of the solar R&D panel of the energy research advisory board, Sept. 2, 1982.
Energy from biological processes, volume Ibiomass resource base, Office of
Technology Assessment, Congress of The United States, Washington, D.C.
Energy projections to the year 2010, Office of Policy, Planning and Analysis, U.S.
Department of Energy, DOE/PE-0029/2, Oct. 1983.
Schiefelbein, G.F., Stevens, D.J. and Gerber, M.A. 1983 annual report: biomass
thermochemical conversion program, PNL-5096, Aug. 1984. National Technical
Information Service, U.S. Department of Commerce, Springfield, VA.
Proceedings of the 16th biomass thermochemical conversion contractors
meeting, May 89, 1984, Portland, OR. CONF-8405157. National Technical
Information Service, U.S. Department of Commerce, Springfield, VA.
Fig. 3. Biomass chars gasify very rapidly compared to peat and coal chars
249
251
Fig. 12. Flowsheet for entrained flow pyrolysls system at Georgia Tech
253
1
INTRODUCTION
Wood-burning research begun in 1976 led to the building and testing of a 375kW wood-burning gas turbine system (Ref. 1). Experience with that system
resulted in the building and testing of a 3000-kW system. The gas turbine had
long been used with liquid and gaseous fuels and operability with wood had been
demonstrated more recently. The main question to be answered, that of turbine
life, could have been answered to some extent by long-term testing of the 375kW system. However, gas turbines differ in their designs. Those differences may
not greatly affect the life of turbines powered by liquid or gaseous fuels, but with
wood there is the possibility of erosion by solid particles, and the number of
turbine stages, gas velocities through the turbine passages, and turbine inlet gas
temperature may make a significant difference in turbine life. Therefore, it was
decided to build a system and test it in a location where power could be
generated and sold during life testing of the turbine. The selection of a gas
turbine, location of the operating site, construction of the system, and
preliminary operating results are discussed in this report.
2
GAS TURBINE SELECTION
Factors that were considered in gas turbine selection were size, adaptability to an
external burner and wood fuel, availability, efficiency, and maintainability. Both
foreign and domestic gas turbines were evaluated.
2.1
Gas Turbine Size
An economic study of the 375 to 500-kW gas turbine quickly eliminated it as a
candidate. The next size turbineone that produces 3,000 to 5,000 kWis
popular for offshore oil drilling rigs, gas pipeline compressors, and emergency
power in the United States and abroad. Its major components can be fabricated at
the factory and moved to the operating site by common carrier. The wood
requirement is only 100 to 150 tons/day, which matches the local wood supply in
many areas. The capital outlay is under $2 million. Use of two, three, or four of
the 3,000 to 5,000-kW systems at a given location provides for continued
operation when one system is shut down for maintenance. The time required for
site preparation and installation is about one year, but movement to a new site
can be accomplished in two months if the new site is prepared before the move.
2.2
Adaptability to External Burner and Wood Fuel
Some gas turbines, especially those of foreign design, employ external burners
for liquid and gaseous fuels. Most U.S. designs are of the through flow type used
on airplane engines. In those designs, a major factor is the ability to modify the
burner section so that air can be ducted away from the compressor to an external
burner, and hot gases can be ducted back to the turbine. The only 3000-kW
engine manufactured in the United States that is readily adaptable to an external
burner is the Allison model 501 K.
The introduction of a higher pressure drop between the compressor and
turbine can adversely affect the matching of these two components. In a gas
turbine engine, the compressor and turbine are sized so that at the operating
speed and pressure of the compressor, the angle of air flow into the compressor
blades matches that of the blade angle within 6 to 8 degrees. If the air flow rate
drops off, the angle of flow into the blades may become large enough to stall the
blades. Continuous operation of a stalled compressor can result in structural failure
of the blades. Thus, detailed performance curves for the compressor and turbine
are needed to determine if changes in the compressor operating point due to
ducting, combustor, and filter losses will result in compressor stall.
With the use of wood there is an additional shift in the operating point. A
greater mass of wood than oil or gas is required for a given gas turbine output
because of the lower heating value of the wood. In addition, the moisture in the
wood results in an increase in the mass flow. Both increase the flow through the
turbine, and consequently there is a higher pressure demand from the compressor
(Ref. 2).
The data required to make an evaluation of engine performance with external
combustion system pressure drop and wood fuel are not easily obtainable from
the gas turbine manufacturers, especially foreign suppliers. Allison supplied
complete data on the 501 K engine. A computer program was written to monitor
the 501 K engine performance, and the results showed that stable operation of
the engine could be expected at design speed.
2.3
Gas Turbine Efficiency
Three foreign gas turbines with external combustors were evaluated as
candidates for the 3000-kW system. All had published efficiencies of 24% or
less, compared to 28.3% for the model 501 K engine. With the increased
pressure loss through the combustor, the computed efficiency for bone-dry wood
is 27.3%. With 35% moisture wood, the efficiency drops to 26.4%.
In Australian experiments (Ref. 3) it was shown that injection of steam into
the hot gas stream ahead of the turbine could significantly increase the power
output and overall efficiency of the gas turbine. GM-Allisons tests on the 501 K
engine showed similar results. Calculations were made for the system
performance on wood with the injection of 4 Ib steam/s at 405F generated by
turbine exhaust gases. By maintaining a turbine inlet temperature of 1700F, an
overall efficiency of 31% can be achieved with a power output of 5000 kW.
2.4
Maintainability
Availability of replacement parts and the speed with which they can be replaced
favors aircraft derivative engines as they are designed for ease of maintenance. The
foreign engines that were evaluated could not be easily replaced and repairing of
the turbine in the field would be lengthy. The turbine section of the 501 K, which
includes the rotating and stationary blades, can be removed and replaced in an
eight-hour shift. With a solids burning system in which there is a chance of
turbine blade erosion, that feature is very attractive and may prove essential to
commercial success.
2.5
Gas Turbine Selection
The results of the evaluation were that none of the foreign built engines could
compare with the Allison model 501 K for this application. GM-Allison provided
a new engine on loan for the application, and the U.S. Air Force provided a T-56
engine on loan as a backup. The Air Force is interested in this test because at
least six Air Force bases have enough wood to provide all of their power needs
on a continuing basis.
3
SITE SELECTION
Two sites were considered for location of the first system. They were Bedford
County, Va., and Macon County, Tenn. The objective in considering the two
sites was to acquire property along high voltage transmission lines where there was
easy access to wood waste. Property was acquired in both places. However, after
purchase of the property in Bedford County, Va., the avoided cost of electricity
paid by the Virginia Electric and Power Company was reduced, making it less
economical to install the system there than in Macon County.
The Tennessee Valley Authority recommended Macon County because it has
a large excess of waste wood, primarily sawdust. Sawmillers in the area have
difficulty in disposing of sawdust because the nearest market of any size is 120
miles away. The selected site, which lies along a 69,000-V transmission line, is
within the city limits of Red Boiling Springs, Tenn. Contracts have been
negotiated with sawmillers in the area for delivery of sawdust for $5 per green
ton. A ten-year contract has been signed with the TVA for generation of up to 14,
000 kW of power to be sold at a rate of 4.645/kWh peak and 2.925/kWh off
peak.
4
SYSTEM CONSTRUCTION
The realistic testing of a gas turbine with wood as the fuel can most
economically be accomplished in a system that converts the fuel to salable
electric power. A system with 3000-kW output requires approximately 100 tons
of green wood per day (50% moisture). The weighing and processing of that
amount of wood requires a sizable outlay of funds for land and equipment. The
combustor and filter are comparable in size to the complete gas turbine generator
set. To deliver the generated power to the TVA requires an electrical substation
complete with switch gear, transformers, oil circuit breakers, metering
equipment, and high voltage line switch.
To design, build, and assemble all of the required equipment requires a total
outlay of approximately $1.8 million. About 22% of the financing of this
beneficial project has been contributed by the Department of Energy. Upon
completion of the research, the system, with some modification, will be useful as
a commercial system. Thus the remaining 78% of the funding has come from
private sources. A brief description of the Red Boiling Springs facility follows.
Scales that have a 50-ton capacity were installed to accommodate tractortrailer loads of sawdust. The fuel is bark-free hardwood sawdust. Moisture
content of random samples is determined for each load as delivered. Some
deliveries will be made in self-dumping trucks and some in trailers, which will
be unloaded with a front-end-loading tractor at an unloading dock.
Approximately one hour is required to unload 25 tons of sawdust. Experience in
the research program has shown that outdoor storage of sawdust is satisfactory
except under the most inclement conditions of snow and ice.
Green sawdust as delivered is fed by a self-unloading enclosed bin to a
fluldized bed dryer that uses exhaust gases from the gas turbine as the drying
agent. A dryer was built and tested for this research. The information derived
from that dryer was furnished to a dryer manufacturer, who designed and
constructed a dryer for the Red Boiling Springs operation. Sawdust with not
more than 30% to 35% moisture leaves the dryer and is fed into a hammermill
with a 7/32-in. screen to produce a uniform particle size. The dried and
pulverized sawdust is fed to a self-unloading van and then to a live bottom
metering bin by means of two elevating augers. Wood is metered from the live
bottom bin into the rotary valve above the combustor (see Fig. 1).
5
PRELIMINARY OPERATING RESULTS
After operation of the system at low speed, three factors emerged: the height of
the firebrick in the secondary combustor, the horsepower requirement during
starting, and a decision on the elimination of the second cyclone. After adequate
operation had been conducted to take action on the first two problems,
preliminary sampling ofthe exhaust gases for particle content was conducted
by Battelle-Columbus Laboratories. As there was no means of dissipating any
significant amount of power generated by the system at Roanoke, it was
necessary to forego testing at rated conditions before moving the system to Red
Boiling Springs. However, the system was operated at approximately 20% wood
feed, maximum operating turbine inlet temperature, and 65% of rated speed, as
discussed later.
5.1
Combustor Performance
Based on the results obtained with the 375-kW system, the burner for the 3000kW system was designed as shown in Fig. 2. In the initial design the primary
combustion chamber was lined from top to bottom with high-density firebrick.
The combustor was designed so that products from the primary chamber swirled
into the secondary chamber and exited at the top. It was estimated that the
burning of any char would be completed within approximately 5 ft of the bottom
of the secondary chamber. That portion was lined with firebrick, and the
remainder of the chamber was lined with RA253MA alloy. During operation it was
determined that some burning at the walls occurred above the 5-ft level, making
it necessary to raise the height of the firebrick. Because of uncertainty as to the
height at which burning ceased, the full height of the chamber was lined with
firebrick. The main side effect of increasing the brick height was a reduction in heat
transfer to the combustion air. As can be seen in Fig. 2, air from the compressor
enters the top of the secondary chamber and moves through an annular passage
down, then across and up to the primary combustion zone. As the air moves, it
cools the firebrick while increasing the air temperature. With the stainless steel
liner above the firebrick, the inlet temperature of the air at the combustion zone
would have been approximately 1000F at operating conditions. Experimental
results to date indicate that with full firebrick lining, the temperature will be 900
F. The air temperature at the primary combustion zone and the condition of the
wood feed determine the completeness of burning in the primary chamber, which
determines the amount of char entering the secondary chamber and the maximum
temperature reached on the walls. That temperature must be kept below
approximately 2300F to prevent slagging of the wood ash, which would require
slag removal at the bottom of the secondary chamber. With 900F air, hardwood
sawdust may have to be dried to below 25% moisture to avoid slagging.
5.2
Starting Requirements
Power from the Oldsmobile starting engine is delivered through the 3000-kW
generator shaft and gear box to the gas turbine. Additional power to the gear box
is provided by driving the generator as a motor. Below approximately 12,000 rpm
in the conventional engine configuration it is necessary to bleed air at the fifth
and tenth stages of the compressor to avoid compressor stall. With the long
ducting, large combustor, and cyclone filter on the wood burning system, the
amount of compressor-generated pressure lost is 8%, compared to 4% for the
conventional engine. As a result, it is necessary to bleed additional air from the
compressor outlet duct to avoid stall. Without the additional bleed, stalling
occurred at 5000 rpm with an engine inlet temperature of 1800F.
The additional bleed together with the greater loss through the external
combustor imposes a required starting horsepower for the engine that is much
greater than is required for conventional starting. Under standard conditions with
JP-4 fuel the conventional 501 KB engine reaches operating speed in under 40 s.
The starter contributes no power after the engine reaches 8022 rpm. Above that
speed the engine is self-sustaining at a turbine inlet temperature of 1450F and
accelerates to 13,146 rpm; all bleed valves then snap shut from compressor
pressure. The fuel is decreased rapidly at that point, dropping the turbine inlet
temperature to 999F. The engine then accelerates to a no-load operating speed of
13,820 rpm. With the wood-fueled system the engine is not self-sustaining at
9300 rpm. It might be possible to partially close the 2-in. bleed valve at the
compressor outlet and achieve sustained speed, but the risk of stalling the
compressor and damaging the blades is too great to allow experimentation. There
was not adequate electrical power at the Roanoke facility to accelerate the engine
to a higher rpm without closing the valve. The Oldsmobile engine is estimated to
be supplying 160 hp; and the 3000-kW generator, running as a motor on 240 V,
is estimated to be supplying 180 hp. At 9300 rpm it is estimated that 340 hp is
wasted by dumping through the 2-in. valve. Therefore, it was decided to sample
the combustion gases for particulate loading at 8000 to 9000 rpm and postpone
complete sampling until after installing the system at Red Boiling Springs, where
adequate power is available.
An advantage to starting with the 3000-kW generator is the degree of control
available. With the unavoidably large volume of hot gases stored in the ducting,
combustor, and cyclone of the wood-fueled system, there is a great risk that the
engine will overspeed if the temperature of the gases exceeds 1000F when the
bleed valves snap shut and the 2-in. valve is closed, even if the wood feed is shut
off. The gas generator model 501 KC engine has a bleed manifold with a
controllable valve that can be fitted to the model 501 KB engine. However, with
the horsepower available with 4160 V, the engine can be accelerated from the
5000 rpm produced by the Oldsmobile engine to 12,000 rpm, where the valves
can all be closed with safety. Therefore, the controllable valve system will be
needed only for starting where distribution grid power is not available, and a
high-horsepower starting engine is used.
5.3
Cyclone Performance
The system was designed and built with two cyclones in series. In early
experiments a negligible amount of ash was collected in the second cyclone.
That led to the consideration of removing it from the system, which also
involved the pressure drop through the system and starting requirements. Each
cyclone in the system imposes a loss equal to approximately 4% of the pressure
generated by the compressor. Eliminating the second cyclone and reducing the
pressure drop would not only increase efficiency but would also improve
matching of the compressor and turbine during start-up, thus reducing the amount
of air to be dumped and the horsepower required for starting. The potential gains
in performance from eliminating the second cyclone were judged to outweigh
any gains from the collection of particulates by that cyclone.
The density and sizes of particles in the hot gas stream entering the turbine
were determined by Battelle-Columbus personnel. The gas turbine speed during
the sampling period varied from 7100 to 8600 rpm, and the wood feed varied
from 800 to 1080 Ib/h with a moisture content of 25%. This is approximately
half the amount needed for idling at the design speed of 14,200 rpm. The air flow
through the combustor at 8600 rpm with all bleed valves open is 25% of that for
design speed. With that air flow rate the wood-feed-to-air ratio approaches that
which will exist at the rated speed of 14,200 rpm and load of 3000 kW. The
sample results showed the total loading to be 60 ppm with no particle larger than
2.6 m. Eighty-six percent of the particles are 0.55 m or less. Allison-GM
engineering sets 5 m as the maximum particle size that can be tolerated by the
turbine.
The 60 ppm loading includes particles of potassium chromate. The potassium
in the wood ash reacts with the chromium in the RA253MA alloy ducting and
cyclones to produce greenish yellow potassium chromate dust particles in the hot
gas stream. In the 385-kW system, the surfaces of the 310 stainless steel reacted
the same way, but formation of the potassium chromate had essentially ceased in
the more than 200 h of operation before the hot gas sampling was performed.
The 3000-kW system had been operated only 48 h before the sampling occurred.
However, in the duct leading from the combustor into the cyclone a piece of eleven
gauge stainless steel, 11 in. x 22 in., believed to be type 304, was mistakenly
used in the fabrication of the ducting and could have contributed heavily to the
production of potassium chromate after the remainder of the surfaces had
become passified. The piece has now been replaced by RA253MA alloy.
Potassium chromate melts at 1775F. At the point in the ducting at which the
sampling took place, the average temperature was 1849F during the sampling
period. During the sampling period, the temperature dropped as low as 1751F
for short periods; at all other times it was above 1830F. The potassium chromate
may have reached that point in the form of an aerosol and condensed in the
sampling stream. The average exhaust gas temperature at the turbine outlet
during the period was 1174F, which is slightly higher than the maximum for a
130F day and 12,400 rpm for the conventional configuration. There were no
provisions for accurately measuring the turbine inlet temperature.
The ambient air temperature varied from 62F at the start to 78F at the finish,
and the ambient pressure was 14.3 psia. After 50 hours of operation there was a
small amount of slag in the bottom of the secondary chamber, which indicated
that the wall temperature had exceeded 2300F. Three major factors determine
the temperature on the walls of the secondary chamber: wood particle size,
moisture content of the particles, and combustor outlet temperature. The more
rapid combustion with smaller particles reduces the amount of char entering the
secondary chamber and, therefore, the wall temperature. Drier particles produce
the same result. With higher combustor outlet temperature there is a
correspondingly higher temperature on the chamber walls. In the effort to
maintain the engine speed of 8000 to 9000 rpm for the sampling run, it was
necessary to maintain the average engine inlet temperature at 1849F. To
produce 3000 kw, the anticipated engine inlet temperature is 1750F. That lower
temperature should drop the wall temperature below the slagging point.
The ash lock hoppers at the bottom of the cyclone showed some caking of fine
particles on the walls. In the work at the Coal Utilization Research Laboratory
(Ref. 4), caking of coal ash occurred on the cyclone walls, an occurrence that has
not been detected with wood. The caking occurs when water condenses on the
walls at start-up. The cyclone is insulated and heats up with the starting gas. The
ash lock hoppers were not insulated and depend on conduction from the bottom
flange of the cyclone tank to heat them during warm-up. While the caked
material flakes off easily upon cooling of the hopper walls, it may prove
necessary to insulate them lightly. The ash particles vary from dust size to
approximately 0.045 in. in average diameter. The ash density is approximately
80 lb/ft3 . Ash is slightly magnetic, indicating the presence of iron oxide Fe3O4.
Wood ash generally contains about 3% iron oxide. There is a tendency for the
particles leaving the cyclone to agglomerate at stagnation points in the ducting
and turbine. The accumulation is soft, and yields to cleaning with milled walnut
hulls. With the wood burning gas turbine, the walnut hulls that go through the
compressor do not reach the turbine. As a result, the compressor and turbine must
7
ACKNOWLEDGEMENT
This work was performed with support from the U.S. Department of Energy
primarily under Contract No. DE-AC0578-ET20058. The DOE contract
managers were Gary F.Schiefelbein and Simon Friedrich.
REFERENCES
1.
2.
3.
4.
Hamrick, J.T. and Hamrick, T.M. Development of wood as an alternative fuel for
large power generating systems part 1: Research on wood burning gas turbines,
Aerospace Research Corporation Final Report DOE/ET/2005872, Sept. 1981.
Hamrick, J.T. and Owen, Nancy H. Gas turbine performance with high moisture
content biomass , Presented at Sixth Symposium on Biotechnology for Fuels and
Chemicals, Gatlinburg, Tenn., May 1984.
Wisdom, J.C. A preliminary examination of a gas turbine cycle with steam
dilution, A.R.L./M.E. Tech. Memo. 188, Nov. 1958.
Roberts, A.G., Barker, S.N., Phillips, R.N., et al. Fluidized bed combustion, NCB
Coal Utilization Research Laboratory Final Report, Department of Energy Report
FE-312115 (a), 1980.
SCALE-UP OF A HIGH-THROUGHPUT
GASIFIER TO PRODUCE MEDIUM-BTU
GAS FROM WOOD
H.F.FELDMANN, M.A.PAISLEY, And H.R.APPELBAUM*
*Battelle Columbus Laboratories
1
THE BATTELLE PROCESS
A schematic flowsheet of the gasification process under development under U.S.
DOEs sponsorship at Battelles Columbus Laboratories is shown in Fig. 1. The
process employs a hot-sand phase as a conveying and heat transfer medium. By
use of such heat transfer methods, it is possible to produce a 500 Btu/SCF fuel
gas from biomass without using oxygen. Wood is fed into the gasifier with no
pretreatment except partial drying to utilize sensible heat present in the flue gas
from the combustor. The wood (or other biomass) is gasified at throughputs up to
2000 lb/h ft of reactor area to produce the fuel gas and a small quantity of char
(typically 20% of the dry wood fed). The char and sand then are removed from
the gas phase and transferred to the combustor, where the char is burned. The
combustion reactions reheat the sand, which returns to the gasifier to provide the
heat for gasification .
Heat recovery from product and flue gases provides both process and export
steam (or other heat sources as might be needed) in addition to drying incoming
wood.
Separation of the gasification and combustion zones allows the following two
major advantages:
Medium-Btu gas can be produced without requiring an oxygen plant, which is
expensive at the scale of most biomass gasification plants.
The heating value of the cooled, cleaned product gas remains constant,
independent of the moisture level of the feed.
2
CURRENT TECHNICAL STATUS
Early testing of the process was conducted in a 250 lb/h process research unit
(PRU) at Battelles Columbus Laboratories (Ref. 1). These tests demonstrated
that the major virtues of this process approach were:
A medium Btu (400 to 500 Btu/ft3) product gas could be produced from wood
without requiring an oxygen plant.
A wide range of feedstocks would be handled without pretreatment.
Constant-heating-value gas can be produced regardless of feed moisture level.
Extremely high throughputs can be achieved, resulting in compact,
economical reactors.
Heat necessary for gasification can be supplied totally from an entrained,
recirculating sand phase.
No by-product char remains after combustion.
Tar production is very low with all feedstocks tested, and since tar is burned
in the combustor no net tar production results.
The success of these initial studies along with favorable economic projections led
to the modification of the PRU to increase the throughput to 12 ton/day by
automating the wood feed system and increasing the gasifier diameter to a 10-in.
internal diameter.
Operation with the larger diameter gasifier in the PRU has been smooth and
trouble-free. R-sults are in agreement with those generated in the smaller gasifier
except that carbon conversion has somewhat increased over that measured in the
smaller gasifier unit.
Total operating time in the PRU has been in excess of 7000 h. Since the
program is funded for single-shift operation and the unit, once heated up, reaches
steady-state conditions in minutes, data gathering runs have ordinarily been
limited to single-shift operating periods. However, one extended around the
clock operation for a period of 4 days was recently completed with a planned
shutdown. This operation demonstrated smooth, stable operation and indicated
that a commercial system should be very suitable for automated control.
3
INDUSTRIAL APPEAL AND APPLICATIONS
The Battelle process is being developed specificall to exploit the special
properties of biomass, namely, its high reactivity and favorable gaseous product
spectrum, to produce a high energy density gas in relatively compact conversion
equipment. In this way it differs from many other biomass conversion processes
that either require oxygen to produce a medium-Btu gas or that produce a lowBtu gas. In addition, unlike single-vessel gasification systems, the heating value
of the gas produced is independent of the moisture level of the feed. This is
illustrated in Fig. 2, where the heating value of the dry product gas at standard
conditions from the Battelle process is compared with that from an air-blown
fluid-bed at varying feed moisture levels.
The high energy density of the product gas allows it to be utilized in place of
natural gas or oil without derating or expensive retrofitting of existing
combustion equipment. Therefore the industrial or utility user of this gas can
maintain natural gas or oil as a backup fuel and use combustion equipment
including gas turbines designed for natural gas. On the other hand, utilization of
low-Btu gas would require extensive derating, retrofitting, and the utilization of
specially designed equipment.
The economics of the process have been discussed in previous publications
(Ref. 1). Briefly, these evaluations indicate that the process can produce mediumBtu gas at a cost competitive with natural gas or oil even in plants as small as
200 ton/day. More important, these economic evaluations indicate that
conversion of biomass to a medium-Btu gas will be cheaper than direct
combustion.
A comparison of cost and performance projections with various direct
combustion-based cogeneration projects indicates a significant advantage for the
Battelle-Columbus gasification process in both investment costs per kilowatt of
capacity and in kilowatts generated per unit of biomass consumed .
4
IDENTIFICATION OF TECHNICAL RISKS
A technical risk is considered to be an unknown that either cannot be determined
from operation of the smaller facility or that is calculated based upon
extrapolation (as opposed to interpolation) of data from the smaller facility.
Technical risks considered important are those that could either affect ultimate
operability of the system or significantly affect the economics of operation.
Obviously, there are risks other than purely technical risks such as estimated
capital costs, interest rates, cost/ availability of raw materials, and many other
issues that can affect the overall economics of a new technology. These
nontechnical risks, though critical, are project specific and will therefore not be
discussed further in this report.
Thermal Performance is defined as the ability of the gasification system to
produce a given yield of product gas of a given composition per unit of dry wood
fed. Thermal performance is, therefore, the single most important factor in
determining the economics of the gasification system. The thermal performance,
upon which existing cost feasibility evaluations were based, was calculated based
on normalizing experimental material balances, published and estimated
thermochemical data, and assumed heat losses. Heat losses in the PRU are
unavoidably high and are compensated for by burning a supplementary fuel
(natural gas) in the fluid-bed combustor. Therefore the thermal performance of
larger, commercial-scale process systems must be based on calculations. Since
the parameters used in the calculations are all subject to error and/or uncertainty,
the thermal performance of a commercial system could deviate from that
predicted by calculations based on PRU data.
System Availability and Reliability are two parameters that are critical in
determining the systems economics. System availability is defined as simply the
time that the system can operate between planned maintenance. Thus, system
Fig. 2. Product gas heating value varies significantly with feed moisture in an air blown
system
LIST OF ATTENDEES
Richard Abbott DAVEY ENVIRONMENTAL SERVICES 1000 Connecticut
Ave., N.W. Washington, D.C. 20036
Betsy Amin-Arsala AGENCY FOR INTERNATIONAL DEVELOPMENT 508
SA-18 Washington, D.C. 20523
David Archer WESTINGHOUSE ELECTRIC CORPORATION 1310 Beulah
Road Pittsburgh, PA 15235
Joseph Asbury ARGONNE NATIONAL LABORATORY 9700 South Cass Ave.
Argonne, IL 60439
Rajai Atalla INSTITUTE OF PAPER CHEMISTRY P.O. Box 1039 Appleton, WI
54912
Allan N.Auclair CANADA FOREST SERVICE OTTAWA-ENVIRONMENTPLACE VINCENT MASSEY Ottawa, Ontario, CANADA K1A 1G5
Don C.Augenstein ELECTRIC POWER RESEARCH INSTITUTE 3412
Hillview Avenue Palo Alto, CA 94303
Suresh P.Babu INSTITUTE OF GAS TECHNOLOGY 3424 South State Street
Chicago, IL 60616
Phillip C.Badger TENNESSEE VALLEY AUTHORITY F4 NFDC Muscle
Shoals, AL 35660
David B.Bancroft COUNCIL OF GREAT LAKES GOVERNORS 122 W.
Washington Ave. Suite 801A Madison, WI 53703
William R.Barclay SERI 1617 Cole Blvd. Golden, CO 80401
John W.Barrier TENNESSEE VALLEY AUTHORITY F4 NFDC Muscle
Shoals, AL 35630
Thomas D.Bath SERI 1617 Cole Boulevard Golden, CO 80401
Don R.Beasley A. T. Kearney Incorporated 699 Prince Street Alexandria, VA
22313
John C.Becker FLOW INDUSTRIES 21414 68th Ave. So. Kent, WA 98032
Steven Beggs ARGONNE NATIONAL LABORATORY 9700 South Cass Ave.
Argonne, IL 60439
John R.Benemann GEORGIA INSTITUTE OF TECHNOLOGY School of
Applied Biology Atlanta, GA 30332
Peter Benson GAS RESEARCH INSTITUTE 8600 W.Bryn Mawr Ave. Chicago,
IL 60631
276
Paul F.Bente, Jr. THE BIO-ENERGY COUNCIL 1815 No. Lynn St., Suite 200
P.O. Box 12807 Arlington, VA 222098807
Beverly Berger DEPARTMENT OF ENERGY Forrestal Bldg. Rm. 5F059
Washington, D.C. 20585
Michael K.Bergman APPA Staff Scientist 2301 M. Street, N.W. Washington,
D.C. 20037 UNIVERSITY/BIOLOGY
Martha D.Berliner VIRGINIA COMMONWEALTH 816 Park Avenue
Richmond, VA 23284)
Kimon Bird GAS RESEARCH INSTITUTE 8600 W. Bryn Mawr Chicago, IL
60631
Paul R.Blankenhorn PENNSYLVANIA STATE UNIVERSITY 310 Forest
Resources Laboratory University Park, PA 16802
Diane Brown GENEX 16020 Industrial Dr. Gaithersburg, MD 20877
Michael D.Brown PACIFIC NORTHWEST LABORATORY P.O. Box 999
Richland, WA 99352
Robert Brown UNIVERSITY OF GEORGIA Chairman: Division of
Agricultural Engineering Athens, GA 30602
Stan Bull SERI 1617 Cole Blvd. Golden, CO 80401
William S.Bulpitt GEORGIA INSTITUTE OF TECHNOLOGY OKeefe Bldg.
Rm. 223 Atlanta, GA 30332
Maria K.Burka National Science Foundation 1800 G. Street, N.W. Washington,
D.C. 20550
Christopher D.Burnett ILLINOIS NATURAL HISTORY SURVEY 607 E.
Peabody Dr. Champaign, IL 61820
James L.Butler USDA SOUTHERN AGRICULTURAL ENERGY CENTER
Coastal Plain Experiment Station Tifton, GA 31793
Robert S.Butner PACIFIC NORTHWEST LABORATORY BATTELLE-N.W.
P.O. Box 999 Richland, WA 99352
Gustavo Calderon INTER-AMERICAN DEVELOPMENT BANK 801 17th St.
N.W. Washington, D.C. 20577
Melvin Calvin UNIVERSITY OF CALIFORNIA Dept. of Chemistry Berkeley,
CA 94708
Susan E.Campbell USAID Washington, D.C. 20523
Craig L.Chase BIOMASS ENERGY SYSTEMS CONSULTANT 12840 S.E.
10th Ct., Suite A4 Bellevue, WA 98006
277
278
279
Ellen G.Carver Univ. of Minnesota Bio-Energy Office 1445 Gortner Avenue St.
Paul, MN 55108
Earle E.Gavett DIRECTOR OFFICE OF ENERGY U.S. DEPARTMENT OF
AGRICULTURE Washington, D.C. 20250
Mark A.Gerber PACIFIC NORTHWEST LABORATORY P.O. Box 999
Richland, WA 99352
Wayne A.Geyer KANSAS STATE UNIVERSITY Forestry Department Call
Hall Manhattan, KS 66506
Mohinder Gill USDA ERS NRED 500 12th Street S.W. Washington, D.C. 20250
Paul Gilman Senate Energy Committee U.S. Senate Washington, D.C. 20510
Barbara H.Glenn SERI 1617 Cole Blvd. Golden, CO 80401
Alberto Goetzl NATIONAL FOREST PRODUCTS ASSOCIATION 1619
Massachusetts Ave. N.W. Washington, D.C. 20036
Barry Goldstein NEW MEXICO SOLAR ENERGY INSTITUTE P.O. Box 3
SOL Las Cruces, NM 88003
Stephen C.Grado PENNSYLVANIA STATE UNIVERSITY 206 Ferguson
Building University Park, PA 16802
Karel Grohmann SERI 1617 Cole Boulevard Golden, CO 80401
Sigmund Gronich DOE Office of Alcohol Fuels Forrestal Bldg. Washington,
D.C. 20585
Suzanne Grundstrom SERI 1617 Cole Blvd. Golden, CO 80401
Marty Gutstein BIOMASS ENERGY TECHNOLOGY DIV. DOE Forrestal
Bldg. 1000 Independence Ave. S.W. Washington, D.C. 20585
David O.Hall KINGS COLLEGE 68 Half Moon Lane London, ENGLAND
Joseph T.Hamrich AEROSPACE RESEARCH CORP. 5454 Jae Valley Rd.
Roanoke, VA 24014
Robert E.Hansen LEGISLATIVE COMMISSION ON MN RESOURCES Room
B46 State Capitol Bldg. St. Paul, MN 55155
Dr. Harold Hanson House Committee on Science 4 Technology 2321 Rayburn
House Office Building Washington, D.C. 20515
Reid Hartsell DOE Forrestal Bldg. Rm. 5-F-044 1000 Independence Ave. S.W.
Washington, D.C. 20585
Sadiq Hasnain NATIONAL RESEARCH COUNCIL DIVISION OF ENERGY
Ottawa, CANADA K1A OR6
280
281
282
283
284
285
286
287
John M.Veigel AEC Pamlico Bldg., Suite 212 P.O. Box 12699 Research
Triangle Park, NC 27709
Robert Vining NEW ENGLAND CONGRESSIONAL CAUCUS 53 D Street,
S.E. Washington, D.C. 20003
George Voss SOLID FUELS, INCORPORATED 4365 Lawn Avenue Western
Springs, IL 60558
Carl J.Wallace ARGONNE NATIONAL LABORATORY 1331 H. St. N.W.
Washington, D.C.
Edward I Wan SCIENCE APPLICATIONS INC. 1710 Goodridge Dr. McLean,
VA 22101
Morris Wayman UNIVERSITY OF TORONTO Toronto, CANADA M5S 1A4
Lloyd Weaver ACCUBURN Box 151A River Road Topsham, ME 04086
Rhonda H.Weaver ACCUBURN Box 151A River Road Topsham, ME 04086
Barry E.Welch TIME ENERGY SYSTEMS INC. 12944 Traviah Rd. Patomic,
MD 20854
Donald H.White UNIVERSITY OF ARIZONA Dept. of Chemical Engr.
Tucson, AZ 85721
Irvin L.White NYSERDA Two Rockefeller Plaza Albany, NY 12223
David Wilson COALITION OF NORTHEASTERN GOVERNORS Washington,
D.C. 20003
Walter H.Winnard BATTELLE MEMORIAL INSTITUTE 2030 M Street N.W.
Washington, D.C. 20036
Frederick E.Wood MERIDIAN CORPORATION 5113 Leesburg Pike, Suite 700
Falls Church, VA 22041
Richard C.Wright ENERGY SYSTEMS DIV. AQUA-CHEM P.0. Box 421
Milwaukee, WI 53201
Charles Wyman 1617 Cole Blvd. Golden, CO 80401
Beverly Yocum DOE Forrestal Bldg. Rm. 5F-059 1000 Independence Ave., S.W.
Washington, D.C. 20585
Jackson Yu ULTRASYSTEMS INCORPORATED 22 Second Street, Third
Floor San Francisco, CA 94105
John I.Zerbe FOREST PRODUCTS LABORATORY P.0. Box 5130 Madison,
WI 53705
288