BUDAPEST UNIVERSITY OF TECHNOLOGY AND ECONOMICS
Department of Agricultural Chemical Technology
PH.D. THESIS
BIOETHANOL PRODUCTION:
PRE-TREATMENT AND ENZYMATIC HYDROLYSIS OF CORN
STOVER
AUTHOR:
VARGA ENIKŐ
SUPERVISOR:
DR. RÉCZEY KATALIN
2003
Varga Enikő
BIOETHANOL PRODUCTION:
PRE-TREATMENT AND ENZYMATIC HYDROLYSIS OF CORN
STOVER
(Ph.D. Thesis)
ACKNOWLEDGEMENTS
Elöljáróban köszönettel tartozom Dr. Réczey Katalin témavezetőmnek. Munkámat,
szakmai fejlődésemet minden elképzelhető szempontból támogatta, és tanulmányaim
során bármikor bizalommal fordulhattam hozzá tanácsért vagy segítségért.
Dr Anne Belinda Thomsen, for her invaluable advice and constructive commands
through the half year spent in her lab. Thank you for teaching me, how to write a
scientific paper and how two see the substance behind the details.
Professor Guido Zacchi, for his never-ending constructive criticism.
Köszönet illeti Babits Miklósnét, felbecsülhetetlen technikusunkat, aki gyakorlati
tapasztalataival megkönnyítette laboratóriumi munkámat.
Szeretnék köszönetet mondani a csoportban dolgozó valamennyi munkatársamnak,
hogy egy kitűnő közösség megteremtésével segítették munkámat.
Végezetül és leginkább szeretnék köszönetet mondani Szüleimnek, akik végtelen
türelemmel és szeretettel álltak mellettem ez alatt a 4 év alatt is és mindenben
támogattak.
LIST OF ACRONYMS
CBU
Cellobiose unit
DM
dry matter
DMC
Direct microbial conversion (enzyme production, hydrolysis and fermentation
in the same stage with special microorganisms).
E “X” fuel
There are two main types of ethanol-blended gasoline in the World: low-level
and high-level ethanol blends. High-level ethanol blends are often
blended in a proportion of 85% ethanol with 15% gasoline, and are called E85.
Low-level ethanol blends are widely available, in proportions of 5-10%
ethanol blended with gasoline (E05 and E10 respectively). Ethanol is a noncorrosive and relatively non-toxic alcohol and it can be used directly as fuel
(most commonly in Brazil), or as an octane - enhancing gasoline additive
(throughout the United States, Canada and Europe). Blends of 5-15% ethanol
with gasoline can be used in all gasoline-powered automobiles, without engine
or carburettor modification.
E 85
E 10
EC
ECC
European Commission
c ⋅V
⋅ 100%
m ⋅ 1.11
where c is the
Enzymatic cellulose conversion,
concentration of D-glucose after enzymatic hydrolysis (g/L), V is the total
volume (L), and m the weight of cellulose before enzymatic hydrolysis (g). The
1.11 factor converts the cellulose concentration to the equivalent glucose
concentration.
ECC =
ECOFIN
The Council of Finance Ministers in the EU
ETBE
fuel additive (ethyl-tertier-butyl-ether)
EU
European Union
FPA
Filter Paper Activity
FPU
Filter Paper Unit
GC
gas chromatography
HPLC
high-performance liquid chromatography
OPEC
Organisation of Petroleum Exporting Countries
Oxygenates These are compounds, such as alcohols and ethers (ETBE), which contain
oxygen in their molecular structure. Oxygenates improve combustion
efficiency, thereby reducing polluting emissions. Many oxygenates, such as
ethanol also serve excellent octane enhancers, when blended with gasoline.
RI
refractive index
SC-CO2
Supercritical carbon-dioxide
SHF
Separate hydrolysis and fermentation
SSF
Simultaneous saccharification and fermentation
TS
TS-1
TS-2
TS-3
TS-4
two-step pre-treatment
1% NaOH & 1% HCl
5% NaOH & 1% HCl
1% Ca(OH)2 & 1% HCl
1% NaOH & 1% H2SO4
Tween 80
surfactant (polyoxyethylene sorbitan monooleate)
IN CONNECTION WITH THE BIOFUELS
CO2
Carbon dioxide, a normal product of burning fuel, is non-toxic, but contributes to the
greenhouse effect (global warming). All petroleum (hydrocarbon) fuels cause
increased atmospheric carbon dioxide levels because they represent the combustion of
fossilized carbon. By contrast, using renewable fuels, such as ethanol, does not
increase atmospheric carbon dioxide levels. The carbon dioxide formed during
combustion is balanced by that utilized during the annual growth of plants used to
produce ethanol.
CO
Carbon monoxide a poisonous gas produced by incomplete combustion. Vehicles
operating at colder temperatures (in winter months, during engine warm-up or in stopand-go traffic) produce significant quantities of this deadly gas, which is of particular
concern in urban areas. Research shows that transportation sources account for over
two-thirds of this pollutant. In the U.S. and in Europe (in Sweden, France, Spain,
Germany) many cities have mandated the use of "oxygenated" gasolines, such as
ethanol blends, to reduce carbon monoxide emissions.
LIST OF PUBLICATIONS
THIS THESIS IS BASED ON THE FOLLOWING PAPERS:
VARGA E., SZENGYEL ZS., RÉCZEY K. 2002. Chemical pre-treatment of corn stover. Appl.
Biochem. Biotech. 98-100:73-87.
VARGA E., SCHMIDT A.S., RÉCZEY K., THOMSEN A.B. 2003. Pretreatment of corn stover
using wet oxidation to enhance enzymatic digestibility. Appl. Biochem. Biotech.
104:37-49.
VARGA E., ZACCHI G., RÉCZEY K. 2003. Optimization of steam pretreatment for corn
stover to enhance enzymatic digestibility, Appl. Biochem. Biotech. In press
VARGA E., KLINKE H.B., RÉCZEY K., THOMSEN A.B. 2003. High solid simultaneous
saccharification and fermentation of wet oxidised corn stover to ethanol, Biotechnol.
Bioeng.. (Manuscript (number: 03-570) submitted: on 10th October 2003.)
OTHER RELATED PUBLICATIONS BY THE SAME AUTHOR:
VARGA E., KÁDÁR ZS., SCHUSTER K.C., GAPES J. R., SZENGYEL ZS., RÉCZEY K. 2002.
Possible substrates for Acetone–Butanol and Ethanol fermentation based on organic
by-products. Hungarian Journal of Industrial Chemistry 30, 19-25.
KÁLMÁN G., VARGA E., RÉCZEY K. 2002. Dilute Sulphuric Acid Pretreatment of Corn
Stover at Long Residence Times. Chemical and Biochemical Engineering Quarterly
16(4):151-157.
KÁDÁR ZS., VARGA E., RÉCZEY K. 2001. New Substrates of Biofuel, Magyar
Mezőgazdaság, V 56. N. 19, pp. 32-33.
POSTER AND ORAL PRESENTATIONS:
VARGA E., KÁDÁR ZS, RÉCZEY K. 1999. Possible substrates for ABE fermentation
International Conference on the Applied Acetone Butanol Fermentation. Krems,
Austria, 16-18. September 1999.
KÁDÁR ZS., VARGA E., RÉCZEY K. 1999. Possible substrates for ethanol fermentation.
3rd European Motor Biofuels Forum, Brussels, Belgium, 10-13. October 1999.
6
VARGA E., RÉCZEY K. 1999. Enzymatic hydrolysis of the rest of fruit juice processing.
Chemist-conference ‘99. 26-28. November, Budapest
VARGA E., RÉCZEY K. 2000. Pretreatment and enzymatic hydrolysis of corn stover.
„Lippay János - Vas Károly” Scientific Conference, Budapest, 6-7. November 2000.
VARGA E., RÉCZEY K. 2001. Pretreatment of corn stover to enhance the enzymatic
digestibility. 23. February, Budapest, Scientific Conference (KÉKI),
VARGA E. 2001. New substrates of the alcohol fermentation - pretreatment of the
lignocellulosic materials. 28. February, Budapest, BUTE „Industrial Open Day”,
Bioenergy, Biofuels symposium
VARGA E., ÁDÁM J., SZENGYEL ZS., RÉCZEY K. 2002. Chemical Pretreatment of corn
stover. 4-9. May, Colorado, USA. 23th Symposium on Biotechnology for Fuels and
Chemicals.
VARGA E., KLINKE H., RÉCZEY K., THOMSEN A.B. 2002. Enzymatic hydrolysis and
fermentation of wet oxidised corn stover. 26-27. April, Floriade, Netherlands,
International Congress & Trade Show Green-Tech® 2002 with European
Symposium Industrial Crops and Products.
VARGA E., KLINKE H., RÉCZEY K., THOMSEN A.B. 2002. Enzymatic hydrolysis and
simultaneous saccharification and fermentation of wet oxidised corn stover.
28. April – 1 May, Gatlinburg, USA 24th Symposium on Biotechnology for Fuels and
Chemicals.
VARGA E., RÉCZEY K. 2002. Pre-treatment of corn stover at high temperature. 8 – 10 April,
Veszprém, Hungary, Conference entitled: Technical Chemical Days’03,
VARGA E., ZACCHI G., RÉCZEY K. 2003. Optimisation of steam pre-treatment for corn
stover to enhance enzymatic digestibility. 4 – 7 May, Breckenridge USA 25th
Symposium on Biotechnology for Fuels and Chemicals.
7
CONTENTS
1. INTRODUCTION ...........................................................................................10
2. BACKGROUND:............................................................................................16
ETHANOL PRODUCTION FROM LIGNOCELLULOSIC SUBSTRATE .........................16
2.1. LIGNOCELLULOSIC BIOMASS ............................................................................. 16
2.2. PRETREATMENT PROCESSES ............................................................................. 19
2.2.1. Biological pretreatment ........................................................................... 20
2.2.2. Physical pretreatment.............................................................................. 21
2.2.3. Physico-chemical pretreatment............................................................... 21
2.2.3.1. Steaming/Steam explosion...................................................................................... 21
2.2.3.2. Wet oxidation ......................................................................................................... 22
2.2.3.3. AFEX process......................................................................................................... 25
2.2.3.4. CO2 explosion......................................................................................................... 25
2.2.4. Chemical pretreatment ........................................................................... 26
2.2.4.1. Acidic pretreatment ................................................................................................ 26
2.2.4.2. Alkaline pretreatment............................................................................................. 27
2.2.4.3. Organosolv process................................................................................................ 27
2.2.4.4. Ozonolysis .............................................................................................................. 28
2.3. HYDROLYSIS PROCESSES ................................................................................... 28
2.3.1. Acid hydrolysis ........................................................................................ 29
2.3.2. Enzymatic hydrolysis................................................................................ 30
2.3.2.1. Surfactant effect in enzymatic hydrolysis ............................................................... 31
2.4. FERMENTATION FOR BIOETHANOL PRODUCTION .............................................. 33
2.4.1. Separate hydrolysis and fermentation (SHF) .......................................... 33
2.4.2. Simultaneous saccharification and fermentation (SHF)......................... 33
2.4.3. Simultaneous saccharification and co-fermentation (SSCF).................. 34
2.4.4. Direct microbial conversion (DMC) ....................................................... 34
3. MATERIALS AND METHODS ..........................................................................36
3.1. RAW MATERIAL ............................................................................................... 36
Analysis of raw material ..................................................................................................... 36
3.2. PRETREATMENTS ............................................................................................. 39
3.2.1. Chemical pretreatments .......................................................................... 39
Alkaline and acidic pretreatments....................................................................................... 39
Combined alkaline and acidic pretreatments (two-step pretreatments) ............................. 39
Pretreatments with dilute base and acid ............................................................................. 39
3.2.2. Steam pretreatment................................................................................. 40
3.2.3. Wet oxidation pretreatment ................................................................... 42
8
3.2.4. Supercritical CO2 pretreatment .............................................................. 43
3.3. ENZYMATIC HYDROLYSIS .................................................................................. 44
3.3.1. Effect of ultrasonic waves during enzymatic hydrolysis........................... 45
3.3.2. Enzyme activity measurement ................................................................. 46
3.4. ANALYSIS OF THE LIQUID FRACTION AFTER PRETREATMENTS ........................... 46
3.4.1. Poly- and monosaccharide analysis of liquid fraction.............................. 46
3.4.2. Carboxylic acids and phenols analysis..................................................... 47
3.5. SIMULTANEOUS SACCHARIFICATION AND FERMENTATION (SSF) AFTER WETOXIDATION ............................................................................................................. 47
3.6. FERMENTATION FOLLOWING STEAM PRETREATMENT ........................................ 48
4. RESULTS AND DISCUSSION ..........................................................................50
4.1. CHEMICAL PRETREATMENTS ON YIELDS AND COMPOSITION .............................. 50
4.1.1. Effect of chemical pretreatments on yield and composition ................... 50
Effect of two-step pretreatments on corn stover.................................................................. 52
4.1.2. Enzymatic hydrolysis................................................................................ 53
4.1.3. Effect of supercritical pretreatment on yield and composition ............... 57
4.1.4. Conclusions on chemical pretreatments ................................................. 58
4.2. STEAM PRETREATMENT .................................................................................... 58
4.2.1. Effect of steam pretreatment ................................................................... 58
4.2.2. Enzymatic hydrolysis................................................................................ 63
4.2.3. Fermentability test................................................................................... 64
4.2.4. Conclusions on steam pretreatments ...................................................... 65
4.3. WET OXIDATION PRETREATMENT .................................................................... 66
4.3.1. Effect of wet oxidation ............................................................................ 66
4.3.2. Enzymatic hydrolysis................................................................................ 71
4.3.3. SSF .......................................................................................................... 74
4.3.3.1. Effect of substrate concentration on ethanol yield................................................. 74
4.3.3.2. Effect of enzyme loading ........................................................................................ 76
4.3.3.3. Conversion of phenols and carboxylic acids by SSF ............................................. 78
4.3.4. Conclusions on wet-oxidation pretreatments ......................................... 79
4.4. COMPARISON OF THE DIFFERENT PRETREATMENT PROCESSES ........................... 80
4.5. ESTIMATE OF PRODUCTION COST OF FUEL ETHANOL ......................................... 83
4.5.1. Production cost of starch based bioethanol............................................. 83
4.5.2. Production cost of lignocellulose based bioethanol................................. 83
5. CONCLUSIONS AND FUTURE POSSIBILITIES ...............................................86
6. REFERENCES ................................................................................................88
9
1. INTRODUCTION
In October of 1973, the Organisation of Petroleum Exporting Countries (OPEC) decreased
the output of oil, which quadrupled the oil prices and tipped the world's leading industrial
powers into crisis. This dramatic increasing in oil prices turned the world’s interest to the
alternative fuels (Mably, 2003). Then the main motivation was to become independent of
petroleum market and to reduce the cost of expensive oil imports. However, the emphasis
today is on reducing pollution and helping to satisfy the Kyoto protocol, established in
1997, by limiting the global net emission of carbon dioxide (CO2).
Energy from the sun heats the earth's surface; in turn, the earth radiates energy back into
space. Atmospheric greenhouse gases (water vapour, carbon dioxide, methane, nitrous
oxide, and ozone) trap some of the outgoing energy, retaining heat somewhat like the glass
panels of a greenhouse. Without this natural phenomenon, which is refereed to as the
“greenhouse effect”, temperatures would be much lower than they are now, and life as
known today would not be possible. However, problems may arise when the atmospheric
concentration of greenhouse gases increases.
Since the beginning of the industrial revolution, atmospheric concentrations of carbon
dioxide have increased nearly 30% from 280 ppm to 365 ppm, methane concentrations
have more than doubled, and nitrous oxide concentrations have risen by about 15% (WMO,
2002). Very powerful greenhouse gases have appeared in the atmosphere, that are not
naturally occurring, include hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and
sulphur hexafluoride (SF6), which are generated in a variety of industrial processes.
Parallel with the increasing green house gas concentration, the Earth's global mean surface
temperature has risen by about 0.6-0.9°C since the late 19th century, but the warming of the
atmosphere is continuing (Schneider 1989). The Word’s nine warmest years all occurred in
the last ten years, among of these 2001 was the warmest year of the world on record
(WMO, 2002), but the year of 2003 could take the lead.
There is new and stronger evidence that most of the warming over the last 50 years is
attributable to human activities, in consequence of the additional release of carbon dioxide,
which is the primary global warming gas.
The OECD countries consume the 57% of the world’s energy (Table 1.1.) and contribute
more than half of the world’s total emission of CO2. The USA is one of the countries that
has the biggest energy demand, and corresponding to this, discharges the most, emitting
about one-fifth of total global greenhouse gases (U.S. Ministry of Environmental, 2003).
Energy consumption and also the CO2 concentration has increased steadily over the last
century and it has increased significantly even in the last 10 years, which is shown in
Table 1.1.
10
Opposite to this trend, energy demand of Hungary has decreased in the last 10 years, but
unfortunately it could be explained with the industrial decline after the political changes in
the latest 1980’s and not with conscious, energy saving behaviour or with new, energy
friendly technologies.
Estimating future emissions is difficult, because it depends on demographic, economic,
technological, policy, and institutional developments, but it is fact, that the world
population has been growing and even more countries have been becoming industrialised.
Figure 1.1. shows the share of the oil consumption in various industrial fields.
Table 1.1. Primary energy consumption by energy source (Mtoe, million tone oil equivalent)
World
OECD
USA
Asia
Europe
EU (15)
FSU
Middle East
Hungary
Oil
3510.6
2189.6
1066.3
972.7
760.2
637.1
169.6
206.4
6.8
Gas
2164.3
1167.2
650.4
274.7
423.0
343.3
493.6
181.3
10.7
2001
Carbon Nuclear
2255.1
601.2
1108.2
518.8
590.9
202.6
1020.7
115.0
344.1
225.0
212.5
201.6
180.4
51.2
8.0
––
3.1
3.2
Water
594.5
291.0
129.7
128.8
142.4
84.4
54.9
1.5
––
Summa
9125.7
5274.8
2639.9
2511.9
1894.7
1478.9
949.7
397.2
23.8
1991
Summa
8147.2
4606.7
2300.7
1886.2
1770.7
1345.6
1375.3
264.3
25.6
One way of reducing both environmental effects and dependence on fossil fuels is to use
alternative, renewable energy sources. In accordance with this trend the European
Commission (EC) has decided to increase the market share of renewable energy to 12%
until 2010.
The rate of renewable energy in the total energy consumption in Hungary was only 3.2% in
2001, mainly low-efficiency direct burning, but the using of alternative fuels in the
transport sector was nearly negligible. In half a year, Hungary will be a member of the EU,
which underlines these problems and pushes to decide to overcome these deficiencies in
the near future.
As the transportation sector is practically 100% dependent on oil and it is responsible for
the greatest and even more increasingly proportion of CO2 emission, the biofuel, including
fuel ethanol, is a topical issue both in the European Union (EU) and in the USA. (Mielenz,
2001). In 2000 the EC has developed a strategy to obtain 20% share in the biofuel market
by the year 2020 (Vermeersch 2002). To reach these aims the EC published an action plan
and two directive proposals in November 2001 to encourage increased use of biofuels in
the transport sector (http://www.platts.com/features/biofuels/europe.shtml).
11
The objectives of the action plan and the directives are to:
• Help reduce the European Union's dependence on external oil supply.
• Contribute to EU greenhouse gas emission reduction targets as agreed in the Kyoto
Protocol.
• Meet the objective of substituting 20% of diesel and gasoline fuels by alternative fuels
in the road transport sector by 2020.
The first directive calls for the establishment of a minimum level of biofuels as a
proportion of fuels sold from 2005, starting at 2% and reaching 5.75% of fuels sold in
2010. Whether the targets of the first directive will be indicative or mandatory is still to be
decided.
The second directive deals with tax issues, seeking to give EU member states the option of
offering tax breaks on pure or blended biofuels used in either heating or motor fuel. The
Council of Finance Ministers (Ecofin) has allowed 100% relief as long as the targets are
indicative, while the EC's draft proposed a 50% duty cut on biofuels.
In Europe, contrary to the US, biodiesel is the dominating biofuel. France, Germany, Italy,
Austria and Belgium together produced 701.6 kton of biodiesel in 2000, according to
EurObserver, the EU research institute on renewable energy.
The European ethanol production is much smaller than biodiesel production. In 2000, the
three main ethanol producting country, France, Spain and Sweden together produced a
total of 216 kton ethanol (Table 1.2.). France is the biggest European producer but does not
use ethanol in its pure form. It transforms the alcohol into fuel oxygenate ETBE (ethyltertier-butyl-ether).
While Spain produced only 80 kton of ethanol in 2000, one Spanish company, Abengoa, is
investing heavily in ethanol production. In the last two years, the company has spent 382
million Euro on its bioethanol production and is now the second biggest producer in the
world with a 15% market share. (http://www.platts.com/features/biofuels/europe.shtml)
Table 1.2. Biodiesel, ethanol and ETBE production in the EU in the year of 2000.
Austria
Belgium
Germany
Italy
France
Spain
Sweden
ETHANOL
91
80
45
BIODIESEL
27.6
20
246
78
328.6
1
12
ETBE (KTON)
193
170
0
In the USA there is also a considerable growth in oxygenated fuel additive improved
gasoline (e.g E10 and E 85 fuels) (Knapp et al., 1998). However, as a result, the US
transportation sector now consumes only about 4540 million litre of ethanol annually,
about 1% of the total consumption of gasoline (Sun and Cheng 2002).
8%
25%
54%
8%
power plant
oil for heating
not for energetic use
5%
petrochemistry
transport
Figure 1.1. The distribution of the world’s oil consumption
The use of fuel ethanol will significantly reduce net carbon dioxide emissions, when it
replace fossil fuels, because fermentation-derived ethanol is already part of the global
carbon cycle.
Apart from a very low net emission of CO2 to the atmosphere, the combustion of
bioethanol in generally results in the emission of low levels of uncombusted hydrocarbons,
carbon monoxide (CO), nitrogen oxides (NOx) and exhaust volatile organic compounds
(VOCs). However, of major environmental concern regarding the increased use of ethanol
fuels is the increased exhaust emission of reactive aldehydes, such as acetaldehyde and
formaldehyde. Thus, a key factor with respect to the possible of ethanol on urban air
quality will be the durability and effectiveness of catalyst system for aldehyde control.
The T model invented by Henry Ford in the early 1900’s was originally developed to run
on ethanol. Today, all cars with a catalyst can be run on a mixture of 90% petrol and 10%
ethanol without engine modification. New cars can even use mixtures containing up to
20% ethanol. There are also new engines available, which can run on pure ethanol, and socalled flexible fuel vehicles (FFV) which are able to use mixtures of 0-85% ethanol in
petrol. Ethanol can also replace diesel fuel in compression-ignition engines, however to be
able to mix diesel with ethanol an emulsifier is needed (Wheals et al., 1999).
Fuel ethanol is used in a variety of ways (Table 1.3.). However, today the major use of
ethanol is as an oxygenated fuel additive. Mixing ethanol and petrol has several
13
advantages. The higher octane number of ethanol (96-113) increases the octane number of
the mixture, reducing the need for toxic, octane-enhancing additives. Ethanol also provides
oxygen for the fuel, which will lead to the reduced emission of CO and uncombusted
hydrocarbons (Mielenz, 2002).
Table 1.3. Common ethanolic motor-fuel formulations
FUEL
Hydrous ethanol (Álcool, in Brazil)
E 85 (North America, Sweden)
Gasoline (Brazil)
E 10 (gasohol, North America, Canada, Sweden)
ETBE (France, Spain)
Oxygenated fuel (USA, Canada)
Biodiesel (EU)
ETHANOL CONTENT (% V/V)
95.5%
85
24
10
7.6
5.7
15
In additional to these points, several positive consequences of an introduction of biofuel
are to be considered as additional assets, e.g. improvement of the trade balance,
employment safeguarding, creation of jobs, social and tax revenues.
Despite the advantages of bioethanol, Brazil and North America are still the only two
countries that produce large quantities of fuel ethanol, from sugar cane and maize,
respectively. Brazil produces 12 million m3 ethanol per year and the USA about half of this
amount. Fuel ethanol production is considerably more modest in the European Union,
where the three dominant producer France, Spain and Sweden together produced a total of
475 000 m3 per year. However, other European countries such as Austria, Italy, Poland and
Portugal have shown interest in the oxygenate ETBE (ethyl tertiary butyl ether) produced
from ethanol.
The efficiency of ethanol production has steadily increased, but tax relief will be required
to make fuel ethanol commercially viable compared with oil. Thus current bioethanol
research is driven by the need to reduce the production cost. Historically, the projected cost
of lignocellulosic biomass-based-ethanol has dropped from about US$ 1.22 / L to about
US$ 0.35 / L because of the continuous improvement in pretreatment, enzymatic
application and fermentation techniques. When using feedstocks such as sugar cane or
maize, the raw material accounts for 40 to 70% of the total production cost (calculation is
showed in the “list of Acronyms”) (Worley et al., 1992, Claassen et al., 1999). However
using high cellulose containing agricultural residues, for example, corn stover, the
production cost could be reduced significantly in the next 15 years (Lynd 1996). The
potential of using lignocellulosic biomass for energy production is even more apparent
when one realises that it is the most abundant renewable organic components in the
biosphere. 1015 kg of cellulose are synthesised and degraded on Earth every year. This
14
resource converted into useful energy at a 65% efficiency could supply all of the world’s
present energy needs (Clausen and Gaddy 1983).
Corn is a major crop in Hungary, produced at nearly 8 million tons in every year.
Assuming the 1:1.3 ratio (depending on the harvesting time) of dry matter of corn grain to
dry matter of corn stover, the equivalent mass dry mass residue was approximately 11
million tons. Corresponding its amount, corn stover is the most abundant agricultural
residue in Hungary. (Statistical Annual Reviews, 2002). Less, then 10% of this amount has
been used as an animal feet, the major portion of corn stover has traditionally been
removed from the field by the practice of open-field burning. This practice plays an
important role in protecting and improving soil quality, however it has a negative influence
on the air quality and it wastes a huge amount of potential energy source. In defend of the
air quality a new law has been brought recently, which prohibit the open air burning in the
fields in Hungary.
The US, which is the main corn producer country in the world, produced approximately
216 million tons of corn grain in 2001. Associated with corn production is a corresponding
annual production of approximately 254 million tons of corn stover (Sokhansanj et al,
2002). In spite of its large quantities, currently only 6% is collected and used for animal
feeding and bedding.
However corn stover consisting of 70% carbohydrates (cellulose and hemicellulose) is a
promising raw material for fuel ethanol (Hahn-Hägerdal 1996). Due to structural features
such as lignin, acetyl groups and cellulose crystalinity, lignocellulosic biomass must be
pre-treated to enhance its digestibility before microbial conversion into liquid fuels (Kaar
and Holtzapple 1998).
The most of the commonly used, effective fractionation methods break down the lignin
structure and solubilise the hemicellulose fraction (Schmidt and Thomsen 1998). However
during these pretreatments, degradation products such as acetic acid, formic acid and
phenol monomers (Klinke et al, 2002) could be formed, which are well known inhibitors
for the fermenting microorganisms, e.g. Saccharomyces cerevisiae.
The absence of these by-products, heterogeneity in feedstock and the influence of different
process conditions on microorganisms and enzymes make the biomass-to-ethanol process
complex.
15
2. BACKGROUND:
ETHANOL PRODUCTION FROM LIGNOCELLULOSIC SUBSTRATE
2.1. LIGNOCELLULOSIC BIOMASS
Lignocellulosic biomass from different sources may appear outwardly quite different, but
the chemical composition is in fact very similar. It primarily comprises three major
fractions: cellulose, hemicellulose and lignin, besides the rest is typically much smaller
amounts of minerals (ash), soluble phenolics and fatty acids and other compounds often
termed extractives. Figure 2.1. A-C shows the three main components of lignocellulose.
The composition of different herbaceous residues are given in Table 2.1. Although the
compositions are quite similar, corn stover generally contains more lignin than other
lignocellulosic materials.
Table 2.1. Percent Composition of lignocellulosic material on a Dry-Wt basis
CORN
WHEAT
A
Glucan
Xylan
Galactan
Arabinan
Mannan
Total glycan
Klason lignin
Acid-soluble lignin
Ash
Acetyl groups
Other
STOVER
STRAW
36.0
19.8
1.3
2.8
59.9
26.9
1.9
7.2
1.4
2.7
38.0
22.4
2.1
3.6
0.7
66.8
14.5
nd
7.2
1.4
10.4
B
SWITCHA
BARLEY
GRASS
STRAW
32.2
20.3
3.7
0.4
56.6
19.5
3.7
7.1
2.35
10.8
37.5
15.0
1.7
3.9
1.3
59.4
13.8
nd
10.8
2.0
14
C
RICE
STRAW
37.2
24.4
1.3
2.8
0.4
66.1
15.0
nd
8.9
4.4
5.6
SORGHUM
D
STRAW
C
32.5
16.9
0.2
2.1
0.8
52.5
14.5
nd
10.1
6.2
16.7
a
Fenske et al. 1998
Bjerre et al. 1996
c
Wilke et al. 1981
d
Vlasenko et al. 1997
b
Cellulose comprises between 35 and 50% of the total dry mass and consists of long chains
of ß-anhydroglycose unit linked by ß1,4-glucoside bonds. The degree of polymerisation
i.e. the number of glucose molecules included in a cellulose chain is normally in the range
of 7500 to 15000 for plant cellulose. The cellulose molecules are organised in elementary
fibrils with a diameter of 2-4 nm. These fibrils are associated through hydrogen and van
der Waals bonds, forming a very rigid, highly crystalline macromolecular structure.
16
Between the elementary fibrils, there are microfibrils contain regions, which are less
ordered, and often called “amorphous regions”. These regions are particularly susceptible
to enzymatic hydrolysis, but about 50-90% of the cellulose in lignocellulosic material
forms crystalline structure. The cellulose is provided by hemicellulose and lignin, which
formed a physical protection of the cellulose hydrolysis.
Figure 2.1. A. Structure of cellulose
Hemicellulose represents to about 25% of total lignocellulosic mass, and, like cellulose, its
monomer units can also be fermented to ethanol. In contrast to cellulose is a branched
polymer of sugars whose units include mostly aldopentoses, such as xylose and arabinose
and some aldohexoses, such as glucose, mannose and galactose. Various substitutes, e.g.
acetyl groups or uronicacid groups are attached to the main chain or the branches and the
DP ranges from 20 to 300. The variety of linkages, branching, and different monomer units
contribute to the complex structure of hemicellulose and thereby its variety of
conformations and function. Within biomass, hemicellulose links covalently to lignin and
through hydrogen bonds to cellulose. However the hemicellulose is much more easily
broken down than crystalline cellulose. (Bringham et al., 1996)
Figure 2.1.B. Structure of hemicellulose
Lignin, the third significant fraction in biomass, is one of the most abundant and important
polymeric organic substance in the plant world. Lignin increases the mechanical strength
to such an extent that even the 100 m height trees remain upright. Although there are a
17
great number of microorganisms, which are able to utilise hemicellulose or cellulose,
relatively few strains have the ability to decompose the lignin effectively. Lignin is a
highly complex, three-dimensional polymer of different phenylpropane units, which are
bound together by ether and carbon-carbon bonds. A few lignin structures have been
elucidated but in general their structures remain unknown. The lignin fraction of biomass
remains as a solid after most hydrolysis methods and can impact fermentation. It is often
burned as boiler fuel because of its high energy-content.
Figure 2.1.C. Structure of lignin
18
ETHANOL PRODUCTION FROM LIGNOCELLULOSIC SUBSTRATE
Ethanol could be produced in various ways according to the various feedstocks. Although
the production of ethanol from cane sugar is a relatively simple process and known since
several hundred years, complexity increases when ethanol is produced from corn or wheat
starch, as these processes require enzymes to hydrolyse starch to glucose prior
fermentation. Production of ethanol from lignocellulosic biomass (Figure 2.2.), which was
investigated in this thesis, requires even more extensive processing to release the
polymeric sugars in cellulose and hemicellulose.
Figure 2.2. Schematic flowchart of the “ethanol from lignocellulosic biomass” process
2.2. PRETREATMENT PROCESSES
Although the most lignocellulosic biomass is rich in carbohydrates, it is an insoluble
substrate with a complex structure. Thus the polysaccharides are not directly available for
bio-conversion as the lignin component and the regular and cross-linked polymers in
lignocellulosics form a very efficient physical barrier. Figure 2.3. shows schematically the
complex structure of lignocellulose. Cellulose fibers are embedded in a sheat of
hemicellulose and lignin and held together by hydrogen and van der Waals bonds, as it was
mentioned above. There are several ways to increase the digestibility of cellulose before it
is exposed to enzymes: physical, chemical and biological processes and combination of
these have been used for pretreatment of lignocellulosic material.
19
The purpose of all pretreatments is to remove lignin and hemicellulose, reduce cellulose
crystalinity and increase the porosity of the materials. All kind of pretreatment must meet
the following requirements:
• Fractionation of the raw material into high quality feedstock for further
processing/conversion,
• Avoid the degradation or loss of carbohydrate,
• Avoid the formation of by-products inhibitory to the subsequent hydrolysis and
fermentation processes,
• Have a minimum consumption of energy and chemicals,
• Be cost-effective with low operating and capital cost.
Figure 2.3. Schematic figure of the complex structure of lignocellulose
2.2.1. Biological pretreatment
The category of biological pretreatments comprises the techniques of applying lignin-,
and hemicellulose-solubilising microorganisms, such as brown-, white- and soft-rot
fungi, to render lignocellulosic materials suitable for enzymatic digestion (Schurz 1978,
Hatakka 1983). According to the results of Fan et al., white-rot fungi are the most
effective basidiomycetes for biological pretreatment of lignocellulosic materials (Sun
20
and Cheng 2002). However the rate of hydrolysis in most biological pretreatment
processes is very low and the most lignin-solubilising microorganisms also solubilise or
consume cellulose (Ghosh and Shingh 1993, Han 1978). In order to prevent the loss of
cellulose, development for a cellulase less mutant is necessary.
2.2.2. Physical pretreatment
Pretreatment techniques, which do not involve chemical application called physical
treatments. Most physical pretreatments, as high energy irradiation, dry, wet, and
vibratory ball milling are mainly reported to be time-consuming, ineffective and energyintensive, therefore expensive (Fan et al., 1982, Chang et al., 1981, Lin et al., 1981 ).
However most pretreatment techniques studied appear to employ particles passing a 3 mm
or smaller rejection screen (Alvo and Belkacemi 1997). This primary size reduction is an
essential step in the conversion of lignocellulosic material to fuel ethanol, thus it is not
considered as a mechanical pretreatment in this work. On the other hand some reviews
indicated that milling alone can result yields near to the theoretical maximum from some
sources of lignocellulose (e.g. straws, woods) (Koullas et al., 1992, Millet et al., 1976).
However it has to be mentioned, that the required particle size was <10-3 mm, which needs
an extremely great portion of energy.
2.2.3. Physico-chemical pretreatment
2.2.3.1. Steaming/Steam explosion
Steaming or steam explosion is an extensively investigated pretreatment method. A vast
amount of literature can be found, treating several types of raw materials (Clark and
Mackie 1987, Eklund et al., 1995). In this method, chipped biomass is treated with highpressure saturated steam and then the pressure is swiftly reduced, which makes the
materials undergo an explosive decompression. Steam explosion is initiated at a
temperature of 160-260°C (corresponding pressure 0.69-4.83 MPa) for several seconds to a
few minutes before the material is exposed to atmospheric pressure. During this treatment
the presence of moisture initiates an auto- hydrolysis reaction catalysed by organic acids,
which are initially formed from acetyl groups, liberated from the biomass. The action
mode of steaming/ steam explosion is therefore similar to that chemical pretreatment with
acid, as described bellow.
Ninety percent efficiency of enzymatic hydrolysis has been achieved in 24 h for poplar
chips treated by steam explosion compared to only 15% hydrolysis of untreated chips
(Grous et al., 1986). Similar result was achieved for steam pre-treated corn stover, (Varga
et al., 2003) however it is reported less effective for softwoods (Stenberg et al., 1998). The
factors that affect steam explosion pretreatment are residence time, temperature, chip size
21
and moisture content (Duff and Murray 1996). Optimal hemicellulose solubilisation and
hydrolysis can be achieved by either high temperature and short residence time (270°C, 1
min) or lower temperature and longer residence time (190°C, 10 min). Recent studies
indicate that lower temperature and longer residence time are more favourable (Wright
1998). Impregnation with sulphuric acid, sulphur-dioxide or carbon-dioxide prior to steam
pretreatment can effectively improve enzymatic hydrolysis, decrease the production of
inhibitory compounds and lead to more complete removal of hemicellulose (Tenborg et al.,
1998) even when softwood are considered.
Limitations of steam explosion include destruction of portion of the xylan fraction,
incomplete disruption of the lignin-carbohydrate matrix, and generation of compounds that
may be inhibitory to microorganisms used in downstream processes.
An advantage of steam pretreatment is that it is one of a very few fractionation methods
that have been tested in pilot scale, and commercial equipment is available. The process
can be performed either in batch reactors, as in the “Masonite” process, or in continuos
reactors, as in the “Stake” process (Heitz et al., 1991, Glasser and Wright 1998).
2.2.3.2. Wet oxidation
Wet oxidation (WO), a reaction involving oxygen and water at elevated temperature and
pressure, was presented in the early 1980’s to pre-treat lignocellulose as an alternative to
the well studied steam explosion (McGinnis et al., 1983a,b). The most important process
parameters of wet oxidation are the reaction temperature and residence time like for
steaming.
In the early studies of wet oxidation, usually the heating and the cooling times of the
reaction have been very long (up to 30 min), and a low reaction temperature was needed to
obtain fractionation. In the temperature range between 120-170°C it is mainly the
hemicellulose fraction that is dissolved, but also the lignin and the cellulose fractions are
affected to a smaller extent. At temperatures higher than 170°C, the reaction becomes
increasingly more oxidative and causes a considerable amount of fragmentation and
oxidation of the biomass, in particular the lignin, leading to the formation of organic acids
and small amounts of neutral organic compounds. Studies using monosaccharides as model
compounds indicate that most of the monosaccharides (88-100%) can be recovered after
wet oxidation at 154°C The oligosaccharides are probably even more stable to the partial
hydrolysis of the biomass, since the glycosidic linkages in the oligosaccharides are more
stable to oxidation than the aldehyde group of the monosaccharide (McGinnis et al., 1983).
Recently, studies of wet oxidation of various herbaceous materials, e.g. corn stover and
wheat straw employ a reactor with very short heating and cooling times, about 1-2 minutes.
Hence, a higher temperature can be used without extensive degradation of the
polysaccharides. At 185°C and at 190°C very high recoveries for cellulose (above 95%)
and reasonable recoveries for hemicellulose (about 60%) were found (Schmidt et al., 1998,
Varga et al., 2002). However, when the temperature approaches 200°C, a drop of the
amount of xylose and xylo-oligomers in the liquid fraction can be observed, due to
22
degradation. The extraction of the lignin fraction depends on the type of the raw material
and the applied chemical during the pretreatment.
The principals both of the wet oxidation and the steam explosion processes are shown in
Figure 2.4. using the plant residue, corn stover as an example. The steam pretreatment is
carried out in generally in the presence of inorganic acid, while wet oxidation uses mainly
alkaline addition. Following steam pretreatment the pressure is swiftly reduced (1-2 sec),
which makes the materials undergo an explosive decompression, but after wet oxidation
the pressure decreases slightly (15-20 min). The major difference between wet oxidation
and steam explosion pretreatment is that the wet oxidation reaction is more complete due
to the presence of oxygen.
Compared to other pretreatment processes, wet oxidation has the advantages that it is
suitable for a wide range of biomasses, including both hardwoods, softwoods and
agricultural residues, mainly for the generation of an easily accessible cellulose fraction
due to the crystalline structure of cellulose is opened during the process. Organic
molecules, including lignin, decompose to CO2, H2O and simpler and more oxidised
organic compounds, mainly to low-molecular-weight carboxylic acids. This method
appears to produce fewer by-products, like furfural and hydroxymethyl-furfural (Bjerre et
al., 1996, Ahring et al., 1999).
Figure 2.4. A typical wet oxidation and the steam explosion concept
Under the conditions of wet oxidation, aliphatic aldehydes and saturated carbon bonds are
very reactive, hence the sugar degradation products, which are known inhibitors of
23
microbial growth, are not expected to be produced at high concentration. The principle of
the treatment is illustrated in Figure 2.5.
Apart from polysaccharides available for fermentation some degradation products were
formed during the treatment (Klinke et al., 1999, 2002). these were identified as phenolic
compounds and low molecular weight carboxylic acids that might be potential inhibitors
during fermentation of the sugar fractions. The inhibitors were produced by decomposition
of the plant constituents (lignin, hemicellulose, cellulose, pectin, wax) due to thermal
degradation and oxidation. By alkaline wet oxidation, the production of furans seemed to
be avoided (Bjerre et al., 1996, Schmidt et al., 1998) due to further oxidation of these
compounds to carboxylic acids. The furans, 2-furfural and 5-hydroxymethil-2-furfural are
common degradation products from steam explosion (Von Sivers and Zacchi 1996) due to
thermal decomposition of xylose and glucose, respectively (also elucidated in Figure 2.5.).
Figure 2.5. A simplified flow diagram of the fractionation and formation of degradation
products by WO of wheat straw
24
2.2.3.3. AFEX process
The ammonia fiber/freezer explosion (AFEX) process, which concept is similar to steam
explosion is another type of physico-chemical pretreatment in which lignocellulosic
materials are exposed to liquid ammonia at moderate temperatures from 25°C to 90°C, and
elevated pressures for a period of time, from 10 to 60 min. (Holtzapple et al., 1991, 1992).
When the reaction is complete, the pressure is explosively released which disrupts the
fibrous structure. After the explosion, to reduce the cost and protect environment, ammonia
must be recycled (Holtzapple et al., 1994). In an ammonia recovery process, a superheated
ammonia vapour with a temperature up to 200°C was used to vapourise and strip the
residual ammonia in the pre-treated biomass and the evaporated ammonia was then
withdrawn from the system by a pressure controller for recovery.
AFEX pretreatment appears to be more effective on agricultural residues and of various
herbaceous crops and grasses, including alfalfa, wheat straw, corn stover and rice straw,
etc. than on substrates derived from wood (Holtzapple et al., 1991.). Although testing on
woody substrates has not been extensively reported. Over 90% hydrolysis of cellulose and
hemicellulose has been obtained after AFEX pretreatment of Bermuda grass, but
hydrolysis for the biomass with high lignin-content (20-25%) such as aspen wood chips
was reported as only 50%. AFEX pretreatment can significantly improve the
saccharification rates
The ammonia pretreatment does not produce inhibitors for the downstream biological
processes, so water wash is not necessary (Dale and Moreira 1982, Mes-Hartree et al.,
1987). The AFEX pretreatment does not significantly solubilise hemicellulose compared to
acid pretreatment or acid catalysed steam explosion. (Mes-Hartree et al., 1988)
Mes-Hartree et al. compared the steam and ammonia pretreatment for enzymatic
hydrolysis of aspen wood, wheat straw and alfalfa steams and found that steam explosion
solubilise the hemicellulose, while AFEX did not.
2.2.3.4. CO2 explosion
Similar to steam and ammonia explosion pretreatment, CO2 explosion is also used for
pretreatment of lignocellulosic materials. It was hypothesized that CO2 would form
carbonic acid and increase the hydrolysis rate. Dale and Moreira (Dale 1982) used this
method to pretreat alfalfa (4 kg CO2/kg fiber at the pressure of 5.62 MPa) and obtained
75% of the theoretical glucose released during 24 h of the enzymatic hydrolysis. The
yields were relatively low compared to steam or ammonia explosion pretreatment, but
high compared to the enzymatic hydrolysis without pretreatment. It was found, that CO2
explosion was more cost-effective than ammonia explosion and did not cause the
formation of inhibitory compounds that could occur in steam explosion.
25
2.2.4. Chemical pretreatment
Chemical pretreatments can be simple; such as soaking the biomass in sodium hydroxide at
room temperature, or more complicated as, for example, treating the material with acidic or
basic catalysis at high-temperature (Stenberg et al., 2000a,b, Kaar et al., 1998, Holtzapple
et al., 1991). Especially these latest methods are very effective. During chemical
pretreatment processes, hemicellulose and/or lignin may be hydrolysed to their monomeric
constituents and lignin-cellulose-hemicellulose interactions are partially disrupted, thus
increasing the enzymatic digestibility of cellulose.
2.2.4.1. Acidic pretreatment
Concentrated acids such as H2SO4 and HCl have been used to pretreat lignocellulosic
materials. Although they are powerful agents for cellulose hydrolysis, concentrated acids
are toxic, corrosive and hazardous and require reactors that are resistant to corrosion. In
addition, the concentrated acid must be recovered after hydrolysis to make the process
economically feasible (Von Sivers and Zacchi 1996).
Dilute acid hydrolysis has been successfully developed for pretreatment of lignocellulosic
materials. The dilute sulphuric acid pretreatment can achieve high reaction rates and
significantly improve cellulose hydrolysis at high temperature (Esteghlalian et al., 1997),
but at moderate temperature the saccharification yields of cellulose is quite poor.
(McMillan 1994).
However, if the hydrolysis of the hemicellulose fraction is the aim of the experiments, low
temperature is more favourable. Recently developed dilute acid (mainly sulphuric acid)
hydrolysis processes use less severe conditions and achieve high xylan to xylose
conversion yields. Achieving high xylan to xylose conversion yields is necessary to
achieve high overall process economics because xylan accounts for up to a third of the
total carbohydrate in many lignocellulosic materials (Hinman et al., 1992).
Wilke et al. (1981) used dilute sulphuric acid pretreatment efficiently for saccharification
of hemicellulose in various herbaceous materials, e.g. wheat straw and rice straw. However
the obtained conversion of cellulose to glucose was only 40% and the overall yield of all
polysaccharides by enzymatic hydrolysis was also quite low, around 60%.
Although dilute acid pretreatment can significantly improve the cellulose hydrolysis,
its cost is usually higher than some physico-chemical pretreatment processes such as
steam explosion or AFEX. A neutralization of pH is necessary before enzymatic
hydrolysis or fermentation processes.
26
2.2.4.2. Alkaline pretreatment
Some bases, e.g. sodium-, potassium-, calcium-, and ammonium-hydroxide are
appropriate chemicals for pretreatment of lignocellulose, but the effect of alkaline
pretreatment depends on the lignin content of the materials. Alkaline pretreatment
techniques are basically delignification processes, however, generally a significant
amount of hemicellulose solubilise as well. During these mechanisms the porosity of
the lignocellulosic materials increases with the removal of the crosslinks of
intermolecular ester bonds between xylan hemicelluloses and other components, for
example, lignin and other hemicellulose (McMillan 1994). Dilute NaOH treatment of
lignocellulosic materials caused swelling, leading to an increase in internal surface
area, a decrease in the degree of polymerization, a decrease in crystallinity, separation
of structural linkages between lignin and carbohydrates, and disruption of the lignin
structure. Compared with acid processes, alkaline pretreatment results in lower
degradation of sugars.
The digestibility of NaOH-treated hardwood increased from 14% to 55% with the decrease
of lignin content from 24-55% to 20%. However, no effect of dilute NaOH pretreatment
was observed for softwoods with lignin content greater than 26% (Millet et al., 1976).
Dilute NaOH pretreatment was also effective for the hydrolysis of straws with relatively
low lignin content of 10-18% (Bjerre et al., 1996).
MacDonald et al. (1983) obtained 77.5% overall conversion applying a dilute sodium
hydroxide pretreatment at high temperature for corn stover. Elshafei et al. (1991) achieved
nearly the theoretical conversion maximum for cellulose, by soaking the corn stover in
1.0 M sodium hydroxide for 24 h at room temperature. Kaar et al. (2000) reported 88.0%
conversion of cellulose to glucose using slake lime as a pretreatment for 4 h at 120°C.
Moreover, lime gives the possibility to recover calcium quite easily, as an insoluble
calcium carbonate.
Ammonia was also used for the pretreatment to remove lignin. Iyer et al. (1996) described
an ammonia recycled percolation process (temperature, 170°C; ammonia concentration,
2.5-20%; reaction time 1 h) for the pretreatment of corn cobs/stover mixture and
switchgrass. The efficiency of delignification was 60-80% for corn cobs and 65-85% for
switchgrass.
2.2.4.3. Organosolv process
In the organosolv process, an organic or aqueous organic solvent mixture alone, or with
addition of an acid (HCl or H2SO4), or rarely alkaline catalyst is used to break the internal
lignin and hemicellulose bonds. Organic solvents, such as methanol, ethanol, acetone,
ethylene glycol, etc. are used in this process (Chum et al., 1988). If the process is
conducted at high temperature (T>185°C), there is no need for acid addition, as it is
27
believed that organic acids released from the wood or from other lignocellulosic material
act as catalysts for the rupture of the lignin – carbohydrate complex. However usually, a
high yield of xylose can be obtained with the addition of acid (Wright 1988). Both the
hemicellulose and the lignin fraction are solubilized, while the cellulose remains as a solid.
The cellulose fraction from organosolv pretreatment is susceptible to enzymatic hydrolysis.
Chum et al. (1988) reported more, than 85% conversion of cellulose to glucose, following
pretreatment at 195°C.
However, because organic solvents are costly, potentially dangerous, and inflammable and
their use requires high-pressure equipment, the process is perceived as complex and
expensive. (Schell et al., 1991.) Solvents used in the process need to be drained from the
reactor, evaporated, condensed and recycled to reduce the cost. Removal of solvents from
the system is also necessary because the solvents may be inhibitory to the growth of
organisms, enzymatic hydrolysis, and fermentation.
2.2.4.4. Ozonolysis
Ozone can be used to degrade lignin and hemicellulose in many herbaceous materials.
Ben-Ghedalia used this technique successfully for wheat straw (Ben-Ghedalia and Miron
1981). The degradation was essentially limited to lignin and hemicellulose was slightly
attacked, but cellulose was hardly affected. The yield of enzymatic hydrolysis increased
from 20% to 57%, following 60% removal of the lignin from wheat straw in ozone
pretreatment. Ozonolysis pretreatment has the following advantages:
(1) it effectively removes lignin;
(2) it does not produce toxic residues for the downstream processes; and
(3) the reactions are carried out at room temperature and pressure.
However, a large amount of ozone is required, making the process expensive.
2.3. HYDROLYSIS PROCESSES
Most processes for the production of ethanol from lignocellulosic materials have similar
designs based on feedstock handling, hydrolysis, fermentation and distillation. Big
difference usually lies in how the hydrolysis step is performed. Therefore, the processes
have been divided into two big groups, according to the design of the hydrolysis steps,
enzymatic and acid hydrolysis processes (Figure 2.6.).
28
Figure 2.6. Schematic flowchart of the “ethanol from lignocellulosic biomass” process
2.3.1. Acid hydrolysis
Acid hydrolysis of plant lignocellulosic biomass has been known since 1819. (Von Sivers
1989). Acid hydrolysis can be performed with several types of acids, including sulphuric,
hydrochloric, hydrofluoric, phosphoric, nitric and formic acid. These acids may be either
concentrated or diluted. Processes involving concentrated acids are operated at low
temperature and give high yields (e.g. 90% of theoretical glucose yield), but the large
amount of acids causes problems both in economical and environmental aspect.
Furthermore, when sulphuric acid is used the neutralisation process produces large
amounts of gypsum. However, the process has attached some new interest due to novel
economic methods for acid recovery proposed.
Dilute acid hydrolysis is fast and easy to perform and it has the advantage of the relatively
low acid consumption, but is hampered by non-selectivity and by-product formation.
Namely, high temperatures are required to achieve acceptable rates of conversion of
cellulose to glucose, but high temperatures increase also the rates of hemicellulose sugar
decomposition and equipment corrosion. Under these conditions, xylose degrades to
furfural and glucose degrades to 5-hydroxymethyl furfural (HMF), both of which are toxic
29
to microorgaisms and can also cause inhibition in the subsequent fermentation stage. The
maximum yield of glucose is obtained at high temperature and short residence time, but
even under these conditions the glucose yield is only between 50% and 60% of the
theoretical.
A two stage acid hydrolysis process enables the hemicellulose and cellulose to be degraded
separately under conditions appropriate for each reaction. In a pre-hydrolysis or
pretreatment stage under rather mild conditions the relatively easily hydrolysed
hemicellulose is removed. These enables the second acid hydrolysis step to proceed under
harsher conditions without degrading the hemicellulose sugars to furfural and other
substances. Using a two-stage dilute acid hydrolysis process, recovery yields of as much as
70-98% of xylose, arabinose and galactose were obtained from corn stover. However the
yield of glucose 60% was still quite low (Wilke et al 1981).
On the top of the Figure 2.6. is the concentrated acid process that uses sulphuric acid to
hydrolysis lignocellulosic biomass prior to fermentation to ethanol. Very efficient
recycling is required for the process to be cost effective. The second part of Figure shows
the two stage acid hydrolysis process prior fermentation. The processes depicted along the
bottom of the figure is the separate enzymatic hydrolysis and fermentation (SHF) and
simultaneous saccharification and fermentation (SSF) processes.
2.3.2. Enzymatic hydrolysis
The biodegradation of lignocellulosics has been investigated since the 1960’s (Philipidis et
al., 1993, 1996). Using cellulase enzymes as catalyst, a very specific conversion of
cellulose is performed. There is wide range of microorganisms, both bacteria and fungi,
which produce enzymes capable of hydrolysing cellulose. Of all these microorganisms,
Trichoderma species have been most extensively studied for cellulase production.
(Coughlan et al., 1992)
During the experimental work of this thesis, a commercial available cellulase (Celluclast
1.5 L from Novozymes) was used in the enzymatic hydrolysis, which is also a cellulase
enzyme complex by Trichoderma reesei.
Cellulases are usually a mixture of several cellulolytic enzymes. At least three major
groups of cellulases are involved in the hydrolysis process (Parisi 1989):
(1) endoglucanases (EG), which are highly active on amorphous cellulose, creating free
chain ends;
(2) exoglucanases or cellobiohydrolases (CBH), which degrade the molecule further by
removing cellobiose units from the free chain-ends, and
(3) ß-glucosidase, which hydrolyses cellobiose to produce glucose (Coughlan et al., 1988).
30
Strictly speaking, “cellulases” is defined as the endo- and exocellulases and does not
include ß-glucosidase. However it has a very important role in hydrolysis, since cellobiose
is highly inhibitory to many cellulases and is unusable by most microorganisms.
ß-glucosidase hydrolyses cellobiose to glucose which is much less inhibitory and highly
fermentable. (Holtapple et al., 1991) On the other hand ß-glucosidase is inhibited by
glucose. Cellulases are inhibited by end-products, thus the build-up of any of these
products decrease the efficiency of hydrolysis.
Because of the specification of cellulases it is considered to have the potential of higher
yields than acid hydrolysis. Other advantages of enzymatic hydrolysis are the mild
operating conditions and the high quality sugar products. The maximum cellulase activity
for most cellulases and ß-glucosidase occurs at 50±5°C and a pH 4.0-5.0.(Saddler and
Gregg 1998) In addition to these, enzymes are naturally occurring compounds which are
biodegradable and therefore environmental friendly. However nowadays the enzymatic
reactions are quite slow and the biomass must be pre-treated to improve the yields and
kinetics.
2.3.2.1. Surfactant effect in enzymatic hydrolysis
Cellulase enzyme loading of 10 – 20 FPU/g cellulose is often used in laboratory studies
because it provides a hydrolysis profile with high level of glucose yield in a reasonable
time (24-72 h) at a reasonable enzyme cost (Gregg and Saddler, 1996).
High cellulose conversion requires high enzyme loading, which has strong negative
effects on the process economy. Thus, methods to increase enzyme effectiveness are
important for reduction of enzyme consumption. Addition of surfactants to enzymatic
hydrolysis of lignocellulose gives a possibility to increase the conversion of cellulose
into soluble sugars, without increasing the enzyme loading.
It is believed, that enzymatic hydrolysis of cellulose consists of three steps: adsorption of
cellulase enzymes onto the surface of the cellulose, the biodegradation of cellulose to
fermentable sugars, and desorption of cellulase. Cellulase activity decreases during the
hydrolysis. The irreversible adsorption of cellulase on cellulose is partially responsible for
this deactivation (Converse et al., 1988).
Adsorption of enzymes to cellulose during hydrolysis has been shown to decrease in the
presence of surfactant (Castanon et al., 1981, Helle et al., 1993). Different mechanisms
have been proposed for the positive effect of surfactant addition in the enzymatic
hydrolysis of cellulose. The surfactant could change the nature of the substrate, e.g. by
increasing the available cellulose surface or by removing inhibitory lignin (Helle et al.,
1993). Based on kinetic analysis, Kaar and Holtzapple (1998) have found indications
that surfactants could promote the availability of reaction sites, which would increase the
31
hydrolysis rate. The surfactant could also increase the stability of the enzymes and thus,
reduce enzyme denaturation during the hydrolysis. Surfactant effects on enzymesubstrate interaction have been proposed, e.g. adsorbed enzymes are prevented from
inactivation by addition of surfactant which facilitates desorption of enzymes from
substrate (Park et al., 1992).
Enhancement of cellulose hydrolysis by adding surfactants to the hydrolysis mixture has
been reported by several authors (Helle et al., 1993, Ooshima et al., 1986, Castanon and
Wilke 1981, Kaar et al., 1998) and different cellulose substrates have been studied.
Castanon and Wilke (1981) showed that conversion of newspaper was increased by 14%
after 48 h hydrolysis by the addition of Tween 80. They also investigated the effect of
Tween additives for enhancing enzymatic saccharification on non-native cellulose and
cellulose
analogs
such
as
newsprint,
microcrystalline
cellulose
and
carboxymethylcellulose (CMC). These researchers postulated that Tween 80 assists the
desorption of enzyme from substrate. Park et al. (1991), also working with newspaper
examined several surfactants and found Tween to be among the best performers, resulted
10% increase in the cellulose conversion. Increased hydrolysis by addition of surfactants
has also been reported for steam-exploded wood (Helle et al., 1993, Erikkson et al., 2002,
Alkasrawi et al., 2003), bagasse (Kurakae et al, 1994), and lime-pre-treated corn stover
(Kaar et al., 1998). Eriksson et al. (2002) found, that addition of non-ionic surfactant
increased significantly the conversion of steam-pretreated spruce. With addition of Tween
20 at 2.5 g L-1 it was possible to lower the enzyme loading by 50% and at the same time
retain cellulose conversion.
Ooshima et al. (1986) compared amorphous cellulose with different types of crystalline
celluloses (Avicel, tissue paper and reclaimed paper). They showed that the higher the
crystallinity of the substrate, the more positive was the effect of the added surfactant. It
was also observed, that the surfactant effect is higher at low cellulase concentration. In
addition to these Eriksson et al. (2002) published, that the effect of the surfactants was
significantly lower, when delignified cellulose substrate was used. They proposed, that
the dominating mechanism for surfactant effects in substrates containing lignin, be found
in the role of surfactant on enzyme interaction with lignin surfaces. Surfactant adsorption
to lignin prevents unproductive binding of enzymes to lignin.
The surfactants used in the enzymatic hydrolysis include non-ionic Tween 20,
(polyoxyethylene sorbitan monolaurate) and Tween 80 (polyoxyethylene sorbitan
monooleate) (Kaar et al., 1998), polyoxyethylene glycol (Park et al., 1992), sophorolipid,
rhamnolipid, and bacitracin (Helle et al., 1993). Inhibitory effects have been observed with
cationic Q-86W at high concentration and anionic surfactant Neopelex F-25, thus non ionic surfactants are therefore believed to be more suitable for enhancing the cellulose
hydrolysis.
32
After the published positive results of the surfactant on enzymatic hydrolysis, it was
suspected that surfactant could increase the enzymatic saccharification of a native
lignocellulose like corn stover. Preliminary experiments were carried out on ground corn
stover improving the enzyme efficiency using Tween 80 supernatant. It was added to the
mixture of corn stover and sodium citrate buffer (pH 4.8), before the loading of
cellulase. The dry matter (DM) content in the mixture was 5% and the enzyme loading
was 25 FPU/ g DM. The recommended Tween loading, according to the results of Kaar et
al (1998) was 0.15 g Tween/g dry biomass.
The achieved conversion of ground corn stover with Tween 80 was 21% compared to
19% without addition of Tween. Thus, contrary to my all expectations the conversion of
cellulose to glucose increased by only 8% after 24 h hydrolysis by the addition of Tween
80.
2.4. FERMENTATION FOR BIOETHANOL PRODUCTION
Several enzyme-catalysed processes have been emphasised for the conversion of
lignocellulosic biomass into ethanol: separate hydrolysis and fermentation (SHF),
simultaneous saccharification and fermentation (SSF), simultaneous saccharification and
co-fermentation (SSCF) and direct microbial conversion (DMC).
2.4.1. Separate hydrolysis and fermentation (SHF)
The main advantage of separate hydrolysis and fermentation (SHF) process is the
ability to carry out both the hydrolysis and the fermentation under optimal conditions, e.g.,
enzymatic hydrolysis at 40-50°C, which is optimal temperature of the cellulases as it was
mentioned, and fermentation at about 30°C, which is required temperature for most ethanol
fermenting microorganisms, e.g. Saccharomyces cerevisae. The major drawback of SHF is
that the released sugars inhibit the cellulases during hydrolysis, therefore the hydrolysis
rate in SHF is strongly affected by end-product inhibition (Alfani et al., 2000)
2.4.2. Simultaneous saccharification and fermentation (SHF)
The simultaneous saccharification and fermentation (SSF) process was published in
1977 by Takagi et al (Takagi et al., 1977, Gauss et al., 1976) and has proven to be a
promising alternative to SHF. The key to the SSF process is its ability to rapidly convert
sugars into ethanol avoiding their build-up in their fermentation broth. In SSF the produced
glucose is immediately consumed by the fermenting microorganism. The ethanol produced
can also act as an inhibitor in the conversion process, but sugars in the enzymatic
hydrolysis of cellulose, are much more inhibitory than ethanol is (Takagi 1984). Thus SSF
process can achieve greater rates, yields, and concentrations than competing process now
known. (Wyman et al., 1992) The SSF process also eliminates expensive equipment and
33
reduces the probability of contamination by unwanted organisms that are less ethanol
tolerant than microbes selected for fermentation. The drawback associated with this
process is that the optimum conditions, especially the optimum temperature for the
cellulases and the microorganism differ (Stenberg et al., 2000a, b). Therefore using
thermotolerant yeast in SSF process have been intensively investigated. The simplified
diagram in Figure 2.6. shows possible processes for ethanol production, including both the
SHF and SSF processes.
2.4.3. Simultaneous saccharification and co-fermentation (SSCF)
In simultaneous saccharification and co-fermentation (SSCF), the fermentation of
hexoses and pentoses is carried out simultaneously with various fermenting microorganism
(Himmel et al., 1997). When the lignocellulosic substrate has a high content of pentoses,
e.g. xylose, a separate pentose fermentation step is required to convert the substrate to
ethanol economically, as the currently used fermentative microorganism in SSF does not
ferment pentoses. The SSCF process option offers the possibility of substantial capital and
operational savings, by reducing the number of required reactors and operating cost
(Padukone 1996).
2.4.4. Direct microbial conversion (DMC)
It is also possible to include enzyme production in the same stage as the hydrolysis and
fermentation, so-called direct microbial conversion (DMC). Today, the ethanol yield in
this process is quite low, several metabolic by-products are formed, and the
microorganisms presently used have a low ethanol tolerance. However genetically
engineered strains may improve the performance in the future (Philipidis et al., 1996).
Biostil is a new technology with continuous alcohol removal, which combines the separate
hydrolysis and fermentation with simultaneous ethanol distillation. The first commercial
scale Biostil plant was established in 1981 in Australia using molasses and cane juice
syrup. In 1982, a Biostil plant using a mixture of molasses and cane syrup, was also built in
Brazil. In this process, concentrated feedstocks (typically 40-55% (w/w) molasses and
sugar cane syrup) are conducted to the fermenter at a constant rate, where a low
concentration of ethanol is maintained at around 5% (w/w) by constant ethanol removal,
thereby the by-product inhibition, the formation of glycerol (main nonvolatile by-product),
and the risk of infection are decreased to a minimum value. The yeast recovery and
recycling reduce the sugar consumption for yeast growth. The fermentation liquid is
pasteurized during recycling, after ethanol distillation, which also reduces the possibility of
contamination. The main benefit of this technology is the reduction in stillage volume.
Since the stillage causes a significant pollution problem as well, its treatment is one of the
major cost factors in the distillation process (Bollók 1999).
34
The most efficient microorganisms for converting glucose into ethanol, e.g. industrial yeast
strains of Saccharomyces cerevisiae or fermenting bacterial strains of Zymomonas mobilis
are not able to utilise xylose and arabinose, the major hemicellulosic derived sugars.
(Ahring et al., 1996)
Xylose-fermenting yeast such as, Pichia stipitis and Candida shehatae, can ferment both
C5 and C6 sugars (Delgenes et al., 1996). However for the glucose fermentation it is
preferable to use well known glucose-fermenting microorganisms like S. cerevisiae,
because the ethanol yield and the productivity from glucose is much better with this
microorganism (Laplace et al., 1993, Du Preez et al., 1986). Recently, recombinant strains
of S. cerevisae, Z. mobilis, and Escherichia coli have received the genes coding for
enzymes for conversion of xylose into ethanol. The cloned genes in recombinant S.
cerevisae were expressed, but the formation of ethanol was limited (Hahn-Hägerdal et al.,
1993). some recombinant E. coli strains had a promising ethanol yield, but were very
sensitive to inhibitors in hemicellulose fraction (Lawford and Rousseau 1992). The aim of
these studies is to find or create yeast species, which would be able to ferment both pentose
and hexose sugars, thus both cellulosic and hemicellulosic derived sugars from
lignocellulose could be expected to be converted to ethanol.
35
3. MATERIALS AND METHODS
3.1. RAW MATERIAL
Corn stover, which is the corn stalk and the leaves without root and corncob was applied as
a feedstock in the investigations presented in this thesis. Corn stover was grown in South
Hungary, and harvested in autumn following the harvest of the mature crop.
Corn stover hemicellulose contains 85-90% xylan and the cellulose fraction is almost
entirely glucan. The lignin content of the feedstock is most generally based on the
gravimetric method of Klason (72% sulphuric acid method) or one of its permutations
which separates lignin based on its insolubility in sulphuric acid.
Table 3.1. Percent composition of corn stover from year to year
Ash
Glucan
Xylan
Lignin
1999
2000
2001
6.3 ± 0.2
41.3 ± 2.3
27.9 ± 1.7
22.1 ± 1.4
5.1 ± 0.2
39.5 ± 2.2
29.9 ± 1.8
23.6 ± 1.4
3.6 ± 0.1
41.6 ± 2.2
27.7 ± 1.7
20.6 ± 1.3
The composition of corn stover varied a little from year to year, because it depends on the
weather (rainy or dry), the time of the harvesting, etc. The compositions of corn stover
(determined with Hägglund’s method) in different years are shown in Table 3.1.
The selected and washed corn stover was air-dried to an average 90% dry matter (DM)
content and ground to approximately to 3 mm particle size. The dried and ground straw
was stored in the dark in paper bags at room temperature.
Analysis of raw material
The compositions of the raw materials and the solid fractions obtained after pretreatments
were analyzed using various analytical methods. One of these was a gravimetric method
by Goering and Van Shoest (1970) and the others were the more or less modified
Hägglund’s method using strong acid hydrolysis.
The strong acid hydrolysis processes (Hägglund 1951, Kaar et al., 1991) are based on
quantification of neutral polysaccharides in biomass (primary cellulose and hemicellulose).
These methods involve the complete acid hydrolysis of the carbohydrate fractions followed
by quantitative measurement of the resulting sugars in the hydrolysate by either liquid or
gas chromatography. The amount of glucose recovered is then adjusted for polymerisation
36
and reported as glucan. In a similar way the amount of xylose recovered in the hydrolysate
is reported as xylan.
The lignin content of the feedstock is most generally based on the gravimetric method of
Klason (72% sulphuric acid method) or one of its permutations which separates lignin
based on its insolubility in sulphuric acid.
In the modification of Hägglund’s method the air-dried material was placed to the oven
at 105°C for 24 h, than ground in a coffee mill. This fine – ground and weighed corn stover
was hydrolysed with 72% (w/w) H2SO4 for 4 h at room temperature, then the hydrolysate
was diluted 11 times with distilled water in two steps and was boiled for 6 h at 100°C
using reflux cooler. The solution was filtered through a G4 glass filter funnel and the lignin
fraction was gravimetric determined, as the weight of washed, and dried filter cake.
For the analysis of the mono – and disaccharides including glucose, xylose, cellobiose and
arabinose in the supernatant, an Aminex HPX-87H liquid-chromatography column was
used at 65°C. The eluent was 5 mM H2SO4 at a flow rate of 0.5 ml/min. with detection by
refractometric index (RI).
The Kaar’s analysis method (1991), similar to the Hägglund’s method, is based on a
strong acid hydrolysis. Approximately 0.15 g DM solid material was weighed and
hydrolysed in 1.5 mL 72% H2SO4 for 1 hour at 30°C. Then 42.5 mL water was added and
the capped Pyrex tubes were autoclaved 1 hour at 121°C. The acid hydrolysate was filtered
and the resulting monosaccharides (glucose, xylose and arabinose) were also quantified by
HPLC with RI-detection.
Conversion factor for dehydration on polymerisation to cellulose was 162/180 for glucose
and to arabinoxylan was 132/150 for xylose and arabinose. The residue was washed, dried
and weighed and reported as Klason lignin.
In the modified gravimetric method described by Goering and Van Soest (1970) the
solid materials were analyzed for their concentration of the different fibres: hemicellulose,
cellulose and lignin, and the non-cell wall material (NCWM) (water-soluble substances or
extractives such as pectin, proteins etc.).
This method is based on the determination of the neutral (NDF) and acid detergent fibres
(ADF) followed by a permanganate procedure and incineration. In the NDF analysis, the
sample material was boiled in a neutral detergent solution for 1 hour, after which the
suspension was filtered quantitatively, washed, dried, and weighed. The solid fraction
was defined as the NDF. The neutral detergent washed and removed soluble components:
lipids, protein, free sugars and water-soluble minerals.
In the ADF analysis, the sample material was boiled for 1 hour in an acid detergent
solution (0.5 M sulphuric acid). The solid fraction after filtration and drying was defined
as the ADF. The acidic detergent washed and removed components of the hemicellulose
37
fraction. The lignin content was determined by treating the ADF residual with potassium
permanganate and acetate-buffer for 90 minutes. The solid residual fraction was then
incinerated. The hemicellulose content was calculated to be the solid removed by the
ADF analysis, the lignin content the solid removed by the permanganate step, and the
cellulose content the solid removed by the incineration step. Additionally, the content of
NCWM was the solid removed in the NDF analysis. All samples were analyzed in
duplicate and results were given as a dry matter percentage.
The material was analysed both with the strong acid hydrolysis and the gravimetric
Goering and Van Soest’s method to compare different analysis methods. The analysis gave
similar results except in determination of lignin content, because “Klason lignin” contains
all components, which were not solubilised during the sulphuric acid hydrolysis. If the
lignin content, according to the gravimetric method, was added to the NCWM content, the
sum was found to be approximately equal to the “Klason lignin” content determined by
strong acid hydrolysis. The compositions after different determination’s method are shown
in Table 3.2.
Lignin Klason ≅ Lignin Gravimetric + NCWM Gravimetri c
(3)
Table 3.2. Percent composition of corn stover (harvested in 2000) determined by different
analysis methods
Ash
Glucan
Xylan
Lignin
NCWM*
A
B
C
5.1 ± 0.2
39.5 ± 2.2
29.9 ± 1.8
23.6 ± 1.4
6.4 ± 0.2
40.6 ± 2.2
28.7 ± 1.8
22.4 ± 1.4
4.9 ± 0.2
41.0 ± 0.5
33.7 ± 0.4
8.7 ± 0.4
11.7 ± 0.2
20.4 ± 0.6
NCWM + lignin
A
Hägglund’s method
Kaar’s method
C
Gravimetric method by Goering and Van Soest
B
To measure the total ash content, approximately 0.5 g sample of corn stover was placed in
a crucible, and ignited at 550°C for 3 hours, cooled in a desiccator, and weighed.
38
3.2. PRETREATMENTS
3.2.1. Chemical pretreatments
Alkaline and acidic pretreatments
The ground corn stover was treated with different chemicals under the same conditions.
Table 3.3. gives the applied chemicals and their concentration. Pretreatments were carried
out in 1-L round-bottomed flasks, which were placed in autoclave and boiled for 1 h under
pressure at 120°C. Time zero for all reactions corresponded with the temperature reaching
120 °C Thirty grams of raw corn stover (27.8g dry matter [DM]) was pre-treated in each
case, and the concentration of the substrate was 10% (w/w) DM; thus, the total amount of
the slurry was 300 g. After pretreatment, the slurry was separated into liquid and solid
fraction, which was thoroughly washed with distilled water. Both the solid and liquid
fraction and also the wash water were analysed using Hägglund's method combined with
high-performance liquid chromatography (HPLC) sugar analysis.
In other cases, the solids were not separated from the liquid phase after acidic pretreatment,
but the whole slurry was hydrolysed after the pH was adjusted to 4.8 with 20% NaOH. The
purpose of this experiment was to gain information whether the enzymes were inhibited
when the whole slurry was used by hydrolysis.
Table 3.3. Applied chemicals and concentrations during pretreatments
ACIDIC PRE-
ALKALINE PRE-
TREATMENT
TREATMENT
([W/W]%)
([W/W]%)
1% H2SO4
5% H2SO4
1% HCl
5% HCl
1% NaOH
5% NaOH
10% NaOH
2% Ca(OH)2
TWO-STEP PRETREATMENTS
note
Soaking base
Acid
([w/w]%)
([w/w]%)
TS 1
TS 2
TS 3
TS 4
1% NaOH
5% NaOH
1% Ca(OH)2
1% NaOH
1% HCl
1% HCl
1% HCl
1% H2SO4
Combined alkaline and acidic pretreatments (two-step pretreatments)
In these pretreatments, the corn stover was first soaked in base for 1-d at room
temperature. Then it was filtered carefully, washed, and finally boiled with acid as
described earlier. The sugar components of the liquid fraction were analysed by HPLC.
Pretreatments with dilute base and acid
After the attractive results of the first series of experiments, dilute sodium-hydroxide
(0.5%) and dilute sulphuric acid (0.5% and 2%), as pretreatment agents were tested in
further investigations. In these pretreatments the conditions were similar to the other, 30 g
39
air dried corn stover was boiled with 270 ml “reagent” (dilute NaOH or H2SO4) for 30, 60
and 90 min, at 120°C and also at 100°C in autoclave. After pretreatment, the slurry was
separated to liquid and solid fractions, and the latter was thoroughly washed with distilled
water to remove all soluble particles prior to analysis and hydrolysis. The slurry following
the pretreatment was also separated into liquid and solid fraction, and the solid fraction
was thoroughly washed with distilled water. The sugar components of the liquid fraction
were analysed by HPLC.
3.2.2. Steam pretreatment
A schematic flow chart of the pretreatment equipment is shown in Figure 3.1. The
pretreatment vessel has an inner diameter of 0.1 m and a volume of 2.4 dm3. Steam is
introduced into the reactor from a boiler which produces saturated steam up to
4000 kPa. Pre-treated material is collected in a cyclone connected to the outlet of the
pressurised vessel. A computer, ABC 800 (Luxor AB, Motala, Sweden), is used to
control valve operation and to record temperature via a thermocouple (Stenberg et al.,
1998).
Figure 3.1. Schematic flow chart of the steam pretreatment equipment
Figure 3.2. shows the schematic experimental procedure used in steam pretreatment
Ground corn stover (180 g DM) was impregnated with 1720 mL diluted sulphuric acid
40
(0.5 or 2%) or with distilled water, and was swollen one night. Then the material was
steam treated in an equipment described above. Three variable factors were investigated:
the concentration of the sulphuric acid (0.5% and 2%), reaction temperature (190, 200 and
210°C) and reaction time (2 and 5 min). Each experiment was performed in duplicate and
he order of experiments was randomized.
Figure 3.2. The schematic experimental procedure used in steam pretreatment
The pretreatment vessel was preheated with steam prior to loading of the impregnated corn
stover. The corn stover was heated by steam to the desired temperature and, when the
preset pretreatment time had elapsed the material was discharged into a flash drum.
The pre-treated material was then separated, by filtration, into a solid fraction and a filtrate
and both fractions were analyzed. The solid residue was washed with distilled water to
remove water-soluble components and the insoluble fiber fraction was enzymatically
hydrolyzed. A portion of the filtrate was hydrolyzed with sulphuric acid to determine the
41
amount of the released sugars in oligomer form during the pretreatment. Another portion of
the filtrate was complemented with glucose and used in a fermentability test with
Saccharomyces cerevisea in the form of compressed baker’s yeast. The sugar yield in
enzymatic conversion and the fermentability were used to estimate the optimal conditions
for pretreatment.
3.2.3. Wet oxidation pretreatment
Wet oxidation was carried out in a specially designed loop autoclave (2 L) constructed
at the Risø National Laboratory (Figure 3.3.) charged with a working volume of 1 L.
The reactor was made of Sandvik Sanicro 28 (27% Cr, 31% Ni, 3.5% Mo, 1% Cu) and
mounted on a rack facilitating the control of temperature by immersing the reactor in an
appropriate heating and cooling bath ( Bjerre and Sørensen, 1992, Bjerre et al., 1996 ).
Due to the excellent heat-transfer conditions, the heating and cooling times were very short
(about 2 min), which made the pretreatment much more controllable and reproducible and
it is suitable also for studies of reaction kinetics.
Figure 3.3. Schematic diagram of wet oxidation equipment
The wet oxidation condition was selected based on previous pretreatment studies on wheat
straw (Schmidt et al., 1998, Klinke et al., 2001, 2003). In “alkaline wet-oxidation” sixty g
(DM) of ground (approximately 3 mm particle size) corn stover was mixed with 2 g
42
Na2CO3 and 1 L water before adding oxygen pressure and heating the suspension. Sodium
carbonate was chosen because it was cheaper than most other water-soluble alkaline
chemicals since lime was not suitable to use in the process In the “acidic pretreatment” 1.9
mL 36.5% (w/w) H2SO4 was added instead of sodium-carbonate. There was also
pretreatment without any chemical addition (“neutral pretreatment”). A statistical partial
factorial design was used to determine the importance of different pretreatment parameters.
Three variable factors were selected: pH, reaction temperature, and reaction time. The
oxygen pressure and the concentration of corn stover were kept constant.
All pretreatments were run within 1 week by a single operator. Each experiment was
performed in duplicate and the order of experiments was randomized.
After the pretreatments, half of each biomass suspension was filtered to separate the solid
cellulose-rich fraction from the liquid hemicellulose-rich fraction. The pH of the liquid
fraction was measured and the solid fraction dried in a climate cabinet at 20°C and 65%
relative humidity and weighed. The composition of each fraction was analyzed and
compared to the untreated, (only ground) corn stover. The filtrate and the non-separated
slurries were stored frozen (-20°C) for further analysis including enzymatic hydrolysis.
3.2.4. Supercritical CO2 pretreatment
For the supercritical CO2 pretreatment 3 g (DM) ground (approx. 3 mm particle size)
corn stover was placed in a high pressure vessel with an effective working volume of
30 mL. Supercritical CO2 was delivered to the vessel, than it was heated to a desired
pretreatment temperature by heating tape. The internal temperature of the high –
pressure vessel was measured by an inserted thermocouple, and was regulated by a
temperature controller. Figure 3.4. shows the schematic experimental procedure used in
supercritical CO2 pretreatment.
Prior to supercritical CO2 pretreatment certain amounts (27 g) of dilute sulphuric acid
(2%) or 2% sodium hydroxide or distilled water was added to corn stover and was swollen
one night. The pretreatment reactor was filled with this slightly wet, impregnated
material. The conditions of the reaction are summarized in Table 3.4.
Table 3.4. The conditions of the supercritical CO2 pretreatment
REACTION CONDITIONS
moisture content
90%
reaction temperature
75°C
pressure
110 bar
reaction time
used chemicals prior pretreatment
60 min
water / 2% NaOH / 2% H2SO4
43
Figure 3.4. Experimental procedure used in supercritical CO2 pretreatment
3.3. ENZYMATIC HYDROLYSIS
To evaluate the efficiency of the pretreatment, the convertibility of the solid cellulose
fraction to fermentable glucose was determined using a mixture of two commercially
available enzyme solutions: the cellulase Celluclast 1.5 L (and Cellubrix after WO
pretreatment) and the ß-glucosidase Novozym 188. The enzyme preparations were kindly
donated by Novo Nordisk A/S, Bagsværd, Denmark. ß-glucosidase was added as a
supplement to the cellulase, at ratio of 1:1 IU of ß-glucosidase to IU cellulase. It was
shown in previous studies (Spindler et al, 1989) that the ß-glucosidase supplementation is
necessary to achieve efficient cellulose conversion.
The activity of cellulase was determined in filter paper unit* (Ghose 1987) and the
activity of ß-glucosidase in CBU** (Berghem and Petterson 1974).
*
1 FPU = 1 µmole glucose /min /mL enzyme. In the filter paper activity (FPA) assay uses a strip
(1 x 6 cm) from Whatman No1. filter paper as a cellulose substrate.
**
1 CBU = 1 µmole glucose /min /mL enzyme. Contrary to the FPA assay a substrate analogue (PNPG,
para-nitrophenil-ß-D-glucopyranoside) is used in the ß-glucosidase assay. Similarly, as the ß-glucosidase
enzyme cuts the cellobiose in two glucose monomer, it is also cuts the PNPG substrate analogue in two
parts, thus releases the glucose.
1 mol cellobiose ↔ 2 mol glucose,
1 mol PNPG ↔ 1 mol glucose + 1 mol PNP
1 ß-glucosidase unit = 1 µmole glucose produced from PNPG under assay conditions.
44
Before hydrolysis the pre-treated, washed, solid materials were diluted to 5% (DM) using
0.05 M acetate buffer (pH 4.8). Hydrolysis, following chemical pretreatment, was
performed in 100 mL shaking flasks containing a working amount of 80 g of suspension.
After steam pretreatment 250 mL shaking flasks were used with a total working weight of
150 g. The hydrolysis was carried out in duplicates at 50°C and the flasks were agitated at
300 rpm in a shaker incubator (LAB-Therm, Kühner, Switzerland and New Brunswick
G24, USA). Hydrolysis of slurries after wet oxidation was performed in triplicates and the
results are presented as the mean value. Following wet oxidation the conditions of the
hydrolysis differed from the previously, in this case the fibrous material was diluted to 2%
(DM) using 0.2 M acetate buffer (pH 4.8) and hydrolysis were performed in 10 mL test
tubes, placed in heating bath at 40, 45 and 50°C, for agitation with magnetic stirrers. For
the hydrolysis of the slurries, the samples were diluted with equivalent amount of 0.2 M
citrate buffer pH 4.8.
The reaction time was usually 72 hours to determine the time required for total hydrolysis,
and samples were taken at the start and after 1, 2, 3, 4, 5, 6 and/or 8, 24, 48 and 72 h. The
reducing sugar concentration was analysed by dinitro-salicilic-acid (DNS) assay (Miller
1959) and the monosaccharide composition by HPLC and results were given as
percentage dry matter of cellulose converted to glucose.
The percentage of cellulose enzymatically converted to glucose (ECC) was calculated as a
quotient of liberated glucose (g) during the hydrolysis and weight of cellulose (g) before
enzymatic hydrolysis. The ECC value based on the glucose concentration measured by
HPLC was calculated as (1):
c ⋅V
ECC =
⋅ 100%
m ⋅ 1.11
(1)
where c is the concentration of D-glucose after enzymatic hydrolysis (g/L), V is the total
volume (L), and m the weight of cellulose before enzymatic hydrolysis (g). The 1.11 factor
converts the cellulose concentration to the equivalent glucose concentration.
3.3.1. Effect of ultrasonic waves during enzymatic hydrolysis
The results of the chemical pre-treated corn stover demonstrated that dilute sodium
hydroxide and especially the combined alkaline and acidic pretreatment rendered the
lignocellulose material digestible by cellulase; however, high sugar yields required very
high enzyme loadings.
To enhance the efficiency of the enzymatic hydrolysis of native corn stover, ultrasonic
waves were used, for help the desorption of cellulase from the cellulose fibres. The
hydrolysates were placed in an ultrasonic wave bath for 10 minutes in every hours, than
placed back to the shaker incubator. The enzymatic hydrolysis was performed in sodium-
45
citrate buffer (pH 4.8) at 50°C for 48 h. Contrary to my expectations no positive effect
could be observed after this ultrasonic treatment.
3.3.2. Enzyme activity measurement
The activity of the cellulytic enzymes was measured as filter paper activity units (FPU).
A 1 × 6 cm strip of a Whatman No. 1 filter paper was added to a total volume of 1.5 mL
enzyme solution containing 0.05M sodium-citrate buffer, pH 4.8. The samples were
incubated for 1 h at 50°C. Reducing sugars were determined after stopping the hydrolysis
by additing 3 mL DNS solution followed by boiling for 5 min. After cooling, 20.0 mL
distilled water was added and the UV - absorbance was read at 540 nm (Ghose 1987).
The ß-glucosidase activity was measured by incubating the enzyme solution with 1 µM pnitrophenyl-ß-D-glucopyranoside and 0.05-M citrate buffer, pH 4.5 at 50°C for 10 min.
The reaction was stopped by addition of 2 mL 0.1 M Na2CO3 and the amount of liberated
p-nitrophenol measured spectrophotometrically at 400 nm. One unit of activity was defined
as the release of 1 µmol of p-nitrophenol per minute (Berghem and Petterson 1974).
3.4. ANALYSIS OF THE LIQUID FRACTION AFTER PRETREATMENTS
3.4.1. Poly- and monosaccharide analysis of liquid fraction
Carbohydrates in the separated liquid fractions following pretreatments were partly
polymers/oligomers. Thus to determine the total glucose, xylose and arabinose
concentration these were hydrolyzed with 4% (w/w) H2SO4 at 121°C for 10 min. The
sulphate anions in 10 mL acidic hydrolysate were precipitated by 0.5 g Ba(OH)2.8H2O.
After separation the supernatant was diluted 1:1 with 4 mM H2SO4 eluent. Average
recovery in this purification procedure for glucose, xylose and arabinose was 86, 83, 86%,
respectively. Glucose, xylose and arabinose in the purified samples were analysed by
HPLC, using an Aminex HPX-87H column with a matching precolumn (Biorad) at 63°C.
The eluent was 4 mM H2SO4 at a flow rate of 0.6 mL/min. with detection by refractive
index (RI).
For the yields calculation the measured amount of xylose and arabinose were used as a
measure of the hemicellulose fraction, while the cellulose fraction was calculated from the
amount of glucose and cellobiose. Conversion factor for dehydration on polymerisation to
cellulose and hemicellulose was 162/180 for glucose and 132/150 for xylose and
arabinose.
46
3.4.2. Carboxylic acids and phenols analysis
Carboxylic acids (acetate, formate, glycolate and malate) were quantified also with the
before mentioned HPLC conditions and RI-detection. The hydrolysates were adjusted to
pH 2 with dilute H2SO4, centrifuged (10,000 rpm, 10 minutes at 4°C) and filtered prior to
analysis.
The phenols were selectively extracted from the liquid fraction by solid phase extraction
on polystyrene divinylbenzene polymer columns: IST Isolute ENV+ 100 mg/1 ml
(International Sorbent Technology Ltd., Mid-Glamorgan, UK) as previously described
(Klinke et al., 2002). For analysis of phenol conversion by yeast, the fermentation broth
was centrifuged at 5,000 rpm, and the supernatant was adjusted to pH 7 or pH 2 prior to
extraction. The phenolic extract was diluted with acetonitrile, dried over Na2SO4 and
silylated with BSTFA at 70°C for 30 minutes (EtOAc-acetonitrile-BSTFA (10:5:3). The
phenols were quantified by gas chromatography (Hewlet-Packard GC 6890, Agilent
Technologies, Palo Alto, CA, USA) with flame ionization detection (FID) using authentic
standards on a HP-5 column (Crosslinked 5% Phenyl Methyl Siloxane, 30 m x 0.32 mm
i.d., 0.25 µm film, Agilent Technologies). The sample (1 µL) was injected splitless and the
temperature at the GC injector and detector were 250°C and 300°C, respectively. Helium
was used as carrier gas at a constant pressure of 74 kPa. The GC oven temperature
program was 80°C at 3 min hold and then increased by 6°C/min to 220°C and by 30°C/min
to 280°C at 1 min hold. The phenols, phenol, guaiacol, syringol, 4-hydroxybenzaldehyde,
vanillin, syringaldehyde, 4-hydroxyacetophenone, acetovanillone, acetosyringone, 4hydroxybenzylalcohol, vanillyl alcohol were quantified from the pH 7 extract. Furoic acid
and the phenol acids, 4-hydroxybenzoic acid, vanillic acid, homovanillic acid, syringic
acid, coumaric acid, and ferulic acid were quantified from the pH 2 extract. Authentic
standards were used for calibration.
3.5. SIMULTANEOUS SACCHARIFICATION AND FERMENTATION (SSF)
WET-OXIDATION
AFTER
SSF runs were performed in 0.5 L blue cap flasks filled with 300-450 g substrate under
semi- sterile conditions in two steps. The flasks were fitted with yeast locks constructed to
vent CO2 through a glycerine trap. The wet oxidised, washed material was diluted with the
separated liquid fraction resulting in final substrate concentration between 8% and 20%
DM. The pH was adjusted with 10% NaOH to pH 4.8. The flasks and the 10% NaOH
solution for pH adjustment were autoclaved, but the wet oxidised corn stover and the liquid
fraction after WO were not sterilised to avoid further high temperature decomposition of
the wet oxidised material (Klinke et al, 2003). The enzyme solutions were not sterilised. In
the first step of SSF the pre-treated corn stover was pre-hydrolysed at 50°C for 24 hours
with 5-10 FPU/g DM cellulase enzyme loading. After the pre-hydrolysis cellulase was
added again - loading 5-20 FPU/g DM cellulase - simultaneously with dried Baker’s yeast
47
(1 g/L). Both in the pre-hydrolysis and in the SSF step ß-glucosidase was added as a
supplement at ratio of 1:1 IUs of ß-glucosidase to IU cellulase. Commercially available
enzymes Celluclast 1.5L and Novozym 188, were applied in the pre-hydrolysis and also in
the SSF step.
The fermentation step was carried out in duplicates at 30°C for 120 h and the flasks were
agitated at 150 rpm. After the SSF the samples were cooled on ice immediately and
centrifuged at 8000 rpm for 10 min. The carbohydrate components of the solid residues
after SSF were analysed by acid hydrolysis (Kaar et al, 1991). Ethanol concentration and
the remaining monosaccharides were determined by HPLC at the previously described
conditions.
The fermentation rate was monitored by weighing the flasks at regular intervals to monitor
CO2 loss. In this case the ethanol yield was calculated from the CO2 loss by multiplication
of the conversion factor 1.045 (e.g. the molar ratio of EtOH/CO2). The formula for
calculating the weight of produced ethanol from CO2 lossis the following (2)
EtOH ( g ) = CO 2 _ loss ( g ) ⋅ 1.045
(2)
The final ethanol concentration was also measured with HPLC. The ethanol yield (YEtOH)
was calculated assuming, that 1 g of glucose present in the liquid theoretically gives 0.511
g of ethanol and 1 g of cellulose gives 1.11 g of glucose. This yield is always less than
100% as a part of the sugars is needed for cell growth and synthesis of other by-products,
such as glycerol and acetic acid.
The conversion of the consumed glucose to ethanol was calculated according to the
following equation (3):
Pr oduced EtOH ( g )
C EtOH (%) =
[CelluloseS ( g ) − CelluloseR ( g )]⋅ 0,568 − Gu cos eR ( g ) ⋅ 0,511
⋅100
(3)
Where factor of 0.568 is the theoretical conversion factor from glucan to ethanol by
Saccharomyces cerevisiae (Sen 1989). “Cellulose S“ is the amount of cellulose in the
substrate (g), and “Cellulose R“ is the amount of cellulose in the residue after SSF (g),
determined by strong acid hydrolysis. “Glucose R“ is the unfermented glucose in the liquor
after SSF, determined directly by HPLC.
3.6. FERMENTATION FOLLOWING STEAM PRETREATMENT
The liquid fractions after pretreatment were fermented using baker’s yeast (Jästbolaget AB,
Rotebro, Sweden) to determine the toxicity of the samples. Fermentation was carried out in
25 cm3 glass flasks, sealed with rubber stoppers and equipped with cannulas for removal of
48
produced broth carbon dioxide. The volume of the fermentation broth was 20 cm3
(18.5 cm3 filtrate, 0.5 cm3 nutrients and 1 cm3 inoculum).
The filtrates were adjusted to pH 5.5 with 10 M NaOH solution and supplemented with
glucose to a total concentration of 30 g/L, and with nutrients to final concentration of
2.5 g/L yeast extract, 0.25 g/L (NH4)2HPO4, 0.025 g/L MgSO4*7H2O, and 0.1 g/L
NaH2PO4. The addition of glucose was made, as the purpose of the fermentation was to
investigate the effect of inhibitors on fermentation. The filtrates were inoculated with
compressed Baker’s yeast to a cell concentration of 3 g DM/L, incubated at 30°C and
stirred using a magnetic stirrer. Reference fermentations, using a pure sugar solution
containing 30 g/L glucose, nutrients and cells, were run for comparison. Samples were
withdrawn from the fermentation broth at the start, before yeast addition, and after 1, 2, 4,
6, 8, 10 hours, and a final sample was taken after 20 h. All samples were analyzed for
glucose and ethanol by HPLC.
49
4. RESULTS AND DISCUSSION
The goal of this work was to evaluate some physical, physico-chemical and chemical
processes for bioethanol production from pre-treated corn stover. Several different pretreated methods, i.e. acid and alkaline pre-hydrolysis, supercritical CO2 extraction,
irradiation, wet oxidation (with and without added alkaline or acid addition), and steaming,
were studied.
The aim was to obtain a solid cellulose-rich fraction accessible for enzyme treatment to
hydrolysis to glucose, without producing inhibitors, which obstacle the enzymatic
hydrolysis and/or the fermentation from glucose to ethanol.
4.1. CHEMICAL PRETREATMENTS ON YIELDS AND COMPOSITION
4.1.1. Effect of chemical pretreatments on yield and composition
The influence of the various acidic and alkaline pretreatments on the composition on the
pre-treated corn stover is summarised in Table 4.1. The acid solubilised lignin fraction, the
proteins, and other non-cell wall materials were not measured in the experiments, that
could explain that the summa were under 100%.
Table 4.1. Components [% w/w) DM] of the pre-treated and untreated corn stover
Type of pretreatment
Lignin
Ash
Cellulose
Untreated corn stover
1% NaOH
5% NaOH
10% NaOH
2% Ca(OH)2
1% H2SO4
5% H2SO4
1% HCl
5% HCl
TS-1*
TS-2
TS-3
TS-4
22.1
4.3
3.8
2.9
18.1
29.5
30.3
27.9
31.2
15.1
13.3
14.7
26.7
6.3
2.2
12.7
18.8
7.2
3.0
2.9
3.0
4.0
3.0
1.1
2.7
1.0
41.3
65.4
63.6
58.9
44.6
50.6
56.0
52.4
54.3
63.3
74.2
64.0
55.9
HemiTotal
summa
cellulose carbohydrate
27.9
69.2
97.6
21.7
87.2
93.6
15.3
78.9
95.4
10.0
68.9
90.6
24.5
69.1
94.4
11.9
62.4
95.0
7.3
63.3
96.5
13.5
65.9
96.7
6.4
60.8
95.9
14.3
77.7
95.7
6.2
80.5
94.9
15.6
79.6
97.1
12.2
68.0
95.7
*
TS: two-step pretreatment
TS-1: 1% NaOH & 1% HCl
TS-2: 5% NaOH & 1% HCl
TS-3: 1% Ca(OH)2 & 1% HCl
TS-4: 1% NaOH & 1% H2SO4
50
In excess of 50% of the total material was solubilised during the alkaline pretreatment, but
the solubilisation during the acidic reaction was a little slighter, between 40.8 – 48.9%
(Table 4.2.). A correlation was observed between the concentration of the applied base or
acid and the quantity of the solubilised solid material.
The solubilisation both of the lignin and hemicellulose were the highest when the most
concentrated base was applied. Following this strong alkaline pretreatment the amount of
the lignin in the solid fraction decreased from the original 22.1 g to 0.9 g, thus the
reduction of the lignin content was 95.7%. However both the hemicellulose and the
cellulose content decreased considerably with 84.7% and 32.9% respectively.
The hemicellulose fraction was also solubilised significant, when more concentrated acid
was used, but the amount of the solubilised hemicellulose seemed to be independent of the
type of applied acid. The highest amount (85.8%) of the solubilised hemicellulose was
achieved, when 5% HCl was used, as a pretreatment agent.
Alkaline pretreatments solubilised more than 90% of the original lignin fraction, and above
60% of the original hemicellulose fraction, which caused an increase in the cellulose
content of the remaining solids ranging from 58.9 to 65.4% of the total pre-treated solids,
compared to 41.3% cellulose for the untreated corn stover. Further increase of the base
concentration resulted higher amount of solubilised cellulose, thus the cellulose content of
the pre-treated solid was lower, than using less concentrated base.
Contrary to the previous expectations the pretreatment with 2% Ca(OH)2 was quite
ineffective. Ca(OH)2 is an inexpensive base, and it is possible to recover calcium from an
aqueous reaction system as insoluble calcium carbonate by neutralizing with inexpensive
carbon dioxide. The calcium hydroxide can subsequently be regenerated using established
lime kiln technology. Although the distinct advantages of using calcium hydroxide as a
pretreatment agent it didn’t seem effective enough, it modified only slightly the
components of the raw material, the lignin content in the solid phase decreased less, than
30%, and the solubilisation of the hemicellulose fraction in the solid was also quite low,
18%.
Table 4.2. shows the amounts of solubilised and remaining components from 100 g (DM)
untreated corn stover following alkaline and acidic pretreatment. Lignin fraction was not
measured in the liquid following pretreatments, it was only identified in the remaining
solid material.
51
Table 4.2. Mass balance of the alkaline and acidic pretreatment from 100 g (DM) corn stover
Type of pretreatment
Amount of Liquid fraction (g)
filter
Cellulose Hemi- Lignin
cake (g) [DM]
cellulose
Untreated
100.0
corn stover
alkaline pretreatment
1% NaOH
43.8
5% NaOH
37.7
10% NaOH
32.7
2% Ca(OH)2
84.5
acidic pretreatment
1% H2SO4
51.1
5% H2SO4
52.7
1% HCl
59.2
5% HCl
54.0
Solid fraction (g)
Ash Cellulose
-
-
22.1
6.3
41.3
Hemicellulose
27.9
8.3
10.7
13.6
1.3
14.5
17.9
20.2
4.8
1.9
1.4
0.9
15.3
0.9
4.8
6.1
6.1
28.6
24.0
19.3
37.7
9.5
5.8
3.3
20.7
10.3
8.1
6.5
7.4
17.5
19.3
15.8
19.8
15.1
16.0
16.5
16.8
1.5
1.5
1.8
2.1
25.8
29.5
31.0
29.3
6.1
3.9
8.0
3.5
Effect of two-step pretreatments on corn stover
Using acidic pre-hydrolysis after one-day-long alkaline soaking the results depended on
the type of the applied chemicals. More than 60% of the total fibre was solubilised during
the two-step pretreatments, when 1% NaOH was used as a swelling agent (TS-1, TS-4).
This result was 40% using Ca(OH)2 (TS-3). The amount of the solubilised lignin, and also
hemicellulose were the highest (77.7% and 88.3% consequently) when 1% HCl was
applied, following the soaking with 1% NaOH (TS-1), In general, in these two-step pretreatment the lignin-solubization was slighter, than in the alkaline pretreatments. Compared
to the acidic pretreatment, the amount of the solubilised hemicellulose in the TS pretreatments was similar, but the amount of solubilised cellulose was lower, than it was
achieved in the acidic treatment. It was extremely low (5.0%) in the case of TS-3, when
corn stover was treated with Ca(OH)2 and HCl consequently (Table 4.3.).
Table 4.3. Mass balance of the two-step pretreatment from 100 g (DM) untreated corn stover
Type of
Amount of
Liquid fraction (g)
prefilter
Cellulose Hemitreatment cake (g) [DM]
cellulose
Untreated
100.0
corn stover
TS1
49.0
7.0
16.9
TS2
37.0
9.2
20.3
TS3
59.3
2.1
14.7
TS4
34.5
14.4
18.9
52
Lignin
Solid fraction (g)
Ash Cellulose
22.1
6.3
41.3
Hemicellulose
27.9
7.4
4.9
8.7
9.2
1.5
0.4
1.6
0.3
31.0
27.4
38.0
20.8
7.0
2.3
9.2
4.2
4.1.2. Enzymatic hydrolysis
The enzymatic degradability of the cellulose in the pre-treated solid material is a critical
point in production of bioethanol. The enzymatic conversion of cellulose to glucose by
hydrolysis gives valuable information about the efficiency of pretreatments. The main goal
of these chemical pretreatments was to enhance the enzymatic digestibility of cellulose in
corm stover.
Figure 4.1. shows the percentage of enzymatic conversions (ECC%) of the solid fraction
following alkaline and acidic pretreatments. The calculation of ECC is shown in chapter of
Materials and Methods.
100
100
75
75
ECC (%)
ECC (%)
The alkaline pretreatment increased the enzymatic hydrolysis of corn stover four times
compared to untreated corn stover. The conversion increased from 17.6% to 65.6, 72.8 and
79.4% using 1, 5 and 10% NaOH, respectively. Although the highest conversion (79.4%)
was achieved, applying 10% NaOH, from the point of view of the amount of released
glucose this pretreatment was the last in line. After 48 h enzymatic hydrolysis, the amount
of released glucose were 20.7, 19.2, 16.8 g from 100 g (DM) untreated material, using 1%,
5% and 10% NaOH.
50
50
25
25
0
0
0
0
24
48
72
time (h)
1% NaOH
5% NaOH
10% NaOH
2% Ca(OH)2
untreated
24
time (h)
5% H2SO4
1% HCl
untreated
48
5% H2SO4
5% HCl
Figure 4.1. Enzymatic conversion (%) of alkaline and acidic pre-treated corn stover during
enzymatic hydrolysis at 50°C, with 25 FPU/ g DM.
53
72
Following acidic pretreatment, the enzymatic conversion (ECC%) of the solid material was
between 32.4% (using 1% HCl) and 46.2% (using 5% H2SO4) of the theoretical limit,
which was only 2-2.5 times higher, than the achieved conversion of untreated control. The
highest enzymatic conversion (46.2%) was observed following pretreatment with 5%
H2SO4. However, because of the great amount of the solubilised sugars during acidic
pretreatment, the amount of the hydrolysis-released-glucose was not significant higher
after acidic pretreatment, than without pretreatment. Using more concentrated acid the
amount of hydrolysis - released glucose were higher, than using less concentrated.
The enzymatic conversions following two-step pretreatments are presented in Figure 4.2.
These results are shown big variety, according to the big variance between composition of
pre-treated materials. TS2 pretreatment (treatment with 1% HCl following soaking in 5%
NaOH) caused the most significant modification in the composition of corn stover.
Although TS2 decreased 77.7% of the lignin and 85% of the hemicellulose content (Table
4.3.), this pretreatment increased only slightly the enzymatic degradability. However the
best enzymatic conversion (95.7%) was achieved following TS4 pretreatment (1% NaOH
and 1% H2SO4), which decreased the lignin content from the original 22.1 g to 9.2 g,
resulted only 58.3% delignification. TS4. pretreatments also decreased the hemicellulose
content considerably, more than 85% of the original hemicellulose were solubilised.
100
TS1
TS2
TS3
TS4
untreated
ECC (%)
75
50
The pretreating agents were the
following:
25
TS-1:
TS-2:
TS-3:
TS-4:
0
0
24
48
1% NaOH & 1% HCl
5% NaOH & 1% HCl
1% Ca(OH)2& 1% HCl
1% NaOH & 1% H2SO4
72
time (h)
Figure 4.2. Enzymatic conversion (%) of two-step pre-treated corn stover during
enzymatic hydrolysis at 50°C, with 25 FPU/ g DM
54
The hydrolysis time was 72 h, but the enzymatic conversions varied less, than 5% after the
first day. The hydrolysis of corn stover seems to be quite fast at 50°C; thus 1 day appears
to be enough for the complete hydrolysis.
To get a real view about the possible use of the pre-treated material an "overall conversion"
was defined and determined, as the total production of monosaccharides via enzymatic
hydrolysis and pretreatment. The calculation overal conversion based on the quantity of
carbohydrates in the untreated corn stover. The highest overall conversion was achieved
following TS4 pretreatment, which gave 89.7% conversion of monosaccharides from
carbohydrates, resulted 68.3 g released sugars from 100 g (DM) untreated corn stover. The
pretreatment with NaOH gave also favourable results, the amounts of total released sugar
were 57.0, 60.3, 63.9 g, using 1, 5 and 10% NaOH respectively. The main disadvantage
applying concentrated base is the large amount of the pretreatment released sugars, which
is solved in the concentrated liquor of pretreatment. The purification of these sugars are
hardly difficult, but very important for further utilisation. In additional, recycling of the
base is also a critical point both in the views of economical and environmental. Table 4.4.
summarises the amounts of pretreatment and hydrolysis released sugars from 100 g (DM)
untreated corn stover.
Table 4.4. Amount of pretreatment and hydrolysis - released sugars from 100 g untreated
corn stover (DM), following different pretreatments and 48 h enzymatic hydrolysis (50°C,
25 FPU/g DM).
AMOUNT OF RELEASED SUGARS (G) FROM 100 G (DM) UNTREATED CORN STOVER
Type of pretreatment Released sugars (g) Released sugars (g)
Total released
from hydrolysis from pretreatment
sugars (g)
untreated corn stover
13.4
13.4
ultrasonic waves
14.1
14.1
Tween 80
14.5
14.5
1% NaOH
27.5
25.8
57.0
5% NaOH
23.8
32.4
60.9
10% NaOH
19.7
38.0
63.7
2% Ca(OH)2
14.3
7.1
21.9
1% H2SO4
16.2
31.5
52.2
5% H2SO4
13.7
31.4
48.7
1% HCl
13.9
25.6
42.4
5% HCl
16.2
31.3
50.8
TS1
12.6
27.4
43.1
TS2
6.8
33.8
44.6
TS3
22.0
19.9
42.8
TS4
24.7
37.3
68.3
55
Although the enzymatic conversions following acidic hydrolysis were not really high, the
overall conversion using 5% H2SO4 or 5% HCl were over 65%, resulted 52.2 and 50.8 g
released sugars respectively from 100 g (DM) untreated corn stover. The acidic
pretreatments have the advantage, that the large part 60-80% of the solubilised sugars is
hemicellulose and during the enzymatic hydrolysis released sugars is mainly (95%)
glucose.
In some experiments, following dilute sulphuric acid pretreatment instead of filtering and
washing, the whole pre-treated slurry was hydrolysed after the pH was adjusted to 4.8 with
20% NaOH. In these cases the total amount of released sugars were between 30.2 – 41.1 g
which was not significant higher, than the amount of pretreatment released sugars.
According to previous results (data not shown) there is no product inhibition until 15 g/l
glucose concentration, thus the high salt concentration or the liberated inhibitors might
have had a negative effect on enzymatic hydrolysis.
Based on the results of this investigation it appears that a good pretreating chemical of corn
stover would be the NaOH. Although the highest overall conversion was achieved using
10% NaOH, considering both economically and environmentally aspects, diluted NaOH
was used for further investigations.
The corn stover was treated with 0.5% NaOH for 30, 60 and 90 min. at 100°C and 120°C.
Table 4.5. shows the result following 48 h enzymatic hydrolysis. Both the highest
enzymatic (80.1%) and overall conversion (83.3%) were achieved following pretreatment
at 120°C for 90 min. These values were very attractive, even higher, than conversions after
pretreatment with more concentrated base for 60 min. It seems, that the longer
pretreatment time increased the accessibility of the corn stover, and gave 50.5, 65.8 and
80.1% enzymatic conversion after pretreatment for 30, 60 and 90 min respectively.
Table 4.5. Enzymatic and overall conversion (%) after 48 h enzymatic hydrolysis at 50°C
with 25 FPU/g DM enzyme loading, following pretreatment with 0.5% NaOH
TIME OF PRETREATMENT WITH 0.5% NAOH
30 min
60 min
90 min
Enzymatic conversion (ECC %)
100°C
46.3%
62.5%
67.0%
120°C
50.5%
65.8%
80.1%
Overall conversion %
100°C
54.8%
58.3%
71.6%
120°C
55.6%
72.7%
83.3%
56
4.1.3. Effect of supercritical pretreatment on yield and composition
Recently, supercritical carbon dioxide (SC-CO2) which has been mostly used as an
extraction solvent (Kim and Hong, 2000 ) is being considered for non-extractive purposes
due to its many advantages. CO2 is inexpensive, clean and environmentally benign, and it
is easy to recover after use. Extraction of pine wood with SC-CO2 was reported to cause no
significant change in microscopic morphology of wood (Ritter and Campbell, 1991 ), and
they concluded that SC-CO2 would not be an effective tool for lignocellulose pretreatment.
On the other Kim and Hong (2001) found that SC-CO2 extraction enhanced nearly three
times the enzymatic conversion of southern yellow pine.
Based on reviewing previous work in this field, it seemed, that no study of SC-CO2
pretreatment of herbaceous material have been reported. In this experiments I
investigated the effect of SC-CO2 on ground corn stover, following 1-night soaking at
room temperature in water or in different solution (2% NaOH, 2% H2SO4). The half of
these swelling samples were not pre-treated with SC-CO2, but were used as a control.
The composition of these pre-treated corn stover didn’t changed significantly (Table 4.6.),
which prompts, that the complex structure of lignocellulose remained untouched.
Only the lignin content has decreased considerably following SC-CO2 pretreatment
combined with alkaline swelling. However the alkaline swelling alone was also enough to
achieve this delignification level.
Table 4.6. The composition of the corn stover following 1 night swelling in water, 2%
NaOH or 2 %H2SO4 solution and pre-treating with SC-CO2
TYPE OF THE PRETREATMENT
COMPOSITION OF THE SOLID FRACTION AFTER
PRETREATMENT
swelling agent
SC-CO2
pretreatment
untreated corn stover
water *
water
2% NaOH*
2% NaOH
2% H2SO4*
2% H2SO4
*
no
no
no
Lignin
(%)
Ash (%)
Hemicellulose
(%)
Cellulose
(%)
22.5
23.0
21.4
14.4
13.7
24.2
24.2
6.0
5.9
6.3
7.9
7.5
3.4
3.4
23.2
22.8
21.2
24.4
23.0
19.2
19.2
46.4
47.5
47.4
52.8
54.0
48.2
51.0
used as a control
Unfortunately the enzymatic conversion of the pre-treated material has also proved the
theory of the untouched structure of lignocellulose, because there was no difference
between the conversion of pre-treated corn stover and the used control.
57
4.1.4. Conclusions on chemical pretreatments
The aim of the chemical pretreatment study was to know what kind of chemicals are the
most efficient to break down the tight association between carbohydrates and lignin, and to
enhance the enzymatic digestibility. Nearly theoretical conversion (95.7%) was achieved
following two-step pretreatment, where dilute sulphuric acid was used after one-day
soaking in dilute sodium hydroxide. To obtain this attractive result, the complete
delignification of the corn stover wasn't necessary, and 24 h seemed to be enough to
enzymatic hydrolysis of corn stover. Using only dilute acids for the pretreatment the
enzymatic conversions were quite low, but these pretreatment appears to have advantage
that mainly the hemicellulose fraction was solubilised, thus this pretreatment seems to be a
good fractionation method. The best overall conversion 89.7% was also achieved applying
1% H2SO4 after one-day long soaking in 1% NaOH. However increasing the reaction time
the enzymatic degradability could be also increased significantly, using less concentrated
base. To increase the reaction time by threefold, in the pretreatment with 0.5% NaOH at
120°C, the overall conversion increased from 55.8% to 83.3%. Considering economically
and environmentally aspects diluted NaOH would to be also much more suitable
pretreating chemical agent, than more concentrated base.
In connection with the supercritical carbon dioxide pretreatment I can conclude, that
supercritical-CO2 pretreatment is not an effective method for lignocellulose pretreatment.
4.2. STEAM PRETREATMENT
The steam pretreatment of corn stover was investigated regarding both the enzymatic
convertibility and the fermentability of per-treated material. The parameters
investigated in the pretreatment step were temperature, residence time and catalyst
concentration. The pretreatment was evaluated by enzymatic hydrolysis and
fermentation, to determine the yields and to investigate how by-products, formed
during pretreatment, affected the fermentability.
4.2.1. Effect of steam pretreatment
After the steam pretreatment the yield of solid material ranged from 37 to 85.4 g per 100 g
and decreased with increased acid concentration at the same temperature, see Table 4.7.
The reduction in solid, fibrous material was due to the solubilisation and/or degradation of
hemicellulose and extractives. However, under more severe conditions a part of the
cellulose was also solubilised.
58
Table 4.7. Amount of solids and yield of various components in the solid fraction after
steam pretreatment, expressed as g per 100 g of dry, untreated corn stover
CONDITION OF
PRETREATMENT
COMPONENTS IN THE SOLID FRACTION (G)
Temp Acid Time
Total
conc.
(min)
amount
of
(°C)
(%)
solid
Raw material
100.0
190
0.5
2
200
0.5
2
210
0.5
2
210
-
Lignin
Glucose
Xylose
20.2±0.2
41.6±0.3
27.7±0.4
Arabi- Ash
nose
3.6±0 4.6±0
1.0
1.7
0.3
1.7
0.1
2.1
0.0
2.2
2
5
2
5
69.4±2.3
62.5±2.1
48.4±2.8
49.1±1.3
18.5±0.5
17.4±0.3
16.6±0.6
19.1±0.2
33.3±0.5
29.6±0.5
24.3±0.4
24.2±0.3
14.3±0.7
11.9±0.6
3.8±0.8
1.4±0.5
2
5
2
5
67.7±1.7
65.3±0.7
50.1±1.1
41.1±0.4
19.5±0.1
19.5±0.4
17.9±0.2
19.1±0.3
32.9±0.2
32.6±0.3
24.1±0.4
20.0±0.1
9.4±0.4
8.5±0.5
4.5±0.3
1.1±0.4
0.0
0.0
0.1
0.0
2.7
3.9
2.8
2.2
2
5
2
5
61.4±1.4
58.2±0.7
41.2±0.9
36.9±0.7
19.2±0.4
18.8±0.2
18.4±0.2
19.4±0.1
30.7±0.3
30.2±0.2
17.4±0.3
13.9±0.1
6.4±1.1
5.7±0.6
2.3±0.2
0.7±0.1
0.0
0.1
0.0
0.0
2.4
2.6
2.8
2.2
5
85.4±1.5
19.7±0.5
38.6±0.2
21.6±0.4
2.7
2.2
Table 4.7. summarizes the composition of the solid fibrous fractions after pretreatment and
Table 4.8. shows the amount of dissolved compounds in the filtrate from 100 g untreated
corn stover. As expected, steam pretreatment preferentially attacked the hemicellulose
fraction; the harsher the pretreatment conditions were the higher amount of hemicellulose
was solubilised. Under the harshest condition (210°C, 2% w/w H2SO4, 5 min) almost all
hemicellulose was removed from the solid fraction, its amount decreased from 31.3 g to
0.7 g. However the solubilisation of hemicellulose was significant about 85%, also at the
lowest temperature (190°C) using 2% sulphuric acid. The concentration of the sulphuric
acid had a higher effect on the decrease of the hemicellulose fraction in the solid residues,
than the temperature had. Without sulphuric acid addition even at 210ºC for 5 min the
pretreatment solubilised only 20% of the hemicellulose, while 0.5% sulphuric acid
decreased the hemicellulose content with 82%. It seems, that acid hydrolysis is the main
mechanism involved, which was also supported by the work of Larsson et al. (1999) on
wheat straw, and Palmqvist et al. (1997) on soft wood pre-treated with 0.5% sulphuric acid
at 185ºC, where 76% of the original hemicellulose content was dissolved. For the
calculation of the amount of hemicellulose, the amount of mono- and disaccharides (xylose
and arabinose) in the liquid, was measured by HPLC, following the strong acid hydrolysis
by Hägglund.
59
The 2% sulphuric acid was also effective in solubilisation of cellulose from the solid
material. Even at the lowest temperature (190°C) the cellulose content in the solid fraction
decreased by 50% from 41.6 to 24.3 g. The solubilisation and/or degradation of the lignin
during steam pretreatment was not significant. The amount of lignin decreased only
slightly from 20.2 g to 18.4 g even under the harshest conditions (210°C, 5 minutes, 2%
sulphuric acid).
Table 4.8. Sugar yields in the liquid after steam pretreatment, expressed as g per 100 g of
dry, untreated corn stover and concentration of the by-products (g/L)
PRETREATMENT
CONDITIONS
Temp Acid Time Cello(°C) conc. (min) biose
(%)
190 0.5
2
0.3
5
0.6
2
2
0.2
5
0.5
200
0.5
2
210
0.5
2
210
-
COMPONENTS IN THE LIQUID FRACTION
Concentration (g/L)
Glucose
Xylose
Arabinose Acetic HMF Furfuacid
ral
3.3±0
6.2±0.2
9.9±0.2
12.8±0.3
9.2±0.2
12.5±0.2
16.6±0.3
18.3±0.4
1.7±0
1.9±0
3.2±0.1
3.0±0.1
0.7
1.0
2.4
1.8
0.1
0.1
0.2
0.1
0.4
0.7
2.7
1.5
2
5
2
5
0.5
0.4
0.5
0.3
3.9±0
2.5±0
7.4±0.1
12.9±0.2
11.8±0.2
10.7±0.2
13.8±0.3
16.3±0.3
1.8±0
1.6±0
2.3±0
2.7±0
1.1
1.5
1.6
2.1
0.1
0.1
0.2
0.3
0.9
0.8
1.5
2.6
2
5
2
5
0.3
0.4
0.3
0.3
3.6±0
4.5±0.1
12.8±0.2
16.2±0.2
12.7±0.2
13.8±0.3
13.3±0.2
13.7±0.2
1.7±0
2.1±0
2.6±0.1
3.1±0.1
1.2
1.7
2.2
3.4
0.1
0.2
0.3
0.5
1.0
1.7
2.6
4.6
5
0.3
1.0±0.05
5.5±0.1
1.1±0
0.7
0.1
0.2
The results were evaluated using the severity factor [Log R0] (Overend et al., 1987) and
the combined severity factor [CSF] (6-7) (Chum et al., 1990).
⎛
⎡ T − Tref
Log (R0 ) = Log ⎜⎜ t ⋅ exp ⎢
⎣ 14,75
⎝
⎤⎞
⎥ ⎟⎟
⎦⎠
(6)
CSF = Log(R0) – pH
(7)
Where t is the residence time (min), T the pretreatment temperature (°C) and Tref the
reference temperature, which was set at 100°C. The use of a severity factor (Log R0)
facilities the comparison of results from different pretreatment conditions in a normalised
60
way, where the temperature and residence time variables are combined into a single
reaction ordinate. The combined severity is used when the effect of pH is also take into
consideration. Implicit in this concept is the assumption that the overall kinetics follows a
first order concentration dependence and that the rate constant has an Arrhenius –type
dependence on the temperature. if the reaction follows the severity correlation, it follows
that it is possible to choose various combination of residence time and temperature such
that equivalent final effects are obtained. Although the degradation of hemicellulose and
cellulose in the pretreatment stage involves a complex series of reactions, the assumption
of a pseudo-first-order reaction gives reasonable results.
100
5
Recovery of sugars (%)
Yield (g/100g corn stover DM)
The amount of the formed by-products (acetic acid, furfural, and hydroxy-methyl-furfural)
during pretreatment increased, with increased temperature and reaction time, thus the
increased value of the combined severity factor. The formation of theses by-products, as a
function of CSF is summarised in Figure 4.3.
4
3
2
1
0
75
50
25
0
0,5
1,5
2,5
3,5
4,5
Combined severity factor
HAC
HMF
0,5
1,5
2,5
3,5
Combined severity factor
glucane
Furfural
pentane
Figure 4.3. The amount of formed by-products during the pretreatment and the recovery of
cellulose and hemicellulose as a function of the combined severity factor.
Figure 4.4. shows the formation of furfural as a function of temperature at different
pretreatment time. The tendency was similar for all by-products.
The recovery both of hemicellulose and cellulose sugars is an important point for a suitable
pretreatment. The recovery of cellulose and hemicellulose was calculated to estimate their
losses during steam pretreatment at different conditions (Table 4.9). The calculation, which
was based on the mass balance was mentioned above (Table 4.7.-4.8.). The amount of
hemicellulose in the solid residue was calculated from the amount of xylose and arabinose
in the filtrate following the Hägglunds hydrolysis.
61
4,5
Yield (g/100 g corn stover DM)
5
4
3
2min
5min
2
1
0
185
190
195
200
205
210
215
Temperature (°C)
Figure 4.4. The amount of formed furfural during pretreatment, as a function of reaction
temperature, and time (2 and 5 min.
Table 4.9. shows, that mainly the hemicellulose was converted and/or degraded in the
steam pretreatment. Around 60% of the original hemicellulose could be recovered, except
when 2% sulphuric acid was combined with high reaction temperature (210°C). In this
case, the hemicellulose recovery was only 55.9%. This relatively low recovery could be
explained with the hemicellulose degradation to other by-products during steam
pretreatment.
Table 4.9. Recovery % of cellulose and hemicellulose
PRETREATMENT CONDITIONS
Acid conc.
Time
Temp.
(%)
(min)
(°C)
190
0.5
2
5
2
2
5
200
0.5
2
5
2
2
5
210
0.5
2
5
2
2
5
210
5
Cellulose
RECOVERY %
Hemicellulose
All sugars
88.8
83.4
89.7
83.1
89.7
85.3
76.9
74.2
83.2
84.2
73.5
73.1
95.9
83.8
80.6
75.7
72.4
73.5
66.5
65.9
64.1
66.5
69.2
58.0
55.9
93.8
86.6
86.5
79.7
82.5
82.8
77.2
72.2
69.9
76.1
77.7
66.8
65.7
97.2
62
Figure 4.3. shows the relation between the CSF and the recovery of cellulose and
hemicellulose. The maximal recovery of hemicellulose (83.8%) was obtained following the
mildest pretreatment at 190°C for 2 minutes using 0.5% sulphuric acid, corresponding to
approximately CSF 2. The harsher the conditions were, the higher the CSF was, the lower
the hemicellulose recovery was reached. The maximum cellulose recovery was 89.7%,
following steam pretreatment at 200°C, but the recovery was sufficient around 75% even at
210°C, which meant in the range of combined severity factor between 2.9-3.2. The
cellulose and hemicellulose exhibit their maximum recovery at different conditions, which
was also found in other studies (Garrote et al., 2002, Tenborg et al., 1998.). However in
the case of softwood the hemicellulose solubilisation needed harsher pretreatment’s
conditions, then corn stover, and the maximum mannose yield from wood was obtained
between CSF 2.5-2.7 (Larsson et al. 1998, between 200-210°C.
The concentration of the sulphuric acid affected the recovery of cellulose significantly.
Using less concentrated acid (0.5%), the recovery of cellulose was approximately 10%
higher, than using 2% sulphuric acid at the same temperature and time.
The overall recovery of all sugars varied from 79% to 86% at 190°C and from 65% to 77%
at 210°C, which was a bit lower than achieved with corn stover using wet oxidation
4.2.2. Enzymatic hydrolysis
The steam pre-treated solid residue was enzymatically hydrolyzed. The hydrolysis at 50°C
with the applied relative high enzyme loading (25 FPU/g DM) seemed to be completed
after 24 hours, as was also found in a previous pretreatment study. The achieved
conversion after 48 h enzymatic hydrolysis is shown in Table 4.10. The enzymatic
conversion of pre-treated cellulose was between 31 and 83% compared to 27% obtained at
enzymatic hydrolysis of the untreated corn stover. To verify the pretreatment effect of
freezing – thawing period, a portion of the raw material was also frozen and than
hydrolyzed. The achieved enzymatic conversion was not significantly higher (29.2%) of
this case.
Considering, that the pretreatment without acid addition has modified only slightly the
composition of corn stover (Table 4.7.), it is not surprising, that the ECC after this
pretreatment was quite poor (31.1%). The highest conversion (83.6%) was achieved with a
pretreatment at 200°C for 5 minutes with 2% sulphuric acid. The ECC conversions in
general were higher following pretreatment using 2% sulphuric acid, than 0.5% acid, but
the amounts of released glucose by enzymatic hydrolysis were still higher using less
concentrated acid. It could be explained with the higher solubilisation of cellulose during
pretreatment. The highest amount of total released glucose (34.3 g per 100 g DM) was
achieved following pretreatment at 190°C for 5 minutes with 2% acid.
63
Table 4.10. The ECC% conversion after 48 h enzymatic hydrolysis and the amounts of
released glucose (g) from 100 g untreated material (DM) during pretreatment and
hydrolysis
CONDITION OF
PRETREATMENT
Temp. Acid Time
(°C) conc. (min)
(%)
Raw material
Freezed raw material
190
0.5
2
5
2
2
5
200
0.5
2
5
2
2
5
210
0.5
2
5
2
2
5
210
5
ECC
(%)
27.3
29.2
47.4
52.2
71.8
81.1
58.4
65.7
65.8
83.6
67.2
74.8
74.0
78.8
31.1
RELEASED
RELEASED
GLUCOSE
( G)
XYLOSE
( G)
RELEASED
SUGARS
( G)
GLUCOSE (G)
during the hydrolysis
during hydrolysis &
pretreatment
12.5
13.4
17.4
17.0
19.2
21.6
21.1
23.6
17.4
18.4
22.7
24.8
14.2
12.0
13.2
0.2
0.2
0.3
0.2
0.2
0.5
0.5
0.5
0.5
0.3
0.3
0.4
0.3
0.3
0.4
TOTAL
12.5
13.5
20.7
23.2
29.1
34.3
25.0
26.0
24.9
31.3
26.3
29.3
27.0
28.3
14.2
TOTAL
12.7
13.9
31.9
35.8
49.1
56.1
39.2
38.8
41.4
50.5
41.0
45.6
43.3
45.3
21.2
Although the most cellulase enzyme complex, including Celluclast, contain xylanase
activity (data not shown), the amount of released xylose after enzymatic hydrolysis was
negligible, especially compare to the amount of the pretreatment released xylose.
The highest overall sugar yield after both pretreatment and enzymatic hydrolysis, 56.1 g
per 100 g DM, was also obtained at 190ºC, for 5 min with 2% sulphuric acid.
4.2.3. Fermentability test
Ethanol production depends not only on the sugar yield, but also on the fermentability of
the solution. To investigate the fermentability of the pre-treated corn stover fermentations
were performed with baker’s yeast. Baker’s yeast has often been proposed as the best
organism for the fermentation of lignocellulosic hydrolysates (Olsson et al., 1993, HahnHägerdal 1991) and has the advantage that it is quite robust and was found to be less
sensitive to inhibitors than cultivated yeast (Stenberg et al., 2000a, b).
Table 4.11. shows the achieved ethanol concentrations after 24 hours fermentation and the
calculated ethanol yield as % of theoretical. After 1-day fermentation the ethanol
concentrations varied from 12.1 to 13.8 g/L corresponding to 78.4 and 90.3% of the
theoretical yield. The glucose was totally consumed after six hours fermentation following
64
pretreatment with 0.5% acid. The maximum ethanol concentration was also reached after 6
hours in these cases, but when 2% sulphuric acid was used in pretreatment, the glucose
was consumed more slowly and the formation of ethanol needed more time. However even
in these cases the achieved ethanol yield were around 85% of the theoretical. These results
were quite similar, to that achieved after fermentation of wet oxidized wheat straw in a
previous study (Klinke et al., 2002).
Table 4.11. Ethanol yield (%) of the theoretical and the achieved ethanol
concentration (g/L) after 24 h fermentation at 30°C with bakers yeast from 30 g/L glucose
solution.
CONDITION OF PRETREATMENT
Acid conc.
Time
Temp.
(%)
(min)
(°C)
ETOH CONCENTRATION.
(g/L)
ETHANOL YIELD
(%)
Control
0.5
12.9
13.2
13.4
13.3
12.1
13.4
13.7
13.0
13.7
13.8
13.8
13.5
13.1
13.6
84.0
86.0
87.5
87.3
78.4
87.8
89.7
85.2
89.4
90.3
90.1
88.4
85.6
89.2
190
2
200
0.5
2
210
0.5
2
210
-
2
5
2
5
2
5
2
5
2
5
2
5
5
4.2.4. Conclusions on steam pretreatments
Steam pretreatment is an efficient method to increase the enzymatic accessibility of the
water-insoluble, cellulose-rich component in corn stover. After pretreatment, the enzymatic
conversion from cellulose to glucose increased nearly four times, compared to the
untreated corn stover.
The best pretreatment conditions for obtaining high conversion of cellulose to glucose was
200°C for 5 minutes after swelling the fibers with 2% sulphuric acid. Most of the
hemicellulose was dissolved during the pretreatment and approximately 60% of the
original hemicellulose could be recovered. The enzymatic hydrolysis at 50°C was
completed after 24 hours and the highest enzymatic conversion from cellulose to glucose
was above 80% corresponding to 18 g glucose per 100 g of untreated solid material.
However the highest overall yield of sugars was 56.1 g from 100 g untreated material DM,
65
corresponding to 73 % of the theoretical, which was achieved following steam
pretreatment at 190ºC, for 5 min with 2% sulphuric acid.
The fermentability of the solution gave good results. The achieved ethanol yields were
about 90%, which were slightly above that obtained with a reference sugar solution,
showing that baker’s yeast could adapt to the pre-treated liquor and ferment the glucose to
ethanol without problems.
4.3. WET OXIDATION PRETREATMENT
In wet oxidation process, three variable factors were selected and the influence of
temperature, time, and pH were studied. To determine the efficiency of the treatment, both
the remaining solid fraction following pretreatment and the unseparated slurry were
enzymatically hydrolysed for 24 hours. The time required for the total hydrolysis at 50°C
and the possibility of reducing enzyme loading and temperature during hydrolysis were
also investigated.
4.3.1. Effect of wet oxidation
The wet oxidation process was investigated to fractionate corn stover to solubilize the
hemicellulose fraction, enhance the enzymatic digestibility of cellulose to glucose, and
maximize the recovery of both polysaccharides.
The influence of the six various pretreatment conditions on the composition of corn stover
are summarized in Table 4.12. At high reaction temperature and longer reaction time, the
solid fraction was enriched in cellulose. The cellulose content ranged from 50.5 to 71.8%
compared to 41% cellulose in untreated stover, the higher concentration resulting from the
removal of lignin and especially hemicellulose. At 195°C and acidic conditions, the
soluibilization of hemicellulose was most effective with 95% of the hemicellulose fraction
dissolved. However, this pretreatment decreased the lignin by only 18.3%, from 8.7 to 7.1
g /100 g raw biomass. At 195°C, alkaline conditions were most efficient for
delignification, where 59.7% of the original lignin was removed. In addition this treatment
also solubilised 57.8% of the hemicellulose. Pretreatments at lower temperatures resulted
in less modification of the composition of the solid portion of the pre-treated corn stover.
At 185°C both under alkaline and acidic conditions, the lignin content was lowered by
approximately 30%, but the solubilisation of hemicellulose was more significant at pH 3.5,
than at pH 9.3.
Reaction temperature, time, and the amount of applied chemicals are all economically and
environmentally important factors in the overall process; therefore, it is promising that
lower temperature (185°C) combined with longer reaction time (15 min) without adding
chemicals resulted in a significant modification of the corn stover composition. The
delignification was 49.4 %, which is nearly as high as at 195°C with alkali addition. The
66
80
80
60
60
Hemicellulose %
Cellulose %
solubilisation of hemicellulose was also high (about 42%). At 195°C and neutral pH,
similar degrees of delignification and hemicellulose solubilisation were achieved, but
unfortunately without adding acid or base the reaction temperature was difficult to control
during the process.
During wet oxidation the material are flowing in the loop reactor. This flow is very
difficult in the beginning of the reaction, because the material is too stick and the diameter
of the loop is relative small (∅ 40 mm). However during the pretreatment corn stover starts
to solubilise and the flowing becomes more and more easier. When no chemical is added,
this solubilisation and also the flowing starts later (only after 2-3 minutes), which results
inhomogeneous temperature profile in the material and insufficient mass transfer. Thus this
pretreatment might be less reproducible. Figure 4.5. shows the cellulose and hemicellulose
content for each condition.
40
40
20
20
0
0
A
B
C
D
E
F untreated
A
Type of pre tre a tme nt
B
C
D
E
F untreated
Type of pre tre a tme nt
A 185°C, 5 min, pH 9.2
B 185°C, 15 min, pH 7.3
C 185°C, 5 min, pH 3.5
D
E
F
195°C, 15 min, pH 9.2
195°C, 5 min, pH 7.3
195°C, 15 min, pH 3.5
Figure 4.5. Cellulose and hemicellulose content % (w/w) following wet-oxidations in the
pre-treated corn stover samples
67
Table 4.12. Material balances for each pretreatment conditions g/100 g raw material
CONDITIONS OF WET-OXIDATION
g/100g raw
material
61
Solid fraction
DM (g)
NCWM* (g)
Hemicellulose (g)
Cellulose (g)
Lignin (g)
Ash (g)
Liquid fraction**
Glucose (g)
Xylose (g)
Arabinose (g)
Native corn
stover
185°C
5 min
pH 9.2
185°C
15 min
pH 7.3
185°C
5 min
pH 3.5
195°C
15 min
pH 9.2
195°C
5 min
pH 7.3
195°C
15 min
pH 3.5
100.0
11.7
33.7
41.0
8.7
4.9
76.4 (100%)
8.5 (11.1%)
21.6 (28.3%)
38.7 (50.7%)
5.8 (7.6%)
2.1 (2.7%)
53.3 (100%)
6.1 (11.4%)
3.3 (6.2%)
37.8 (70.9%)
4.4 (8.3%)
1.5 (2.8%)
65.1 (100%)
8.2 (12.6%)
9.2 (14.1%)
39.8 (61.1%)
5.9 (9.1%)
2.1 (3.2%)
48.6 (100%)
5.7 (11.7%)
2.6 (5.3%)
34.9 (71.8%)
3.5 (7.2%)
1.9 (3.9%)
60.0 (100%)
7.3 (12.2%)
6.9 (11.5%)
39.2 (65.3%)
5.0 (8.3%)
1.6 (2.7%)
53.7 (100%)
7.4 (13.8%)
0.9 (1.7%)
36.2 (67.4%)
7.1 (13.2%)
2.1 (3.9%)
1.6 (0.05)
3.2 (0.16)
1.2 (nd)
2.8 (0.3)
16.4 (1.67)
3.0 (0.07)
1.1 (0.03)
14.3 (0.6)
1.3 (nd)
3.4 (0.35)
16.5 (2.3)
3.0 (0.6)
1.5 (0.1)
12.5 (0.9)
3.2 (0.08)
2.3 (0.28)
17.1 (2.48)
3.2 (0.9)
* Non-cell-wall-material: water soluble substances and extractives such as pectin, protein, etc.
** The number in bracelets shows the amount of pretreatment “direct” released mono-, and disaccharides (without sulphuric acid hydrolysis)
nd Non detected
68
The pretreatment process enlarges the inner surface area partly by hemicellulose
solubilisation and lignin degradation. The maximal recovery of hemicellulose and cellulose
sugars is important for an ideal pretreatment. The recovery of cellulose and hemicellulose
was calculated to estimate their losses during wet oxidation at the different conditions
(Table 4.13.). The calculation was based on the mass balance. The equation below shows
the calculation for the recovery of cellulose (4) and hemicellulose (5).
C re cov ery %( w / w) =
[Cellulose in filter cake ( g ) + ( glu cos e in filtrate ( g )) / 1.11]
[Cellulose in untreated corn stover ( g )]
(4)
HC re cov ery %( w / w) =
*
[HC in
filter cake ( g ) + ( xylose arabinose in filtrate ( g )) / 1.14
HC in untreated corn stover ( g )
[
]
C = cellulose
HC= hemicellulose
(5)
**
In wet oxidation, mainly the hemicellulose was converted and/or degraded. Around 60% of
the original hemicellulose could be recovered, in all experiments. This relatively low
recovery was probably due to hemicellulose oxidation to other products, such as carboxylic
acids, CO2. and H2O. The recovery of cellulose at 185°C was nearly 100%, whereas the
cellulose recovery at 195°C was still above 90%. An overall recovery of 80%
carbohydrates was obtained, which was similar to the results achieved with wet oxidation
of wheat straw (Schmidt et al. 1998).
Table 4.13. Recovery % (w/w) of different components for each pretreatment conditions
Components
Hemicellulose
Cellulose
Overall recovery
185°C
5 min
pH 9.2
76.0
98.0
88.1
CONDITIONS OF WET-OXIDATION
185°C
185°C
195°C
195°C
15 min
5 min
15 min
5 min
pH 7.3
pH 3.5
pH 9.2
pH 7.3
60.3
69.5
60.1
64.9
98.7
99.6
92.6
98.3
81.4
86.0
78.0
83.2
195°C
15 min
pH 3.5
57.2
93.4
77.1
Regarding the results both of the recovery of the polysaccharides and the enzyatic
digestibility of the cellulose in the pre-treated corn stover, two pretreatments, one alkaline
and one acidic were selected for further investigations. Both wet oxidation were carried out
at 195°C for 15 min, the differences were the applied chemical: 2 g/L Na2CO3 in alkaline
0.5 mL 96% H2SO4 in the acidic pretreatment.
Similar to the previously, after treatment, the material were separated into a solid and a
liquid fraction and both fractions were analysed.
69
]
The hydrolysate after alkaline WO (Table 4.14.) consisted of soluble hemicellulose
(10.1 g/L), quantified as monomers 8.5 g/L xylose, 1.2 g/L arabinose and 1.7 g/L glucose.
After acidic pretreatment the concentration of the soluble hemicellulose was 10.6 g/L,
quantified as monomers 8.6 g/L xylose, 1.6 g/L arabinose and 1.8 g/L glucose.
The origin of the solubilised glucose is unclear, as it may originate from cellulose or as
minor component in hemicellulose, which main component is xylan in monocots (Hon and
Shirashi, 2001).
The carboxylic acids formed by WO of corn stover were acetic acid, formic acid and
glycolic acid (Table 4.14.). Alkaline WO resulted in higher concentrations of carboxylic
acids (5.9 g/L) than acidic WO (4.6 g/L). During WO pretreatment at alkaline and neutral
conditions, carboxylic acids are more stable towards degradation due to resonance
stabilisation of the deprotonated carboxylgroup (Bjerre and Sørensen, 1992). Therefore,
the observed lower carboxylic acid concentration in acidic WO hydrolysate, is explained
by the unstability of the protonated acids and resulting degradation by decarboxylation and
oxidation.
Table 4.14. Composition (g/L) and pH of corn stover liquid fraction after pretreatment by
wet oxidation (WO).
LIQUID FRACTION
pH
Hemicellulose3
(Xyl: Ara: Glu)
Xylose4
Arabinose4
Glucose4
Glycolic acid
Formic acid
Acetic acid
ALKALINE WO1 (G/L)
3.9
8.7
(10: 0.7: 2)
0.8
0.6
0.1
1.3
2.0
2.6
ACIDIC WO2 (G/L)
2.7
3.8
(10: 0.7: 3)
5.5
1.4
0.8
1.1
1.8
1.7
1
Pretreatment conditions: 195°C, 15 min, 12 bar O2, 2 g Na2CO3.
Pretreatment conditions: 195°C, 15 min, 12 bar O2, 0.5 mL 96% H2SO4.
3
Quantified by substracting the free sugars from the total sugars released by acid hydrolysis, and
corrected for loss of water on polymerisation. Ratio of the individual sugars is shown in
parenthesis.
4
Free sugars released by pretreatment
2
The sum of monomeric phenolic lignin degradation products from corn stover was quite
low (Table 4.18.), both after alkaline WO (0.45 g/L) and acidic WO (0.62 g/L). The main
phenols in both hydrolysates were guaiacol, 4-hydroxybenzaldehyde, vanillin,
syringaldehyde, 4-hydroxybenzoic acid, vanillic acid, syringic acid and coumaric acid
occurring in the concentration range 28-91 ppm (mg/L).
70
4.3.2. Enzymatic hydrolysis
To enhance the enzymatic susceptibility of the cellulose, applying specific pretreatment
process is essential. The goal of the pretreatment is to disrupt the lignocellulosic matrix to
make the substrate more accessible to the enzymes. The efficiency of the wet oxidation
process was investigated in enzymatic hydrolysis. Table 4.15. shows the percentage of
enzymatic convertible cellulose (ECC%) of the filter cakes after 24 h hydrolysis at 50°C.
The enzymatic conversion of pre-treated cellulose in the remaining solid was between 52
and 83% compared to 18% enzymatic conversion for the native (untreated) corn stover.
The highest enzymatic conversion (83.1%) was achieved following wet-oxidation at 195°C
for 15 minutes at alkaline pH. This was expected, because this pretreatment modified most
significantly the composition of the solid residue following wet oxidation compared to the
untreated stover. The combination of high temperature and acidic pH also gave highly
convertible cellulose, resulted 73.7% conversion.
Table 4.15. Cellulose conversions (%) and the saccharification yields (% of theoretical)
after hydrolysis (24 h at 50ºC, 25 FPU/g DM) of filter cakes and slurries
Untreated 185°C
corn
5 min
stover
pH 9.2
Saccharification yield (%)[YieldDNS %]
18.9
52.2
Filter cake at 50°C
50.5
Slurry at 50°C
38.1
Slurry at 40°C
CONDITIONS OF WET-OXIDATION
185°C 185°C 195°C 195°C
15 min 5 min 15 min 5 min
pH 7.3 pH 3.5 pH 9.2 pH 7.3
195°C
15 min
pH 3.5
73.9
61.6
50.7
70.3
66.1
48.2
93.0
71.8
64.4
68.3
70.0
52.2
78.7
61.4
52.7
Enzymatic Cellulose Conversion (%) ECC
18.1
52.0
62.0
Filter cake at 50°C
55.9
59.1
Slurry at 50°C
39.5
44.8
Slurry at 40°C
61.7
61.5
47.0
83.1
63.5
55.2
63.0
60.3
43.6
73.7
47.2
46.2
The wet oxidation at high temperature (195°C) generally increased more the ECC%, than
pretreatments at lower temperature (185°C), especially if the pretreatment was carried out
at acidic or alkaline pH. This could be explained by either lignin and hemicellulose
removal or by decreased crystalinity of cellulose due to the hydrolysis catalysed by harsh
pertreatment conditions. During pretreatment at 185°C with alkali addition, the pH
decreased from 9.2 only to 5.7 and both the hemicellulose and lignin content decreased
only slightly, which explained the poorest ECC% value, of about 50%.
Although pretreatment at neutral pH gave similar enzymatic conversion at 185°C and
195°C (62 and 63% respectively), the conversions of the replicates samples showed wide
variance (63, 72%) (Figure 4.6.-E) corresponding to the poor reproducability, observed in
composition analysis (Figure 4.5.-E).
71
80
ECC %
60
40
20
0
A
B
C
D
E
F Untreated
Type of pretreatment
A
B
C
185°C, 5 min, pH 9.2
185°C, 15 min, pH 7.3
185°C, 5 min, pH 3.5
D
E
F
195°C, 15 min, pH 9.2
195°C, 5 min, pH 7.3
195°C, 15 min, pH 3.5
Figure 4.6. Conversion of cellulose (ECC %) of the remaining solids, following 24 h
enzymatic hydrolysis at 50°C
In a high-temperature pretreatment, furfurals, e.g., hydroxymethyl-furfural, and other byproducts can be produced, which are known inhibitors of microorganisms and affect
adversely the action of cellulases. To investigate the inhibitory effects of these byproducts, the whole slurry was also hydrolyzed. The hydrolysis yields based on the DNS
total reducing sugar analysis are presented as a saccharification yield in Table 4.15.
The saccharification yield (%) (YieldDNS %) based on the total reducing sugar measured by
DNS was calculated as (6):
c DNS ⋅ V
⋅ 100%
Yield DNS =
mcellulose ⋅ 1.11 + mhemicellulose ⋅ 1.14
(6)
where cDNS is the reducing sugar (g/L) measured by DNS, V is the total volume (L), the
mcellulose and the mhemicellulose are the weights of cellulose and hemicellulose before
enzymatic hydrolysis (g), which was calculated in Table 4.12. The 1.11 and the 1.14
factors convert the polymer concentrations to the equivalent monomer concentrations.
In the hydrolysis of slurries, the combination of high temperature (195°C) and alkaline pH
(9.3) resulted both the highest enzymatic conversion (63%) and the highest saccharification
yield (71.8%) as it was also observed in hydrolysis of filter cake. Compared to the
hydrolysis of filter cake and the slurry, the achieved conversions following hydrolysis of
slurries were usually lower at high temperature (195°C), but were similar at lower
temperature (185°C), except at acidic conditions. In this case the enzymatic conversion of
filter cake was significant higher 73.7%, than in the case of slurry (47.2%).
72
The steps following pretreatment, i.e., hydrolysis and fermentation can be run separately
(SHF) or simultaneously (SSF). The main benefit of the SSF is, that the yeast immediately
consume the produced glucose, thus the strong inhibitory effect of the glucose in
hydrolysis could be avoided. Using thermotolerant yeast (Klyveromyces marxianus)
(Olsson and Hahn-Hägerdal 1993, Palmqvist et al., 1997, 1999) the temperature
maximum of the SSF is 40°C, hence, hydrolysis of the slurries was also performed at
40°C.
The hydrolysis at 40°C after 24 hours gave about 20% lower conversions, than at 50°C
(Figure 4.7.). The hydrolysis at 50°C seemed to be completed after 48 hours, but at 40°C
the cellulose conversion increased even after 48 hours. At 50°C, the enzymatic reaction
seemed to be was faster. The optimal temperature for enzymatic hydrolysis using T. reesei
cellulase is usually considered to be 50°C, but is for short reaction times, such as the filter
paper assay. However, for a longer hydrolysis, the denaturation and inactivation of the
enzymes should be considered, which has been reported by Kaar et al (2000).
100
ECC %
75
FC 40°C
FC 50°C
50
Slurry 40°C
Slurry 50°C
25
0
0
24
48
72
time (h)
Figure 4.7. Time curve of hydrolysis of filter cake and slurry (pre-treated at 195°C, for 15
minutes, on pH 9.2) at 40°C and 50°C
To enhance the commercial competitiveness of bioethanol, its production cost must be
reduced. When considering the cost of hydrolysis, the enzymes are the greater part. For
that reason, it is important to use enzymes efficiently by creating a favourable environment
in the hydrolysis. The effect of the enzyme loading at 40 and 50°C are shown in
Figure 4.8. The highest enzymatic conversion was achieved when the highest FPU was
applied, both for hydrolysis of slurries and filter cakes. However, in the hydrolysis of filter
cakes at 50°C, the conversion was only 15% lower when the enzyme loading was
decreased from 25 to 5 FPU per gram dry biomass. In hydrolysis of slurry, decreasing the
enzyme loading by five fold, decreased cellulose conversion by 50 %.
73
100
ECC (%)
75
FC 50°C
FC 40°C
slurry 50°C
slurry 40°C
50
25
0
0
5
10
15
FPU/ g DM
20
25
Figure 4.8. Cellulose conversion for different enzyme loading after 24 h hydrolysis of
filter cake and slurry at 50°C and 40°C. Pretreatment conditions: 195°C, 15 min, 9.2 pH.
4.3.3. SSF
4.3.3.1. Effect of substrate concentration on ethanol yield
In several studies it was found, that for conventional fermentation of lignocellulosics, the
content of solids is limited to about 10%, resulting in a maximum ethanol concentration of
4 (v/v)%. However, if higher solids levels could be fermented, it might be possible to
achieve higher ethanol concentration (Spindler et al, 1990, Mohagheghi et al, 1990).
To maximise the ethanol concentration, the substrate concentrations were increased from
8% to 20% DM, corresponding to cellulose concentrations of 4% to 14%. In our
preliminary studies of SSF a mixing problem was encountered in the shaking flasks filled
with pre-treated corn stover. When the dry matter content was increased to 12% DM no
fermentation products could be seen at all (data not shown). Stenberg et al (2000a)
observed similar problems. With steam-pre-treated softwoods, as well as Spindler et al.
(1989, 1990) had problems fermenting pre-treated herbaceous crops and wheat straw at
high DM contents. To avoid this problem the wet oxidised material was added to the
reaction in 3 portions during enzymatic pre-hydrolysis. The second portion was added in
the fifth hour of the hydrolysis and the final portion in the tenth hour. The hydrolysis rate
in the beginning of the reaction was high, as it was shown above, thus the pre-treated
fibrous material was sufficiently liquidised in 5 hours. In this way the dry matter content
could be increased efficiently to 17% DM. The effect of substrate concentration on the
ethanol yield at 120 h SSF using alkaline wet oxidised corn stover is shown in Table 4.16.
Increasing the dry matter content to 20%, the achieved ethanol yield decreased
74
significantly. The ethanol yield showed a maximum around 10% substrate concentration,
but it resulted in only 63% ethanol yield, which was about 25% lower, than it was
calculated from the disappeared glucose (CEtOH). The inefficient conversion of glucose to
ethanol could be explained by non-adequate anaerobic conditions.
Table 4.16. Ethanol yield and ethanol concentration (g/l) after 120 h SSF of alkaline wet
oxidised (195°C/ 15 min/ 9.2 pH) corn stover with the final enzyme loading of
30 FPU/ g DM. (The filling level of the bottles was 50%.)
SUBSTRATE CONCENTRATION
(W/W % DM)
8%
10%
12%
14%
17%
20%
YEtOH
(%)
57.3
63.3.
58.9
53.6
50.7
0.41
CEtOH
(%)
65.5
70.2
68.2
60.2
58.5
5.7
To make the fermentation more anaerobic, the amount of the fermentation broth was
increased from 300 g to 450 g in the 0.5 L flasks, which diminished their headspace of air.
The 0.5 L flasks containing 450 g hydrolysate obtained superior ethanol yields compared
to the flasks containing less amount of fermentation liquor. Further experiments were
carried out with the optimal 90% filling level.
To study the effect of substrate concentration on the ethanol yield acidic wet oxidised corn
stover was also investigated. Four different DM concentrations were tested: 10, 12, 15 and
20% (w/w DM). The maximum ethanol yield of 85% was achieved at 12% DM. This
substrate concentration was also chosen for SSF study on acidic pre-treated wheat straw
(Spindler et al, 1989). Figure 4.9. shows, that the ethanol yield was markedly reduced at a
substrate concentration of 20% DM compared to 10-15% DM, probably due to insufficient
mass transfer caused by stirring hindrance. At high substrate concentrations the enzymes
couldn’t liquefy the pre-treated fibrous material, and the poor saccharification rate resulted
low ethanol yield of around 5%.
75
YEtOH (%)
100
75
50
25
0
10%
12%
15%
20%
substrate concentration (% w/w DM)
Figure 4.9. Effect of the substrate concentration (% w/w DM) to the ethanol yield after
120 h SSF using acidic pretreatment corn stover with 30 FPU/g DM enzyme loading.
4.3.3.2. Effect of enzyme loading
Considering the cost of the bioethanol process, the initial pretreatment of the biomass and
the costly cellulase enzymes are critical targets for process and cost improvements
(Mielenz et al, 2001). For that reason, it is important to minimise the addition of enzymes
by using optimal pretreatment conditions producing easy digestible carbohydrates e.g.
cellulose. A substrate level of 12% DM alkaline WO corn stover was chosen for cellulase
concentration study. Figure 4.10. demonstrates the effect of increasing enzyme loading on
the ethanol yield. With 10 FPU/g DM, ethanol production was quite low and resulted in
only 62% ethanol yield. Increasing the enzyme loading from 10 to 15 FPU/g DM the
achieved ethanol yield was more satisfactory around 75%. For further lift of the enzyme
loading increased slightly the ethanol yield, which was also found by Wyman (1994).
However, the highest ethanol yield, 84% of theoretical, was achieved when the highest
enzyme loading (30 FPU/g DM) was applied. According to Figure 4.10. lower enzyme
loading (10 FPU/g DM) required a longer reaction time. Using 10 FPU/g DM the ethanol
yield nearly doubled from the second to the fifth day, but applying higher enzyme loading
the fermentation seems to be finished after 72 h, the ethanol yield increased only with 5%
in the last 2 days.
76
100
YEtOH (%)
75
5+5 FPU
5+10 FPU
10+10 FPU
10+20 FPU
50
25
0
0
50
100
150
time (h)
Figure 4.10. Effect of the enzyme loading on ethanol yield (YEtOH) as a function of time
during SSF at 30ºC with 12 % substrate concentration, after alkaline pretreatment
The results including the ethanol yields calculated from the ethanol concentration
determined by HPLC (YEtOH_HPLC) and measured from the weight loss (YEtOH_CO2) are
presented in Table 4.17. The comparison of the two ethanol yields (YEtOH_HPLC and
YEtOH_CO2) showed no significant difference, thus monitoring the ethanol production by
weight loss was found to be acceptable to follow the fermentations. SSF with 17% DM of
alkaline wet oxidised corn stover resulted the highest ethanol concentration of 52.3 g/l,
corresponding to 78% ethanol yield of the theoretical. However, lower substrate
concentration (12% DM) gave also favourable result of 43 g/L final ethanol concentration.
Table 4.17. Non isothermal simultaneous saccharification and fermentation (SSF) results
for experiments with Saccharomyces cerevisiae
DM
(%)
Enzymes
Glucose
(FPU/
(g/l)
g DM)
Alkaline WO corn stover
12
10
0.2
12
20
0.2
12
30
0.1
17
30
0.3
Acidic WO corn stover
12
30
0.3
15
30
0.3
20
30
51.8
EtOH (g/l)
YEtOH_HPLC
(%)
YEtOH_CO2
(%)
CEtOH
(%)
18.3
35.1
43.2
52.3
63.0±0.9
76.3±1.0
83.8±1.1
78.0±2.2
64.2±1.8
76.8±1.1
81.5±1.5
79.7±2.5
77.6±1.2
83.4±1.3
96.4±1.2
85.8±1.4
39.9
38.9
0.3
85.0±1.8
75.1±2.4
5.7±2.1
86.7±2.1
76.0±2.4
8.4±2.8
93.6±1.6
88.4±1.8
5.7±1.5
YEtOH_HPLC: ethanol yield calculated from EtOH concentration measured by HPLC
YEtOH_CO2: calculated ethanol yield from the measured weight loss.
77
Similar results 47-52 g/L ethanol concentration were achieved by Mohagheghi et al. (1992)
on acid pre-treated wheat straw after six days SSF fermentation. The achieved ethanol
concentration after 120 h SSF of acidic wet oxidised corn stover using 30 FPU/g DM
enzyme loading, was also around 40 g/L both in the case of using 12 and 15% substrate
DM, corresponding to a yield of 85 and 75%, respectively. The lowest ethanol
concentration and also the lowest ethanol yield was obtained, when the highest substrate
concentration (20%) was applied of acidic pre-treated material.
The limit of substrate concentration was found to be about 17% using alkaline and 15%
using acidic WO pre-treated corn stover, respectively.
4.3.3.3. Conversion of phenols and carboxylic acids by SSF
During SSF of alkaline WO and acidic WO corn stover, a slight decrease of the total
phenol monomer concentration was observed (Table 4.18.). This was mainly due to the
assimilation of phenol aldehydes and homovanillic acid from the fermentation broth.
Phenol aldehydes (4-hydroxybenzaldehyde, vanillin, syringaldehyde) were assimilated and
partly recovered in the fermentation broth as the corresponding phenol alcohol (4hydroxybenzyl-, vanillyl-, syringyl alcohol). The change in phenol concentration of WO
corn stover hydrolysates during fermentation resembled earlier studies with S. cerevisiae
model fermentations of phenols (Klinke et al., 2003). The phenol acids recovered after
fermentation differed in the same way, e.g. the concentration of 4-hydroxy benzoic acid
was unaltered, however, only 71-84% of vanillic acid and syringic acid was recovered by
the end of fermentation. Acetovanillone was the phenol ketone formed in highest amount
and most efficiently assimilated during fermentation.
The arabinoxylan from primary cell walls of gramineous monocots is esterified with ferulic
and p-coumaric acids (Hon and Shirashi, 2001). The solubilised hemicellulose of WO corn
stover was mainly arabinoxylan, hence enzymatic hydrolysis caused an increase of
coumaric acid and ferulic acid during SSF.
78
Table 4.18. The concentration (ppm=mg/L) of phenolic degradation products in corn
stover WO hydrolysates before and after 120 h combined enzymatic hydrolysis and
fermentation (SSF).
Compounds
Phenol
Guaiacol
Syringol
Total phenols
4-hydroxybenzaldehyde
Vanillin
Syringaldehyde
Total phenol aldehydes
4-hydroxy benzylalcohol
vanillyl alcohol
syringyl alcohol
Total phenol alcohols
4-hydroxyacetophenone
Acetovanillone
Acetosyringone
Total phenol ketones
2-furoic acid
4-hydroxy benzoic acid
vanillic acid
homovanillic acid
syringic acid
coumaric acid
ferulic acid
Total phenol acids
Sum of phenols
After alkaline WO
Hydrolysate After SSF
6
4
12
9
5
4
23
17
54
0
57
0
20
0
131
0
0
33
0
48
0
18
0
99
8
6
13
6
18
13
40
25
7
8
51
49
55
46
12
0
28
23
32
42
7
32
192
202
385
342
After acidic WO
Hydrolysate
After SSF
4
2
13
7
5
0
22
9
102
0
74
0
38
0
215
0
0
44
0
46
0
11
0
101
8
4
58
28
0
7
65
38
4
0
54
55
57
44
28
11
42
30
35
31
9
15
228
186
530
335
4.3.4. Conclusions on wet-oxidation pretreatments
Wet oxidation was an efficient process for breaking the tight association between lignin
and polysaccharides in corn stover. After the best pretreatment the enzymatic accessibility
of the cellulose in corn stover increased four times. The best conditions for obtaining high
convertible cellulose was 195°C for 15 minutes with added Na2CO3 giving about 85%
conversion to glucose with enzyme loading of 25 FPU/g DM. Major part of the
hemicellulose was dissolved, and approximately 36-45% of the original hemicellulose
could be identified as saccharides in the separated liquid fraction following pretreatment.
79
The enzymatic hydrolysis at 50°C was completed after 24 hours, however the enzyme
activity decreased by 40% after 1 day at 50°C. A temperature of 40°C needed longer time
for the complete hydrolysis, but after 72 hours even better results of cellulose conversion
was obtained compared to the conversions at 50°C after 24 hours. The enzyme loading
could be decreased from 25 FPU/g DM to 15 FPU/g DM efficiently in the hydrolysis of
filter cakes. The wet oxidation (WO) process at 195ºC, for 15 min, 12 bar oxygen both at
acidic and alkaline pH was able to produce a cellulosic fraction that was efficiently
fermented in WO hydrolysate by S. cerevisiae (dried Baker’s yeast) to ethanol. The results
indicate that it is possible to obtain over 80% of theoretical ethanol yield, based on the
cellulose available in the pre-treated corn stover with SSF in 5 days. Phenols and
carboxylic acids were present at sub-inhibitory levels. After 72 hours fermentation, the
phenol aldehydes were assimilated, while other phenols were neither converted nor
assimilated.
An increase in cellulase concentration resulted in an increase in the production rate and
higher ethanol yield during the fermentation period investigated. An enzyme loading of 30
FPU/g DM was sufficient to complete SSF in 5 days, lower enzyme loading (10 FPU/g
DM) required additional 2 days SSF. However, the enzyme loading could be decreased
from 30 FPU/g DM to 15 FPU/g DM, lowering the final ethanol yield by only 10%.
The substrate concentration could be increased efficiently up to 15 % DM, but above this
substrate level the ethanol yield decreased significantly, due to mixing problems. The
highest ethanol yield 85% of the theoretical was achieved, using 12% DM of acidic pretreated corn stover. The highest ethanol concentration using SSF process was above 50 g/l
(6 Vol. % ethanol), which exceeds the technical and economical limit of the industrialscale alcohol distillation.
The fermentation was monitored by weighing the fermentation bottles at regular intervals
and the amount of produced ethanol was calculated from the CO2 resulted weight loss.
Comparison of the calculated final ethanol yield with HPLC measurements, showed no
significant differences, thus this easy monitoring system could be applied to monitor
ethanol production by SSF.
4.4. COMPARISON OF THE DIFFERENT PRETREATMENT PROCESSES
The main purpose of this study was to investigate the different pretreatment process mainly
in the view of enzymatic hydrolysis.
The enzymes are of vital importance in the breakdown of cellulose to glucose, however the
pretreatment is essential to make cellulose accessible to the enzymes. In Chapter 4,
detailed results following different type of pretreatments are presented, and it seemed, that
many of them were able to increase significantly the enzymatic convertibility of cellulose
to glucose. The 4.19. summarises the promising pretreatment methods. The numbers
marked with blue show the best results following different pretreatment process.
80
Table 4.19. Summarzing table: Results following the best pretreatment methods. The mass balance based on 100 g (DM) untreated corn stover
(containing 41 g cellulose, 28 g hemicellulose and 22 g lignin).
74
XYLOSE (G) CELLULOSE (G)
ECC (%)
Released from
in the solid
Enzymatic
the pretreatment
fraction
convertibility
following
pretreatment
UNTREATED CORN STOVER
41
18.1
CHEMICAL PRETREATMENT 1: ALKALINE PRETREATMENT
1% NaOH
14.5
79.5
28.6
5% NaOH
17.9
24.0
82.0
10% NaOH
19.3
20.2
84.4
CHEMICAL PRETREATMENT 2: ACIDIC PRETREATMENT
1% H2SO4
17.5
25.8
51.9
5% H2SO4
29.5
38.4
19.3
CHEMICAL PRETREATMENT : TWO-STEP PRETREATMENT
1% NaOH and 1% H2SO4
18.9
18.5
95.7
STEAM EXPLOSION
190°C/2% H2SO4/5min
24.2
81.1
21.3
200°C/2% H2SO4/5min
19.0
20.0
83.6
210°C/2% H2SO4/5min
16.8
13.9
78.8
WET OXIDATION
195°C/15min/alkaline
19.5
34.9
83.1
195°C/15min/acidic
73.7
20.3
36.2
81
RELEASED GLUCOSE (G)
RECOVERY (%) OF
From
From hydrolysis Hemicellulose
Cellulose
hydrolysis
and
pretreatment
8.2
25.0
21.6
17.9
33.3
32.3
31.5
85.9
85.1
84.1
89.4
84.0
79.7
14.7
12.5
25.1
20.5
84.7
83.0
87.5
90.9
19.5
36.7
82.9
79.6
21.6
18.4
12.0
34.3
31.3
28.3
72.0
64.0
55.1
83.0
74.0
73.0
31.9
29.3
35.3
31.6
60.1
57.2
92.6
93.4
The highest enzymatic cellulose conversion, nearly theoretical (95.7%), was achieved
following two-step chemical pretreatment, where 1% NaOH was used after 1% H2SO4.
Although the convertibility of the pre-treated cellulose is very important in characterisation
of the pretreatment, but a good fractionation method is also able to retain the cellulose in
the pre-treated solid fraction, which makes the further processing more easier. During this
two-step chemical pretreatment nearly half of the original cellulose and around 75% of the
original hemicellulose was solved. The purification and/or utilisation of the solved sugars
from the acidic and the alkaline solution and also from the washing-water is a difficult
technical problem and makes a shock in process economy. The separation and the washing
following both steps of this pretreatment are necessary to remove the chemicals from the
fibers, because it has a negative effect to the enzymes.
Satisfactory enzymatic conversions were also achieved using concentrated, 5 and 10%
NaOH, resulting 73 and 79% conversions respectively. The disadvantages of this chemical
pretreatment are the same as in the case of the two-step pretreatment. Acidic chemical
pretreatment, even when concentrated acid was used, at the applied relative low
temperature (100-120°C) gave only poor enzymatic conversion. Thus it is not proposed for
pretreatment of lignocellulosic materials, however it could be used effective when the aim
of the work is the hemicellulose solubilization.
To make the pretreatment more environmental friendly and find an economically feasible
method, steam pretreatment and wet oxidation process were tested for treating of corn
stover. Both processes run at higher temperature and with fewer amounts of applied
chemicals, compared to previous mentioned chemical pretreatments.
Steam pretreatment and wet oxidation showed several common features. Both processes
increased the enzymatic conversion significantly, the highest conversions was 83.1
following wet oxidation and 83.6% after steam pretreatment. The selected “best”
pretreatment conditions in the view of enzymatic conversion, were 195°C, 15 min, alkaline
addition in wet oxidation and 200°C, 5 min, 2% H2SO4 in steam pretreatment. The
distinctive feature of an ideal pretreatment, is the high recovery of both polysaccharides. In
comparison of the hemicellulose recovery of this two physico-chemical pretreatments, the
results were also the same around 60%, however cellulose recovery was 15% lower
following steam pretreatment. The wet oxidation was also much better in retaining
cellulose in the solid fraction after reaction, than steam pretreatment. Following the above
selected pretreatment conditions the cellulose content in the solid fraction was 30% higher
after wet oxidation, than after steam pretreatment. In lignin content the difference was even
bigger. The original lignin content decreased by 75% following wet oxidation while it was
only 10% after steam pretreatment.
To obtain high ethanol yield, the efficient enzymatic hydrolysis is necessary, however the
fermentability of the pre-treated material is also essential. The wet oxidised corn stover
was investigated in simultaneous saccharification and fermentation and no inhibitory effect
could be observed. The achieved ethanol yield was above 80% after wet oxidation at
82
195°C, for 15 min, both at alkaline and acidic conditions. It is worth to mention, that while
alkaline wet oxidation resulted the highly digestible cellulose, the highest ethanol yield
was achieved after acidic wet oxidation. The fermentability of steam pre-treated material
was also tested, and these results were also attractive, the achieved ethanol yield was above
80%.
Based on the comparison of wet oxidation and steam pretreatment, it could be concluded,
that both processes are able to enhance significantly the enzymatic conversion, and the pretreated material is highly fermentable. However because of the higher cellulose recovery,
and the higher cellulose content of the pre-treated material, wet oxidation process (under
the above selected conditions) is proposed for pretreatment of corn stover.
4.5. ESTIMATED PRODUCTION COST OF FUEL ETHANOL
4.5.1. Production cost of starch based bioethanol
For a rough cost calculation, approximately 3.4 kg corn is needed for production of 1 L
EtOH. While the production cost of corn is on average 73-98 $/t, the total production cost
of 1L starch based ethanol is 0.25-0.35 $ in general.
The use of corn for producing EtOH in the world shows significant regional variations with
respect to the energy consumed. The major differences are to be found in the use of
fertilizers and in the irrigation techniques employed. In most cases, the energy inputs are
coal, natural gas and fuel oil, i.e., all of fossil origin. Energy is consumed both in the corn
production process (primarily in the form of energy used for the production of fertilizers
and as fuel for the agricultural machines) and in the process of converting corn into ethanol
(grinding, enzymatic hydrolysis, fermentation and distillation).
4.5.2. Production cost of lignocellulose based bioethanol
A detailed engineering and economical analysis of a process employing steam explosion as
pretreatment followed by enzymatic hydrolysis, performed by Nysrtom et al. (1985) gave
an ethanol production cost of 1.02 $/L, employed a capacity of 410 metric tons of dry hard
wood per day.
Another cost estimate made by Clausen and Gauddy (1983) for a dilute acid hydrolysis
process resulted in an ethanol production cost of 0.24 $/L. The process capacity was about
630 metric tons of dry oak per day.
Lambert et al. (1990) have presented data for hardwoods and corn stover as feed stock for
dilute and concentrated acid hydrolysis processes, based on pilot-scale studies. Both
83
processes were based on 500 metric tons of raw material per day, treated in two hydrolysis
steps. The ethanol production cost was estimated to 0.45 $/L in the concentrated acid
process, and 0.48 $/L in the dilute acid process.
A flow sheet program and an economic evaluation program were used for simulations and
economics evaluations in the cost estimate study of Söderström (2004). The production
plants considered were designed to utilise spruce as raw material with a capacity of 550
tons per day, employing steam explosion as pretreatment followed by enzymatic
hydrolysis. Based on these assumptions, the ethanol production cost was 0.57 $/L.
The variation in the production cost can be explained by different assumptions made in the
technical and economic calculations. Differences are found not only in the raw materials
used (softwood, hardwood, straw, corn stover, etc.) and the type of process utilised
(enzymatic hydrolysis, dilute or concentrated acid hydrolysis), but also in the design of the
process (separate hydrolysis and fermentation, simultaneous saccharification and
fermentation, pentose fermentation, etc.) and the assumptions concerning yields, capacity
and costs (Von Sievers and Zacchi, 1996). There are other factors that have to be taken in
consideration; e.g. collection cost of corn stover would depend on the amount of stover
collected per unit area, the number of operations, machine efficiency in each operation and
bulk density, but it also depends on the delivering distance from a harvested field
(Sokhansanj et al. 2002).
The differences of the estimators are, to a certain extent, unavoidable, as there are no fullscale plants from which information on yields and other crucial data are available. The
accuracy in the calculations also varies between different investigators. A rough estimate
of the poduction cost on starch and cellulose based bioethanol is shown in Figure 4.19.
Fuel ethanol cost (%)
100
Variable operating
costs
75
By-products
50
Labor, supplies
and overhead
25
Feedstock
0
starch base d
ce llulose base d
Figure 4.19. Comparison of the starch and cellulose based bioethanol cost
84
Figure 4.20. shows the contribution of major cost elements to overall ethanol
production cost (Wyman, 1999). The feedstock in this case is poplar and the applied
pretreatment method is a dilute acid hydrolysis at elevated temperature. This figure
shows that the feedstock is still the single most costly element, at approximately 39%
of the total, but as mentioned above, it is diffcult to impact feedstock costs
substantially for eventual large-scale bioethanol production. However, the costs of the
processing steps can be reduced further, and the most expensive of these steps is for
pretreatment, representing almost one-third of the total processing costs. The second
most costly operation is the SSF process, accounting for approximately 28% of the
total. Thus, a better pretreatment technology could have an impact both by lowering
the cost to break down hemicellulose and by improving the rates and yields in the
SSF process. The third most costly operation is product recovery, but at a far lower
12.6% of the total processing cost with a similar contribution by the remaining
process steps, including waste recovery. The costs for pentose conversion and
cellulase production are about half the cost of distillation.
Percent total cost
40.0
30.0
20.0
10.0
io
n
en
t
at
SS
F
fe
rm
tre
Pr
e
at
en
t
m
k
sto
c
Fe
ed
D
ist
ill
O
at
th
io
er
n
pr
oc
Pe
es
sin
nt
os
g
ec
on
Ce
ve
rs
llu
io
la
n
se
pr
od
uc
tio
n
Po
w
er
cy
cl
e
0.0
Figure 4.20. Contribution of major cost elements to overall ethanol production costs.
85
5. CONCLUSIONS AND FUTURE POSSIBILITIES
The present study experiments have been performed to enhance the enzymatic digestibility
of corn stover using various pretreatment techniques.
The following conclusions could be drawn from the experiments:
•
•
•
•
•
•
•
•
•
•
•
Nearly theoretical enzymatic conversion (95.7%) was achieved following two-step
chemical pretreatment (1% NaOH combined 1% H2SO4).
The main drawback of this method was, that a big portion of the hemicellulose and also
the cellulose have lost because of the purification’s problem of the liquid fraction.
Using only dilute acids for the pretreatment the enzymatic conversions were quite low,
but these pretreatment appears to have advantage that mainly the hemicellulose fraction
was solubilised.
The steam pretreatment study justified the hypotheses, that steam explosion is an
efficient method to increase the enzymatic accessibility of the water-insoluble,
cellulose-rich component in corn stover.
After steam pretreatment, the enzymatic conversion from cellulose to glucose increased
nearly four times, from 25% to above 80% following the “best pretreatment” (200°C, 5
min, impregnation with 2% sulphuric acid).
The highest overall yield of sugars was 56.1 g from 100 g untreated material DM,
corresponding to 73 % of the theoretical, which was achieved following steam
pretreatment at 190ºC, for 5 min with 2% sulphuric acid.
The achieved ethanol yields following steam pretreatment were about 80%, showing
that baker’s yeast could adapt to the pre-treated liquor and ferment the glucose to
ethanol without problems.
Similar results were observed after wet oxidation, which was also an efficient process
for breaking the tight association between lignin and polysaccharides in corn stover.
The best conditions for obtaining high convertible cellulose was 195°C for 15 minutes
with added Na2CO3 giving about 85% conversion to glucose with 25 FPU/g cellulase
loading 50°C.
The recovery of cellulose was around 85% following steam pretreatment, and a little
higher around 90% following wet oxidation. The recovery of hemicellulose was in both
cases around 60%. The harsher the conditions of pretreatment were, the lower the
recovery were.
Hydrolysis experiments at 40°C following wet oxidation resulted, that the lower
temperature needed longer reaction time, but after 72 hours even better results of
86
•
•
•
•
•
•
cellulose conversion was obtained compared to the one-day conversions at 50°C after
24 hours.
Wet oxidation at acid conditions was the most efficient pretreatment process for
production of fermentable sugars from corn stover in the high gravity SSF set-up.
The highest ethanol yield of 85% of the theoretical was achieved, using 12% DM of
acidic pre-treated corn stover with an ethanol concentration of 50 g/l (6 (v/v)%
ethanol), which exceeds the technical and economical limit of the industrial-scale
alcohol distillation.
An enzymatic pre-hydrolysis at 50oC (10 FPU/g DM) was needed to perform the
following SSF at high fibre consistency. However 15% was the upper limit due to
mixing problems.
Increased enzyme loadings accelerated the enzyme reaction resulting in higher ethanol
yield during the same fermentation period. An enzyme loading of 30 FPU/g DM was
sufficient to complete SSF in 5 days, however lower enzyme loading (10 FPU/g DM)
required additional 2 days SSF.
Potential fermentation inhibitors (phenols and carboxylic acids) were produced at subinhibitory levels. After 120 hours the phenol aldehydes were assimilated while other
phenols were unaffected.
The fermentation was monitored by weighing the fermentation bottles at regular
intervals and the amount of produced ethanol was calculated from the CO2 resulted
weight loss. Comparison of the calculated final ethanol yield with HPLC
measurements, showed no significant differences, thus this monitoring system could be
applied to show the ethanol production by SSF in an easy way.
Corn stover is a favourable substrate for the ethanol production in Hungary, because of its
huge amount and its high sugar content. The polysaccharides in this herbaceous material is
highly convertible to monomeric sugars following pretreatment at high temperature, such
as wet oxidation or steam explosion. These pretreatment processes used a minimal amount
of chemicals (sulphuric acid or sodium-carbonate), but the achieved enzymatic cellulose
conversions were generally four times higher compared to the untreated corn stover.
Thanks to the sub-inhibitory levels of the potential fermentation inhibitors following
pretreatments, the hydrolysis released glucose was convertible to ethanol using baker’s
yeast.
87
6. REFERENCES
AHRING B.K., JENSEN K., NIELSEN P., BJERRE A.B., SCHMIDT A.S. 1996. Pretreatment of
wheat straw and conversion of xylose and xylan to ethanol by thermophilic anaerobic
bacteria, Bioresour. Technol. 58:107-113.
AHRING B.K., LICHT D., SCHMIDT A.S., SOMMER P., THOMSEN A.B. 1999. Production of
ethanol from wet oxidised wheat straw by Thermoanaerobacter mathranii.
Bioresour. Technol. 68:3-9.
ALFANI F., GALLIFOUCO A., SAPOROSI A., SPERA A., CANTARELLE M. 2000. Comparison
of SHF and SSF processes for the bioconversion of steam-exploded wheat straw, J.
of Industrial Microbiology and Biotechnology. 25:184-192.
ALKASRAWI M., ERIKSSON T., BÖRJESSON J., WINGREN A., GALBE M., TJERNELD F.,
ZACCHI G. 2003. The effect of Tween-20 on simultaneous saccharification and
fermentation of softwood to ethanol. Enzyme Microb. Technol. 33:71-78.
ALVO P., BELKACEMI K. 1997. Enzymatic Saccharification of milled timothy (Phleum
preatense L.) and alfalafa (Medicago sativa L.), Bioresour. Technol. 61:185-198.
BALLESTEROS I., OLIVA J.M., NEGRO M.J., MANZANARES P., BALLESTEROS M. 2002.
Enzymic hydrolysis of steam exploded herbaceous agricultural waste (Brassica
carinata) at different particule sizes, Process Biochemistry 38:187-192
BEN-GHEDALIA D., MIRON J. 1981. The effect of combined chemical and enzyme
treatment on the saccharification and in vitro digestion rate of wheat straw.
Biotechnol. Bioeng. 23:823-831.
BERGHEM L.E.R., PETTERSON L.G. 1974. The mechanism of enzymatic cellulose
degradation. Isolation and some properties of a ß-glucosidase from Trichoderma
viride. Eur. J. Biochem. 46:295-305.
BJERRE A.B., OLESEN A.B., FERNQVIST T., PLÖGER A., SKAMMELSEN A.S. 1996.
Pretreatment of wheat straw using combined wet oxidation and alkaline hydrolysis
resulting in convertible cellulose and hemicellulose, Biotechnol. Bioeng. 49:568-577.
BJERRE A.B., SØRENSEN E. 1992. Thermal decomposition of dilute formic acid solutions.
Ind. Eng. Chem. Res. 31:1574-1577.
BOLLÓK M. 1999. Studies on ethanol production on lignocellulosics: SSF and cellulase
production. PhD-Thesis. Budapest University of Technology and Economics,
Department of Agricultural Chemical Technology.
88
BRIGHAM J.S., ADNEY W.S., HIMMEL M.E. 1996. in Handbook on bioethanol production
and utilzation, Wyman C., ed., Taylor and Francis, Washington, DC, pp. 119-141.
CASTANON M, WILKE C.R. 1981. Effects of the surfactant Tween 80 on enzymatic
hydrolysis of newspaper. Biotechnol. Bioeng. 23:1365-72.
CHANG M. M., CHOU T. Y. C., TSAO G. T. 1981. Structure, pretreatment and hydrolysis of
Cellulose. Biochem. Eng. 20:15-42.
CHUM H.L., JHONSON D.K., BLACK S.K., OVEREND R.P. 1990. Pretreatment-catalyst
effects of the combined severity parameter, Appl. Biochem. Biotechnol. 24/25:1-14.
CHUM H.L., JOHNSOON D.K., BLACK S. 1988. Organosolv pretreatment for enzymatic
hydrolysis of poplars: I. enzyme hydrolysis of cellulosic residues. Biotechnol.
Bioeng. 31:643-649.
CLAASSEN P.A.M., SIJTSMA L., STAMS A.J.M., DE VRIES S.S., WEUSTHUIS R.A. 1999.
Utilisation of biomass for the supply of energy carriers. Appl. Microbiol. Biotechnol.
52(6):741-755.
CLARK T.A., MACKIE K.L. 1987. Steam explosion of the softwood Pinus radiata with
sulphur dioxide addition. I. process optimisation J. Wood Chem. Technol. 7(3): 73403
CLAUSEN E.C., GADDY J.L. 1983. Economic analysis of bioprocess to produce ethanol
from corn stover. Biotechnol. Bioeng. Symp. No. 13:495-510.
CONVERSE A.O:, MATSUNO R., TANAKA M., TANIGUCHI M. 1988. A model for enzyme
adsorption and hydrolysis of mycrocrystalline cellulose with slow deactivation of the
adsorbed enzyme. Biotechnol. Bioeng. 32:38-45.
COUGHLAN M.P. 1992. Enzymatic hydrolysis of cellulose: an overview. Bioresour.
Technol. 39:107-115.
COUGHLAN M.P., LJUNGDAHL L.G. 1988. Comparative biochemistry of fungal and
bacterial cellulolytic enzyme system. In: Biochemistry and genetics of cellulose
degradation. Eds. Aubert J.P., Beguin P., Millet J., pp.11-30.
DALE B.E., MOREIRA M.J. 1982. A freeze-explosion technique for increasing cellulose
hydrolysis. Biotechnol. Bioeng. Symp. 12:31-43.
DELGENES J.P., MOLETTA R., NAVARRO J.M. 1996. Effects of lignocellulose degradation
products on ethanol fermentation of glucose and xylose by Saccharomyces
cerevisiae, Zymomonas mobilis, Pichia stipitis, and Candida shehatae, Enzyme
Microb. Technol. 19:220-225.
89
DU PREEZ J.C., BOSCH M., PRIOR B.A. 1986. The fermentation of hexose and pentose
sugars by Candida shehatae and Pichia stipitis. Appl. Microbiol. Biotechnol.
23:228-233.
DUFF S.J.B., MURRAY W.D. 1996. Bioconversion of forest products industry waste
cellulossics to fuel ethanol: a review. Bioresour. Technol. 55:1-33.
EKLUND R.M., GALBE M., ZACCHI G. 1995. Optimization of temperature and enzyme
concentration in the enzymatic saccharification of steam-pretreatead willow. Enzyme
Microb. Technol., 12:225-228.
ELSHAFEI A.M., VEGA J.L., KLASSON K.T., CLAUSEN E.C., GADDY J.L. 1991. The
saccharification of corn stover by cellulse from Penicilium funiculosum. Bioresour.
Technol. 35:73-80.
EPA, 2003. http://yosemite.epa.gov/oar/globalwarming.nsf/content/climate.html
E RIKSSON T., B ÖRJESSON J., T JERNELD F. 2002. Mechanism of surfactant effect in
enzymatic hydrolysis of lignocellulose. Enzyme Microb. Technol. 31:353-364
ESTEGHLALIAN A., HASHIMOTO A.G., FENSKE J.J., PENNER M.H. 1997. Modeling and
optimization of the dilute sulphuric-acid pretreatment of corn stover, poplar and
switchgrass. Bioresour. Technol. 59:129-136
FAN L.T., LEE Y., GHARPURAY M.M. 1982. The nature of lignocellulosic and their
pretreatments for enzymatic hydrolysis. Biochem. Eng. 23:157-187.
FENSKE J.J., HASHIMOTO A., PENNER M.H. 1998. Relative Fermentability of
Lignocellulosic dilute-acid prehydrolysates application of a Pichia stipitis-based
toxicity assay. Appl. Biochem. Biotechnol. 73:145-157.
GARROTE G., DOMINGUEZ H., PARAJÓ J.C. 2002. Interpretation of deacetylation and
hemicellulose hydrolysis during hydrothermal treatments on the basis of the severity
factor Process Biochemistry 37:1067-1073.
GAUSS W.F., SUZUKI S., TAKAGI M. 1976. Manufacture of alcohol from cellulosic
materials using plural ferments. U.S. Pat. No.3,990,944, November 9, 1976.
Ghose T.K. 1987. Measurement of cellulase activities. Pure Appl. Chem. 59:257-268.
GHOSH P., SINGH A. 1993. Physiochemical and biological treatments for enzymatic/
microbial conversion of lignocellulosic biomass. Appl.. Microbiol. 39:295-333.
GLASSER W.G., WRIGHT R.S. 1998. Steam-assisted biomass fractionation II. Fractionation
behavior of various biomass resources, Biomass and Bioenergy 14:219-235.
90
GOERING H.K., VAN SOEST P.J. 1970. Forage fiber analyses (apparatus, reagents,
procedures, and some applications). - In: Agricultural handbook No. 379, pp. 1-20.
Agricultural Research Service - United States Department of Agriculture. USDA,
Washington, DC.
GREGG D.J., SADDLER J.N. 1996. Factors affecting cellulose hydrolysis and the potential of
enzyme recycle to enhance the efficiency of an integrated wood to ethanol process.
Bioethanol. Bioeng. 51:375-383.
GROUS W.R., CONVERSE A.O., GRETHLEIN H.E. 1986. Effect of steam explosion
pretreatment on pore size and enzymatic hydrolysis of poplar. Enzyme Microb.
Technol., 8:274-280.
HÄGGLUND E. 1951. The wood components and their chemical properties. – In: Chemistry
of Wood. Academic Press: New York, pp. 375-389.
HAHN-HÄGERDAL B. 1996. Ethanolic fermentation of lignocellulose hydrolysates, a
minireview. Appl. Biochem. Biotechnol. 57/58:195-199.
HAHN-HÄGERDAL B., HALLBORN B., JEPPSSON J., OLSSON L., SKOOG K., WALFRIDSSON M.
1993. Pentose fermentation to alcohol. In: Bioconversion of forest and agricultural
plant residues, ed. Saddler J.N., CAB International, Wallingford UK, pp. 231-290.
HAHN-HÄGERDAL B., LINDÉN T., SENAC T., SKOONG K. 1991. Ethanolic fermentation of
pentoses in lignocellulosic hydrolysates. Appl. Biochem Biotechnol 28/29:131-141.
HAN Y. W. 1978. Microbial utilization of straw (a review). Appl. Microbiol. Biotechnol.
23:119-153.
HATAKKA A.I. 1983. Pretreatment of wheat straw by white-rot fungi for enzymatic
saccharification of cellulose. Appl. Microbiol. Biotechnol. 18:350-357.
HEITZ H.E., CAPEK-MÉNARD P.G., KOEBERLE J., GAGNÉ E., CHORNET R.P., OVEREND J.D.,
TAYLOR J.D., YU E. 1991. Fractionation of Populus tremuloides at the plant scale:
optimization of steam pretreatment conditions using the STAKE II. Technology.
Bioresour. Technol., 35:23-32.
HELLE S.S., DUFF S.J.B., COOPER D.G. 1993. Effect of surfactants on cellulose hydrolysis.
Biotechnol. Bioeng. 42:611-617.
HIMMEL M.E., ADNEY W.S., BAKER J.O., ELANDER R., MCMILLAN J.D., NIEVES R.A.,
SHEEHAN J.J., THOMAS S.R., VINZANT T.B. AND ZHANG M. 1997. Advanced
Bioethanol Production Technologies. A perspective. – In: Fuels and chemicals form
biomass. BB.D. Saha, J. Woodward (ed.) ACS Symp. Ser 666, American Chemical
Society, Washington, DC, Chapter 1:1-45.
91
HINMAN N.D., SCHELL D.J., RILEY C.J., BERGERON P.W., WALTER P.J. 1992. Preliminary
estimate of the cost of ethanol production for SSF technology. Appl. Biochem.
Biotechnol. 34/35:639-649.
HOLTZAPPLE M.T., JUN J.H., ASHOK G., PATIBANDLA S.L., DALE B.E. 1991. The ammonia
freeze explosion (AFEX) process: a practical lignocellulose pretreatment. Appl.
Biochem. Biotechnol. 28/29:59-74.
HOLTZAPPLE M.T., LUNDEEN J.E., STRUGIS R. 1992. Pretreatment of lignocellulosic
municipal solid waste by ammonia fiber explosion (AFEX). Appl. Biochem.
Biotechnol. 34/35:5-21.
HOLTZAPPLE M.T., RIPLEY E.P., NIKOLAOU M. 1994. Saccharification, fermentation, and
protein recovery from low-temperature AFEX-treated coastal Bermudagrass.
Biotechnol. Bioeng. 44:1122-1131.
HON D.N.S., SHIRASHI N. 2001. Wood and cellulosic chemistry. 2nd ed. New York: Marcel
Dekker Inc. 914 p.
http://www.afdc.doe.gov/pdfs/4898.pdf
http://www.agecon.ucdavis.edu/outreach/crop/cost-studies/2000CornYolo.pdf
http://www.platts.com/features/biofuels/europe.shtml
http://europa.eu.int/comm/energy_transport/atlas/htmlu/lbtech.html
http://www11.myflorida.com/Construction/fuel&bit/fuel&bit.htm
IYER P.V., WU Z.W., KIM S.B., LEE Y.Y. 1996. Ammonia recycled percolation process for
pretreatrment of herbaceous biomass. Appl. Biochem. Biotechnol. 57/58:121-132.
KAAR W.E., GUTIERREZ C.V., KINOSHITA C.M. 1998. Steam explosion of sugarcane
bagasse for conversion to ethanol. Biomass and Bioenergy 14:277-287.
KAAR W.E., HOLTZAPPLE M. 1998. Benefits from Tween during enzymic hydrolysis of
corn stover. Biotechnol. Bioeng. 59:419-27.
KAAR W.E., HOLTZAPPLE M.T. 2000. Using lime pretreatment to facilitate the enzymic
hydrolysis of corn stover. Biomass Bioenergy 18:189-199.
KAAR W.E.; COOL L.G.; MERRIMAN M.M., BRINK D.L. 1991. The Complete Analysis of
Wood Polysaccharides Using HPLC. J. Wood. Chem. Technol. 11:447-463.
KIM K.H., HONG J. 2000. Dynamic extraction of spearmint oil components by using
supercritical CO2. Sep. Sci. Technol. 35:315-322.
KIM K.H., HONG J. 2001. Supercritical CO2 pretreatment of lignocellulose enhances
enzymatic cellulose hydrolysis. Bioresour. Technol. 77:139-144.
92
KIM M.H., LEE S.B., RYU D.D.Y. 1982. Surface deactivation of cellulase and its
prevention. Enzyme Microb. Technol. 4:99-103.
KLINKE H.B, OLSSON L, THOMSEN A.B, AHRING B.K. 2003. Potential inhibitors from wet
oxidation of wheat straw and their effect on ethanol production of Saccharomyces
cerevisiae: Wet oxidation and fermentation by yeast. Biotechnol. Bioeng. 81:738747.
KLINKE H.B, THOMSEN A.B, AHRING B.K. 2001. Potential inhibitors from wet oxidation of
wheat straw and their effect on growth and ethanol production by
Thermoanaerobacter mathranii. Appl. Microbiol Biotechnol. 57:631-638.
KLINKE H.B., AHRING B.K., SCHMIDT A.S., THOMSEN A.B. 2002. Characterization of
degradation products from alkaline wet oxidation of wheat straw. Bioresour.
Technol. 82:15-26.
KNAPP K.T., STUMO F.D., TEJADA S.B. 1998. J. Air Waste Manage Assoc. 48: 646-653.
KOULLAS D.P., CHRISTAKOPOULOS P., KEKOS D., MACRIS B.J., KOUKIOS E.G. 1992.
Correlating the effect of pretreatment on the enzymatic hydrolysis of straw.
Biotechnol. Bioeng. 39:113-116.
KURAKAKE M., OOSHIMA H., KATO J., HARANO Y. 1994. Pretreatment of bagasse by
nonionic surfactant for the enzymatic hydrolysis. Bioresource Technol. 49:247-251.
LAMPERT R.O. MOORE-BULLS M.R., BARRIER J.W. 1990. An evaluation of two acid
hydrolysis processes for the conversion of cellulosic feedstock to ethanol and other
chemicals. Appl. Biochem. Biotechnol. 24/25:773-778.
LAPLACE J.M., DELGENES J.P., MOLETTA R., NAVARRO J.M., 1993. Ethanol production
from glucose and xylose by separated and co-culture processes using high cell
density systems. Process. Biochem. 28:519-525.
LARSSON S., PALMQUIST E., HAHN-HÄGERDAL B., TENBORG C., STENBERG K., ZACCHI G.,
NILVEBRANT N.O. 1999. The generation of fermentation inhibitors during dilute acid
hydrolysis of softwood. Enzyme Microb. Technol. 24:151-158.
LAWFORD H.G., ROUSSEAU J.D., 1992. Fuel ethanol from corn residue prehydrolysate by a
patented ethanologenic Escherichia coli. Biotechnol. Lett. 14:421-426.
LIN K.W., LADISH M.R., SCHAEFER D.M., NOLLER C.H., LECHTENBERG V., TSAO G.T.
1981. Review on effect of pretreatment on digestibility of cellulosic materials.
AIChE Symp. 207(77):102-106.
LYND L. 1996. Overview and evaluation of fuel ethanol from cellulosic biomass:
technology, economics, and policy. Annu. Rev. Energy. Environ. 21:403-465.
93
MABLY R. 2003. The Arab embargo - from oil crisis to
http://www.metimes.com/issue98-42/opin/the_arab_embargo.htm
OPEC
crisis,
MACDONALD D.G., BAKHSHI N.N, MATHEWS J.F., ROYCHOWDHURY A., BAJPAI P., MOOYOUNG M. 1983. Alkali treatment of corn stover to improve sugar production by
enzymatic hydrolysis. Biotechnol. Bioeng. 25:2067-2076.
MANDELS M., ANDREOTTI R., ROCHE C. 1976. Measurement of saccharifying cellulases.
Biotechnol. Bioeng. Symp. 6, 21-33.
MCGINNIS G.D., WILSON W.W., MULLEN C.E. 1983a. Biomass pretreatment with water
and high-pressure oxygen. The wet oxidation process. Ind. Eng. Chem. Prod. Res.
Dev. 22:352-357.
MCGINNIS G.D., WILSON W.W., PRINCE S.E., CHEN C.C. 1983b. Conversion of biomass
into chemicals with high-temperature wet oxidation. Ind. Eng. Chem. Prod. Res.
Dev. 22:633-636.
MCMILLAN J.D. 1994. Pretreatment of lignocellulosic biomass. In: Himmel, M.E., Baker
J.Q., Overend R.P. (Eds.) Enzymatic Conversion of biomass for fuels production.
American Chemical Society, Washington, DC pp. 292-324.
MES-HARTREE M., DALE B.E., CRAIG W.K. 1988. Comparison of steam and ammonia
pretreatment for enzymatic hydrolysis of cellulose. Appl. Microbiol. Biotechnol.
29:462-468.
MES-HARTREE M., HOGAN C.M., SADDLER J.N. 1987. Recycle of enzymes and substrate
following enzymatic hydrolysis of steam pretreated aspenwood. Biotechnol. Bioeng.
30:558-564.
MIELENZ J.R. 2001. Ethanol production from biomass: technology and commercialization
status. Curr Op Microbiol. 4:324-329.
MILLER G.L. 1959. Use of dinitrosalicylic acid reagent for determination of reducing
sugar. Anal. Chem. 31:420-428.
MILLET M.A., BAKER A.J., SCATTER L.D. 1976. Physical and chemical pretreatment for
enhancing cellulose saccharification. Biotechnol. Bioeng. Symposium 6:125-153
MOHAGHEGHI A., TUCKER M., GROHMAN K., WYMAN C.E. 1992. High solid simultaneous
saccharification and fermentation of pretreated wheat straw to ethanol. Appl.
Biochem. Biotechnol. 33:67-81.
NYSTROM J., GREENWALD C.G., HAGLER R.W., STAHR J.J. 1985. Technical and economic
feasiblity of enzyme hydrolysis for ethanol production from wood. NYERDA Report
85-89.
94
OLSSON L., HAHN-HÄGERDAL B. 1993. Fermentative performance of bacteria and yeasts in
lignocellulose hydrolysates. Proc. Biochem. 28: 249-257.
OOSHIMA H., SAKATA M., HARANO Y. 1986. Enhancement of enzymatic hydrolysis of
cellulose by surfactant. Biotechnol. Bioeng. 28:1727-1734.
OVERENED R.P., CHORNET E. 1987. Fractionation of lignocellulosics by steam-aqueous
pretreatments, Phil. Trans. R. Soc. London, Series A, 321, 523-536.
PADUKONE N. 1996. Advanced process options for bioethanol production. - In: Handbook
on bioethanol: Production and utilization. C.E. Wyman (ed.) Taylor and Francis,
Washington, DC, Chapter 14:315-327.
PALMQVIST E., HAHN-HÄGERDAL B., GALBE M., ZACCHI G. 1999. The effect of watersoluble inhibitors from steam-pretreated willow on enzymatic hydrolysis and ethanol
fermentation, Enzyme and Microbial Technology, 19(6):470-476.
PALMQVIST E., HAHN-HÄGERDAL B., SZENGYEL ZS., ZACCHI G., RÉCZEY K. 1997.
Simultaneous detoxification and enzyme production of hemicellulose hydrolysates
obtained after steam pretreatment. Enzyme Microb. Technol. 20:286-293.
PARISI F. 1989. Advances in lignocellulosaic hydrolysis and in the utilisation of the
hydrolysates. Adv. Biochem. Eng. 38:53-87.
PARK J.W., TAKAHATA Y., KAJIUCHI T., AKEHATA T. 1992. Effect of non-ionic surfactant
on enzymatic hydrolysis of used newspaper. Biotechnol. Bioeng. 39:117-120.
PHILIPIDIS G.P. 1996. Cellulose bioconversion technology. In: Handbook on bioethanol:
Production and utilization. Ed. Wyman C.E., Taylor & Francis, Bristol, PA, pp 253285.
PHILIPIDIS G.P., SMITH T.K., WYMAN C.E. 1993. Study of the enzymatic hydrolysis of
cellulose for production of fuel ethanol by the simultaneous saccharification and
fermentation process. Biotechnol. Bioeng. 41:846-853.
RITTER D.C., CAMPBELL A.G. 1991. Supercritical carbone dioxide extraction of southern
pine and ponderosa pine. Wood Fiber Sci. 23:98-113.
SADDLER J.N., GREGG D.J. 1998. Ethanol production from forest product wastes. In: Forest
products biotechnology. Eds. Bruce A., Palfreyman J.W., Taylor & Francis, London,
UK, pp. 183-207.
SADDLER J.N., RAMOS L.P., BREUIL C. 1993. Steam pretreatment of lignocellulosic
residues. In: Bioconversion of Forest and Agricultural Plant Residues. C.A.B.
International, Wallingford, UK, pp. 73-91.
95
SCHELL D.J., TORGET R., POWER A., WALTER P.J., GRIHMANN K., HINMAN N.D. 1991. A
technical and economical analysis of acid-catalyzed steam explosion and dilute
sulfuric acid pretreatments using wheat straw or aspen wood chips. Appl. Biochem.
Biotechnol. 28/29:87-97.
SCHMIDT A.S., THOMSEN A.B. 1998. Optimization of wet oxidation pretreatment of wheat
straw. Bioresour. Technol. 64:139-151.
SCHNEIDER S.H. 1989. The Changing Climate, Scientific American, n 3, September/1989.
SCHURZ J. 1978. In Ghose T.K. (Ed.) Bioconversion of cellulosic substances into energy,
Chemicals and Microbial Protein Symposium Proceedings, IIT, New Delhi, pp.3757.
SEN D.C. 1989. Ethanol fermentation, In: Kitani O, Hall CW, eds Biomass Handbook,
New York, Gordon and Breach Science Publishers, p. 254-270.
SHEVCHENKO S.M., CHANG K., ROBINSON J., SADDLER J.N. 2000. Optimisation of
monosaccharide recovery by post-hydrolysis of the watersoluble hemicellulose
component after steam-explosion of sofwood chips, Bioreseor. Technol. 72:207-211.
Post-hydrolysis
SOKHANSANJ S., TURHOLLOW A., CUSHMAN J., CUNDI J. 2002. Engineering aspects of
collecting corn stover for bioenergy, Biomass and Bioenergy 23:347 – 355.
SÖDERSTRÖM J. 2004. Optimisation of two-step steam pretreatment of softwood for
bioethanol production. PhD Thesis, Department of Chemical Engineering, Lund
University, Sweden, Chapter 7: Economic Evaluation pp:57-67.
SPINDLER D., WYMAN C.E., GROHMAN K., MOHAGHEGHI A. 1989. Simultaneous
saccharification and fermentation of pretreated wheat straw to ethanol with selected
yeast strains and b-glucosidase supplementation. Appl. Biochem. Biotechnol.
20/21:529-540.
SPINDLER D.D., WYMAN C.E., GROHMAN K. 1990. Evaluation of pretreated herbaceous
crops for the simultaneous saccharification and fermentation process. Appl.
Biochem. Biotechnol. 24/25:275-286.
STATISTICAL ANNUAL REVIEWS of the Hungarian Agriculture 1996-1999. Hungarian
Central Statistical Office, Budapest
STATISTICAL ANNUAL REVIEWS of the Hungarian Agriculture, 1999-2002. Hungarian
Central Statistical Office, Budapest
STENBERG K., BOLLÓK M., RÉCZEY K., GALBE M., ZACCHI G. 2000a. Effect of substrate
and cellulase concentration on simultaneous saccharification and fermentation of
steam-pretreated softwood for ethanol production. Biotechnol. Bioeng. 68:204-210.
96
STENBERG K., GALBE M., ZACCHI G. 2000B. The influence of lactic acid formation on the
simultaneous saccharification and fermentation (SSF) of softwood to ethanol.
Enzyme Microb. Technol. 26:71-79.
STENBERG K., TENBORG C., GALBE M., ZACCHI G. 1998. Optimisation of steam
pretreatment of SO2-impregnated mix softwoods for ethanol production , Appl.
Biochem. Biotechnol. 71:299-308.
SUN Y., CHENG J. 2002. Hydrolysis of lignocellulosic materials for ethanol production,
Bioresour. Technol. 83:1-11
TAKAGI M. 1984. Inhibition of cellulase by fermentation products. Biotechnol. Bioeng.
26:1506-1507.
TAKAGI M., ABE S., SUZUKI S., EMERT G.H., YATA N. 1977. A method for production of
alcohol directly from cellulose using cellulase and yeast. Proc. Biocon. Symp. ITT.
Delhi, India, 551-571.
TENBORG C., STENBERG K., GALBE M., ZACCHI G., LARSSON S., PALMQUIST E., HAHNHÄGERDAL B. 1998. Comparison of SO2 and H2SO4 impregnation of softwood prior
to steam pretreatment on ethanol production, Appl. Biochem. Biotechnol. 70-72:315.
THAMMASOUK K., TANDIJO D., PENNER M.H., 1997. Influence of extractives on the
analysis of herbaceous biomass. J. Agric. Food Chem. 45:437-443.
U.S. MINISTRY OF ENVIRONMENTAL. 2003. U.S Greenhouse Gas Emissions Inventory
http://www.sierraclub.org/globalwarming
VARGA E., SCHMIDT A.S., RÉCZEY K., THOMSEN A.B. 2003. Pretreatment of corn stover
using wet oxidation to enhance enzymatic digestibility. Appl. Biochem. Biotechnol.
104:37-49.
VERMEERSCH.G. 2002. Biofuel: a topical issue, Eur. J. Lipid Sci. Technol. 104:360-375.
VLASENKO E.YU., DING H., LABAVITCH J.M., SHOEMAKER S.P. 1997. Enzymatic
hydrolysis of pretreated rice straw. Bioresour. Technol. 59:109-119.
VON SIVERS M., ZACCHI G. 1996. Ethanol from lignocellulosics: a review of the economy.
Bioresour. Technol. 56, 131-140.
WHEALS A.E., BASSO L.C., ALVES D.M.G., AMORIM H.V. 1999. Fuel ethanol after 25
years. Trends Biotechnol. 17(12):482-487.
WILKE C.R., YANG R.D., SCIAMANNA A.F., FREITAS R.P. 1981. Raw materials evaluation
and process development studies for conversion of biomass to sugars and ethanol.
Biotechnol. Bioeng. 23:163-183.
97
WMO, 2002. The World Meteorological Organization (WMO) World Climate News,
Global Climate System Reviews, 2002 www.wmo.org
WORELY J.W., VAUGHAN D.H., CUNDIFT J.S. 1992. Energy analysis of bioethanol from
sweet shorgum. Bioresour. Technol. 10:263-273.
WRIGHT J.D. 1998. Ethanol from lignocellulose: an overview. Energy Prog. 8:71-98.
WRIGHT J.D., WYMAN C.E., GROHMANN K. 1988. Simultaneous saccharification and
fermentation of lignocellulose: process evaluation. Appl. Biochem. Biotechnol.
18:75-105.
WYMAN C.E. 1994. Ethanol from lignocellulosic biomass: technology, economics and
opportunities. Bioresour. Technol. 50:3-16.
WYMAN C.E., SPINDLER D.D., GROHMANN K. 1992. Simultaneous saccharification and
fermentation of several lignocellulosic feedstocks to fuel ethanol. Biomass and
Bioenergy 3/5:301-307.
WYMAN C.E. 1999. Biomass Ethnaol: Technical progress, opportunities and commercial
challenges. Annu. Rev. Energy Environ. 24:189-226.
98