Biological delignification of wood and straw for ethanol production via solid state culture

S BiolojiicaL

662.6f>9 dellgnlf ica tlon of
.V7biw wood and straw for

1989 ethanoL production

via solid state

cul ture

T RENEWABLE ENERGY

I.-

STATE DOCUMENTS COLLECTION

REPORT LIBRARY , ,,

MONTANA STATE LIBRARY

1515 E. 6th AVE.
HELENA, MONTANA 59620

BIOLOGICAL DELIGNIFICATION OF WOOD
AND STRAW FOR ETHANOL PR0DUCTI0r4
h via SOLID STATE CULTURE

NOVEMBER 1989

PLEASE RETURN

Prepared for
!*• MONTANA DEPARTMENT of NA TURAL RESOURCES and CONSERVA TION

AUG 2 2 1990

MAY 1 ^ 1991
JUL 1 7 1991

MONTANA STATE LIBRARY

S 6«2 M9 N7ba" 1 9»9 c 1

3 0864 00066314 9

S

662. 66;:^

1989

vl* solid state
culture

FINAL REPORT

DNRC GRANT AGREEMENT NO. RAE-86-1066

Biological Deliqnif ication of Wood and Straw

for Ethanol Production

via Solid State Culture

Submitted to:

Montana Department of Natural Resources and Conservation

1520 East 6th Avenue, Third Floor

Helena, Montana 59620

Prepared by:

Clifford Bradley

Pauline Wood

Robert Kearns

Bill Black

of

Mycotech Bioproducts , Inc .

P.O. Box 4113, 630 Utah Ave

Butte, MT 59702

Phone: (406)782-2386

November 15, 19 89

NOTICE

This project was undertaken in part with government support
under contract no. DE-FG79-82BP35776 awarded by the Bonneville
Power Administration and a grant awarded by the Montana
Department of Natural Resources and Conservation. Such support
does not constitute an endorsement by EPA or DNRC of the views
expressed in this work, and the user agrees to hold EPA and DNRC
harmless against any direct or consequential damages claimed by
the user or their parties arising from or related to use or
interpretations of this report.

RAE-8«-1066; BT601 11

TABLE OF CONTENTS

Introduction 1

Research Methodology 5

Laboratory and Pilot Plant Equipment 6

Organisms 9

Analytical and Process Analysis Procedures 9

Direct culture pretreatment methods 11

Ligninase pretreatment methods 12

Results and Discussion 17

Description of Process Options 17

Direct Culture Pretreatment 18

Ligninase Pretreatment 20

Whole-cell ligninase 21

Cell-free ligninase 22

Overall Process Description of Lignocellulose

to Ethanol 26

Preliminary Economics 27

Results and Conclusions 28

Appendices 31

1) Work Schedule 31

2) Budget 31

3) Technology Transfer/Commercialization

Activities 33

4) References 33

RAE-86-1066: BT601 HI

INTRODUCTION

Lignocellulosic materials are a promising feedstock for
alcohol fuel production. Lignocellulosic materials are abundant
and renewable. As an alternative energy source, low-value or
waste lignocellulosic material can be used without affecting
supplies of food and fiber.

Millions of tons of low-value or waste lignocellulosic
material are available in Montana as byproducts of the
agriculture and forest products industries. Much of this
material is discarded, which creates solid waste disposal
problems. With efficient low-cost conversion technology, wastes
which are now an economic liability can be turned into an asset
as an energy source.

Lignin, a complex polymer of phenylpropanoid molecules in
nonlinear random linkages, is the major structural component of
plants. Cellulose is encased in a lignin matrix which must be
degraded before the cellulose can be enzymatically hydrolyzed to
fermentable sugars .

Most research in lignocellulose conversion has concentrated
on mechanical and chemical pretreatments to remove lignin and has
emphasized enzymatic hydrolysis of cellulose fractions.
Pretreatments often involve heat and are capital and energy
intensive. Chemical solvent extractions also have a potentially
adverse environmental impact.

RAE-86-1066: BT601

A report by the Solar Energy Research Institute contains a
detailed description and economic analysis of lignocellulose
mechanical/chemical pretreatment followed by enzymatic hydrolysis
and fermentation of the cellulose. Two of the most limiting
economic factors in alcohol production from lignocellose are
pretreatment and cellulase costs.

In research supported by the U.S. Department of Energy under
the Small Business Innovation Research Program (SBIR), Mycotech
adapted solid state culture (SSC) technology to production of a
low-cost cellulase for use in biomass conversion.^ This work
successfully developed a cellulase preparation that is about one-
fifth the cost of liquid culture preparations for equivalent
activity levels. The potential to use Mycotech 's SSC technology
to produce lignin degrading enzymes led to the concept of a
completely biological process for producing alcohol from
lignocellulosic feedstocks.

White rot fungi are a class of wood decaying organisms that
enzymatically degrade lignin.^ One organism, Phanerochaete
chrysosporium has been the subject of considerable research.''
The lignin degrading enzyme system produced by this organism in
nitrogen limited, stationary liquid culture systems has been
purified. White rot fungi have high oxygen requirements'^'''® and
unusual responses to nitrogen and carbon concentration in liquid
growth media. '^° As a result it has been difficult to develop
efficient liquid culture systems that can be scaled up

RAE-86-1066; BT601

economically for ligninase' production or lignocellulose

pretreatment. White rot fungi produce ligninases when grown on

solid substrates; however, commercial application is seen as

infeasible because of long culture times and because solid

culture technology is generally regarded by the U.S. fermentation

industry as unconventional and difficult to apply on a large

scale .

With support from DNRC^^ and USDOE,^ Mycotech developed

innovative solid culture technology that is:

low in capital and operating cost

suitable for large scale operations

widely adaptable to fungi that perform poorly in liquid

culture

This technology has been developed on a small scale for a
pilot project to produce amylase and cellulase enzyme
preparations. It is based on innovations in substrate
characteristics, culture reactor design, and monitoring and
control systems. Computerized monitoring and control systems
were developed with support from DNRC."

Mycotech 's proposal to develop a process for biological
delignif ication arose out of the company's SSC capability, the
development of a low-cost cellulase, and the availability of
sophisticated SSC monitoring and control systems. The proposal
was submitted in March 1986, and the grant was awarded June 1986.

• The term ligninase will be used throughout this report to refer to the multiple enzymes involved in
degradation or modification of lignin.

RAE-86-1066: BT601

During this project, two other contracts were obtained for
work with white rot fungi. The first was a U.S. Department of
Energy phase I SBIR project to evaluate solid culture for
production of cell-free ligninase enzyme preparations. The
second was a contract with a private European company to produce
ligninase preparations from a variety of white rot fungi for
application testing in improving paper pulp. These projects
provided a much wider range of enzyme samples for analysis in the
DNRC project than would otherwise have been available.

RAE-86-1066, 8T601

RESEARCH METHODOI/XSY

The overall objective of the project was to find a process
for biological delignif ication that could be integrated with SSC
technology to produce and use a low-cost cellulase on Montana-
specific lignocellulosic materials to produce ethanol. Meeting
this objective would result in a low-cost biological technology
which integrates lignin removal, cellulose hydrolysis, and
fermentation to convert lignocellulose to alcohol fuel. Work
emphasized the use of low-value or waste material from forestry
and agriculture products.

A literature search revealed that none of the existing
processes are commercially promising. The literature on liquid
culture ligninase sometimes conflicts with solid culture results,
particularly over nutrient and oxygen responses and whether
liquid culture ligninases will completely depolymerize

1 ,• ^^i ^ 12,13,14,15

iignin .

The literature does suggest two principal alternative routes
to developing a biological delignif ication process. The first
alternative is a delignif ication pretreatment process by direct
culture of the fungus on the fermentation feedstock. This
approach is suggested by studies which evaluated cultures of
white rot fungi on plant residue^* and on wood and kraft wood
pulp as a bleaching step.^^'^® This alternative is referred to
throughout this report as direct culture pretreatment. The
second alternative would use SSC for culture of white rot fungi

RAE-86-1066: BT601 5

to produce a commercial ligninase preparation. The ligninase
preparation would be used to pretreat lignocellulose prior to (or
simultaneously with) enzymatic cellulose hydrolysis. The
feasibility of this approach is suggested by RTI ' s work with
cellulase production and the studies^**'^ describing production of
extracellular ligninase by white rot fungi. This procedure is
referred to throughout this report as ligninase pretreatment . A
diagram of alternative process and options evaluated in the
project is shown in Figure 2 in the next section.

Much of the work was devoted to developing methodologies and
analytical procedures as a basis for evaluating the process
options. This section will describe laboratory and pilot plant
equipment, laboratory processes used in evaluating direct culture
and ligninase pretreatment, and analytical procedures.

Laboratory and Pilot Plant Equipment

Three types of experimental solid culture systems were used
in the project. These were: the 10-culture test stand with
computer monitoring and control developed as part of previous
DNRC sponsored work' (Contract RAE 85 1055); pilot reactors
constructed as part of the DOE, SBIR phase II cellulase project^;
and a new laboratory test stand built for this project. The new
test stand was developed without sophisticated monitoring and
control systems as a low-cost way to increase capacity. This was
necessary because of long culture times, the need to screen

RAE-86-1066; BT601

substrates, and the need to evaluate a wide range of variables in
direct culture pretreatment processes. The test rack was built
to hold 15, 250 cc polycarbonate culture tubes, each with
individual manual control for humidified air flow. Two gases,
typically air and oxygen, could be blended. Temperature was
controlled by an air heater in the enclosed test stand. This was
a modification of the proposed design to allow more flexibility
in handling cultures . Figure 1 is a photograph of the test
stand.

Figure 1. Laboratory SSC test stand,

RAE-86-1066: BT601

Pilot reactors were 6-inch X 30-inch glass columns holding
about 2 kg of dry weight substrate. Reactors were modified by
changes in air systems. In some experiments, reactors were
operated horizontally with 500 grams of substrate in a 3-inch
deep bed on a tray. In later experiments, columns stood
vertically and were filled to about 28 inches deep. Reactors
were equipped with compressor or fan driven air systems and
humidifiers. Air flow and pressure could be regulated either at
the inlet or outlet of the reactor. Monitoring included air flow
rate and pressure, temperature, humidity level, and oxygen
concentration in exhaust gas.

An air shaker with a capacity of 25, 125 ml flasks or 16,
250 ml flasks was also purchased for the project to increase
capacity for hydrolysis and fermentation experiments.

A variety of Mycotech's general laboratory support equipment

was used in the project. Key equipment included:

- Blenders

Centrifuge and filter systems used in enzyme extraction

Shimadzu 260 recording spectrophotometer used in enzyme

and colormetric reducing sugar assays and in evaluating

lignin solubilization in pretreatments

Yellow Springs Instrument Model 23A enzymatic glucose

analyzer used in hydrolysis studies

Varian model 3700 gas chromatograph used for ethanol

determinations

Clean rooms used for maintenance and transfer of white

rot fungus cultures

RAE-86-1066: 8T601

Organisms

Organisms tested in the project were selected from
literature review^' and because they were being used in related
contract work. Selection criteria included such factors as known
production of different types of ligninases and reports of tests
in solid culture. Five organisms were tested in various process
alternatives: Phanerochaete chrysosporium , ''^ Pleurotus
gstreatus./' Phlebia (Merulius) tremellosus /'' Trametes
versicolor .^°'^^ and Bierkandera adusta.*^^ Cultures were obtained
from public culture collections and private companies. Cultures
were maintained on agar slants with periodic transfer to fresh
media. Inoculum cultures were fresh agar slants or liquid
cultures in either carbon nitrogen limited salts media or
glucose, peptone, yeast extract media. Bierkandera inoculum
cultures were grown with agitation, while other organisms were
generally grown in stationary culture.

Analytical and Process Analysis Procedures

Many methods have been used to directly measure lignin
biodegradation: release of radiolabeled carbon dioxide from ^'c
lignin, high performance chromatography of soluble lignin
degradation products, UV absorption scans of soluble lignin
degradation products, and determination of substrate lignin
content using digestion methods .^''■^° The common disadvantage of
all these methods is difficulty in correlating values of lignin
degradation with the bioconversion efficiency of pretreated

RAE-86-1066; 8T601 9

lignocellulose feedstocks. As a result, the principal method to
evaluate delignif ication in the project was to measure enzymatic
cellulose hydrolysis of pretreated lignocellulose. This is an
indirect measure of delignif ication but does provide a direct
measure of the end product of an enzymatic delignif ication and
bioconversion process - namely glucose or ethanol. With
hydrolysis conditions held constant, variation in the rate and
extent of cellulose hydrolysis is a measure of delignif ication
efficiency.

Hydrolysis was assayed by glucose or total reducing sugar
assay. In most cases, glucose assay of hydrolysis was used to
evaluate delignif ication processes. However, in some cases yeast
was added to the system and results determined as ethanol. A
commercial strain of Saccharomyces cerevisciae (Gist Brocades
Fermipan) , previously used in cellulose hydrolysis and
fermentation, was used in these experiments.

In general, cellulose hydrolysis was carried out using
either Mycotech's solid culture cellulase or a commercial
cellulase, Genencor 150L, at a dose of 1 percent w/w (0.1 g
cellulase per 1 g of dry weight lignocellulose). Hydrolysis was
supplemented with a cellobiase preparation (Novo 188) to ensure
that soluble cellodexdrins were quantitatively converted to
glucose. When hydrolysis was carried out as a distinct process
step, conditions were pH 4.8 in 0.4 M acetate buffer at 45° C for
24 hours. Conditions were varied when hydrolysis was carried out
simultaneously with ligninase treatment and/or fermentation.

RAE-S6-1066; BT601 10

Some of the analytical methods employed were specific to one
of the two processes - direct culture pretreatment and ligninase
pretreatment - evaluated in the project.

Direct culture pretreatment methods

In the direct culture process, white rot fungi were grown on a
lignocellulose feedstock. After a suitable culture period the
material was washed, suspended in buffer, and tested for
cellulose hydrolysis. In some cases, the initial wash was made
quantitatively and delignif ication was evaluated by UV scan (320-
190nin) to indicate the level of soluble, UV absorbing, lignin
degradation products.

Variables affecting the direct culture process include
organism, inoculum, substrate treatment, culture nutrients, time,
temperature, and aeration/Oj enrichment. Lignocellulose
feedstocks tested were barley straw, pine and fir sawdust, wood
chips (pine), brown paper bag, and two types of wood pulp (kraft
process hardwood and softwood pulp from the International Paper
Company) . Organisms tested were Phanerochaete , Phlebia,
Pleurotus and Trametes (these experiments were conducted prior to
receipt of Bjerkandera) . In general, a liquid culture started
from a slant was used as an inoculum, although spores from slant
cultures were also employed. Lignocellulose materials were
wetted to 60 - 70 percent moisture content with a nutrient
solution, autoclaved, inoculated, and transferred to 250 cc
columns. Aeration was generally at 25 cc per minute with air or
with air supplemented to about 60 percent oxygen. Culture

RAE-86-1066; ST601 11

temperature was 30° C for all organisms except Phanerochaete
which was grown at 37° C. Nutrient solutions were varied. A
basic salts solution using low levels of inorganic nitrogen was
used for nitrogen limited conditions. The nutrient solution was
supplemented with varying levels of peptone, yeast extract, and
glucose for high nitrogen/carbon conditions. Veratryl alcohol at
0.7 g/1 in nutrient solution served as an inducer in some
experiments .

Ligninase pretreatment methods

In ligninase pretreatment, white rot fungi were grown under
conditions (determined in related contract work) to produce
ligninase. These enzyme production cultures were used either as
solid, whole-cell preparations or as cell-free liquid extracts in
pretreatment of lignocellulose feedstocks. Barley straw milled
to pass through a 20-mesh screen (0.850mm) was used in all
ligninase pretreatment experiments. In some cases, straw was
water washed and dried prior to use.

For whole-cell preparations, whole wet culture was simply
mixed as is, at varying levels, in buffer suspensions of barley
straw and incubated for 12 hours to 7 days. Tests included
whole-cell preparations from Phanerochaete , Trametes, Phlebia .
and Bjerkandera.

Cell-free enzyme preparations were made by extracting enzyme
production cultures in water or salt solutions followed by
centrifuging and filtering through 0 . 8u polypropylene filters.
Two organisms, Trametes and Bjerkandera , were used for cell-free

RAE-66-1066; BT601 12

ligninase preparations. Two ligninase preparations from liquid
culture of Phanerochaete were obtained from another company and
used for comparison.

Pretreatment reactions with peroxide and manganese dependent
ligninase activities were supplemented as appropriate. Reaction
buffers were acetate pH 4.5, tartrate pH 3.5, tartrate pH 4.0,
malanate pH 3.5, malanate pH 4.0, or phosphate pH 5.7. After
treatment, straw was either removed, washed, and resuspended in
cellulose hydrolysis buffer, or buffer and cellulase were added
directly to the reaction flask. Wash liquid was evaluated by UV
scans .

Different enzyme assay procedures were used to evaluate
enzyme preparations depending on the organism and type of
ligninase activity. Enzyme assays were based on oxidation of
lignin model compounds - veratryl alcohol,* phenol red,^' and anis
alcohol. ^^ Veratryl alcohol and anis alcohol are oxidized to the
aldehyde with a corresponding increase in optical density at
310nm for veratryl aldehyde and 285nm for anis aldehyde. In the
phenol red assay, oxidation was carried out at acidic pH (3.5 -
5.5). When the reaction is stopped by changing pH to alkaline
conditions, the degree of oxidation is determined by a change in
absorption of 600nin.

RAE-86-1066: BT601

13

The choices of organism and culture conditions were used to

produce four different types of ligninase activities from solid

culture:

Peroxidase: Hydrogen peroxide dependent oxidation of
phenol red or veratryl alcohol.

Mn Peroxidase: Hydrogen peroxide and manganese
dependent oxidation of phenol red or veratryl alcohol.

Mn dependent Oxidase: Manganese dependent oxidation of
phenol red,

Oxidase/Laccase: Oxidation of phenol red, veratryl alcohol,
or anis alcohol without hydrogen peroxide or manganese.

Each organism produced one or more of the four types of

ligninase. However, the different types of enzyme from Trametes

and Bierkandera solid culture and Phanerochaete liquid culture

showed significant differences in oxidation of the different

lignin model compounds. This indicates that the same type of

enzyme (i.e., peroxidase) from the different organisms and

culture systems functions differently.

By varying culture conditions, Trametes cultures could be

manipulated to selectively produce each of the four types of

activities. Trametes enzyme would oxidize phenol red but not

veratryl or anis alcohol. Culture extracts were assayed at pH

4.5 in acetate buffer with all combinations of the presence or

absence of peroxide and manganese to distinguish the different

types of activity. Bierkandera cultures could be manipulated to

selectively produce either a manganese dependent peroxidase or an

oxidase. Both enzymes oxidized anis alcohol and veratryl

alcohol .

RAE-66-1066; BT601 14

The two Phanerochaete liquid culture preparations were: a
peroxidase (not requiring manganese) that oxidized veratryl
alcohol but only poorly oxidized phenol red and anis alcohol, and
a laccase that oxidized phenol red but no other compounds.

Assays did not distinguish whether the non hydrogen
peroxide, non manganese dependent Trametes activity was an
oxidase or a laccase. The Bierkandera enzyme was an oxidase,
requiring molecular oxygen for activity and generating H2O2 as a
reaction product. ^^ The Phanerochaete liquid culture was
described by the donor as a laccase.

These enzyme preparations provided a wide spectrum of
ligninase activities. They were tested singly and in various
combinations in the pretreatment of barley straw. Four different
delignif ication process options were tested using cell-free
ligninase preparations as shown in Figure 2 in the next section.
In sequential process steps, reaction conditions of pH, buffer,
and temperature could be varied for ligninase treatment and then
changed to optimum conditions for cellulase. In simultaneous
treatment steps, compromise conditions were employed. Generally,
Trametes enzyme was used in acetate buffer pH 4.5 - 5.0,
Bierkandera enzyme in phosphate buffer pH 5.0 - 5.7, and
Phanerochaete liquid culture preparations in tartrate pH 3.5 -
4.5. However, a large number of variations were tested.
Hydrogen peroxide and manganese were added when using
peroxidases .

RAE-86-1066: BT601 15

Controls were run with all experiments. The basic procedure
was to run controls without ligninase through the entire
pretreatment (duplicating incubation times, different buffers,
etc.) and through the hydrolysis procedure being tested. This
provided a background hydrolysis level for straw in each
treatment variation tested.

RAE-86-1066: BT601 16

RESULTS AND DISCUSSION

The two overall process and the options within each process
that were tested during the project are described below and shown
in Figure 2 .

Description of Process Options

1) Direct Culture Pretreatment

Delignif ication by direct culture of white rot fungi on
lignocellulose feedstock. Two process options for
hydrolysis and fermentation of pretreated feedstock were
tested.

A. Sequential cellulose hydrolysis and fermentation
in two separate process steps

B. Simultaneous cellulose hydrolysis and fermentation
in a single process step.

2) Ligninase Pretreatment

Delignif ication by treatment with ligninase enzyme
preparations. Fungus is grown using substrates and culture
conditions for optimal ligninase production. Six process
options were tested.

Whole-cell Ligninase: Whole culture used as ligninase
preparation.

A. Sequential ligninase pretreatment and cellulose
hydrolysis in two separate process steps.

B. Simultaneous ligninase pretreatment and cellulose
hydrolysis in one process step.

Cell-free Ligninase: Ligninase production culture extracted
and filtered to remove cells and culture solids.

C. Sequential ligninase pretreatment, cellulose
hydrolysis, and fermentation in three separate
process steps.

D. Sequential ligninase pretreatment, and
simultaneous cellulose hydrolysis and fermentation
in two process steps .

RAE-86-1066; BT601 17

0)

E
cS

0)
0)

3

o

o

0)

m

C
0)

E

0)

"S

o

CD

c
c

CN

^

I i

lU

03

E

>^

N

C

LU .^

cu

(/) >-

=3 =3

D) ^

c ,=^

^ o

u- c

o o

QC o

Q) -^

.-^ TD

J= O

^ ti

Z

o

H

<

O CO

1^5

z2

cvj

55S:

CD

rH o

D

LU

Q C/)

. cn 1

ICAL
OCE

(3 CC

O Q-

^

O

CD

E. Simultaneous ligninase pretreatment and cellulose
hydrolysis followed by fermentation in two process
steps .

F. Simultaneous ligninase pretreatment, cellulose
hydrolysis, and fermentation in a single process
step.

Direct Culture Pretreatment

Direct culture pretreatment experiments evaluated four
different fungi grown on seven different representative
lignocellulosic materials. Steam-rolled barley was used as a
control with known solid culture characteiistics .

Best cell growth was obtained using straw with 1-5 percent
ground paper bag. Very high cell densities were obtained in 7-
20 day culture times. Wood pulp also supported good growth, and
in samples returned to International Paper, showed reduction in
lignin content as measured by brightness and Kappa number
(residual lignin content). Fir sawdust showed no growth and pine
sawdust minimal growth.

Experiments were repeated using straw as the substrate with
variations in culture time, nutrients, oxygen enrichment, and
organism. In the majority of experiments, subsequent hydrolysis
with cellulase showed no significant difference in glucose yield
or extent of hydrolysis compared with untreated controls. In
most cases glucose yield was less than in controls. With some
Phanerochaete and Pleurotus treated straw, there was about a 10
percent increase in glucose yield. In these experiments, the
extent of hydrolysis of untreated straw controls was about 9
percent, and the best level achieved with direct culture

RAE-86-1066; BT601 18

pretreatment was about 10 percent. This compares with about 50
percent hydrolysis obtained with straw chemically delignified by
autoclaving in 1 percent NaOH, followed by washing and
neutralization.

In treatments showing increased cellulose hydrolysis the
amount of glucose produced was to small to be practical. The
decreased hydrolysis seen in most experiments was probably due to
the use of cellulose by the fungus thereby reducing the cellulose
available for subsequent enzymatic hydrolysis. Assays of
Pleurotis and Phanerochaete culture showed cellulase activity.
The cellulase concentration was probably sufficient to hydrolyze
readily accessible cellulase in the straw with the resulting
glucose used by the fungus .

Several experiments were conducted in which glucose or
starch was added to the cultures in an attempt to repress
cellulase production. No repression was found. Enrichment of
the culture atmosphere up to 100 percent oxygen had no effect on
subsequent hydrolysis results.

Water extracts of culture material did show much higher
levels of soluble lignin than controls as shown by UV absorption.
Often straw would be visibly lighter in color after direct
culture pretreatment. Both of these results indicated extensive
delignif ication of the straw; however, there was never a
significant increase in cellulose hydrolysis.

RAE-86-1066: BT601 19

Ligninase Pretreatment

Pretreatment with a ligninase preparation has a wider range
of process options than direct culture as shown in Figure 2. in
this process, the fungus is grown under conditions for optimal
production of specific ligninase enzymes. The whole culture may
be used as an enzyme preparation, or the ligninase can be
extracted to produce a cell-free liquid enzyme preparation. A
whole-cell preparation might be the best option for on-site
enzyme production and biomass utilization, while a cell-free
enzyme could be produced and marketed from a central location.
Whole-cell preparations could contain enzyme activities bound to
culture substrate that would be lost in extraction. Cell-free
enzyme preparations may not contain inhibitors or undesirable
activities that could be present in whole-cell culture
preparations. With each type of enzyme preparation, the
enzymatic ligninase and cellulase reactions and fermentation
steps were tested both sequentially and simultaneously.
Sequential processes have the advantage that each reaction can be
run at optimal conditions. Simultaneous processes have the
advantage of simplicity in process design and equipment.

Whole-cell enzyme preparations were evaluated in sequential
and simultaneous delignif ication and cellulase hydrolysis. Cell-
free enzyme preparations were tested in both sequential and
simultaneous enzyme steps and in enzyme steps coupled with
fermentations. Fermentations were run as a third sequential

RAE-86-1066; BT601 20

step, simultaneously with the cellulase hydrolysis step and
simultaneously with both enzyme reactions.

A number of variables were evaluated for each step in each
of the process options. These included:

- Straw concentration

- Ligninase source organism

- Ligninase type (peroxidase, oxidase, or mixtures)

- Ligninase concentration and reaction time

- Buffer system and pH

- Control of bacterial and fungal growth

- Cellulase preparation and concentration

- Fermentation conditions

Whole-cell ligninase

In experiments with whole-cell ligninase preparations, straw
concentration was a slurry of 10 percent straw or a moist solid
at 30 percent straw. In sequential cellulose hydrolysis, straw
concentration was 5 percent after all additions of buffer and
enzyme. Trametes preparations were cultures with peroxidase,
manganese-dependent peroxidase, or oxidase activities.
Bjerkandera preparations contained manganese dependent peroxidase
or laccase, and Phlebia contained peroxidase. Trametes and
Bjerkandera enzymes were tested singly and in numerous ^
combinations. Liquid culture Phanerochaete peroxidase and
laccase preparations were added to whole-cell preparations in
some experiments. Enzyme concentrations of 10 - 50 percent of
the weight of straw were tested with reactor times of 12 hours to
seven days. Acetate and tartrate buffers and unbuffered systems
were tested. Straw was used either sterilized in buffer or
unsterilized with ammonium bifluoride or antibiotics added in

RAE-86-1066, BT601 21

some cases to suppress bacterial growth. Cellulase was a
commercial preparation from Genencor.

Results showed delignif ication with solxd culture Trametes
and liquid culture Phanaerochaete preparations as measured by UV
absorption of lignin degradation products in wash water from
treated straw. Inactivated whole culture ligninase preparations
were used in these experiments to control for lignin degradation
products introduced with the crude enzyme preparations. In some
cases, washed treated straw was visibly lighter in color. These
indications of delignif ication did not correlate with significant
increases in cellulose hydrolysis. Best results were about 10
percent conversion of total straw weight compared with about 8 -

9 percent in controls. Both the extent of conversion and
relative increase compared with controls were not significant.

Cell-free ligninase

Cell-free enzyme preparations were tested extensively using
solid culture extracts with peroxidase and oxidase from Trametes
and Bierkandera and the liquid Phanerochaete preparations.
Ligninase preparations were used singly and in all possible
combinations in both sequential and simultaneous cellulase
hydrolyses .

An extensive set of tests were conducted to follow stability
of ligninase in these reactions. The first experiments evaluated
binding to straw. With this baseline, delignif ication reactions
were sampled and assayed for enzyme activity over time. All
enzyme preparations showed some binding to straw. Stability

RAE-86-1066; BT601 22

assays showed that Trametes enzyme could not be recovered from
reaction mixtures after a few hours. This activity, particularly
the oxidase, is either unstable or becomes irreversibly adsorbed
to straw. Both Bjerkandera and Phanerochaete enzymes could be
recovered even after three days of reaction, indicating good
stability and equilibrium binding conditions.

A variety of buffer systems and pH profiles were analyzed.
Optimum conditions for each ligninase were used for single enzyme
reactions . In sequential processes with mixed ligninase
preparations and in simultaneous cellulase hydrolysis, compromise
conditions were tested. Enzyme assays were run on each enzyme
preparation and type of ligninase activity to establish buffer
and pH range for activity. In mixed ligninase systems,
conditions were found in which all components of the mixture
showed at least a significant fraction of optimal activity.
Cellulase preparations have a pH optimum of about 4.8; however,
the activity is retained over a broad pH range. Therefore,
conditions favorable for ligninase activity could be employed in
simultaneous ligninase/cellulase systems.

Reaction times from 12 hours to seven days were run in
different experiments with liquid ligninase concentrations as
high as 50 percent by volume of reaction mixtures.

In experiments with long culture times, it was necessary to
control bacterial contamination and fungal growth from ligninase
and cellulase preparations. A series of experiments were
conducted to test the effects of several different chemical and

RAE-W-1066; BT601 23

antibiotic control agents on ligninase and cellulase activities.
Combinations were found that did not affect enzyme activity and
maintained uncontaminated reactions for up to a week.

Three different cellulases were employed in these
experiments. The liquid, commercial cellulase is high in
specific (endo- and exo- ) cellulase activities, but very low in
"side" activities, particularly hemicellulose degrading xylanase.
Reasoning that hemicellulose degradation may be important in
pretreatment , we tested two cellulase preparations v/ith high
levels of xylanase. These were a preparation from Amano and
Mycotech solid culture cellulase preparations.

Ferinentations were run using buffer and bacterial
suppression that did not affect yeast. Fermentation controls
were run using hydroxide delignified straw in addition to straw
without ligninase treatment.

Results of experiments with cell-free ligninases (extracts)
were similar to other process experiments. Straw showed evidence
of delignif ication, but economical levels of conversion to
glucose or ethanol were not obtained. Best results were obtained
in experiments using combinations of peroxidase and oxidase type
ligninase activities as a pretreatment followed by simultaneous
hydrolysis and fermentation of the cellulose. (Figure 2, option
D.) In some experiments, alcohol concentrations of 0.26 - 0.37
mg/ml were obtained compared with 0.15 - 0.16 mg/ml in controls.
Although twice the alcohol concentration was obtained, overall
conversion of straw was only about 10 - 12 percent. This

RAE-86-1066; BT601 24

compares poorly with 50 percent overall conversion of straw that
is hydroxide pretreated and hydrolyzed.

RAE-86-1066: BT601 25

nvT?RAT.T. PRfX:ESS DESCRIPTION OF LIGMOCELLULOSE TO ETHANOL

The overall description of alternative processes for
conversion of lignocellulose to ethanol was shown in Figure 2 in
the previous section. None of the process alternatives we tested
showed economically feasible conversion rates. The best results
were obtained with delignif ication using a mixed cell-free
ligninase enzyme preparation followed by simultaneous cellulase
hydrolysis and fermentation. (Process option D, Figure 2.) A
schematic of this type of process is shown in Figure 3.

An overall mass and energy balance can not be established
from the available data. Overall conversion was too low to
establish ligninase dose response or lignin removal. Alcohol
concentrations in fermentations were only about 0.2 - 0.4 mg/ml
which is too low to economically distill without a concentration
step. Evaluation of the mass and energy balance of a dilute beer
concentration step were beyond the scope of this project.

RAE-86-1066; BT601 26

o

3

•a
o

a.

»

■

>

>

c
o

o
= o

O V

(0 cc

c
o

g >, c

Ip

0) >. V

U X u.

g
o

"O

o
qI
O

CO

en

5

<

oc;

1

0)

(5

M

oi5

C

i:

0)

3

cn

5

O

z

C/)

oO

C/)

1P>^

c c .

.2 Q

o 00

CT3 '-

§ u.

UJ

C '

d)

o

tf)

oducti

ndera
Oxida

k_

lO

Q_

■:« C~-

><- "■=

o

CO
CO

CQ ^

_l

o

I

z

<

I

t^

o3

i-u!^

^H:|(/)<

coGoi;ja:

0) OujO

3 =Jo<

LU Q.

o

o

z

o

_I

PRELIMINARY ECONOMICS

Best results with enzymatic delignif ication showed about a
10 percent overall conversion of straw to ethanol on a weight
basis. The alcohol yield and value for conversion of one ton of
straw is shown below.

2000 lbs straw X 0.10 conversion = 200 lbs glucose

200 lbs glucose X 0.51 theoretical molar fermentation
efficiency = 102 lbs ethanol

102 lbs ethanol -r 6 . 6 lbs ethanol per gallon = 15.4
gallons

15.4 gallons X $1.55/gallon selling price for ethanol =
$25.41.

In southwest Montana, the cost for straw collected and
delivered to a central point is in the range of $25 - $30 a ton.
Conversion efficiencies obtained in the project would not cover
feedstock cost for alcohol production. Results did not justify
further in-depth economic analysis.

RAE-86-1066: BT601 27

RESULTS AND CONCLUSIONS

The project did not demonstrate economic feasibility of a
completely biological conversion of lignocellulose to ethanol.
Results consistently showed increased cellulose hydrolysis of
ligninase pretreated straw compared with controls; however, the
increase fell far short of conversion efficiencies necessary for
economic ethanol production.

Clearly, straw was delignified, but the question remains:
why did this not correlate with increased cellulase hydrolysis.
The extent of delignif ication may have been insufficient to
expose a significant fraction of the cellulose for enzymatic
hydrolysis. Alternatively, there may be other factors critical
to delignif ication in which the enzymatic ligninase and cellulase
preparations were not effective. These could include secondary
structures such as linkages between cellulose, hemicellulose, and
lignin, or tertiary structure limiting access of the enzyme.

Delignif ication might be increased by the use of higher
enzyme concentrations, longer reaction times, or different types
of activities. Economics of these steps would be questionable as
the project tested a very wide range of ligninase sources,
concentrations, and long reaction times without achieving good
yields .

There may be some promise in testing a pretreatment that
combines enzymatic, chemical, and mechanical processes. The cost
and efficiency of chemical/mechanical delignif ication might be

RAE-86-1066; BT601 28

improved sufficiently by an enzymatic step to justify enzyme
cost. A mild chemical treatment to disrupt secondary structure,
or milling to disrupt tertiary structure, could improve enzymatic
delignif ication enough for a conversion process to be economical.
If further research were to be conducted based on this project, a
combination of mechanical, chemical, and enzymatic pretreatment
is one recommended approach.

One other important factor may have been the choice of
lignocellulose feedstock. Barley straw was the primary feedstock
for this project because it is low in cost and readily available
in southwest Montana. Other feedstocks, such as, aspen, poplar,
or other agricultural residues, may be more amenable to enzymatic
pretreatment. This is supported by recent work in Canada^^ using
aspen wood delignified by direct culture treatment with Phlebia
tremellosus . After a 12 week culture time, cellulose hydrolysis
and fermentation of delignified aspen wood yielded 0.116 g
ethanol per g of wood - an overall conversion efficiency of about
23 percent of theoretical. There may also be partially
delignified waste material such as waste paper in which an
enzymatic delignif ication would be a cost effective conversion
step. These types of alternatives were too wide ranging to be
effectively addressed in this project.

In conclusion, the project did make a significant, if
negative, contribution to work in this field. Because of
integration with two other projects related to white rot fungi,
Mycotech was able to test a large number of different enzyme

RAE-86-1066; BT601 29

preparations under a very broad range of conditions. A
comprehensive evaluation of different types of ligninases from
five different fungi does provide a good basis for further work,
and probably indicates that factors other than ligninase need to
be addressed in future work on enzymatic pretreatment .

RAE-86-1066; BT601 30

APPENDICES

1) Work Schedule

The time line of the original contract was extended
considerably from April 30, 1988 to September 30, 1989. The
principal reason for this was that initial experiments with the
direct culture process route were unsuccessful and efforts were
focused on using ligninase preparations. Mycotech received two
other projects that included work to produce and evaluate
ligninase enzyme preparations from different organisms.
Extending the DNRC project provided a unique opportunity to test
enzyme produced for these other projects in the DNRC biomass
conversion project. The increased time allowed testing of a much
wider range of organisms and enzyme preparations, and to have
more completely characterized these enzyme preparations, than
would have been possible within the scope of the DNRC project
alone.

2) Budget

Budgeted and actual expenses by project milestone are shown
below. The major changes in budgeted vs. actual expenses were in
subcontracts and associated travel expenses. Results did not
justify subcontractor tests of ligninase preparations. As a
result, the subcontract and the supporting travel budget were
transferred to salaries and indirect costs for additional
experimental work on alternative process options and enzyme
preparations .

RAE-86-1066; BT601 31

Budget Summary

Budgeted

MSI

16556.50

MS2

9033.46

MS 3

10589.55

MS4

13895.79

MS5

12593.05

MS5A

13268.64

MS5B

14887.20

Sub Total

90825. 19

MS 6

Salaries

3040.00

Fringe Benefits

1051.84

Travel

650.00

Misc. & Ind.

3191.47

Sub Total

7933.31

Project Total

98758.50

Actual Billed

15608.45 15608.45
8633.46 8633.46

11011.40 11011.40

13895.75 13895.79

13512.46 13512.46
13269.64 13269.64

8237.20 8327.20

84168.40 84168.40

7090.00

2453.14

0.00

7443.65

16986.79 14590.00

101155.19 98758.50

In-Kind Contribution 2396.69

Two other projects on production of ligninase by white rot
fungi provided enzyme preparations for tests of straw conversion
in the DNRC project. An exact matching value cannot be
determined as the other projects did not have biomass conversion
as a direct project objective. However, the production and
evaluation of cell-free enzyme preparations from these projects
was a considerable contribution. Total amounts of related white
rot fungus contracts were:

US Dept of Energy SBIR Phase I $ 49,780

Private Company Contract 219,500

$269,280

RAE-86-1066: BT601 32

I

3) Technoloqry Transfer/Commercialization Activities

The project did not demonstrate an economically feasible
process for enzymatic delignif ication of Montana sources of
lignocellulose. As a result, Mycotech has not initiated any
commercialization activities from this project.

The original work plan included provisions for tests to be
subcontracted to SERI (or TVA) using Mycotech ligninase
preparations. Because results did not show significant
pretreatment , the outside tests were dropped in order to conduct
additional tests of alternative process options in house.
Because results did not show significant conversion efficiencies,
and SERI was not involved in tests, the report was not sent for
outside review.

4 ) References

1. Isaacs, S. H. (1984). Ethanol production by Enzyrnatic
Hydrolysis; Parametric Analysis of a Base Case Process. Solar
Energy Research Institute, Golden, Colorado, U.S. DOE Contract
#EG-77-C-01-4042.

2. Bradley, C, Black, W. , Kearns , R., Wood, P. (1987). Solid
State Cultures of Trichoderma reesei for Cellulase Production.
SBIR Phase II Final Report, U.S. DOE Contract #DE-AC03-
84ER80188. Renewable Technologies, Inc.

3. Hatakka, A., Tervika-wilo, A. (1986). Ligninases of white rot
fungi in Proceedings Soviet Finnish Seminar on Microbial
Degradation of Lignocellulose Materials. USSR Academy of
Sciences, Puschino, pp. 65-73.

4. Tien, M. (1987). Properties of ligninase from Phanerochaete
chrysosporium and their possible applications. CRC Critical
Reviews in Microbiolocry ; 15 (2), 141-168.

5. Tien, M., Kirk, T. K. (1984). Lignin-degrading enzyme from
Phanerochaete chrysosporium; Purification, characterization
and catalytic properties of a unique hydrogen peroxide

RAE-86-1066; BT601 33

requiring oxygenase. Proc . National Academy of Science 8 ; 2280-
2284.

e.Leisola, M., Ulmer, D., Fiechter, A. (1983). Problem of oxygen
transfer during degradation of lignin by Phanerochaete
chrysosporium. Eur J Appl Microbiol Biotechnol 17:113-116.

7. Bar-Lev, S., Kirk, T. (1981). Effects of molecular oxygen on
lignin degradation by Phanerochaete chrysosporium. Biochem &
Biophys Res Comm 99 ( 2 ): 373-378 .

8. Reid, I., Seifert, K. (1982). Effects of an atmosphere of
oxygen on growth, respiration and lignin degradation by white
rot fungi. Canadian J Botany 60:252-260.

9. Jeffries, T., Suki, C, Kirk, T. (1981). Nutritional
regulation of lignin degradation by Phanerochaete
chrysosporium. Appl and Environ Micro 42 ( 2 ): 290-296 .

lO.Leatham, G., Kirk, T. (1983). Regulation of ligninolytic
activity by nutrient nitrogen in white rot basidiomycetes .
FEMS Microbiol Lett 16:65-67.

11. Bradley, C, Black, W., Runnion, K. (1987). Commercial
Development: Ambient Temperature Starch Hydrolysis. Final
report to Montana Dept of Natural Resources and Conservation.
Contract #5 RAE 82-1007 and RAE 84-1044. Renewable
Technologies (Mycotech) Inc.

12. Reid, I. (1983). Effects of nitrogen supplements on
degradation of aspen wood lignin and carbohydrate components
by Phanerochaete chrysosporium. Appl and Environ Micro
45(3) :830-837.

13. Buswell, J. et al . (1984). Lignolytic enzyme production by
Phanerochaete chrysosporium under conditions of nitrogen
sufficiency. FEMS Microbiol. Letters 25:295-299.

14. Mycotech Inc. (1989). Unpublished results.

15. Leatham, G. (1989). Personal communication. USDA Forest
Products Laboratory, Madison, WI .

16. Zadrazil, F., Brunnert, H. (1982). Solid-state fermentation
of lignocellulose containing plant residues with Sporotrichum
pulverulentum Nov. and Dichomitus squalens (Karst) Reid. Eur
J. Appl Microbiol Biotechnol 16:45-51.

RAE-86-1066; BT601 34

IT.Reid, I., Deschamps, A. (1986). Biological delj.gnif ication
of aspen wood by solid state fermentation with the selective
lignin degrading fungus Phlebia tremellosus. Biotechnology
in the Pulp and Paper Industry. The Third International
Conference. Swedish Forest Products Laboratory, 49-51.

18. Kirk, T., Yung, H. (1979). Partial delignif ication of
unbleached kraft pulp with lignolytic fungi. Biotech Letters
1(9) :347-352.

19. Waldner, R., Leisola, M., Fiechter, A. (1988). Comparison of
lignolytic activities of selected white rot fungi. Appl
Microbiol and Biotechnol 29:400-407.

20. Forrester, I. et al . (1989). Manganese, Mn dependent
peroxidases and the biodegradation of lignin. Biochem
Biophys Res Comm. In press.

21. Jonsson, L. et al . (1987). Purification of ligninase
isozymes from the white rot fungus Trametes versicolor. Acta
Chemica Scandinavica 41:766-769.

22.Muheim, A. et al. (1989). An extracellular aryl-alcohol

oxidase from the white rot fungus Bierkandera adusta. Enzyme
and Microbial Technology. In press.

23. Janshekar, H., Haltmeier, T., Brown, C. (1982). Fungal
degradation of pine and straw alkali lignins. Eur J Appl
Microbiol and Biotechnol 14:174-181.

24. Brown, B. (1967). Determination of Lignin in Methods of Wood
Chemistry Volume II pp. 785-823. John Wiley & Sons.

25 . Mes-Hartree, M. et al. (1987). Suitability of aspen wood
biologically delignified with Phlebia tremellosus for
fermentation to ethanol or butanediol. Appl Microbiol
Biotechnol 26:120-125.

RAE-86-1066; BT601 35