MEMORANDUM

Date:		November 17, 2010

Subject:	Assessment of Advanced Biofuel Technologies

From:		Lester Wyborny

		Chemical Engineer

To:		EPA Air Docket Number EPA-HQ-OAR-2010-0133

	1.	Background Information of Cellulosic Biofuel Production

	The biochemical conversion of cellulosic biomass to fuels requires a
detailed understanding of the chemistry of the cellulosic biomass. 
Unlike grain feedstocks where the major carbohydrate is starch,
lignocellulosic biomass is composed mainly of cellulose (40-60 %) and
hemicellulose (20-40 %).  The remainder consists of lignin, a complex
polymer which serves as a stiffening and hydrophobic (water-repelling)
agent in cell walls.  Cellulose and hemicellulose are made up of sugars
linked together in long chains called polysaccharides.  Once hydrolyzed,
they can be fermented into ethanol or other biofuels.  Currently, lignin
cannot be fermented into ethanol, but could be burned as a by-product to
generate electricity or converted by the other technologies into
biofuels as discussed in this document.    

	Both starch (e.g. corn grain) and cellulosic feedstocks must be
hydrolyzed prior to fermentation.  Structural differences at the
molecular level make it far more difficult, and therefore more costly,
to hydrolyze cellulosic biomass than it is to hydrolyze starch. 
Glucose, C6H12O6, the repeating monomer in both starch and cellulose, is
a six-sided ring, similar in conformation to the classic ‘chair’
conformation of cyclohexane or benzene, except one carbon atom in the
ring is replaced by an oxygen atom.  For uniformity (and ease) of
discussion, it is generally assumed that the first carbon atom next to
the oxygen, is carbon #1; the numbering, 2-5, continues around the ring
with oxygen in the 6th position; one of the four bonds of the fifth
carbon atom is attached to the oxygen atom to complete the ring, one is
attached to hydrogen atom and the fourth to a -CH2OH group.  Thus, a
glucose molecule/monomer is a six-sided molecule, but not a six-carbon
ring (although there are six-carbon molecules present, one of which is
in the –methylhydroxy group).  

e made up of β-glucose monomers, strung together through β-linked
1,4-glucosidic bonds.  In starch with the (-conformation, the hydroxyl
group on carbon #1 is in the axial or (-position, which causes the
-OH’s on each successive glucose monomer to end up on the same side of
the polymer.  There are also 1,6-linked glucose branches that occur
irregularly on approximately one in twenty-five glucose units.   The -OH
groups on the same side of the polymer, along with the randomly attached
1,6-glucose branches, leaves starch polymers relatively weak, flexible,
and able to easily wrap and twist together to form tiny granules ( e.g.,
common, everyday corn starch).

oluble in water and resistant to chemical attack.  The β-conformation
and the resulting hydrogen bonds stabilize the glucose chair structure
to help minimize the polymer’s flexibility (which hinders hydrolysis)
and to add to its strength.  

	The second cellulosic component is called hemicellulose.  It consists
mainly of a random mixture of highly branched and heavily substituted
five- and six-carbon rings.  The five-carbon residues are usually
D-xylose and L-arabinose; the six-carbon residues are usually
D-galactose, D-glucose, and D-mannose, and uronic and acetic acid. 
Hemicellulose is not as rigid or strong as cellulose, but does
contribute additional strength and helps protect the plant cell wall
against attack by microbes or water.  Hemicellulose is relatively easy
to hydrolyze, due to its highly branched, somewhat random or non-uniform
structure. 

	Lignin, the third principle component, is a complex, cross-linked
polymeric, high molecular weight substance derived principally from
coniferyl alcohol by extensive condensation polymerization.  Covalently
bonded to the hemicellulose, it is essentially a glue-like polymer that
covers the cellulose and hemicellulose polymer cell walls and helps hold
them together, provides additional strength, helps resist microbial
decay, and perhaps most importantly, for this discussion, inhibits
hydrolysis.  Its molecular weight is around 10,000.  While both
cellulose and hemicellulose contribute to the amount of fermentable
sugars for ethanol production, lignin does not, but can be combusted to
provide process energy in a biochemical plant or used as feedstock to a
thermochemical process.  

	A significant part of the reason it is more difficult and more costly
to produce ethanol from cellulosic feedstocks, has to do with the
differences in the molecular structures of simple starches and oils and
those of cellulosic plant matter.  That is, as a plant grows, glucose
monomers are added to the polysaccharide chains of the plant cell walls
through condensation reactions.  In general, condensation is a chemical
process by which two molecules are joined together to make a larger,
more complex molecule, and a molecule of water is a byproduct of the
reaction.  In the formation of polysaccharides, and enzyme catalyzes the
reaction wherein the -OH group on carbon #1 of one monomer, or glucose
residue, reacts with the -OH on carbon #4 or #6 of another residue.  An
H-OH (H2O or water) molecule is removed leaving an -O- that links the
monomers together to form the polysaccharide chain.  Again, depending on
the direction of the –OH group at carbon 1, it may be called an alpha
(as in starch) or a beta (as in cellulose) linkage.  

 	Hydrolysis is the reverse reaction.  The -H from an H-OH (water)
molecule is added to one monomer and the remaining -OH is added to its
pair, e.g., to the next monomer on the chain, to regenerate separate
glucose monomers.  During starch hydrolysis, water and water borne
hydrolyzing enzymes can easily penetrate the randomly formed polymers
(the tiny granular particles or bundles) in order to break the bonds to
release glucose monomers.  However, the cellulosic or glucan polymers
formed in tightly packed, dense, rigid microfibrils are especially
resistant to water and hydrolyzing enzymes.  Xylan, the main constituent
of hemicellulose, is more easily hydrolyzed than cellulose, but not
easily fermented.  Cellulose is not easily hydrolyzed, but readily
ferments into alcohol.  These are two of the major problems that must be
satisfactorily resolved for biochemical conversion of cellulosic
feedstocks.

While the biochemical process primarily uses enzymes or acids to
disassemble the cellulose and hemicellulose followed by fermentation,
the thermochemical process operates substantially differently.  In the
thermochemical process, biomass is gasified at temperatures that are
above 850°C.  The gasification process partially oxidizes all biomass
carbonaceous feed stock into gas, which is high in energy and consists
of CO, CO2, CH4, H2 and H2O, along with other minor constituents.  The
CO and hydrogen are the main building block components for making liquid
fuels via conversion through downstream synthesis reactions.  

Because of the thermal oxidation that occurs, all of the cellulosic
biomass feedstock, including the lignin component, can be used to make
renewable fuels.  Due to its ability to convert the lignin into fuels,
the thermochemical process may be more feedstock flexible than the
biochemical process.  In general, the lignin fraction content in woody
feedstocks is higher than fractions found in herbaceous biomass (e.g.,
grasses, straw, corn stalks).  Therefore, thermochemical technologies
may be more compatible than biochemical cellulosic biofuel technologies
for processing wood-based feedstocks.  Wood products generally have
higher heating values than other biomass stocks, such as stover, grasses
etc which also makes them better feedstocks for a thermochemical unit. 
Higher energy cellulosic feeds yield more CO and hydrogen synthesis gas,
resulting in higher liquid yields per ton of feedstock.

	2.	Emerging Technologies

	When evaluating the array of biofuel technologies which could produce
one or more fuels that could qualify under RFS2, we found that it is
helpful to organize them into fuel technology categories.  Organizing
them into categories eases the task of understanding the costs and life
cycle impacts of these technologies because like technologies likely
have similar cost and life cycle impacts.  The simplest organization is
by the fuel produced.  However, we frequently found that additional
subdivisions were also helpful.  Table 2.1 provides a list of
technologies, the cellulosic fuels produced and a list of many of the
companies which we learned are pursuing the technology (or something
very similar to the technology listed in the category).

Table 2.1

List of Biofuel Categories, the Fuels Produced and the Companies
Pursuing Cellulosic Biofuel Technologies

Technology Category	Technology	Fuels Produced 	Companies

Biochemical	Enzymatic Hydrolysis	Ethanol	Abengoa, AE Fuels,
DuPont/Danisco, Florida Crystals, Gevo, Poet, ICM, Iogen, BPI, Energy, 
Fiberight, KL Energy.

	Acid Hydrolysis	Ethanol	Agresti, Arkenol, Blue Fire, Pencor, Pangen,
Raven Biofuels

	Dilute Acid, Steam Explosion of Cellulose	Ethanol	Verenium, BP, Central
Minnesota Ethanol Coop.

	Consolidated Bioprocessing (one step hydrolysis and fermentation) of
Cellulose	Ethanol	Mascoma, Qteros

	Conversion of Cellulose via carboxylic acid	Ethanol, Gasoline, Jet
Fuel, Diesel Fuel	Terrabon, Swift Fuels

	One step Conversion of Cellulose to distillate	Diesel, Jet Fuel or
Naphtha	Bell Bioenergy, LS9

Thermochemical 	Thermochemical/Fischer Tropsch	Diesel Fuel and Naphtha
Choren, Flambeau River Biofuels, Baard, Clearfuels, Gulf Coast Energy,
Rentech, TRI.

	Thermochemical/Fischer Tropsch	DME	Chemrec, New Page.

	Thermochemical/Catalytic conversion of syngas to alcohols	Ethanol	Range
Fuels, Pearson Technologies, Fulcrum Bioenergy, Enerkem, and Gulf Coast
Energy.

Hybrid	Thermochemical w/ Biochemical catalyst	Ethanol	Coskata, INEOS Bio

	Acid Hydrolysis of cellulose to intermediate; hydrogenation using
Thermochemical syngas from non-cellulose fraction	Ethanol, Other
alcohols	Zeachem

Depolymerization	Catalytic Depolymerization of Cellulose	Diesel, Jet
Fuel or Naphtha	Cello Energy, Covanta, Kior, University of Massachusetts

	Pyrolysis of Cellulose	Diesel, Jet Fuel, or Gasoline	Envergent
(UOP/Ensyn), Dynamotive, Petrobras

Other	Catalytic Reforming of Sugars from Cellulose	Gasoline	Virent



	Of the technologies listed above, many of them are considered to be
“second generation” biofuels or new biofuel technologies capable of
meeting either the advanced biofuel or cellulosic biofuel RFS standard. 
The following sections describe specific companies and the new biofuel
technologies which the companies have developed or are developing.  This
summary is not meant to be an unabridged list of new biofuel
technologies, but rather a description of some of the more prominent or
interesting of the new biofuel technologies that serve to provide a
sense of the technology categories listed above.  The process technology
summaries are based on information provided by the respective companies.
 EPA has not been able to confirm all of the information, statements,
process conditions, and the process flow steps necessary for any of
these processes and companies.  

	3.	Biochemical Process for the Production of Cellulosic Biofuels

	Biochemical conversion refers to a broad grouping of processes that use
biological organisms or enzymes from such organisms to convert
cellulosic feedstocks into biofuels.  While no two processes are
identical, many of these processes follow a similar basic pathway to
convert cellulosic materials to biofuel.  The general process of most
biochemical cellulosic biofuel processes consists of five main steps,
feedstock handling, pretreatment, hydrolysis, fermentation/fuel
conversion, and distillation/separation.  The feedstock handling step
reduces the particle size of the incoming feedstock and removes any
contaminants that may negatively impact the rest of the process.  In the
pretreatment step the structure of the lignin and hemicellulose is
disrupted, usually using some combination of heat, pressure, acid, or
base, to allow for a more effective hydrolysis of the cellulosic
material to simple sugars.  In the hydrolysis stage the cellulose and
any remaining hemicellulose is converted into simple sugars, usually
using an enzyme or strong acid.  In the fermentation or fuel conversion
step, the simple sugars are converted to the desired fuel by a
biological organism.  In the final step the fuel that is produced is
separated from the water and other byproducts by distillation or some
other means.  A basic diagram of the biochemical conversion process can
be found in Figure 3.1.

Figure 3.1

	While this diagram shows the production of ethanol from cellulosic
biomass, it is possible to use the same process to produce other fuels
or specialty chemicals using different biological organisms.  The
following sections will discuss each of these steps in greater detail,
discuss some of the variations to this general process, and discuss some
of the advantages and disadvantages of the biochemical process of
producing biofuel from cellulosic materials as compared to other fuel
production processes.

	Three of the five companies that EPA believes may produce cellulosic
biofuel in 2011 plan to use a biochemical process to produce biofuels. 
All three of these companies, Dupont Danisco Cellulosic Ethanol,
Fiberight, and KL energy, plan to use an enzymatic hydrolysis, while a
variety of other companies are exploring other production methods.  The
main reason for the dominance of biochemical technologies in 2011 is the
relatively low capital costs of these projects compared to other
cellulosic biofuel facilities.  Biochemical projects also benefit less
from economies of scale, making smaller and less capital intensive
commercial facilities more feasible.  The following sections provide
more information on the biochemical processes being pursued by the
majority of the companies we expect to produce cellulosic biofuels and
make them commercially available in 2011, as well as many other
companies planning to begin production in later years.

	3.1	Feedstock Handling

	The first step of the biochemical conversion process is to insure that
the biomass stream can be utilized by the rest of the conversion
process.  This most often takes the form of size reduction, either by
grinding or chipping as appropriate for the type of biomass.  While this
is a relatively simple process it is essential to allow the following
steps of the process to function as designed.  It is also a potentially
energy intensive process.  It may be possible for biofuel producers to
purchase cellulosic material that is already of the appropriate size,
however we believe that in the near term this is unlikely and most
biofuel producers will have to invest in equipment to reduce the size of
the material they receive as needed for their process.  In coming years,
as the market for cellulosic materials expands, purchasing feedstock
that has already been ground or chipped may be possible and cost
effective, as these processes increase the density of this material and
may reduce transportation costs.

	In addition to size reduction, steps must also be taken to remove any
material from the feedstock that might be detrimental to the fuel
production process.  At best, contaminants in the feedstock would travel
through the fuel production process unchanged, perhaps resulting in a
slight reduction in fuel production volume.  Depending on the type of
contaminant they may also be converted to undesired byproducts that must
be separated from the fuel.  At worst, these contaminants could be toxic
to the biological organisms being used to convert the sugars to fuel,
necessitating a shut down and restart of the plant.  Either of these
scenarios would result in a significant cost to the fuel producer. 
Feedstocks such as agricultural residues, wood chips, or herbaceous or
woody energy crops are likely to contain far fewer contaminants than
more heterogeneous feedstocks such as municipal solid waste (MSW).

	3.2	Biomass Pretreatment

	The purpose of the biomass pretreatment stage is to disrupt the
structure of the cellulosic biomass to allow for the hydrolysis of the
cellulose and hemicellulose into simple sugars.  The ideal pretreatment
stage would allow for a high conversion of the cellulose and
hemicellulose to simple sugars, minimize the degradation of these sugars
to undesired forms that reduce fuel yields and inhibit fermentation, not
require especially large or expensive reaction vessels, and be a
relatively robust and simple process.  No single biomass pretreatment
method currently exists that meets all of these goals, but rather a
variety of options are being used by various cellulosic fuel producers,
each with their own strengths and weaknesses.  Dilute acid pretreatment
and alkaline pretreatment are two methods currently being used that
attack the hemicellulose and lignin portions of the cellulosic biomass
respectively.  Other methods, such as steam explosion and ammonia fiber
expansion, seek to use high temperature and pressure, followed by rapid
decompression to disrupt the structure of the cellulosic biomass and
allow for a more efficient hydrolysis of the cellulose and hemicellulose
to simple sugars.  The cost and characteristics of the cellulosic
feedstock being processed is likely to have a significant impact on the
pretreatment process that is used.  

	3.2.1	Dilute Acid Pretreatment

	The dilute acid pretreatment uses a dilute acid solution, generally
~1%, sulfuric, hydrochloric, or nitric acid, at a high temperature,
usually greater than 160° C to hydrolyze the hemicellulose to simple
sugars.  The dilute acid pretreatment produces very high yields of
xylose, the most important sugar in the hemicellulose, and requires very
short reaction times, often as low as several minutes.  It also allows
for high conversion rates of the cellulose to glucose in the later
stages of the process.  Using a dilute acid pretreatment, however, is a
relatively expensive method.  Corrosive resistant materials must be used
for the reaction vessels due to the presence of the acid.  The acid must
also be neutralized prior to the cellulose hydrolysis stage if enzymes
are to be used, and before the fuel production stage as it is harmful to
the biological organisms used for the fuel production.  Lime can be used
to neutralize the acid, however this produces gypsum that must be
removed from the process.  Other compounds, such as ammonia, can be used
that do not produce solids, but the higher costs for these compounds
generally negates any savings from the lack of equipment required to
separate the gypsum.

	3.2.2	Alkaline Pretreatment

	The alkaline pretreatment process is designed to dissolve and disrupt
the structure of the lignin portion of the lingocellulosic materials
rather than the hemicellulose.  This method exposes the feedstock to an
alkaline solution, usually a solution of 1-2% sodium hydroxide, calcium
hydroxide, or ammonia.  The reaction time is approximately one hour,
however this can vary depending on the feedstock being used.  High
temperatures are not required for this process, however they are
sometimes used to decrease the reaction time or increase the sugar
conversion of the cellulose and hemicellulose in later stages of the
process.  If an alkaline pretreatment is used a different combination of
enzymes are required for the subsequent hydrolysis step, as now the
hemicellulose must be hydrolyzed in addition to the cellulose.  Alkaline
pretreatment has lower capital costs than dilute acid pretreatment as
cheaper materials are able to be used for the reaction vessels.  The
alkaline solution can also be recovered and recycled rather than
neutralized which decreases the cost of chemical usage and waste water
treatment.  The main disadvantage of the alkaline pretreatment process
is that yields of the sugars from the cellulose and hemicellulose tend
to be lower than those for other processes.  This must be weighed
against the lower capital costs when considering using an alkaline
pretreatment process.  DuPont Danisco Cellulosic Ethanol is one company
planning on using an alkaline pretreatment in the production of
cellulosic biofuels.

	3.2.3	Steam Explosion

	In the steam explosion pretreatment process the lingocellulosic
feedstock is treated with high temperature and pressure saturated steam.
 The feedstock is held at high temperature and pressure, up to 50 bar
and 290° C for a period of time ranging from several seconds to ten
minutes, and then undergoes a sudden decompression to atmospheric
pressure.  This decompression causes a disruption of the lignin and
hemicellulose structure allowing for hydrolysis of the remaining
hemicellulose and cellulose in later steps.  This process is
advantageous in that it uses no acids or bases for the pretreatment
resulting in decreased costs and environmental impacts.  In the rapid
decompression, however, compounds can be formed that can inhibit the
later hydrolysis and fuel production processes.  The pretreated biomass
will likely have to be washed to remove these compounds.  This washing
can also result in the loss of soluble hemicellulose and ultimately
simple sugars and fuel produced.  Some steam explosion processes add
small amounts of acid with the saturated steam to increase the
hydrolysis of hemicellulose and cellulose in later steps and to decrease
the production of undesired compounds that inhibit simple sugar and fuel
production.  The addition of acid, however, increases the capital costs
and adds acid neutralization costs as discussed in section 3.2.1 above. 
Companies using steam explosion to produce cellulosic biofuels include
IOGEN and Verinium.

	3.2.4	Ammonia Fiber Explosion

	Ammonia Fiber Explosion (AFEX) operates much like steam explosion. 
Lignocellulosic feedstock is treated with liquid ammonia at a high
temperature (~90° C) and pressure.  These conditions are held for much
longer than with steam explosion, approximately 30 minutes, before being
rapidly decompressed to atmospheric pressure.  As with the steam
explosion process where acid has been added, compounds are not produced
that would interfere with later hydrolysis and fuel production steps ,
so a wash step to remove these compounds is not necessary.  In order for
the AFEX pretreatment process to be economically feasible the ammonia
must be recovered and reused.  This process produces a treated feedstock
with high rates of simple sugar production in the hydrolysis step as
long as the lignocellulosic feedstocks have relatively low lignin
content.  The AFEX process is effective at treating herbaceous energy
crops and agricultural residues such as corn stover and wheat and barley
straw.  AFEX is less effective as a pretreatment process for feedstocks
with high lignin content such as wood chips or newspaper.  If the AFEX
pretreatment is used for feedstocks with high lignin content the result
will be low sugar production in the hydrolysis stage.

	3.3	Hydrolysis

	In the hydrolysis step the cellulose and any remaining hemicellulose
are converted to simple sugars.  There are two main methods of
hydrolysis, acid hydrolysis and enzymatic hydrolysis.  Acid hydrolysis
is the oldest technology for the conversion of cellulosic feedstock to
ethanol and can only be used following an acid pretreatment process.  An
alternative method is to use a mixture of different enzymes to perform
the hydrolysis after the biomass has been pretreated.  This process is
potentially more effective at hydrolyzing pretreated biomass but in the
past has not been economically feasible due to the prohibitively high
cost of the enzymes.  The falling cost of these enzymes in recent years
has made the production of cellulosic biofuels using enzymatic
hydrolysis possible.  The lignin is largely unaffected by the hydrolysis
and fuel production steps but is carried through these processes until
it is separated out in the fuel separation step and burned for process
energy or sold as a co-product.

	3.3.1	Acid Hydrolysis

	Acid hydrolysis is a technique that has been used for over 100 years to
convert cellulosic feedstocks into fuels.  In the acid hydrolysis
process the lignin and cellulose portions of the feedstock that remain
after the hemicellulose has been dissolved, hydrolyzed, and separated
during the dilute acid pretreatment process is treated with a second
acid stream.  This second acid treatment uses a less concentrated acid
than the pretreatment stage but at a higher temperature, as high as
215° C.  This treatment hydrolyzes the cellulose into glucose and other
6 carbon sugars that are then fed to biological organisms to produce the
desired fuel.  It is necessary to hydrolyze the hemicellulose and
cellulose in two separate steps to prevent the conversion of the pentose
sugars that result from the hydrolysis of the hemicellulose from being
further converted into furfural and other chemicals.  This would not
only reduce the total production of sugars from the cellulosic
feedstock, but also inhibit the production of fuel from the sugars in
later stages of the process.  

	The acidic solution containing the sugars produced as a result of the
hydrolysis reaction must also be treated so that this stream can be fed
to the biological organisms that will convert these sugars into fuel. 
In order to operate an acid hydrolysis process cost effectively the acid
must be recovered, not simply neutralized.  Methods currently being used
to recover this acid include membrane separation and continuous ion
exchange.  The advantages of using an acid hydrolysis are that this
process is well understood and capable of producing high sugar yields
from a wide variety of feedstocks.  Capital costs are high however, as
materials compatible with the acidic streams must be extensively
utilized.  The high temperatures necessary for acid hydrolysis also
result in considerable energy costs, and profitability is highly
dependent on the ability to effectively recover and reuse the acid.

	3.3.2	Enzymatic Hydrolysis

	The enzymatic hydrolysis process uses enzymes, rather than acids, to
hydrolyze the cellulose and any remaining hemicellulose from the
pretreatment process.  This process is much more versatile than the acid
hydrolysis and can be used in combination with any of the pretreatment
processes described above, provided that the structure of the
lignocellulosic feedstock has been disrupted enough to allow the enzymes
to easily access the hemicellulose and cellulose.  After the feedstock
has gone through pretreatment a cocktail of cellulose enzymes is added. 
These enzymes can be produced by the cellulosic biofuel producer or
purchased from enzyme producers such as Novozymes, Genencor, and others.
 The exact mixture of enzymes used in the enzymatic hydrolysis stage can
vary greatly depending on which of the pretreatment stages is used as
well as the composition of the feedstock.

	The main advantages of the enzymatic hydrolysis process are a result of
the mild operating conditions.  Because acid is not used, special
materials are not required for the reaction vessels.  Enzymatic
hydrolysis is carried out at relatively low temperatures, usually around
50° C, and atmospheric pressure, and therefore has low energy
requirements and incurs lower capital costs.  These conditions also
result in less undesired reactions that would reduce the production of
sugars and potentially inhibit fuel production.  Enzymatic hydrolysis
works best with a uniform feedstock, such as agricultural residues or
energy crops, where the concentration and combination of enzymes can be
optimized for maximum sugar production.  If the composition of the
feedstock varies daily, as can be the case with fuel producers utilizing
MSW or other waste streams, or even seasonally, it would make it more
difficult to ensure that the correct enzyme cocktail is being used to
carry out the hydrolysis as efficiently as possible.  The main hurdle to
using an enzymatic hydrolysis has been, and continues to be, the costs
of the enzymes.  Recent advances by companies that produce enzymes for
the hydrolysis of cellulosic materials have resulted in a much lower
purchase price for these enzymes.  If, as many researchers and
cellulosic biofuel producers expect, the cost of these enzymes continues
to fall, it is likely that enzymatic hydrolysis will be a lower cost
option than acid hydrolysis, especially for cellulosic biofuel producers
utilizing uniform feedstocks.

	3.4	Fuel Production

	After the cellulosic biomass has been hydrolyzed to simple sugars this
sugar solution is converted to fuel by biological organisms.  In some
biochemical fuel production processes the sugars (principally five
carbon) produced from the fermentation of the hemicelluloses are
converted to fuel in a separate reactor and with a different set of
organisms than the sugars (principally six carbon) produced from the
hydrolysis of cellulose.  Others processes, however, produce fuel from
the five and six carbon sugars in the same reaction vessel.

	A wide range of biological organisms can be used to convert the simple
sugars into fuel.  These include yeasts, bacteria, and other microbes,
some of which are naturally occurring and others that have been
genetically modified.  The ideal biological organism converts both five
and six carbon sugars to fuel with a high efficiency, is able to
tolerate a range of conditions, and is adaptable to process sugar
streams of varying compositions that may result from variations in
feedstock.  Many cellulosic biofuel producers have their own proprietary
organism or organisms optimized to produce the desired fuel from their
unique combination of feedstock, pretreatment and hydrolysis processes,
and fuel conversion conditions.  Other cellulosic fuel producers license
these organisms from biotechnology companies who specialize in their
discovery and production.  

	The many different biological organisms being considered for cellulosic
biofuel production are capable of producing many different types of
fuels.  Many cellulosic biofuel producers are working with organisms
that produce ethanol.  In many ways this is the most simple fuel to
produce from lignocellulosic biomass as the production of ethanol from
simple sugars is a well understood process.  Others intend to produce
butanol or other alcohols that have higher energy content.  Butanol may
be able to be blended into gasoline in greater proportion to ethanol and
therefore has a potentially greater market as well as higher value due
to its higher energy content.  Yields for butanol, however, are
currently significantly lower per ton of feedstock than ethanol.  Some
of the fuel producers who plan to produce alcohols are considering
purchasing and modifying already existing grain ethanol plants.  This
would potentially have significant capital cost savings as many of the
units used in a grain ethanol process are very similar to those required
by the biochemical fuel production process and could be used with
minimal modification.

	Other cellulosic biofuel producers intend to produce hydrocarbon fuels
very similar to gasoline, diesel, and jet fuel.  These fuels command a
higher price than alcohols, have a greater energy density, and are
potentially drop-in fuels that could be used in conventional vehicles
without strict blending limits.  They could also be transported by
existing pipelines and utilize the same infrastructure as the petroleum
industry.  Some of the processes being researched by fuel producers
produce a single compound, such as iso-octane, that would need to be
blended into petroleum gasoline in order to be used while others produce
a range of hydrocarbons very similar to those found in gasoline or
diesel fuel refined from petroleum and could potentially be used in
conventional vehicles without blending.  While the prospect of producing
hydrocarbon fuels from cellulosic feedstock is promising, the current
yields of fuel produced by these organisms are significantly lower than
those that are producing ethanol and other alcohols.  Improvement in the
yields of these organisms will have to be realized in order for
cellulosic hydrocarbon fuels produced via a biochemical process to
compete with cellulosic ethanol, and ultimately petroleum based fuels.

	3.5	Fuel Separation

	In the fuel separation stage the fuel produced is separated from the
water, lignin, any un-reacted hemicellulose and cellulose, and any other
compounds remaining after the fuel production stage.  The complexity of
this stage is highly dependent on the type of fuel produced.  For
processes producing hydrocarbon fuels this stage can be as simple as a
settling tank, where the hydrocarbons are allowed to float to the top
and removed.  Recovering the ethanol is a much more difficult task.  To
recover the ethanol a distillation process, nearly identical to that
used in the grain ethanol industry, is used.  The ethanol solution is
first separated from the solids before being sent to a distillation
column called a beer column.  The overheads of the beer column are fed
to a second distillation column, called a rectifier for further
separation.  The rectifier produces a stream with an ethanol
concentration of approximately 96%.  A molecular sieve unit is then used
to dehydrate this stream to produce fuel grade ethanol with purity
greater than 99.5%.  The distillation of ethanol is a very energy
intensive process and new technologies, such as membrane separation, are
being developed that could potentially reduce the energy intensity, and
thus the cost, of the ethanol dehydration process.  The neat ethanol
which is produced is blended with natural gasoline at a ratio of 2 %
natural gasoline to 98 % ethanol to denature the ethanol before it is
stored.   The solids which were separated from the ethanol solution,
which contains the unused lignin, are dried and either burned on site to
provide process heat and electricity or sold as a byproduct of the fuel
production process.  The waste water is either recycled or sent to a
water treatment facility.

	3.6	Process Variations

	While the process described above outlines the general biochemical
process used by many cellulosic biofuel producers, there are several
prominent variations being pursued by prospective biofuel producers. 
These variations usually seek to simplify the biochemical fuel
production process by combining several steps into a single step or
using other means to reduce the capital or operating costs of the
process.  Simultaneous Saccharification and Fermentation (SSF),
Simultaneous Saccharification and Co-Fermentation (SSCF), Consolidated
Bio-Processing (CBP), and Single Step Fuel Production are all production
methods being developed by various biofuel production companies to
combine two or more of the steps outlined above.  These modifications
are usually enabled by a proprietary technology or biological organism
that makes these changes possible.

	3.6.1	Simultaneous Saccharification and Fermentation (SSF)

	In simultaneous saccharification and fermentation (SSF), the cellulose
hydrolysis and fuel conversion steps are combined.  After the
lingocellulosic biomass has been pretreated cellulose enzymes and fuel
conversion organisms, usually yeast, are added simultaneously.  As the
cellulase enzymes convert the cellulose to simple sugars the yeast
immediately converts the sugars to ethanol.  There are several benefits
of the SSF process.  The first is the potential lowering of capital
costs because only a single reaction vessel is required for both the
hydrolysis of the cellulose and the fuel production from the resulting
sugars.  In addition, the products of the cellulose hydrolysis, the
glucose and other simple sugars, are inhibitors to the enzymes
performing the hydrolysis.  By converting them to ethanol as they are
produced the sugar concentration is kept low and the conversion rate of
the cellulose into simple sugars is maximized.  There are several
drawbacks to the SSF process.  The rate of ethanol production is now
limited by the enzyme activity.  The ideal conditions for cellulose
hydrolysis and the fermentation of the resulting sugars are not the
same.  This can result in slower overall reaction rates, and the fuel
yields using this technology are not currently as high as when these two
steps are performed separately.

	3.6.2	Simultaneous Saccharification and Co-Fermentation (SSCF)

	Simultaneous saccharification and co-fermentation (SSCF) is very
similar to SSF.  This process, however, combines the hydrolysis of the
cellulose and the fermentation of the resulting six carbon sugars with
the fermentation of the five carbon sugars produced form the hydrolysis
of the hemicellulose.  The advantages and disadvantages of this process
are also very similar to SSF, with further reductions in capital cost
possible.  Conditions also move further away from the optimal conditions
for each of the three reactions and slower reaction rates and lower
yields can result.

	3.6.3	Consolidated Bio-Processing (CBP)

	Consolidated Bio-Processing (CBP) further increases the simplicity of
the process from SSCF.  CBP uses genetically engineered organisms that
are capable of producing cellulase enzymes and fermenting the simple
sugars to ethanol.  By combining all these steps into one process the
fuel production process is greatly simplified and the cost of fuel
production is potentially reduced.  The simple process and mild
conditions could result in both capital cost reductions as well as
operating cost reductions.  While CBP is a promising technology, there
are currently no organisms that can efficiently produce cellulase
enzymes and ferment sugars to ethanol.  Organisms have been found in
nature that possesses one or the other of these qualities, but not both
together.  Significant research is being done to produce such an
organism, usually by starting with an organism that naturally expresses
one of the desired traits and inserting genetic material to allow for
the expression of the other desired trait.  The commercial viability of
CBP is dependent on the discovery of an organism that can perform both
these tasks efficiently.

	3.6.4	Single Step Fuel Production

	Further simplification of the biochemical cellulosic biofuel process
would include the elimination of the pretreatment stage.  Bell
Bio-Energy claims to have developed such a process.  In this process the
lingocellulosic feedstock is first ground to reduce the size and then
the feedstock is immersed into water containing genetically engineered
bacteria .  The bacteria begin to digest the cellulose after only
several hours, but require 30 to 60 days to fully digest the cellulose. 
The produced fuel is constantly removed from the reaction vessel, and a
significant amount of organic material is also produced which will be
marketed as potting soil.  The process is expected to produce 30 to 40
gallons of renewable fuel per ton of feedstock and the simplicity of the
process results in low capital costs per volume produced.

	While this process may appear very similar to those discussed above
(SSF, SSCF, and CBP) it differs in several important ways.  The fuel
produced by the process is a diesel-like fuel that would need to be
upgraded before being used as a transportation fuel, rather than
ethanol.  Bell Bio-Energy is currently working with a University to
develop a process for upgrading the fuel they produce.  The amount of
time required for the organisms to produce fuel from the lingocellulosic
feedstock is also much longer, 30 – 60 days rather than hours.  The
yields of fuel per ton of biomass are much lower in this process, 30 –
40 gallons per ton versus 80 – 100 gallons per ton with most
biochemical processes.  Finally, the byproduct of this process is a
potting soil, which Bell Bio-Energy has found avenues to sell, rather
than lignin.  Despite the disadvantages listed above, the extremely low
capital and operating costs of this process and high value byproduct may
allow for commercial success.  As this technology has not been studied
as extensively as many of the alternative methods at a small scale,
there is greater technical uncertainty associated with this process.

	3.6.5	Gravity Pressure Vessel

 

	Another variation of the general biochemical cellulosic fuel production
process is the use of a gravity pressure vessel for the pretreatment and
hydrolysis of the lingocellulosic biomass.  Using this process the
feedstock material is pumped deep underground, to a depth of
approximately 2000 feet.  At this depth the pressure is near 60 bar and
the solution is heated to approximately 250° C.  Under these conditions
the lignin is dissolved and oxygen is added to combust the lignin to
provide heat to maintain this high temperature.  Next a weak acid is
added to hydrolyze the remaining hemicellulose and cellulose into simple
sugars.  Once the hydrolysis is complete the acid is neutralized, the
solution is returned to the surface, and the sugars are fed to the fuel
producing organisms.  

	The gravity pressure vessel allows the extreme conditions necessary for
the break down and hydrolysis of the lingocellulosic material to be
achieved with relatively little energy inputs, taking advantage of the
naturally high pressure deep underground and the combustion of the
lignin to provide the required heat.  The two main downsides of using
the gravity pressure vessel are high capital costs and no byproduct
recovery.  The gravity pressure vessel is an expensive piece of
equipment, and because the lignin is combusted to provide process heat
for the hydrolysis it cannot be recovered and used to provide process
heat for fuel separation or sold as a potentially valuable byproduct. 
Agresti Biofuels is planning to use a gravity pressure vessel to produce
ethanol from MSW at a facility in Pike County, Kentucky.

	3.7	Current Status of Biochemical Conversion Technology

	The biochemical cellulosic fuel production industry is currently
transitioning from an industry consisting mostly of small scale research
and optimization focused facilities to one capable of producing fuel at
commercial scale facilities.  Companies such as Iogen, DuPont Danisco
Cellulosic Ethanol, Fiberight, and KL Energy are just beginning to
market the fuel they are producing at their first small scale commercial
fuel production facilities.  By 2011 we expect several other cellulosic
fuel production facilities using biochemical processes to come online. 
Many other facilities, including some large scale facilities capable of
producing tens of millions of gallons of fuel, are planned to come
online starting in 2012.

	There are many factors that are likely to continue to drive the
expansion of the cellulosic biofuel industry.  The high price of
petroleum fuels and the mandates put into place by the RFS2 program have
created a large demand for cellulosic biofuels.  The biochemical
production process also has several advantages including relatively low
capital costs, highly selective fuel production, flexibility in the type
of fuel produced, and the promise of future production cost reductions.

	While the poor worldwide economy and tight credit markets has had a
negative impact on the biofuel industry as a whole the cellulosic
biofuel producers utilizing biochemical processes have not been as hard
hit as many others in the industry.  This is partially due to the
relatively low capital costs of biochemical production plants as a
result of the relative simplicity and mild operating conditions of these
plants.  Several companies have been able to purchase distressed grain
ethanol plants and are in the process of modifying them to produce
cellulosic ethanol, further reducing the capital costs of their initial
facilities.  Once biochemical fuel production facilities have been
constructed another advantage they have over other fuel production
processes is that their high selectivity in the fuels they produce. 
Unlike chemical catalysts, which often produce a range of products and
byproducts, biological organisms often produce a single type of fuel,
which leads to very high fuel production rates per unit sugar.  Finally,
there is a large potential to further decrease the production costs of
cellulosic biofuels using the biochemical processes.  Unlike other
production methods such as gasification which are relatively mature
technologies, biochemical production of fuels is a young technology. 
One of the major costs of the biochemical fuel production processes
currently are the enzymes.  Great strides have been made recently in
reducing the cost of these enzymes, and as the price of enzymes
continues to fall so will the operating costs of biochemical fuel
production processes.

	3.8	Path to Commercialization

	While there are many promising qualities of the biochemical fuel
production process we have identified several different aspects of the
process which can be further improved.  The pretreatment processes can
be improved to speed the conversion of cellulose and hemicellulose to
simple sugars and to minimize the production of other undesired
compounds, especially those that may inhibit the fuel production
process.  The ability of the biological fuel production organisms to
process a wide range of both five and six carbon sugars can also be
improved.  Both these improvements will increase the fuel yield per ton
of cellulosic feedstock, reducing the operating costs of the process. 
Finally, the enzyme production processes can be further optimized which
would lower the price for enzymes and improve the economics of
hydrolyzing cellulose to sugars.

	Another opportunity for improvement would be the profitable utilization
of the lignin portion of the cellulosic feedstock.  Unlike some of the
other cellulosic biofuel production processes, the biochemical process
does not convert the lignin to fuel.  Cellulosic feedstock can contain
up to 40% lignin, depending on the type of feedstock used, so the
effective utilization of this lignin is an important piece of the
profitability of the biochemical process.  One option for the use of the
lignin is to burn it to provide process heat and electricity, as well as
excess electricity that could be sold to the grid.  While this would
provide good value for the lignin, it would require fairly expensive
boilers and turbines that increases the capital cost of the facility. 
If the lignin cannot be used as part of the fuel production process it
may be able to be marketed as a solid fuel with high energy density and
low carbon intensity.  These various improvements to the cellulosic
biofuel plants would make biochemical processes more cost-competitive
with petroleum and other cellulosic biofuels.

	While decreases in production costs are not a given there is a wide
consensus within the cellulosic biofuels industry that future
technological developments and economy of scale benefits from commercial
scale production will result in significant production cost reductions. 
An analysis conducted by the National Renewable Energy Laboratory in
2008 projected the cost of cellulosic ethanol production via the
biochemical pathway decreasing from $2.43 per gallon in 2010 to $1.21
per gallon in 2022.  A table from this report, found below, shows how
the operating costs of different parts of the cellulosic ethanol
production process are expected to change over time.

Table 3.1

Changes in Operating Cost of Cellulosic Ethanol via the Biochemical
Process Over Time

	In considering NREL’s analysis of changes in the cost of production
of cellulosic biofuels via the biochemical pathway over time, as well as
studies conducted by other parties, there are three main areas from
which the majority of the cost improvements are expected; feedstock
cost, the cost of enzymes, and yields per ton of feedstock.  The cost of
feedstock is a large contributor to the cost of ethanol, responsible for
approximately 40% of the cost of a gallon of ethanol in all three cases
above.  Any reductions in the cost of cellulosic feedstocks to the
biofuel producers will have a direct and potentially significant impact
on the overall cost of cellulosic biofuel production.  Efforts to
decrease the cost of cellulosic feedstocks are taking place on a wide
variety of fronts.  Work in the area of plant breeding and genetics is
underway to increase the efficiency with which plants utilize water and
nutrients in order to decrease the cost of growing plants used by
cellulosic ethanol facilities.  Another area of much interest is
increasing the annual growth rate of crops that can potentially be use
for biofuel production.  This would decrease the number of acres needed
to supply a cellulosic biofuel production facility as well as decrease
the transportation cost for the feedstock by decreasing the area from
which a production facility would have to import feedstock.  New
harvesters are being developed to increase the efficiency of harvesting
cellulosic biomass.  These are just some of the many different ways in
which the cost of feedstocks for cellulosic biofuel producers is likely
to decrease in the future.

	A second area of significant potential cost savings for cellulosic
biofuel producers is the cost of enzymes.  Enzyme cost has long been a
major hurdle to the economic production of cellulosic biofuel via the
biochemical process, and it continues to be a major factor in the
overall production cost.  The recent trends in the cost of enzymes
necessary to produce cellulosic biofuel, however, are promising.  In
February 2010 Novozymes announced that it had reduced the cost of
enzymes require for cellulosic ethanol production to less than $0.50 per
gallon of ethanol produced, an cost reduction of 80% in the past two
years alone.  It is likely that the costs of the enzymes required for
cellulosic biofuel production, and thus the overall cost of cellulosic
biofuel production via the biochemical process, will continue to
decrease.

	A third way in which the cost of cellulosic biofuel production is
expected to decrease is by increasing the yield of fuel per dry ton of
biomass.  As with the feedstock costs, there are many different ways in
which cellulosic biofuel yields may be increased.  Feedstocks are being
developed with lower lignin content and higher cellulose and
hemicellulose content to enable a larger portion of the biomass to be
converted into biofuels by the biochemical process.  More effective
pretreatment methods and enzymes will likely convert more of the
cellulose and hemicellulose into simple sugars.  More versatile
biological organisms may be able to convert more of the simple sugars
into fuel.  The cost impacts of increasing the yield of biofuel per ton
of feedstock are similarly diverse.  Greater yields mean a lower
feedstock cost on a per-gallon of fuel basis.  They also mean lower
capital costs due to relatively smaller production facilities to produce
the same amount of biofuel.  Relatively smaller biofuel production
facilities mean lower costs for other raw materials and energy inputs. 
The combination of these many effects would have a significant impact on
the cost of cellulosic biofuel production.  While it is impossible to
precisely determine the extent of the progress in each of these areas
over time, it highly likely that progress in some or all of these areas
will contribute to future decreases in cellulosic biofuel production
costs via the biochemical process .

	4.	Thermochemical Process for the Production of Cellulosic Biofuels

Thermochemical conversion involves biomass being broken down into
intermediates using heat and upgraded to fuels using a combination of
heat and pressure in the presence of catalysts.  For the sake of the
discussion here, we are defining Thermochemical to mean gasification
processes (partial oxidation in the presence of a gasifying agent,
usually air, oxygen, and/or steam) that convert the biomass to a
syn-gas, but not including other processes such as pyrolysis (discussed
below) which occurs in the absence of oxygen and at lower temperatures. 
The thermochemical process is also applicable to other feedstocks (e.g.,
coal or natural gas); the main difference is that a renewable feedstock
is used (i.e. biomass) to produce cellulosic biofuel.  A thermochemical
unit can also complement a biochemical processing plant to enhance the
economics of an integrated biorefinery by converting lignin-rich,
non-fermentable material left over from high-starch or cellulosic
feedstocks conversion.   

Thermochemical conversion processes using gasification can be designed
to either produce primarily ethanol (alcohols) or other petroleum like
fuels such as diesel and naphtha, which is commonly referred to as a
biomass-to-liquids (BTL) thermochemical process.  The processing steps,
however, used to produce either ethanol or BTL fuels in a thermochemical
unit are similar until the fuel synthesis and product separation stages,
as discussed below.  

Compared to corn ethanol or biochemical cellulosic ethanol plants, the
use of biomass gasification may allow for greater flexibility to utilize
different biomass feedstocks at a specific plant.  Mixed biomass
feedstocks may be used, based on availability of long-term suppliers,
seasonal availability, harvest cycle, and costs.  Agricultural residue,
energy crops, wood residues, and municipal solid waste are all being
considered as potential feedstocks.  Geographic location, availability
of biomass, the existence of biomass suppliers, and costs would all
likely influence the mix of biomass feedstocks utilized.  The general
steps of the gasification thermochemical process include: feedstock
handling, gasification, gas cleanup and conditioning, fuel synthesis,
and separation.  Refer to Figure 4.1 for a schematic of the
thermochemical cellulosic ethanol production process through
gasification.  For greater detail on the thermochemical mixed-alcohols
route refer to NREL technical documentation.  Refer to Figure 4.2 for a
schematic of a block diagram of a BTL Thermochemical process.  

Figure 4.1

  Cellulosic Ethanol Thermochemical Gasification Process

Figure 4.2

  Biomass to Liquids (BTL) Thermochemical Gasification Process

4.1	Feedstock Handling 

	The particle size requirement for a thermochemical process is around
10-mm to 100-mm in diameter.  Once the feedstock is ground to the proper
size, flue gases from the char combustor and tar reformer catalyst
regenerator dry the feed from the as-received moisture level of around
30% to 50% moisture to the level required by the gasifier.  

4.2	 Gasification  

Ideally the gasification process would completely convert biomass to CO
and H2, although the process is not completely efficient as some CO2,
water, methane and coke/ tars are also produced.  The gasification of
biomass feedstock is carried out in the presence of a low concentration
of oxygen, with oxygen levels that are only about 30% of that required
for complete combustion.  Tars and coke are also formed as biomass is
gasified, which is primarily influenced by the gasifier operating
conditions, gasifier residence time, gasifier design and operating
temperature, with less influence based on the cellulosic feedstock type.
 The tar yield can vary by a wide range based on these parameters. 
There are many types of gasifiers, many designs resulting in different
performance and conversion of biomass.  For most gasifiers, the overall
efficiency for converting biomass into synthesis gas of CO and hydrogen,
is only about 70 to 80% efficient, as some of the biomass energy is
converted into CO2, H2O and tars.  Additionally, there is heat loss and
other energy losses in the gasifier, that also lower efficiency. 
Downstream of the gasifier, additional energy is also lost when
converting the synthesis gas of CO and H2 into ethanol and liquids,
which lowers the overall process efficiency to about 42-57% (amount of
biomass energy converted to liquid fuels) for currently explored
thermochemical designs.

The amount of carbon, hydrogen and oxygen present within the cellulosic
biomass feedstock can vary from feedstock to feedstock or even within a
feedstock.  As the relative amount of carbon and hydrogen in the
feedstock varies, the amount of CO and hydrogen produced during
gasification would vary as well which ultimately affects the amount of
fuel which can be produced.  If the oxygen content of the feedstock
varies, the amount of air or oxygen which is added to the gasifier to
help oxidize the biomass and combust any char being formed, can also be
varied to balance the overall quantity oxygen needed for gasification. 
An elemental chemical analysis is conducted on the feedstock to the
thermochemical unit to identify the amounts of carbon, oxygen, hydrogen,
sulfur, nitrogen and other inorganic elements present to indicate the
potential volume of liquid fuels that can be produced.  However, not all
of the available energy in biomass feedstock is converted to fuels, due
to inefficiencies in the thermochemical process.

The presence of moisture in feed stocks, though, has a detrimental
impact on the process yields and process efficiency of a gasification
unit.  This is because any moisture in the feedstocks must be vaporized
in the gasifier, which diverts energy away from the production of liquid
fuels.  So, a feedstock with a higher moisture content will have lower
overall process efficiency than that same feedstock with lower moisture
content.  Wood has higher moisture content, with values around 50%, than
other cellulosic stocks that have dried in the field (around 15%)vi. 
Wood’s high heating value, though, outweighs the negative effects of
its higher moisture content versus other cellulosic feedstocks.

	There are two general classes of gasifiers.  First, partial oxidation
(POx) gasifiers (directly-heated gasifiers) use the exothermic reaction
between oxygen and organics to provide the heat necessary to
devolatilize biomass and to convert residual carbon-rich chars.  In POx
gasifiers, the heat to drive the process is generated internally within
the gasifier. 

The second general class, called indirect gasification, uses steam
gasifiers to accomplish gasification through heat transfer from a hot
solid or through a heat transfer surface.  Either the byproduct char
and/or a portion of the product gas can be combusted with air (external
to the gasifier itself) to provide the energy required for gasification.
 Most steam gasifiers operate at low pressure and therefore require
product gas compression for downstream purification and synthesis unit
operations. ,  

For both the direct or indirect types of gasifiers, it advantageous to
use pure oxygen instead of air to the gasifier, as pure oxygen does not
add nitrogen into the synthesis gas which results in cost savings from
smaller equipment sizes for downstream equipment.  A disadvantage of
using pure oxygen as a feedstock is that oxygen production is expensive
and typically requires large plant sizes to improve economics.  

	There are different subcategories of gasifiers which are either
directly or indirectly heated.  One subcategory is termed a bubbling
fluidized bed gasifier and it employes a bubbling fluidized bed of inert
material and the reactant (biomass) is also bubbled through the
fluidized bed.  A second variant is the circulating fluidized bed
gasifier which is similar to the bubbling fluidized bed reactor except
that a high feedstock and air flow rate circulates the fluidized bed out
of and back into the reactor.  For the fluidized bed, the bed material
may either be inert alumina or sand which helps the heat transfer. 
There are also fixed bed reactors which either feed the reacting gas
(oxygen or air) upward or downward through a fixed bed of the reactant
(biomass).  Because of the tar formed when using biomass as a feedstock,
a second reactor is sometimes added which solely targets converting the
tar to syn-gas.  If the biomas feedstock is ground to a sufficiently
small particle size, or liquefied, the biomass is considered to be
“entrained” in the reactor, and the reactor is defined as an
entrained flow reactor.  

	The indirect gasification process begins as the biomass is fed to the
reactor containing a heat transfer media, such as sand, and is partially
reacted with air (or oxygen) which is introduced to the bottom of the
reactor.  The air serves as the carrier-gas and as the oxidant for
partially oxidizing the biomass to syn-gas, carbon monoxide and
hydrogen.  In addition to the syngas produced, char and coke are also
formed.  The heat for the endothermic gasification reactions is supplied
by circulating heat transfer media (e.g. sythetic sand) between the
gasifier and the char combustor.  The heat generated by the combustion
of the char and coke heats the heat transfer media to over 1800°F.  The
syngas is separated from the sand and ash and sent to gas cleanup. 

4.3	Gas Cleanup & Conditioning  

	Once the biomass is gasified and converted to syngas, the syngas must
be cleaned and conditioned.  Inorganic materials and minerals, such as
sulfur, sodium, silica and potassium are contained within the cellulosic
material and are also accumulated from dirt and the environment as the
biomass is collected from the field and transported to the plants.  The
minor entrained contaminants such as tars, sulfur, nitrogen oxides,
alkali metals, and particulates have the potential to negatively affect
the syngas conversion steps.  Most of this material is deposited in the
gasifier as a molten slag and is removed as a waste byproduct. 
Additionally, ash content in agriculture residues (from fertilzers,
dirt) and woods place a burden on the process, as ash must be scrubbed
out of flue gas that exits the process, though woody feedstocks have
less ash than agriculture resides.  The scrubbing of the ash from the
syn-gas consumes power and requires expensive capital for a flue gas
scrubber system.  Sulfur in biomass is converted into gaseous H2S in the
gasifier which is poisonous to the Fischer Tropsch process as it
deactivates the downstream reformer and Fischer Tropsch synthesis
catalyst.  H2S is therefore removed from the gasifier synthesis gas via
an acid gas scrubbing section prior to entry into the reformer catalyst.
 Wood has low sulfur levels with a typical values estimated around 10
ppm, while energy crops, such as corn stover and switch grass have much
higher values with estimates around 500 ppm.  The gasifier also converts
some of the nitrogen in biomass to NH3 and HCN, which is also removed in
the downstream acid gas scrubbing section, as these are poisonous to the
Fischer Tropsch catalyst.  If scrubbers are being used anyways,
thermochemical technologies may be better suited to process municipal
solid waste which can contain a significant amount of contaminants.  The
benefit to using MSW is that it is priced lower than other feedstocks
and the lower feedstock price would help to offset the thermochemical
plant’s much higher capital costs.  The gasification reactors and the
downstream scrubbers typically used in thermochemical processes reduces
the chance that contaminants contained in the feedstock make it into the
final fuel.  Gas conditioning steps include sulfur polishing to remove
trace levels of H2S and water-gas shift to adjust the final H2/CO ratio
for optimized fuel synthesis.   

4.4	Fuel Synthesis and Product Separation

	After cleanup and conditioning, the “clean” syngas is comprised of
essentially CO and H2.  The syngas is then converted into a liquid fuel
by either a catalytic process or through the use of a microorganism. 
The fuel producer has the choice of producing diesel fuel or alcohols
from syngas by optimizing the type of catalyst used and the H2/CO ratio.
 Diesel fuel has historically been the primary focus of such processes,
as it produces a high quality distillate product, however, with the 45
cent tax subsidy currently available for ethanol production, it may be
economically advantageous for fuel producers to convert syngas to
ethanol instead of to diesel fuel.  Below, we discuss the processing
steps for a plant configured to make either ethanol or BTL fuels.

4.5	Heat & Power 

	For either an ethanol or BTL plant configuration, a carefully
integrated conventional steam cycle produces process heat and
electricity (excess electricity is exported).  Pre-heaters, steam
generators, and super-heaters generate steam that drives turbines on
compressors and electrical generators.  The heat balance around a
thermochemical unit or thermochemical/biochemical combined unit must be
carefully designed and tuned in order to avoid unnecessary heat losses.
Any unreacted carbon monoxide and hydrogen and any gaseous hydrocarbon
material from the process are burned to produce electricity in a
turbine.  The waste heat from the gas turbine along with the steam
created to cool the syn-gas, may be sent to steam turbines to produce
additional electricity.  Most of the electricity would be used within
the thermochemical plant, however, some could be sold to raise
additional revenues.

4.6a	Ethanol Production

	Conceptual designs and techno-economic models have been developed for
ethanol production via mixed alcohol synthesis using catalytic
processes.  The proposed mixed alcohol process produces a mixture of
ethanol along with higher normal alcohols (e.g., n-propanol, n-butanol,
and n-pentanol).  The by-product higher normal alcohols have value as
commodity chemicals and fuel additives.  Typically the mixed alcohol
products are high in methanol, but contain a wide distribution of
several different alcohols.  One concept proposed in literature is to
completely recycle this methanol in order to increase the production of
ethanol and higher alcohols which are generally more valuable.  This
concept was modeled by NREL for the thermochemical production of ethanol
for the year 2012, using a Molydenum Sulfide type catalyst.  Total mixed
alcohol yield was 94.1 gallons per dry ton, in which 85% of the total
alcohol product was ethanol.  This was made possible through the
addition of an almost complete recycle of methanol within the process.
For the Renewable Fuels rulemaking, we worked with NREL to develop the
thermochemical mixed-alcohols model for the 2015 and 2022 timeframe . 

	The liquid rundown from the low-pressure separator is dehydrated in
vapor-phase molecular sieves, producing the dehydrated mixed alcohol
feed into a methanol/ethanol overhead stream and a mixed, higher
molecular weight alcohol bottom stream.  The overhead stream is further
separated into a methanol stream and an ethanol stream.

	Range Fuels produces cellulosic biofuel via a two step thermochemical
process.  Their technology converts biomass to syngas followed by
catalytic conversion of the syngas to alcohols.  Range claims that their
technology is capable of producing more ethanol than other cellulosic
technologies based on yields per energy input.  They utilize a two step
process which can use many forms of non food biomass, such as
agriculture waste, wood, and corn stocks.  Additionally, the technology
can process feed stocks with variable water content.

	Range has operated a pilot plant for over 7 years using over 20
different nonfood feedstocks.  Range built its first commercial
demonstration plant which started late in  2010.  This plant is located
in Soperton, Georgia and is partially funded from proceeds of DOE and
USDA grants.  The plant uses wood, grasses, and corn stover as
feedstocks.  In its initial phase, the Range plant will be  producing 4
million gallons per year of methanol.  Once the company is confident in
its operations, Range intends to add additional reaction capacity to
convert much of the methanol to ethanol and then expand the plant.  

Enerkem is another company pursuing cellulosic ethanol production via
the thermochemical route.  The Canadian-based company was recently
announced as a recipient of a $50 million grant from DOE to build a 10
MGY woody biomass-to-ethanol plant in Pontotoc, MS.  The U.S. plant is
not scheduled to come online until 2012, but Enerkem is currently
building a 1.3 MGY demonstration plant in Westbury, Quebec.  According
to the company, plant construction in Westbury started in October 2007
and was completed in December 2009.  Since then, the plant has been in
the commissioning and start-up phase with methanol production projected
to occur in the fall of 2010.  Later, the plant will switch from
methanol production to ethanol. While this fuel would likely be eligible
to generate RINs under the RFS program if imported, Enerkem has
indicated that they plan to use all the fuel locally and do not plan to
export any fuel to the United States.  

4.6b	BTL Production 

In this configuration, the syngas is fed to a Fischer-Tropsch (FT)
reactor to primarily produce diesel fuel and some naphtha.  Usually the
ratio of carbon monoxide to hydrogen is too high not allowing maximum
fuel production volumes with the syngas,  If that is the case, the
syngas must first be sent to a water shift reactor where some of the
carbon monoxide is used to free up hydrogen from water which is also
present in the reactor.  The cleaned and water-shifted syngas is sent to
the FT reactor where the carbon monoxide and hydrogen are reacted over a
FT catalyst.  The FT catalyst is either iron-based or cobalt-based.  The
cobalt catalyst is more expensive, although it does not require a
recycle, while the less expensive iron catalyst does require a recycle. 
The FT reactor creates a syncrude, which is a variety of hydrocarbons
that boil over a wide distillation range (a mix of heavy and light
hydrocarbons).  The syncrude from the FT reactor is sent to a
distillation column where it is separated into various components based
on their vapor pressure, mainly liquid petroleum gas (LPG), naphtha,
distillate and wax fractions.  The heavier compounds are hydrocracked to
maximize the production of diesel fuel.  The distillate boiling
compounds have high cetane and thus are of high quality for blending
into diesel fuel.  Conversely, the naphtha material is very low in
octane thus, it would either have to be upgraded, or blended down with
high octane blendstocks (i.e., ethanol), or be upgraded to a higher
octane blendstock to have much value for use in gasoline.  The naphtha
could also be sold as feedstock for the petrochemical market for
manufacturing chemical products such as ethylene and benzene.

	Choren has a technology called Carbo-V, which is a Fischer-Tropsch
process that can be used to make diesel fuel.  The process can process a
wide variety biomass and recycled material materials as feedstocks.  The
process converts agriculture biomass, forestry biomass, biogenic waste
and recycling substances into a synthesis gas which can be further
converted to a diesel fuel using a Fischer-Tropsch reactor.  The Carbo-V
process can also be configured without the Fischer-Tropsch hydrocracking
technology, so as to produce electricity, heat and power, methanol, and
other chemical feedstocks.

	The principal aspect of the Carbo-V Process is a three-stage
gasification process consisting of low temperature gasification, high
temperature gasification and endothermic entrained bed gasification.  In
the first stage, biomass is partially oxidized with air or oxygen at
temperatures between 400 and 500 °C.  This breaks down the feedstock
into a gas containing tar and solid carbon.  In the second stage, the
tar is oxidized at temperatures higher than the tar’s melting point,
converting the tar into a synthesis gas.  In the third stage, solid
carbon is mechanically pulverized and blown into the hot gasification
stream.  The fluidized carbon endothermicly reacts with the gasification
stream and is converted into a synthesis gas.  In the next
Fischer-Tropsch stage of the process, the synthesis gas (CO and H2)
reacts with the aid of a catalyst to form hydrocarbons.  The resulting
heavy hydrocarbons produced from the three stages can then be sent to a
hydrocracking process to primarily produce diesel fuel. 

	Choren will be building a commercial Plant in Freiberg/ Saxony Germany
that is expected to be operational in 2011 or 2012.  Initially, the
plant will use biomass from nearby forests, wood waste from the
wood-processing industry and straw from farmland.

4.7 	Pathways to Commercialization 

	Although there are examples of thermochemical gasification processes
which process coal and natural gas, a thermochemical plant which must
gather, pretreat and gasify cellulosic biomass has not yet been
demonstrated on a commercial scale.  The capital costs for
thermochemical plants are much higher per volume of fuel produced than
other cellulosic biofuel technologies, due to the complexity of their
design which requires many process stages, expensive equipment pieces
and high catalyst costs.  The first investors in these thernochemical
plants that process cellulosic biofuels are expected to be those which
are willing to tolerate the somewhat higher risk of investing in a new
application of an existing technology.  Once the first several
thermochemical plants which produce cellulosic biofuels are built, more
investors are expected to be confident that these plants will have long
term reliable performance and that the product yields will be
satisfactory for economic payoff and invest in this technology. 
Additionally, optimization and improvements to the gasification stage,
catalyst synthesis stages and overall configuration of the plant will
help to lower the capital and operating costs to further improve the
competitiveness of this technology.

	

	5.	Hybrid Thermochemical/Biochemical Processes

	Hybrid technologies include process elements involving both the
gasification stage of a typical thermochemical process, as well as the
fermentation stage of a typical biochemical process and therefore cannot
be placed easily into either category.  For more specific information
regarding either biochemical processes or thermochemical, please see
Sections 3 and 4 of this technical memo respectively.  Currently, there
are several strategies for the production of ethanol through hybrid
processes; these strategies are differentiated by the order in which the
thermochemical and biochemical steps take place within the process, as
well as how the intermediate products from each step are used.  The
following subsections describe these different processes; specifically
how they differ from processes that may use only one technology type
(either thermochemical or biochemical).

	5.1 “Thermochemical First” Process

	The ‘thermochemical first’ strategy involves the gasification of
all feedstock material to syngas before being processed into ethanol
using a fermenter.  Hence, the first steps in this process most closely
resemble a typical thermochemical process, while the following steps
most closely resemble that of a typical biochemical process.

	Stage 1 – 3: Feedstock Handling, Gasification and Cleanup

	The first stages do not differ in any significant way from the
respective stages found in a traditional thermochemical process. 
Gasification converts carbonaceous material under high heat and pressure
to a mixture of mainly carbon monoxide and hydrogen gasses.  This
mixture is commonly referred to as syngas.  This gas may also contain
other compounds such as metallics or sulfur; these compounds can be
detrimental or poisonous to steps farther downstream in the process and
therefore must be removed before further treatment of the syngas.  More
information regarding these steps can be found in Section 4 of this
technical memo.

	Stage 4: Fuel Synthesis

	After gasification, the syngas stream is cooled and bubbled into a
fermenter containing genetically modified microorganisms.  Unlike
traditional fermentation, where C5 and C6 sugars are digested to
ethanol, these microorganisms are engineered to convert the carbon
monoxide and hydrogen contained in the syngas stream directly into
ethanol.  Using a fermenter in this hybrid process replaces the catalyst
needed in a thermochemical process.  Traditional thermochemical
catalysts require a very specific ratio of carbon monoxide to hydrogen
in the syngas stream in order to reach efficient conversions.  This
requirement necessitates the inclusion of an additional water-shift step
for thermochemical fuel production.  A water-shift reaction is used to
balance the ratio of carbon dioxide and hydrogen gas to that required by
the traditional thermochemical catalyst.  The microorganisms used in a
hybrid process allow for more variation of the incoming syngas stream
ratio, therefore avoiding the water-shift reaction preceding typical
thermochemical conversion using chemical-based catalysts.  While
thermochemical catalysts require high pressures and temperature,  the
microorganisms are able to operate at significantly lower temperatures
and pressures than those required for a traditional catalytic conversion
to ethanol.  Microorganisms, unlike a catalyst, are also self-sustaining
and do not require periodic replacement.  They are; however, susceptible
to bacterial and viral infections which requires periodic cleaning of
the fermentation reactors.

	Stage 5: Product Separation

	After fermentation, the effluent water/ethanol stream from the
fermenter is separated similarly to a biochemical process; usually using
a combination of distillation and molecular sieves.  The separated water
can then be recycled back into the fermentation stage of the process. 
Typical yields of ethanol are predicted in the 100-130 gallon per ton
range, similar to that of a traditional thermochemical ethanol process. 
For more information regarding product separation, refer to Section 3.5
biochemical product separation.

	Coskata has a gasification-based technology which produces ethanol from
biomass and other forms of carbon through a biofermentation route.  A
wide variety of feedstocks can be used including municipal solid waste,
agriculture waste and other carbonaceous containing material.  Since
this process uses combustion and biofermentation, we classified it as a
hybrid technology.  This process requires that the biomass or
carbonaceous material be processed to a small particle size and then it
is injected into a gasifier.  

	In the Coskata process, the gasifier combusts any dry carboneous feed
stocks into syngas.  The syngas produced is fermented in a reactor by
micro-organisms, which convert the carbon monoxide and hydrogen directly
into ethanol.  The micro-organisms are low cost and can process a wide
range of carbon monoxide and hydrogen molar ratios in the syngas,
providing feedstock processing flexibility.  No other enzymes are
required by this process for producing ethanol, providing significant
cost savings over current cellulosic and corn based fermentation
production methods.  The Coskata micro-organism reactor is operated at
low temperatures and pressures, which offers savings on capital and
energy costs.  Additional energy savings can be realized by employing
membrane technology to separate ethanol from the reactor decant liquid. 
This technology uses gravity and filtration to recover ethanol,
resulting in significant savings on distillation capital and energy
costs used in other cellulosic and corn based ethanol production
methods.  According to Coskata, initial ethanol production cost
estimates are lower than the biochemical and thermochemical cellulosic
technologies.

 

	For woody biomass, Coskata estimates that each ton of this feedstock
would generate about 100 gallons of ethanol and small amounts of ash
which would be burned to supply energy needs for the process.  Corn
stover is expected to provide similar ethanol yields as woody biomass
feed stocks, though details about yields from the various feed supply
stocks are not yet public.

 		  

	Coskata has a bench scale pilot plant in Warrenville, IL, and its
larger 40,000 gallon per year pilot plant became operational in 2009 in
Madison, Pennsylvania.  Coskata is targeting to design and build a 55
million gallon per year commercial demonstration plant that it expects
to be operational in 2012.

INEOS Bio also has been developing a thermochemical first hybrid biofuel
production technology.  INEOS Bio (along with partner New Planet Energy)
has recently been selected for a $50MM DOE grant for the construction of
an 8 MGPY plant in River County, Florida; predicted to finish
construction in late 2011.

	5.2 “Biochemical First” Process

	A ‘biochemical first’ process ferments the cellulose and
hemicellulose like a typical biochemical process, but the typically
unfermentable feedstock fraction (lignin) is gasified concurrently with
a fermentation step.  The syngas created from the lignin is then reacted
with the product of the fermented stream. 

	Stage 1 – 2: Feedstock pretreatment and hydrolysis

	Feedstocks are prepared for this process in the same fashion as a
traditional biochemical plant.  Incoming feedstock is pretreated and
hydrolyzed to separate fermentable sugars from non-fermentable compounds
such as lignin.  More information on these steps can be found in
Sections 3.2 and 3.3 of the biochemical technology Section of this
technical memo.

	Stage 3a: Fermentation

	Stage 3a and 3b in this process take place concurrently; before
fermentation, the unfermentable portion of feedstock (lignin, ash and
other residue) is fractioned and sent to a gasifier, see stage 3b.  The
remaining fraction of feedstock is fermented using an acetogen
microorganism, in a manner similar to that of a traditional biochemical
process. The acetogen converts the sugars and mixed carbohydrates of the
feedstock to acetic acid.  This reaction creates no carbon dioxide,
unlike a typical fermentation using yeast, preserving a higher amount of
carbon in the finished fuel.  Acetogens are also naturally occurring and
therefore more robust against infection or contamination versus
microorganisms that would be genetically engineered.  The acetic acid
stream is then stabilized to ethyl acetate.  Ethyl acetate is both
easier to store, as well as forming a precursor to many other chemical
compounds that can be created at a later stage in the process.

	Stage 3b: Gasification of unfermentable feedstock

	Meanwhile, the unfermentable fraction separated in stage 3a is fed to a
gasification reactor similar to that found in a traditional
thermochemical process.  Due to the reduced volume of feedstock going to
this reactor, the size of the gasifier is greatly reduced compared to
that of a thermochemical-only process.  The syngas stream produced from
the gasification of lignin and other residue is then separated into its
carbon monoxide and hydrogen components.  The hydrogen is fed to the
next stage of the process, while excess carbon monoxide is combusted,
producing enough energy to meet the needs of the other process stages.

	Stage 4: Product Production

	The hydrogen stream from stage 3b is reacted with the ethyl acetate
stream from stage 3a to form ethanol.  It is also possible, by changing
reaction parameters, to form other chemical compounds from the ethyl
acetate in order to meet the needs of changing market conditions; acetic
acid and ethyl acetate form the precursors to many other chemical
compounds and therefore may also be sold in addition to ethanol.  It is
predicted that this process can achieve conversions of 130 - 150 gallons
of ethanol per ton of feedstock.

A company currently pursuing a biochemical first process technology is
Zeachem Inc.  Zeachem is currently constructing a 250 thousand gallon
per year demonstration plant in Boardman, Oregon.  They have received a
$25MM DOE grant and expect to have a full commercial production facility
operational in 2013.  

	5.3 	Path to Commercialization

	Since these hybrid process technologies include steps from both
biochemical and thermochemical processes, they also have some of the
same commercialization issues as both of these technology types. 
‘Thermochemical first’ processes carry the same high capital costs
as traditional thermochemical operations, as well as the complexity of
heat transfer between stages required for these processes.  However,
because of the use of a biochemical stage instead of a catalyst, initial
catalyst cost and tuning can be eliminated. This is offset by initial
microorganism cost and first batch startup of the microorganisms. The
addition of a biological step also increases the need for sterile
procedures in that part of the plant, adding costs that would not be
present in a typical thermochemical operation. ‘Biochemical first’
processes also share many of the commercialization issues of a typical
biochemical process.  Feedstock pretreatment and hydrolysis steps need
to be improved on a large scale to achieve efficient production.  The
addition of a thermochemical processing step for lignin and
unfermentable residue solves the problem of wasted feedstock; however,
it adds both complexity and capital cost with the addition of a
thermochemical step.  Both hybrid process pathways are able to achieve
higher conversion rates than either of the sole technology processes,
with the tradeoff of added complexity.  This added complexity adds to
startup time and capital costs, which may lengthen construction times.

6.	Pyrolysis and Catalytic Depolymerization

Pyrolysis and catalytic depolymerization encompass a group of
technologies which are capable of creating biofuels from cellulose by
either only relying on heat, or by relying on a combination of heat with
catalysis.  Pyrolysis technologies are usually thought of being solely
or primarily thermal technologies, however, pyrolysis technologies are
being developed integrate some catalysts into the technology.  These are
all unique processes, typically with single companies developing the
technologies, so they are discussed separately.  

6.1 	Pyrolysis 

	Pyrolysis oils, or bio-oils, are produced by decomposing cellulosic
biomass at lower temperatures (400 to 800 degree Celsius) than the
gasification process, thus producing a liquid bio-oil instead of a
synthesis gas.     The reaction can occur either with or without the use
of catalysts, but it occurs without any additional oxygen being present.
 The bio-oil produced varies in oxygen content or viscosity according to
the feedstock used and the reactor conditions.  Typically, to maximize
the production of liquid fuels, the reactor temperature is controlled at
400 to 500 degrees Celsius and the residence time is only 2 seconds –
hence the name “Fast pyrolysis.”   The fast pyrolysis process uses
sand (not catalyst) in a fluidized bed to transform bio-fuels into a
product named bio-oil.  The sand is used to enable the quick heat
transfer needed to keep the reaction time to 2 seconds.  Three products
are produced by the fast pyrolysis process, bio oil (60-75% by weight),
char (15-20% wt.) and non-condensable gases (10-20% wt.).  The char
produced is similar to coke and can be used as fuel by other industries
while the gases yielded from the process can be used to supply about 75%
of the energy requirements of the pyrolysis process.  Once the reaction
is complete, the oil must have particulates and ash removed in
filtration to create a homogenous product.  One common way to conduct
the separation of particulate matter and ash from the bio-oil is through
the use of a cyclone.  

The bio-oil is typically high in oxygen content (i.e., 20% – 30%)
which makes this product immisible with petroleum, but enables it to be
miscible with water.  Bio-oil is also very low in cetane and has about
half the energy content per gallon of petroleum-based diesel fuel. 
Because of the high oxygen content and the immersed water, the bio-oil
has a density more like that of water.  Also the acid compounds in the
bio-oil cause the bio-oil to be acidic which requires the metals in
contact with the bio-oil to be acid resistant stainless steels.  This
makes bio-oil very challenging to use in internal combustion engines. 
If a use can be found for the bio-oil, the production cost is very low
due to the very low capital and operating costs for these plants.  

If pyrolysis is to be widely used to generate as transportation fuels or
for other uses, the bio-oil will have to be upgraded to hydrocarbon
fuels.  The upgrading process would likely involve one or more reactor
vessels which likely would contain a catalyst that reduces the acidity,
reduces the viscosity and reduces the density of the bio-oil.  One
likely technology for this upgrading step would be hydrotreating.    
The Pacific Northwest National Laboratory (PNNL) has modeled
hydrotreating as an upgrading strategy for bio-oils sourced from
pyrolysis.  PNNL projected that two hydrotreating reactors would be
needed.  The first hydrotreater is projected to operate at a higher
space velocity and lower temperature than conventional hydrotreaters - 
the lower temperature is used to avoid coke formation.  The purpose of
the first reactor would be to deoxygenate the more reactive acid and
aldehyde oxygen-containing hydrocarbon species, and saturate the more
reactive olefins.  After the first reactor, the bio-oil would likely no
longer be miscible with water and be blendable with hydrocarbons, which
would allow any entrained water to be separated from the bio-oil.  

The second hydrotreater would operate at a temperature and space
velocity more like a conventional refinery hydrotreater and its primary
purpose would be to drastically remove the oxygen contained within the
hydrocarbon molecules.  The resulting fuel would still contain a small
fraction of the oxygen originally contained within the cellulose, but it
would be much more like hydrocarbon compounds sourced from crude oil. 
The downside for the hydrotreating route is that the hydrogen
consumption from the two upgrading hydrotreating reactors would be a
very high – on the order of 4600 standard cubic feet per barrel of
feed.  This level of hydrogen consumption is an order of magnitude
higher than that typically experienced with distillate hydrotreating.   


Instead of solely relying on hydrotreating as the upgrading technology,
the first reactor after the pyrolysis reactor could contain a catalyst
which upgrades the bio-oil without the use of hydrogen.  The somewhat
upgraded bio-oil would still contain a significant amount of oxygen, but
it may be sufficiently stable to store for a second upgrading step that
could involve hydrotreating.  Because the bio-oil is stable, it perhaps
could be transported to a refinery with extra hydrotreating capacity to
hydrotreat the bio-oil.  An important advantage of employing a
nonhydrotreating catalyst as a first step would be to significantly
reduce the amount of hydrogen required.  The second advantage of this
approach is the two upgrading duties can be optimized separately
depending on the cellulosic biomass location and best location for fuels
upgrading.  For example, the pyrolysis reactors and upgrading reactor
(first step) could be located close to where the cellulosic biomass is
located avoiding the long distance shipments of the biomass (biomass is
a solid which is more costly to handle, and it contains about 50%
oxygen, which is deadweight because oxygen does not add any energy
value).  Once the biomass has been “liquefied” and at least
partially deoxygenated, transportation is made easier, perhaps by
barges, ships and pipelines.  The second upgrading step can be located
further away from the biomass harvesting locations where fewer but
larger, more cost-efficient upgrading equipment could be utilized.  The
upgrading step can even occur using the existing capital at today’s
refineries. 

There is another means other than using catalysts to stabilize bio-oil. 
The National Renewable Energy Laboratory (NREL) has a proprietary
technology named “hot filtration” that apparently was able to
stabilize bio-oil samples for a very long period of time (years). 
NREL’s pyrolysis reactor uses sand, but the resulting bio-oil after
their hot filtration will contain a much lower amount of contaminant
metals and is stable.  So, it will be easier to solve the second
problem, the bio-oil upgrading. 

	Dynamotive Energy Systems Corporation employs a fast pyrolysis
technology that uses medium temperatures and oxygen free reactions to
convert dry waste biomass and energy crops into specialty products.  The
process is flexible on the types of biomass feedstock’s that can be
processed.  The bio-oil produced by their process contains significant
amounts of water, thus it is not directly useable as fuel in
conventional vehicles and would have to be converted via another
catalytic conversion processing step.  The additional catalytic step
Dynamotive envisions for upgrading their bio-oil for this would combust
the material into a synthesis gas which would then be converted into
diesel fuel (the BTL process).  If upgraded, the diesel fuel produced is
expected to be compatible with existing petroleum diesel fuels.  

	Dynamotive has two small demonstration plants.  One demonstration plant
is located in Guelph, Ontario, Canada and its capacity is 66,000 dry
tons of biomass a year with an energy output equivalent to 130,000
barrels of oil per year.  Dynamotive also has a demonstration plant in
West Lorne, Ontario Canada.  The West Lorne plant has a capacity to
convert 130 tonnes of biomass into bio-oil per day which, if
proportional to the Guelph plant, translates to an energy-equivalent of
84,500 barrels of oil per year.  This plant started operation in early
2005 using waste sawdust as a feedstock.  Dynamotive does not have any
contracts in place to supply U.S. based clients with transportation fuel
in year 2010 or 2011.  Additionally, Dynamotive has not announced plans
for building a bio-oil upgrading facility.

	Envergent is a company formed through a joint venture between
Honeywell’s UOP and the Ensyn Corporation.  Although Ensyn has been
using fast pyrolysis for more than a decade to produce specialty
chemicals, UOP is relying on its decades of experience developing
refining technologies to convert the pyrolysis oils into transportation
fuels.  Envergent is also working with Federal laboratories to further
their technology.  Based on their current technology and depending on
the feedstock processed, about 70 % of the feedstock is converted into
liquid products.  The gasoline range products produced are high in
octane, while the diesel fuel products are low in cetane.  Envergent is
licensing this technology as well as working with a U.S. oil company to
test out this technology in a commercial setting here in the U.S. 

Petrobras is working on developing a pyrolysis technology.  A United
States and Brazil Memorandum of Understanding to advance the cooperation
on biofuels was signed by United States’ Secretary of State and
Brazil’s External Relations Minister on March 9, 2007.  A second
Memorandum of Understanding was signed by Petrobras and the National
Renewable Energy Laboratory (NREL) on September 2008 aiming at
collaborating to maximize the benefit of their respective institutional
interests in second generation biofuels.  Petrobras is working with NREL
to develop a pyrolysis technology to produce biofuels from agricultural
wastes such as sugar cane bagasse, wood chips or corn stover.  NREL and
Petrobras intend to use their combined expertise to develop a two-step
pyrolysis route to biofuels production.   

Petrobras is also working to develop a one-step pyrolysis technology,
although the company is somewhat skeptical that even if an effective
one-step pyrolysis catalyst can be developed, it may not be economically
viable for use in rural areas.  The Cooperation Agreement with NREL
covers the Life Cycle Assessment and Technoeconomic Analysis of both the
two-step and one-step routes.

6.2	One Step Catalytic Pyrolysis and Catalytic Depolymerization

It is possible to use a sophisticated catalyst (instead of sand) to heat
and depolymerize cellulose and at the same time produce a stable bio-oil
or even produce transportation quality fuels.  Some papers in literature
address these problems by using ZSM-5 zeolite.  The resulting bio-oil is
very aromatic, but theoretically it could be used directly as
transportation fuel and the posterior upgrading would be no longer
necessary, since fuel properties can be improved by decreasing the
amount of oxygen in the bio-oil through deoxygenation and cracking to
form H2O, CO and CO2.  The challenge of this route is that many of the
available catalysts which are candidates are relatively expensive and
they degrade quickly due to the metals and water which are present. 

There are at least two ways to address the cost impacts of using
expensive catalysts which degrade over time from the contaminants.  One
solution would be to find a catalyst formulation which is less
expensive, but also robust enough to handle the high level of water and
metal contaminants present in biomass, namely sodium, potassium,
magnesium, iron.  Otherwise, the process will not be economically
viable.  Zeolytic catalysts are likely not candidates because they
deactivate from being exposed to most of these metals.  A second
approach would be to develop a new catalyst which may still be
expensive, but which is resistant to the contaminants.  Alternatively,
one could use a susceptible catalyst, but somehow protect the catalyst
from the contaminants, such as a guard bed.  A guard bed would be placed
above the rest of the catalyst and its role would be to adsorb the
contaminants preventing the more expensive one-step pyrolysis catalysts
located lower down in the reactor from being exposed to the
contaminants.  

There seem to be several companies which have found catalysts which can
function in the environments present when converting cellulosic biomass
to biofuels.  Since their processes are proprietary, some of the
companies will not share what catalysts they are using, nor do we know
the cycle length of the catalysts they are using.  

Covanta is currently operating 44 energy-from-waste facilities which
annually convert 20 million tons of municipal solid waste materials into
9 million megawatt hours of electricity and 10 billion pounds of steam,
which is sold to a variety of industries.  Covanta has secured license
rights to a catalytic depolymerization technology developed by AlphaKat
GmbH.  Covanta constructed an AlphaKat demonstration plant in West
Wareham, Massachusetts designed to process 45 tons of waste per day into
renewable diesel fuel.  If successful, the total liquid fuel production
capacity of this demonstration plant will be 1 million gallons per year.
 This plant started up in mid-2010 and after experimenting with the
technology to further understand its capabilities, Covanta expects to
use the liquid distillate fuel produced from this demonstration plant
within its own plant as heating oil and nonroad diesel fuel.  

The AlphaKat process licensed by Covanta catalytically depolymerizes
cellulosic feedstocks at moderate temperatures and low pressure to
produce diesel fuel.  The proposed feedstock is municipal solid waste
(MSW) or other waste material such as animal waste, non-recyclable
plastics, agriculture residue, woody biomass and sewage waste.  The
feedstock must be prepared by shredding it to a size smaller than 2
inches, and removing both ferrous and non-ferrous metals.  The prepared
feedstock is fed to the process along with a powdered zeolite catalyst
and some lime which serves as a neutralizing agent, where they are mixed
with a non-reactive carrier oil, which provides a liquid medium for the
reactions to take place.  This carrier oil slurry is then circulated
through a friction turbine reactor, which heats the slurry to about 280
degrees Celsius and turbulently mixes the feedstock and catalyst.  In
the reactor the waste feedstock is catalytically converted to diesel
fuel, which is recovered in a distillation column.  Depending on its
use, the diesel product may require further processing to remove sulfur.
 The process is capable of producing approximately 60 gallons of diesel
product per ton of waste feedstockI Inorganic material that cannot be
converted will be removed from the process via a purge stream, which may
be further processed to recover catalyst and carrier oil to be returned
to the process.  Carbon dioxide and water are also formed in the
catalytic depolymerization reactions and are released from the process.
  

	The Cello-Energy process is also a catalytic depolymerization
technology.  At moderate pressure and temperature, the Cello-Energy
process catalytically removes the oxygen and minerals from the
hydrocarbons that comprise finely ground cellulose.  This results in a
mixture of short chain (3, 6 and 9 carbon) hydrocarbon compounds.  These
short chain hydrocarbon compounds are polymerized to form compounds that
boil in the diesel boiling range, though the process can also be
adjusted to produce gasoline or jet fuel.  The resulting diesel fuel
meets the ASTM standards, is in the range of 50 to 55 cetane and
typically contains a very low concentration of sulfur.  

	

	The Cello process is reported to be on the order of 82 % efficient at
converting the feedstock energy content into the energy content of the
product, which is very high compared to most of today's biochemical and
thermochemical processes which are on the order of 50 % efficient or
less.  Because of the simplicity of the process, the capital costs are
very low.  A 50 million gallon per year plant is claimed to only incur a
total cost of $45 million.  Because of its high efficiency in converting
feedstocks into liquid fuel, the production and operating costs are also
estimated to be very low. 

	In December 2008, Cello completed construction of a 20 million gallon
per year commercial demonstration plant.  However, they are still
working to resolve process issues that have arisen upon scaleup from
their pilot plant.  However, we are doubtful that Cello will be able to
produce any volume of cellulosic biofuel in 2011.

	Greenpower has a process similar to the Alphakat process licensed by
Covanta.  Greenpower completed construction of a demonstration plant
located in Fife, Washington in March of 2008.  Greenpower is working on
obtaining additional funding and an air permit through the State of
Washington Environmental Office.  While we do not expect that Greenpower
will have its plant operational in 2011, it is possible that outstanding
issues could be resolved to allow this company to produce renewable fuel
that could help refiners comply with the cellulosic biofuel volume
standard for 2011. 

6.3	Pathways to Commercialization 

Pyrolysis technologies produce very low quality products which cannot be
used in today’s vehicles and engines.  Therefore, the bio-oil must be
upgraded to improve the stability and the usability of the fuel and
continued effort will likely be invested in improving the efficiency and
reducing the overall cost of upgrading the bio-oil.  For those working
with catalysts for combining with a pyrolysis process, or for a
catalytic depolymerization process, the companies will likely continue
to improve the effectiveness and lower the costs of their catalyst.

7.	Catalytic Reforming of Sugars to Gasoline

	Virent is pursuing a process called “Bioforming” which functions
similarly as the gasoline reforming process used in the refining
industry.  While refinery-based reforming raises natural gasoline’s
octane value and produces organic chemicals, benzene, xylene and toluene
as a byproduct, Bioforming reforms biomass-derived sugars into
hydrocarbons for blending into gasoline and diesel fuel.  The process
operates at moderate temperatures and pressures.  The Bioforming process
which initially was conceived at the University of Wisconsin, is being
developed through a partnership between Virent, Shell, Cargill and
Honda.  Virent currently has 16 benchscale pilot plants in operation,
each of which is capable of producing about one gallon of product per
day.  In March of 2010, Virent announced that they had begun operating a
larger pilot plant capable of about 30 gallons per day. 
Commercialization of the Virent process will happen sometime after 2011.
 

	Biomass feedstocks for the Bioforming process are sugar feeds, such a
corn syrup, sucrose, glycerol, sorbitol, xylose, glucose, cellulose and
hemi cellulose.  These are primarily converted into gasoline and diesel
fuel, though other hydrocarbons such as jet fuel, LPG, benzene, toluene,
xylene, hydrogen and natural gas can also be produced.  Water is also
produced, as the reforming process removes oxygen from the sugar feeds. 
The resulting properties and energy content of gasoline and diesel
produced, though, are physically comparable to those yielded from
refining industry.  The variable operating costs of the Virent process
are low because no distillation equipment is needed to separate the
produced gasoline, diesel and other hydrocarbons, as these separate
naturally from the aqueous solutions generated in the reforming process.
 It appears that Bioforming is a promising technology, as production
cost estimates are low in comparison to many other renewable and biomass
production processes while the products are compatible with traditional
petroleum stocks so they don’t need special handling.   

	For this technology to become a cellulosic biofuel technology, it will
be necessary to link this reforming technology with a technology which
breaks cellulose down into starch or sugars.  In parallel with its
Bioreforming work, Virent is working on a  technology to break down
cellulose into sugars upstream of its technology which reforms sugars to
gasoline.  

 Image From: 
http://www.afdc.energy.gov/afdc/ethanol/production_cellulosic.html

 DOE. “Biomass Program: ABC’s of Biofuels”. Accessed at: 
http://www1.eere.energy.gov/biomass/abcs_biofuels.html#content

 Starch Hydrolysis; http://www.bio-link.org/pdf/starch.pdf

 Glossary of Biomass Terms, National Renewable Energy Laboratory,
Golden, CO.

   HYPERLINK "http://www.nrel.gov/biomass/glossary.html" 
http://www.nrel.gov/biomass/glossary.html 

 Glossary of Biomass Terms, National Renewable Energy Laboratory,
Golden, CO.

   HYPERLINK "http://www.nrel.gov/biomass/glossary.html" 
http://www.nrel.gov/biomass/glossary.html 

  Condensation and Hydrolysis;   HYPERLINK "http://www.biotopics" 
http://www.biotopics .co.uk/as/disaccharideformation.html 

 Aden, Andy and Ling Tao.  “Technoeconomic Modeling to Support the EPA
Notice of Proposed Rulemaking (NOPR)”.  November 3, 2008.

 Ibid

 Novozymes Press Release.  “New Enzymes Turn Waste into Fuel.” 
February 16, 2010.  Accessed at: 
http://www.novozymes.com/en/MainStructure/PressAndPublications/PressRele
ase/2010/New+enzymes+turn+waste+into+fuel.htm

 US. DOE. Technologies: Processing and Conversion. Accessed at:  
HYPERLINK
"http://www1.eere.energy.gov/biomass/processing_conversion.html on
October 28" 
http://www1.eere.energy.gov/biomass/processing_conversion.html on
October 28 , 2008

 EERE, DOE, Thermochemical Conversion, & Biochemical Conversion, Biomass
Program Thermochemical R&D . 
http://www1.eere.energy.gov/biomass/thermochemical_conversion.html
http://www1.eere.energy.gov/biomass/biochemical_conversion.html

 Aden, Andy, Mixed Alcohols from Woody Biomass – 2010, 2015, 2022,
National Renewable Energy Laboratory (NREL), September, 23 2009.

 Lin Wei, Graduate Research Assistant, Lester O. Pordesimo, Assistant
Professor

Willam D. Batchelor, Professor, Department of Agricultural and
Biological Engineering, 

Mississippi State University, MS 39762, USA, Ethanol Production from
Wood: Comparison of Hydrolysis Fermentation and Gasification
Biosynthesis, Paper Number: 076036, Written for presentation at the 2007
ASABE Annual International Meeting. Minneapolis Convention Center,
Minneapolis, MN, 17 - 20 June 2007

  Ryan Davis, National Renewable Energy Laboratory, Golden, Colorado
80401-3393, “Techno-Economic Analysis of Current Technology for
Fischer Tropsch Fuels Production”, August 14, 2009.

vi 

 J. Phillips, “Different Types of Gasifiers and Their Integration with
Gas Turbines,” EPRI/Advanced Coal Generation,
http://www.netl.doe.gov/technologies/coalpower/turbines/refshelf/handboo
k/1.2.1.pdf, October 30, 2008.

  Ciferno, Jared P., Benchmarking Biomass Gasification Technologies for
Fuels, Chemicals and Hydrogen Production, National Energy Technology
Laboratory, Department of Energy, June 2002.

 A. Aden, National Renewable Energy Laboratory, Golden, Colorado
80401-3393,  “Feedstock Considerations and Impacts on Biorefining” 
December 10, 2009

 S. Phillips, A. Aden, J. Jechura, and D. Dayton, National Renewable
Energy Laboratory, Golden, Colorado 80401-3393, T. Eggeman, Neoterics
International, Inc., Thermochemical Ethanol via Indirect Gasification
and Mixed Alcohol Synthesis of Lignocellulosic Biomass, Technical
Report, NREL/TP-510-41168, April 2007

 S. Phillips, A. Aden, J. Jechura, and D. Dayton, National Renewable
Energy Laboratory, Golden, Colorado 80401-3393, T. Eggeman, Neoterics
International, Inc., Thermochemical Ethanol via Indirect Gasification
and Mixed Alcohol Synthesis of Lignocellulosic Biomass, Technical
Report, NREL/TP-510-41168, April 2007

 Aden, Andy, Mixed Alcohols from Woody Biomass – 2010, 2015, 2022,
National Renewable Energy Laboratory (NREL), September, 23 2009.

  Davis, Ryan; Techno-economic analysis of current technology for
Fischer-Tropsch fuels production, National Renewable Energy Laboratory
for EPA, August 14, 2009.

 DOE EERE Biomass Program. “Thermochemical Conversion Processes:
Pyrolysis”   HYPERLINK
"http://www1.eere.energy.gov/biomass/thermochemical_processes.html" 
http://www1.eere.energy.gov/biomass/thermochemical_processes.html ,
November 6, 2008.

 Venderbosch, RH, Prins, W; Fast Pyrolysis Technology Development,
Biofuels, Bioprod. 4: 178-208 (2010).

 Holmgren, Jennifer; Converting Pyrolysis Oils to Renewable Transport
Fuels:  Processing Challenges and Opportunities; AM-08-81; Presentated
at the National Petrochemical and Refiners Association Annual Meeting,
March 9 – 11, 2009.

 Saffron, Christofer M.; Biomass Fast Pyrolysis; Presentated to the
Michigan Agri-Energy Conference, March 30, 2009.

 Bridgwater, AV: Thermal Conversion of Biomass and Waste:  The Status.

 Stewart, Gerald W.; Bio-Oil Commercialization Plan; Prepared for the
New Hampshire Office of Energy and Planning; July 2004.

 Chiaramonti, David; Power Generation using Fast Pyrolysis Liquids from
Biomass; July 26, 2005.

 Bridgwater, Anthony V.; Biomass Fast Pyrolysis.

 Brown, Robert C., Holmgren, Jennifer; Fast Pyrolysis and Bio-Oil
Upgrading.

 Elliot, Douglas C.; Historical Developments in Hydroprocessing
Bio-Oils; Energy and Fuels 2007, 21: 1792-1815.

(

F

X

œ

Ê

Ê

ò

7

8

9

;

<

=

h= 

h= 

h= 

{

{

{

{

{

{

{

{

{

{

{

{

{

{

{

j»

ü

萑ː葠ː摧䂕D

—

û

ü

×

Ü

ã

ä

è

h

h

h

ht

摧氷0

ᄀ킄㜂$␸䠀$葠ː摧氷0

Jones, SB; Production of Gasoline and Diesel from Biomass via Fast
Pyrolysis, Hydrotreating and Hydrocracking:  A Design Case; Pacific
Northwest National Laboratory (PNNL), February 2009.

 Marinangeli, Richard; Opportunities for Biorenewables in Oil
Refineries; Final Technical Report from UOP, NREL, PNNL and MTI to DOE;
December 12, 2005.

 Dynamotive Brochure.

 Goodfellow, Randal; Biomass to Green Hydrocarbons; November 6, 2008.

 E-mail from Andrea Pinho, Senior Researcher FCC Catalysts, Petrobras.

 Huber, George W.; Synergies between Bio- and Oil Refineries for the
Production of Fuels from Biomass; Angew Chem Int Ed 2007, 46: 7184-7201.

 Carlson, Torren C., Green Gasoline by Catalytic Fast Pyrolysis of Solid
Biomass Derived Compounds; ChemSusChem 2008, 1: 397-400.

 Huber, George H.; Breaking the Chemical and Engineering Barriers to
Lignocellulosic Biofuels:  Next Generation Hydrocarbon Biorefineries;
Based on the June 25-26 2007 Workshop, Washington DC, Published March
2008.

