UPTEC X 08 047
Examensarbete 30 hp November 2008
Thermochemical pretreatment and enzymatic saccharification
of lignocellulose for biofuel production
Jesper Svedberg
Molecular Biotechnology Programme
Uppsala University School of Engineering
UPTEC X 08 047 Date of issue 2008-11
Author
Jesper Svedberg
Title (English)
Thermochemical pretreatment and enzymatic
saccharification of lignocellulose for biofuel production
Title (Swedish) Abstract
The efficiency of five different thermochemical pretreatment methods has been evaluated for efficiency and suitability for laboratory scale use with the purpose of facilitating enzymatic saccharification of aspen sawdust and oat straw. The ground substrates were first pretreated, while process conditions such as temperature, time, catalyst concentration and substrate loading were varied. The pretreated substrates were then saccharified using a commercially available enzyme mixture and finally the composition and concentration of solubilized sugars were determined using a high performance anion exchange chromatography system with pulsed amperometric detection (HPAE-PAD).
The most efficient pretreatment method appears to be dilute acid and least efficient hot water, but problems achieving stable measurements in the analysis step limits the quality of the results and further studies are needed.
Keywords
Lignocellulose, ethanol, cellulases, thermochemical pretreatment Supervisors
Jerry Ståhlberg and Mats Sandgren
Dept. of Molecular Biology, Swedish University of Agricultural Sciences Scientific reviewer
Sherry Mowbray
Swedish University of Agricultural Sciences
Project name Sponsors
Language
English
Security
ISSN 1401-2138 Classification
Supplementary bibliographical information
Pages
49
Biology Education Centre Biomedical Center Husargatan 3 Uppsala
Box 592 S-75124 Uppsala Tel +46 (0)18 4710000 Fax +46 (0)18 555217
Thermochemical pretreatment and enzymatic saccharification of lignocellulose for biofuel production
Jesper Svedberg
Populärvetenskaplig sammanfattning
Den etanol som används som biobränsle idag tillverkas till stor del av jordbruksgrödor som innehåller stora mängder socker eller stärkelse. Ett råmaterial som dock finns tillgängligt i betydligt större mängder är den cellulosa som är en del av växters cellväggar och som kan brytas ner till jäsbar glukos. Detta kan göras med hjälp av särskilda enzymer som finns naturligt i vissa bakterier och svampar, men för att då uppnå en effektiv nedbrytning krävs först att man luckrar upp strukturen på
cellväggarna med hjälp av höga temperaturer och korrosiva kemikalier. Denna process kallas för termokemisk förbehandling.
Syftet med detta projekt var att undersöka hur effektiva fem olika
förbehandlingsmetoder (förbehandling med utspädd syra, kalk, alkalisk väteperoxid, hett vatten samt ångexplosion) var på att underlätta enzymatisk nedbrytning av havrehalm och aspspån. Halmen och spånen maldes och förbehandlades med de olika metoderna, varefter de behandlades med enzym i upp till 72 timmar. Slutligen
bestämdes mängden och sammansättningen av det socker som lösts ut med hjälp av ett HPLC-system.
Förbehandling med utspädd syra verkar ge de bästa sockerutbytena och hett vatten de sämsta, men problem med att få stabila mätvärden vid analysen av sockermängderna innebär tyvärr att ytterligare studier för att bestämma mer tillförlitliga värden krävs.
Examensarbete 30hp
Civilingejörsprogrammet Molekylär bioteknik
Uppsala universitet November 2008
Contents
CONTENTS ... 1
INTRODUCTION ... 4
L IGNOCELLULOSE ... 4
T HE ETHANOL - FROM - LIGNOCELLULOSE PROCESS ... 5
R ESIDUAL MATERIAL ... 7
T HERMOCHEMICAL PRETREATMENT ... 7
Dilute acid ... 8
Lime ... 8
Alkaline Peroxide ... 9
Hot water ... 9
Steam explosion ...10
Ammonia Fiber Explosion (AFEX) ...10
Organosolv...10
Biological pretreatment ...11
C ARBOHYDRATE ANALYSIS ... 11
Colorimetry ...11
HPLC/HPAE-PAD ...11
O BJECTIVES ... 12
MATERIALS AND METHODS ... 12
L IGNOCELLULOSE MATERIALS ... 12
P RETREATMENT ... 12
Dilute acid ...12
Lime ...13
Hot water ...13
Alkaline peroxide ...13
Steam explosion ...14
E NZYMATIC S ACCHARIFICATION ... 14
F ILTRATION ... 15
A NALYSIS ... 15
DNS ...15
PAHBAH ...16
GOD/POD...16
HPLC/HPAE-PAD ...16
RESULTS AND DISCUSSION ... 17
T HERMOCHEMICAL PRETREATMENT ... 17
Alkaline hydrogen peroxide pretreatment ...17
Lime pretreatment...20
Hot water pretreatment...21
C RITICAL ANALYSIS OF THE RESULTS ... 23
K INETICS OF THE ENZYMATIC SACCHARIFICATION ... 24
S UGAR CONTENT OF THE A CCELERASE ENZYME SOLUTION ... 25
T ABULATED SACCHARIFICATION RATES AND ETHANOL POTENTIALS ... 25
Colorimetric sugar measurements ...26
L ESSONS LEARNED , IMPROVEMENTS AND REMAINING PROBLEMS ... 27
F UTURE PERSPECTIVES ... 27
REFERENCES ... 29
ACKNOWLEDGEMENTS ... 31
APPENDIX A ... 32
APPENDIX B ... 39
APPENDIX C ... 41
Introduction
Over the last decade ethanol has increased dramatically in importance as a fuel for motorized vehicles. Today it is one of the most widely used biofuels and because of present concerns over climate change and the diminishing reserves of fossil fuel, there are widespread efforts to increase production even further.
The ethanol produced today is largely made from crops containing high levels of sugar and starch, such as sugarcane, maize and wheat. Production of ethanol from such crops is a mature process, which can be established relatively easily on an industrial scale using commercially available technology. Despite this, there are reasons to look for alternative raw materials. Concerns have, for instance, been raised over the fact that this type of ethanol production competes with the food and feed industries for the same raw materials, and so an increased ethanol production may in turn lead to an increase in food prices and potentially starvation. Also, when producing ethanol from sugar and starch, only a part of the energy stored in the plant through photosynthesis is
transferred into the fuel, making the process inefficient. However, if the parts of the plants that consist of lignocellulose are used as well, it is actually possible to take advantage of a much larger part of that stored energy.
Lignocellulose
Plant cell walls are the most common organic material in the biosphere. At an estimated annual production of 5-25 billion tons, it accounts for approximately 50% of the total biomass production in the world [Claassen et. al., 1999]. These cell walls consist primarily of three types of polymers, cellulose, hemicellulose and lignin, which form a tight and resilient structure. This complex is often referred to as lignocellulose [Claassen et. al., 1999].
Cellulose is a polysaccharide that consists of D-glucose bound in chains by β-(1,4)- glycosidic bonds. These chains bind to each other with hydrogen bonds to form strong, crystalline microfibrils. Hemicellulose is collective name for other polysaccharide components present in plant cell walls; it is a heterogeneous and complex mixture that contains many different types of monomeric sugars, such as glucose, xylose, mannose, galactose and arabinose. Lignin is a hydrophobic polymer consisting of various organic moieties - such as aromatic rings - which are polymerized in a complex and
unpredictable manner. Some plants also contain a significant amount of pectin, which is another kind of heterogeneous carbohydrate polymer [Taherzadeh and Karimi, 2007].
The composition of the lignocellulose varies between plant species, but a general ratio is 30-50% cellulose, 10-30% hemicellulose and 10-30% lignin. A more detailed
breakdown of the composition for a few important plant species can be seen in Table 1.
In order to use lignocellulose as a raw material for ethanol production the different sugars bound in the cellulose and hemicellulose must first be released as monomers, which makes them accessible to the microorganisms that ferment them into ethanol.
Hexoses - or six carbon sugars - such as glucose and mannose can be fermented using
normal baker’s yeast (Saccharomyces cerevisiae), but in order to utilize the different
pentose sugars found in the hemicellulose, other forms of microorganisms, which carry
the necessary metabolic enzymes, will be needed. Such organisms exist, but have yet to be used on a scale comparable to that of baker’s yeast [Taherzadeh and Karimi, 2007].
Table 1
Dry weight percentage of the main components of some model lignocellulosic materials.
Material Cellulose Hemicellulose Lignin
Corn stover 37.5 22.4 17.6
Pine wood 46.4 8.8 29.4
Poplar 49.9 17.4 18.1
Wheat straw 38.2 21.2 23.4
Switch grass 31.0 12.4 17.6
Percentages do not add up to 100%, since minor components have not been listed.
All data taken from Mosier et. al., 2005.
The process in which monosaccharides bound in the cellulose and hemicellulose are released is referred to as saccharification. It can be performed using concentrated or dilute acids at high temperatures, but a method that has received a greater amount of attention in recent years is enzymatic saccharification. Here enzymes found in some bacteria and fungi are used to break down the cellulose and hemicellulose polymers.
[Taherzadeh and Karimi, 2007].
In order to achieve an efficient hydrolysis of a lignocellulosic material it is necessary to use a mixture of several different types of enzymes that work synergistically on the cellulose and hemicellulose. These include cellulases, hemicellulases and -glucosidases.
There are two main types of cellulases which degrade cellulose in different ways:
endoglucanases, which can make a cut in the middle of a cellulose chain and
exoglucanases, which cut off cellobiose units (a disaccharide consisting of two glucose units) at the end of the cellulose chain, or at the points where the endoglucanase has cut the chain. The presence of both types of enzyme will speed up the hydrolysis to a degree that further addition of just one of the types alone would not. The hemicellulases are a large group of different enzymes (among others, endo-1,4- -D-xylanases, exo-1,4- -D- xylanases, -glucuronidases and acetyl xylan esterases) that are needed to degrade the very complex hemicellulose polymers. Finally, -glucosidases are necessary to break the glucoside bond in cellobiose and release the two glucose molecules [Taherzadeh and Karimi, 2007].
There are many different species of bacteria and fungi that produce cellulases and hemicellulases, but the one that has been the subject of most research is the fungus originally designated Trichoderma reesei (or Hypocrea jecorina as it should be called now). This is also the microorganism that is primarily used to produce cellulases on an industrial scale [Taherzadeh and Karimi, 2007].
The ethanol-from-lignocellulose process
When going from a lignocellulosic raw material to ethanol, a process consisting of
several steps is necessary for an efficient degradation and conversion into ethanol. At
the most basic level the substrate is treated with degrading enzymes, the monomeric
sugars then released are put it into a fermentation vessel where microorganisms use it
to produce ethanol, which is finally separated from the culture solution by for instance
distillation. In reality a more complex process is required (Figure 1).
Since the lignocellulose in the plant cell walls has a compact structure, which is highly resistant to degradation, it is necessary to make the cellulose and hemicellulose more accessible to the enzymes used in the
saccharification. This is done first through size reduction, where the size of the substrate particles is reduced by cutting, chipping or milling the material (this also makes the material easier to handle in an industrial process) and then through thermochemical pretreatment, where high temperatures and corrosive chemicals are used to open up the structure of the lignocellulose.
The particle size can vary from a sub- millimeter size up to a few centimeters.
Generally smaller particles mean larger surface area and a more easily degradable substrate, but even though a smaller particle size may lead to a higher degree of hydrolysis, it might still be economically sound to use larger particles, since milling often is a highly energy intensive process.
The part of the process with the greatest potential impact on maximizing the final sugar yield is probably the thermochemical pretreatment. Where an untreated
substrate may produce a sugar yield of less than 20% of the theoretical maximum after
enzymatic saccharification, the leading pretreatment methods regularly give rise to yields of 70-90% [Mosier et. al., 2005]. Many different pretreatment methods, where reaction time, temperature and chemical mechanisms differ, have been tested over the last thirty years; some of the more important alternatives are discussed here.
From a process perspective, the enzymatic saccharification is a relatively
straightforward procedure. Substrate, enzymes and water are mixed in a stirred vessel where temperature and pH are kept on a controlled level (30-60 C, pH 4-5). The enzymes are allowed to act upon the substrate for a period of 1-3 days, after which almost all sugars that can be recovered from a particular substrate will be solubilized.
When fermenting hexose sugars (such as glucose and mannose) into ethanol, baker’s yeast can be used. This is done when ethanol is produced from sugar or starch
containing crops and this method can also be used to ferment the hexose sugars found in lignocellulosic crops. However, since such crops also will contain pentose sugars found in the hemicellulose, baker’s yeast will not be able to fully realize the ethanol potential of the crop. Researchers have therefore looked for, or tried to create, microorganisms that can ferment pentose sugars to ethanol and today there are several alternatives where pentose hydrolyzing metabolic pathways have been introduced in well known
microorganisms, such as Escherichia coli and Zymomonas mobilis. While these designed
Figure 1: Schematic representation of
the lignocellulose-to-ethanol process.
organisms show promising results, they have yet to be tried and proven useful in large- scale ethanol production projects [Galbe and Zacchi, 2002].
A somewhat different procedure compared to the one outlined above is simultaneous saccharification and fermentation (SSF), where the enzymatic saccharification and the fermentation takes place together in the same vessel. The main advantage of this
procedure is that the continuous metabolization of the enzymatically released sugars by the fermenting microorganisms reduces product inhibition in the enzymatic process (it has been reported that –glucosidase loses 75% of its activity at glucose levels of 3 g/l [Philippidis and Smith, 1995; Philippidis et. al., 1993]). The main disadvantage is the necessity to keep the temperature and pH at levels that are acceptable to the fermenting microorganisms, which are not necessarily ones that are optimal for the enzymes. The enzymatic saccharification should optimally be performed at 50-60 C, whereas an organism like S. cerevisiae prefers a working temperature of 30-35 C. Experiments have shown that 38 C is the optimal compromise and since product inhibition may be severe the catalytic efficiency of the enzymes in SSF still shows a superior efficiency, compared to a separated saccharification and fermentation, even at such a low temperature. Usage of thermotolerant bacteria and yeast, such as Candida acidothermophilum and
Kluyveromyces marxianus as fermenting microorganisms may in the future allow for higher working temperatures in an SSF process, but the ongoing work aimed at minimizing product inhibition of cellulases by directed mutagenesis and protein engineering might instead lessen the need for an SSF process [Taherzadeh and Karimi, 2007].
Residual material
After the saccharification and fermentation there will still be some material left that has not been metabolized into ethanol or CO 2 by the fermenting microorganisms. The lignin in the lignocellulosic raw material cannot be utilized in the process and there will also be undigested cellulose and hemicellulose polymers left since it is generally very difficult to achieve a complete saccharification of these. In order to utilize the raw material to its full capacity, this residual material can be treated in several different ways. Incineration is an attractive alternative, since it is a simple process that generates products (heat and electricity) for which there is a constant demand. Another
alternative is to utilize the residual material in an anaerobic digestion process where biogas is produced. A third is to use the material to produce some other product of higher value, such as plastics that can be derived from lignin [Galbe and Zacchi, 2002].
The most efficient way to utilize the raw material will depend on what type of material it is, the location of the processing plant and other variations in economical conditions. It is also necessary to choose pretreatment methods appropriate to the usage the leftover products are intended for. If the lignin is intended to be refined for production of other high value chemicals, it is unsuitable to use a pretreatment method where the lignin is oxidized [Taherzadeh and Karimi, 2007].
Thermochemical pretreatment
Due to the recalcitrant nature of lignocellulose a pretreatment step is necessary before
any lignocellulosic material can be enzymatically hydrolyzed with an adequate level of
efficiency. The pretreatment should break up the tight structure of the plant cell walls
and expose the cellulose and hemicellulose to the enzymes, either by dissolving a large
part of the material or by “loosening up” the lignin-cellulose-hemicellulose matrix and thereby creating space for the enzymes to access carbohydrate polymers [Galbe and Zacchi, 2002].
During the last twenty to thirty years, large efforts have been put into developing efficient pretreatment methods, and today there exists a plethora of different alternatives. These range in mechanism from simple grinding of the lignocellulosic biomass to the usage of specialized microorganisms for delignification of the feedstock, but the majority of the methods that appear to produce the best results either use high temperatures, corrosive chemicals, or more often a combination of the two, when preparing the substrates for enzymatic saccharification. This is why one often speaks of
“thermochemical” pretreatment [Taherzadeh and Karimi, 2007].
Dilute acid
Dilute acid is, together with steam explosion, the pretreatment method most popular among the many commercial lignocellulosic ethanol projects that have recently started or reached the planning stages [Taherzadeh and Karimi, 2007]. The lignocellulosic biomass is mixed with water and acid is added to a low concentration (0.5-2% v/v) in a pressurized reactor. The reaction mixture is heated to 140-240 C and is kept there for a time period of a few minutes up to an hour. Sulphuric acid (H 2 SO 4 ) is the most
commonly used catalyst, but others such as hydrochloric acid and phosphoric acid have also been investigated [Mosier et. al., 2005].
The basic catalytic function of dilute acid pretreatment is hydrolysis of hemicellulose.
This not only releases sugars that can be fermented to ethanol, but also breaks up the structure of the lignocellulose and makes the cellulose fibers more accessible to the cellulases. It is a fairly efficient and cheap method, but there are a few problems. Most significantly, the degradation of the hemicellulose does not limit itself to the release of the monomeric sugars, but these are in turn further degraded into HMFs
(hydroxymethylfurfurals) and furfurals, which both act as inhibitors in the fermentation step [Mosier et. al., 2005]. It is therefore often common to wash away the liquid fraction from the solid biomass and only use the solids in the enzymatic step [Taherzadeh and Karimi, 2007]. Unfortunately this not only removes the inhibitory compounds but also the sugars that have been released from the hemicellulose, thus limiting the final
amount of ethanol produced. Beyond this, the acid is also a corrosive agent, which leads to higher equipment costs; it is also necessary to neutralize it before the
saccharification, which is an extra cost as well [Mosier et. al., 2005].
Lime
Lime (calcium hydroxide, “slaked lime” or Ca(OH) 2 ) is also a popular pretreatment catalyst. It is added to the reaction mixture to about 10% of the weight of the biomass (w/w) in water; further addition of lime does not seem to significantly increase its catalytic efficiency. The lime pretreatment is generally performed at a lower
temperature than when acid is used, and a range between 80 C and 120 C can be found in the literature; accordingly the pretreatment time is also significantly longer at 1-24 hours [Chang et. al, 1997].
Lime pretreatment works primarily by two mechanisms: firstly it solubilizes lignin and
breaks up the structure of the lignocellulose; secondly it removes acetyl and uronic acid
substitutions from the hemicellulose, which facilitates the access of the cellulases to the hemicellulose and cellulose. Lime pretreatment works better on substrates with a lower lignin content (such as straw materials and corn stover), but if oxygen is added during the process, lignin is more easily depolymerized and more lignin-rich substrates, such as wood, can also be efficiently pretreated [Mosier et. al., 2005; Chang et. al., 1997].
Other alkaline catalysts, such as sodium hydroxide, ammonia and urea will also affect lignocellulose in a similar manner, but lime has a few advantages that give it a
comparative edge. The biggest one is the price; lime is significantly cheaper than other alkalis. The other main advantage is the possibility of recovering the lime by bubbling CO 2 through the substrate mixture; this will cause the lime to react with the CO 2 and form calcium carbonate, which can then be turned back into calcium hydroxide using classical limekiln technology [Chang et. al, 1997]. Compared to other leading
pretreatment methods, lime shows good efficiency and a low price, but it has yet to be tried on a large scale.
Alkaline Peroxide
Using hydrogen peroxide (H 2 O 2 ) in an alkaline environment (pH 11-13) is another efficient way of pretreating lignocellulose. Using a H 2 O 2 concentration of about 2% (v/v) and pretreatment temperatures as low as 35-50 C, after incubation times of 12-24 hours, this method can contribute to a production of monomeric sugars after enzymatic saccharification at a level very close to the theoretical limit.
The primary mechanism of H 2 O 2 action derives from its ability to oxidize lignin, but it also solubilizes hemicellulose to some extent. The oxidation of lignin is a highly efficient way of loosening up the structure of the lignocellulose, but unfortunately it also means that the lignin can no longer be used for further conversion into high value end products.
This, together with the fact that hydrogen peroxide is a comparatively expensive chemical makes this method better suited for laboratory work than for industrial scale usage [Saha and Cotta, 2006; Fang et. al., 1999].
Hot water
It is possible to use plain water, without any corrosive chemicals added, when
pretreating lignocellulose. When heating water to very high temperatures (200-230 C), the pH of the water will decrease and it will, together with compounds such as acetyl groups released from the hemicellulose, act upon the lignocellulose in a similar manner to dilute acid. Both the catalytic mechanism (primarily depolymerization of
hemicellulose) and the unwanted formation of inhibitory compounds (HMFs and
furfurals) are the same in hot water pretreatment as in dilute acid pretreatment, even if it is noticeable to a lesser extent in both cases. Hot water pretreatment has the added advantages of being cheaper and more environmentally friendly compared to dilute acid pretreatment and the products do not need to be neutralized before the saccharification step [Mosier et. al., 2005; Taherzadeh and Karimi, 2007].
If oxygen is continuously added during the hot water pretreatment, a process known as
wet oxidation takes place. In this process the pretreatment works by oxidizing the lignin
instead of acting on the hemicellulose, and it is therefore better suited for more lignous
substrates [Taherzadeh and Karimi, 2007].
Steam explosion
In the previously described pretreatment methods the lignocellulosic raw material is mixed with a significantly larger amount of water (a substrate load of 5-10% of the water content is common) and the mixture is then heated to the final temperature. An alternative procedure is to limit the amount of water (to about 50% of the total
substance amount), and then heat the water and substrate by adding steam, which also raises the pressure. The temperature and pressure are kept at a high level for a short period of time (seconds to minutes), during which the water, or added catalysts such as sulphuric acid or sulphur dioxide acts upon the lignocellulose. At this point the
pressurized vessel is opened and an explosive depressurization occurs, which in turn leads to the expulsion of the material into another vessel.
This procedure is known as steam explosion and it is perhaps the pretreatment method that is most popular today. As with dilute acid and water pretreatment, the high
temperature (160-260 C) and a high pressure, acting together with a catalysts or without it, hydrolyzes the hemicellullose and thus breaks up the structure. The main difference is the more limited use of water, which also makes it possible to limit the use of catalysts and lower the total price and energy usage. The explosive decompression is often said to contribute to a further loosening of the structure, but it appears to be of a more limited importance compared to the removal of the hemicellulose [Mosier et. al., 2005; Galbe and Zacchi, 2002].
Both H 2 SO 4 and SO 2 are efficient catalysts and the optimal choice may differ between different substrates. H 2 SO 4 is more corrosive and more difficult to apply to the substrate, but SO 2 is a highly toxic gas, which might be difficult to handle. However, there are already industrial processes where SO 2 is used, so this may not be a major problem [Galbe and Zacchi, 2002].
Steam explosion is a method that is commonly used in the production of fiber boards (the Masonite method) and it has been the subject of very intensive research regarding its use as a pretreatment method. All of this, together with a reasonably high efficiency and a relatively low cost, contributes to its considerable popularity [Mosier et. al., 2005].
Ammonia Fiber Explosion (AFEX)
Ammonia fiber explosion is basically the same method as steam explosion, but with ammonia is as the catalytic agent. It works at lower temperatures compared to steam explosion and the pretreatment mechanism is primarily delignification and the removal of lignin-hemicellulose bonds. This is a very efficient pretreatment method for
agricultural byproducts and other materials with a lower lignin content, which makes it an attractive option, despite the relatively high process costs associated with the price, and necessary recycling, of ammonia [Taherzadeh and Karimi, 2007].
Organosolv
In the organosolv process, the various components in the lignocellulosic material are
separated into different fractions using organic solvents and catalysts that solubilize the
cellulose and hemicellulose. The lignin will be collected in the organic fraction and the
cellulose and hemicellulose in the water fraction, which may then be separated further.
The process can be performed from ambient temperature up to over 200 C, depending on which organic solvents and catalysts are used.
Organosolv is a potentially very attractive pretreatment alternative. Some versions show a high level of efficiency, and the fractionation of the different lignocellulosic parts makes the process easier to manage downstream. For instance, the relatively pure lignin in the organic fraction will be much easier to refine for further use. Industrial scale trials have yet to be performed though, which makes it difficult to assess its economic
feasibility [Taherzadeh and Karimi, 2007].
Biological pretreatment
Biological pretreatment methods have also been considered and tried. Here, the idea is to utilize microorganisms such as white rot fungi, which have the natural ability to degrade lignin, in order to make the cellulose and hemicellulose more accessible. The procedure is very simple: the lignocellulosic material is mixed with spores from the fungus and is kept for a few weeks or months until a sufficient amount of lignin has been removed [Taherzadeh and Karimi, 2007].
Carbohydrate analysis
Many different methods for carbohydrate analysis exist. Here the two main types of analytic methods that were used for measuring sugar content during this project will be described.
Colorimetry
There are several different types of colorimetric methods, but they all share the same basic principle: the sample is mixed with a chemical substance that changes color in the presence of sugar, and the intensity of the color, which should be proportional to the sugar content, can then be measured using a spectrophotometer. Three different colorimetric analysis methods were evaluated: DNS (where the active chemical is
dinitrosalicylic acid) [Miller, 1959], PAHBAH (p-hydroxy benzoic acid hydrazide) [Lever, 1972] and GOD/POD (glucose oxidase/peroxidase) [McCleary and Codd, 1991]. DNS and PAHBAH methods measure the total amount of reducing sugar, whereas GOD/POD is able to measure only the glucose using the glucose specific enzyme glucose oxidase in combination with peroxidase in order to cause a change in color of another substance.
Colorimetric methods have the advantage that they are simple, fast and easy to use on large sample sets, but they are limited by their ability to measure only total reducing sugar or, in the case of GOD/POD glucose, and the presence of other redox-active
compounds may influence the response. If one has a complex carbohydrate mixture and wants to quantify the components individually, other methods are needed.
HPLC/HPAE-PAD
HPLC (High Performance Liquid Chromatography) is a form a column chromatography performed at higher flow rates and pressures than with standard liquid
chromatography. Dionex (Sunnyvale, CA, USA) has developed a type of high
performance anion exchange chromatography (HPAE) that together with a pulsed
amperometric detector (PAD) can separate mono- and oligosaccharides with high
resolution and detect and quantify them down to a nanomolar scale [Dionex, 2006].
Neutral sugars can be separated on an anion exchange column when a strongly alkaline mobile phase is used, due to the fact that they are weak acids and will become partially or completely ionized at a high pH. Detection is performed by measuring the electrical currents generated by the oxidation of carbohydrates on the surface of a gold electrode, a process that is also facilitated in an alkaline environment.
Objectives
One long-term goal for the project described in this master’s thesis is to make biofuel production from lignocellulose more efficient. In order to achieve this goal it is necessary to develop methods for optimization of pretreatment and saccharification conditions for different lignocellulose substrates. Specific goals for this study were to:
Establish efficient laboratory-scale routines for thermochemical pretreatment and enzymatic saccharification of lignocellulose substrates, as well as
measurement of fermentable sugar yields.
Make a literature survey of pretreatment methods that has been tried and suggest which methods that may be best suited for laboratory-scale pretreatments of lignocellulosic plant material.
Perform selected pretreatments of cellulose model substrates (aspen sawdust and oat straw) followed by enzymatic saccharification and sugar yield
measurements.
Evaluate the five pretreatment methods for efficiency and suitability for laboratory scale use, and compare the yields with methods developed for industrial scale usage.
Materials and Methods
Lignocellulose materials
Oat straw from the region around Sala was provided by Sala-Heby Energi AB (Sala, Sweden). Aspen sawdust, with a particle size of up to a few millimeters, was received from a small sawmill outside of Uppsala. The sawdust came from planks consisting mainly of highly lignified aspen heartwood, which had been sawed with the specific intention of creating sawdust. The oat straw was cut up into 5-15 cm pieces. Both materials were dried at 60 C for 24 hour in preparation for milling. The milling was performed in a hammer mill (provided by The Department of Animal Nutrition and Management at the Swedish University of Agricultural Sciences in Uppsala) until all material could pass through a 1 mm sieve (Figure 2). Most of the particles of both oat straw and aspen were significantly smaller than 1 mm after the milling.
The dry weight was measured by weighing the samples before and after placing them samples in an oven at 105 C for 24 hours.
Pretreatment
Oat straw and aspen sawdust was pretreated with the following pretreatment procedures: dilute acid, lime, alkaline peroxide, hot water and steam explosion.
Dilute acid
2.5 g or 5.0 g of lignocellulosic substrate was put into 100 ml Pyrex bottles. Water and concentrated sulphuric acid was added to a final volume of 50 ml and a final acid
Figure 2: The hammer mill used for sample preparation.
concentration of 0.5 or 1.0 %. The bottles were put in a Certoclave tabletop autoclave and heated to 140 C for four different time periods ranging from 15 to 90 minutes.
These times refer to the time spent at 140 C; it generally took about 20 minutes to reach this temperature and about 40 minutes to cool down enough to open the autoclave. The bottles were then cooled down to room temperature in a water bath and the pH was set to 5 using 10 M NaOH before they were frozen at -20 C, awaiting enzymatic
saccharification.
Lime
5 g of substrate was put into 250 ml Pyrex bottles together with 0.5 or 0.75 g of lime (calcium hydroxide, Ca(OH)2). 50 ml of 90-100 C water was added and the bottles were then put in a shaking incubator at 80 or 95 C. Bottles were then removed after 1, 3, 6 or 24 hours and cooled down to room temperature in a water bath. The pH was set to 5 using 25 % hydrochloric acid after which the bottles were finally frozen at -20 C.
Hot water
Hot water pretreatment was performed in the same manner as lime pretreatment, with the exception that no lime was added and that the pH was not changed before freezing.
Alkaline peroxide
5 g or 10 g of substrate was added to 250 ml bottles, together with water and 7.5 ml
30 % hydrogen peroxide. The pH was set to 11.5 using 10 M NaOH and the volume was
adjusted to 100 ml. The samples were kept in a shaking incubator at 35 C for 24 hours,
after which the pH was set to 5 using 25 % hydrochloric acid and the bottles were frozen.
Figure 3: Oat straw and aspen, after pretreatment using sulphuric acid.
Steam explosion
Steam explosion was performed at the Department of Chemical Engineering at Lund University, Sweden, where they have the necessary equipment for the procedure. The
“steam gun” used in the pretreatment may clog if the substrate particles are too small and the hammer-milled material could therefore not be used. Instead the oat straw was cut into 3-5 cm pieces using a household mixer and the aspen sawdust was used directly.
The dry weight of the substrates was measured. The oat straw was sprayed with diluted sulphuric acid to a final dry weight of 50 % and an acid concentration of 0.2 % and was left to soak overnight at room temperature. The aspen sawdust was then put in a plastic bag and infused with 2.5 % of sulphur dioxide (SO 2 ) for 30 minutes.
Approximately 1 kg of material was put into the reactor of the steam gun and the temperature was raised to 200 C for the aspen sawdust and 190 C for the oat straw.
After 5 minutes for the aspen and 10 minutes for the oat straw, a valve was opened into a collecting vessel, which caused the explosive expulsion of the substrate from the reactor. The substrate was then collected and the dry weight measured again.
Enzymatic Saccharification
For the enzymatic saccharification a commercially available enzyme preparation called Accelerase 1000 was provided as a kind gift by Genencor - A Danisco Division (Palo Alto, USA), consisting of biomass degrading enzymes from Hypocrea jecorina,
supplemented with additional -glucosidases. The accompanying product statement
declared that it had a measured cellulase activity of 2707 CMCU/g (1 CMCU defined as 1
µmol of reducing sugar released per minute when hydrolyzing carboxymethylcellulose
at 55 C and pH 4.8) and a -glucosidase activity of 403 U/g (U defined as 1 µmol of nitrophenol liberated from para-nitrophenyl-B-D-glucopyranoside in 10 minutes at 50 C and pH 4.8). Saccharification of the pretreated lignocellulosic samples was
performed according to the recommendations of the manufacturer, at 55 C, pH 4.5 and with an enzyme loading of 0.25 g of Accelerase solution per g of substrate (dry weight).
This is an enzyme loading at the upper limit of the recommended loading range.
Bottles with pretreated lignocellulosic substrate were taken from the freezer and defrosted. New bottles were weighed empty and the contents of each bottle with pretreated material were transferred into the new bottle. The old bottles were rinsed with a small amount (~30 ml) of water, which was also transferred into the new bottles in order to recover as much as possible of the pretreated material. Citrate buffer was added to a final concentration of 50 mM, the pH was set to 4.5 using NaOH or
hydrochloric acid and water was added to a total weight of 100 g (excluding the bottle).
0.5 ml samples were taken in triplicate from each bottle and saved. These are the zero- hour saccharification samples. The bottles were put in a shaking incubator at 55 C for at least one hour in order to reach the process temperature. Enzymes were then added in varying amount, depending on dry weight and initial amount of substance added, and the bottles were put back in the incubator.
0.5 ml samples were taken in triplicate at 30 minutes, 1, 3, 6, 24, 48 and 72 hours for carbohydrate analysis. The samples taken were loaded on 1.5 ml microtiter plates, which were sealed using silicone lids or adhesive film and put into a water bath set at 95-100 C. This was done in order to inactivate the cellulases and stop the hydrolysis of the cellulose and hemicellulose. The microtiter plates containing the samples were then frozen at -20 C, awaiting filtration.
The zero-hour samples were also put in a 95-100 C water bath and a small amount of enzymes was added, corresponding to the enzyme concentration in the saccharification bottles. This control was done because the enzyme solution contains a small, but
potentially significant, amount of sugars, which may need to be subtracted from all samples in order to accurately estimate the increase caused by the hydrolysis of the substrate. The enzymes were added to the hot samples in order to destroy them immediately, before they could act upon the substrate.
After 72 hours the bottles were removed from the incubator, boiled for twenty minutes and frozen.
Filtration
Before any type of analysis could be performed on the saccharification samples, the remaining substrate particles were removed using a vacuum manifold filtration unit for microtiter plates, with 1 µm glass fiber filters. After filtration, the samples were either frozen again or immediately prepared for further analysis.
Analysis DNS
DNS (dinitrosalicylic acid) reagent was prepared according to three slightly differing
protocols from Methods in Enzymology 160, Wang [Wang, 2008], and the Accelerase
manual. Equal amounts of DNS reagent and sample were added to a 10 ml Falcon tube,
which was then boiled for 10 minutes. After cooling, the boiled sample was measured in a spectrophotometer at 575 nm and the reducing sugar concentrations quantified using a calibration curve made with glucose standards.
PAHBAH
PAHBAH (p-hydroxybenzoic acid hydrazide) reagent was prepared according to instructions from Megazyme [Megazyme, 2008]. Reagent and sample were mixed in a ratio of 5 to 1 and then boiled for 10 minutes. The absorbance was measured at 410nm using a spectrophotometer; reducing sugar concentrations were quantified by
comparison to glucose standards of known concentrations.
GOD/POD
A GOD/POD assay kit was bought from Megazyme (K-GLUC) and reagents were prepared according to the manual. The reagent was mixed with the sample and incubated at 50 C for 20 minutes and absorption was measured at 510 nm.
Figure 4: The Dionex ICS 3000 HPAE-PAD carbohydrate analysis system. From left to right: autosampler, column oven and detector compartment, and pumps. On top: a fraction collector and eluent bottles.
HPLC/HPAE-PAD
A Dionex (Sunnyvale, CA, USA) ICS-3000 HPAE-PAD system (Figure 4) was used to
separate arabinose, galactose, glucose, xylose, mannose and cellobiose and measure
their concentrations in the samples taken from the enzymatic saccharification. A
CarboPac PA10 (4x250mm) (Dionex) anion exchange column was used at 30 C and
detection of the separated sugars was performed on a pulsed amperometric detector
(PAD), using the “Standard Quad” waveform.
The flow rate through the column was 1.0 ml/min. Arabinose, galactose, glucose, xylose and mannose were eluted isocratically using 100% water, and a gradient up to 200 mM NaOH and 70 mM sodium acetate was then initiated in order to elute the cellobiose and other oligosaccharides that might otherwise remain on the column. Table 2 shows a detailed description of all gradient steps. 300 mM NaOH at 0.5 ml/min was added after the column, in order to keep the pH sufficiently high for an efficient detection by the PAD detector.
Table 2
Analytical program for the Dionex HPEA-PAD system.
Time (minutes) Eluent composition
-12 to -7 Regeneration: 200 mM NaOH and 70 mM sodium acetate isocratically.
-7 to -6 Gradient down to 100% water.
-6 to 17 100% water isocratically.
0 Injection of samples.
17 to 35 Gradient up to 200 mM NaOH and 70 mM sodium acetate.
Enzymatically treated samples were diluted 200 or 400 times in water before analysis.
Quantification was performed by comparing peak areas of sugars in the samples to those of five different standard solutions containing all six sugar types, with known concentrations ranging from 2 to 80 mg/l. The injection volume was at all times 25 µl.
Calculations of the sugar concentrations of the samples were done automatically using the Chromeleon software controlling the HPLC machine and data were later compiled and processed further using Microsoft Excel.
Results and discussion
The dry weight of the aspen and oat straw substrates was determined to be 96.4 % and 94.5 % of the initial weight, respectively.
Thermochemical pretreatment
The comparative efficiency of different pretreatment methods and different conditions of a particular pretreatment method was determined by measuring the sugar content and sugar composition of samples taken after saccharification of the pretreated material. In order to compare the full potential of the different pretreatment methods, sugar composition of the 72-hour samples was analyzed using the Dionex HPAE-PAD equipment for all saccharifications. The impact of the pretreatment on the kinetics of the enzymatic saccharification was also investigated on a smaller number of samples, by analyzing sugar composition of samples taken at all time points (0-72 hours).
Alkaline hydrogen peroxide pretreatment
Alkaline hydrogen peroxide pretreated samples showed saccharification yields of 0.25-
0.50 g total sugar per gram of pretreated substrate (wet weight) for aspen (Figure 5)
and 0.37-0.40 g sugar per g substrate for oat straw (Figure 6) with substrate load as the
variable pretreatment condition. Aspen gave a significantly lower saccharification yield
for a 10 % (weight of substrate/volume of liquid) substrate load compared to a 5 % substrate load. This might be due to the fact that aspen has a high lignin content and there might not be enough H 2 O 2 available in the pretreatment mixture for a sufficient oxidation of the lignin to effectively break up the lignocellulose matrix at a substrate load of 10 %. In contrast, oat straw shows a much smaller variation in saccharification rate between 5 and 10 % substrate load. This is unsurprising if the ratio of H 2 O 2 to lignin is the determining factor of catalytic efficiency, since straw materials have a lower lignin content than hard woods.
Aspen, hydrogen peroxide pretreatment
0,000 0,100 0,200 0,300 0,400 0,500 0,600
Hydrogen peroxide, 10% s Hydrogen peroxide, 5% s
g s u g a r / g s u b s tr a te
Glucose Xylose Other sugars
Figure 5: Alkaline hydrogen peroxide pretreatment of aspen, where substrate loading has been varied between 5 and 10 %. The bars show the average sugar amounts based on triplicate samples; the error bars show the highest and lowest total amount of sugars of the triplicate samples.
Oat straw, hydrogen peroxide pretreatment
0,000 0,050 0,100 0,150 0,200 0,250 0,300 0,350 0,400 0,450
Hydrogen peroxide, 10% s Hydrogen peroxide, 5% s
g s u g a r / g s u b s tr a te
Glucose Xylose Other sugars
Figure 6: Alkaline hydrogen peroxide pretreatment of oat straw, where substrate loading has been
varied between 5 and 10 %.
Lime pretreatment
The impact of varying incubation time, temperature and amount of lime was
investigated for lime pretreatment. Lime pretreated aspen shows saccharification yields of 0.075-0.18 g sugar / g substrate (Figure 7). It is difficult to find a pattern in the
saccharification rates, when varying the above-mentioned factors. When time was
Aspen, lime pretreatment
0,000 0,050 0,100 0,150 0,200 0,250 0,300
1%
lime, 10%
1h, 80°C
1%
lime, 10%
24h, 80°C
1%
lime, 10%
6h, 80°C
1.5%
lime, 10%
s, 24h, 80°C
1.5%
lime, 10%
s, 24h, 95°C
1.5%
lime, 10% s, 3h, 80°C 1.5%
lime, 10% s, 3h, 95°C 1.5%
lime, 10% s, 6h, 80°C
g s u g a r / g s u b s tr a te
Glucose Xylose Other sugars
Figure 7: Lime pretreatment of aspen. Substrate loading, lime loading, time and temperature have been varied.
Oat straw, lime pretreatment
0,000 0,050 0,100 0,150 0,200 0,250 0,300 0,350 0,400
1%
lime, 10%
s, 1h, 80°C
1%
lime, 10%
s, 24h, 80°C
1%
lime, 10%
s, 6h, 80°C
1%
lime, 10%
s, 3h, 80°C
1.5%
lime, 10% s, 24h, 80°C 1.5%
lime, 10% s, 24h, 95°C 1.5%
lime, 10%
s, 3h, 80°C
1.5%
lime, 10%
s, 3h, 95°C
1.5%
lime, 10%
s, 6h, 80°C
g s u g a r / g s u b s tr a te
Glucose Xylose Other sugars
Figure 8: Lime pretreatment of oat straw. The same parameters have been varied as with aspen.
varied from 1 to 24 hours, no major differences can be seen and in fact both the lowest and highest saccharification yields were obtained in samples pretreated for 24 hours.
The variation of lime from 1 % to 1.5 % and of temperature from 80 to 95 C shows no clearer patterns either. The variations in the sugar analysis seem to be larger than any variations caused by different pretreatment conditions.
Oat straw gave saccharification yields of 0.15-0.35 g sugar / g substrate (Figure 8), though apart from one sample (1 % lime, 10% substrate, 3 hours at 80 C) no samples gave yields below 0.25 g/g. The same issues with interpreting the results as with aspen apply here and it is difficult to draw any conclusions regarding optimal conditions. The general levels of saccharification are higher for oat straw compared to aspen. This is in line with results found in the literature, where a generally higher catalytic efficiency of lime pretreatment on substrates with a lower lignin content has been reported [Mosier et. al., 2005], but the total rates are still comparatively low.
Hot water pretreatment
Pretreatment time and temperature were varied when saccharification rates were measured for hot water pretreatment. Pretreated aspen samples gave yields of 0.040- 0.045 g sugar / g substrate and oat straw 0.10-0.12 g sugar / g substrate (Figures 9 and 10). The highest conversion rates were found after 24 hours pretreatment at 95 C for both aspen and oat straw, but these were not significantly higher than pretreatment carried out at 80 C or for 3 hours. The total conversion levels are noticeably lower than for the other methods investigated in this study and it is likely that much higher
temperatures (from 150 C and upwards) are necessary in order to reach competitive levels of efficiency using hot water pretreatment.
Aspen, hot water pretreatment
0,000 0,010 0,020 0,030 0,040 0,050 0,060
Water, 10% s, 24h, 95°C Water, 10% s, 3h, 80°C Water, 10% s, 3h, 95°C
g s u g a r / g s u b s tr a te
Glucose Xylose Other sugars
Figure 9: Hot water pretreatment of aspen sawdust.
Oat straw, hot water pretreatment
0,000 0,020 0,040 0,060 0,080 0,100 0,120 0,140 0,160
Water, 10% s, 24h, 95°C Water, 10% s, 3h, 80°C Water, 10% s, 3h, 95°C
g s u g a r / g s u b s tr a te
Glucose Xylose Other sugars
Figure 10: Hot water pretreatment of oatstraw shows somewhat higher saccharification rates compared to aspen sawdust, but both substrates show significanly lower rates when treated with water compared to other pretreatment methods.
Diluted sulphuric acid pretreatment
Pretreatment time, substrate loading and acid loading were investigated. Acid pretreatment produced saccharification yields of 0.35-0.55 g sugar / g substrate of aspen and 0.25-0.90 g/g of oat straw (Figures 11 and 12). In this case it is also difficult to draw any conclusions regarding how the variable factors influence the
saccharification rates. 1 % acid seems to lead to higher rates than 0.5 %, but the
differences are not significant. The oat straw sample producing a saccharification rate of 0.9 g/g (0.5 % acid, 5 % substrate, 30 minutes) is unreasonably high, since it is probably impossible the reach such high levels due to lignin content etc. It is probably the result of a faulty measurement, but even so it seems clear that acid pretreatment and
particularly acid pretreatment of oat straw produces the highest final sugar
concentrations.
Aspen, sulphuric acid pretreatment
0,000 0,100 0,200 0,300 0,400 0,500 0,600 0,700
0.5%
acid, 10%
s, 30min
0.5%
acid, 10%
s, 60min
0.5%
acid, 10%
s, 90min
0.5%
acid, 5% s, 15min 0.5%
acid, 5% s, 30min 0.5%
acid, 5% s, 60min 0.5%
acid, 5% s, 90min 1%
acid, 10%
s, 30m in
1%
acid, 10%
s, 60m in
1%
acid, 5% s, 15m in
1%
acid, 5% s, 30m in
1%
acid, 5% s, 60m in
1%
acid, 5% s, 90m in
g s u g a r / g s u b s tr a te
Glucose Xylose Other sugars
Figure 11: Dilute sulphuric acid pretreatment at 140 C of aspen sawdust. There are only small
differences between the varying conditions, but the general level is higher than for other pretreatments of aspen.
Oat straw, sulphuric acid pretreatment
0,000 0,200 0,400 0,600 0,800 1,000 1,200
0.5%
acid, 10%
s, 30 min
0.5%
acid, 10%
s, 60 min
0.5%
acid, 10%
s, 90 min
0.5%
acid, 5% s,15 min 0.5%
acid, 5% s,30 min 0.5%
acid, 5% s,60 min 0.5%
acid, 5% s,90 min 1%
acid, 10%
s, 30 min
1%
acid, 5% s, 15 min 1%
acid, 5% s, 30 min 1%
acid, 5% s, 60 min
g s u g a r / g s u b s tr a te
Glucose Xylose Other sugars
Figure 12: Dilute sulphuric acid pretreatment at 140 C of oat straw appears to be very efficient.
However saccharification rates above 0.6-0.7 are generally not possible, since the total amount of cellulose and hemicellulose does not exceed 60-70 % of the dry weight of most lignocellulosic materials.
Critical analysis of the results
Figure 13 shows a compilation of all pretreatment methods investigated, including
steam explosion, for both aspen sawdust and oat straw. The conditions producing the
highest saccharification rates were chosen in order to visualize the comparative
catalytic potential. For both oat straw and aspen sawdust, sulphuric acid pretreatment
produced the highest conversion rates, but with the inhibitory compounds that may be formed during acid pretreatment this method might still prove to be an inferior option.
For aspen sawdust, both alkaline hydrogen peroxide and steam explosion pretreatment show comparable results, and they may prove to be better options. For oat straw, alkaline peroxide and steam explosion pretreatment show somewhat lower final sugar levels compared to aspen, but lime shows higher conversion rates. Saccharification yields above 0.60-0.75 g sugar/ g substrate are unreasonable, since there is not that much cellulose and hemicellulose available in a substrate to reach such conversion levels (see Table 1) and some of the values reported for acid pretreatment of oat straw therefore appear to be inherently untrustworthy. Hot water pretreatment gives
conversion levels that are significantly lower than other methods and they do not appear to be much higher than reported values for untreated materials.
Most of the pretreatment methods tested in this study show lower rates of
saccharification compared to values reported in the literature for similar substrates [Saha et. al., 2005; Saha and Cotta, 2006; Chang et. al., 1997; Galbe and Zacchi, 2002;
Taherzadeh and Karimi, 2007; Mosier et. al, 2005]. There are many possible
explanations for this; equipment used for pretreatment and saccharification could differ, with for instance better mixing during the different incubation steps, the enzymes could be more active or the analytical methods could differ.
Comparision of pretreatment methods for oat straw and aspen
0,000 0,100 0,200 0,300 0,400 0,500 0,600 0,700 0,800 0,900
1.5%
lime, 10% s, 24h, 80°C Hydrogen peroxide, 10%
s
Water, 10%
s, 24h, 95°C
0.5%
acid, 5% s,60 min Steam
explosion, 10%
s
1%
lime, 10%
24h, 80°C
Hydrogen peroxide, 5% s Water, 10%
s, 24h, 95°C 1%
acid, 5% s, 60m in
Steam explosion, 10%
s
g s u g a r / g s u b s tr a te
Glucose Xylose Other sugars
Figure 13: Summary of the most efficient example of each pretreatment method for oat straw and aspen. Dilute sulphuric acid appears to be the most efficient method for oat straw, but for aspen steam explosion and hydrogen peroxide are comparable.
Kinetics of the enzymatic saccharification
Figures A1 to A19 in Appendix A show complete curves of the saccharification, with samples taken at 0, 0.5, 1, 3, 6, 24, 48 and 72 hours. A majority of the curves show that the final sugar concentrations are reached after 6-24 hours and that sugar levels remain
Oat Straw Aspen
basically constant after that. The exception is oat straw, pretreated with 1.5% lime for 24 hours at 95 C (Figure A7), which appears to show a constantly increasing sugar concentration up to 72 hours. It is difficult to say whether this is caused by variations in the analysis or by an actual increase in sugar levels, but a repeated saccharification of this pretreated substrate with samples taken up to five days (120 hours) would hopefully clarify the situation.
Certain saccharification curves appear to reach a clear maximum after 3 to 6 hours of incubation and then decrease (examples can be seen in Figures A4 and A14). This is an unreasonable result, since there is no obvious reason why sugar should disappear from the solution, but there are a few possible explanations. The first is a bacterial infection, which would cause sugars to be consumed during the saccharification. This explanation can be probably disregarded though, since the bacteria in all likelihood would consume the glucose first and then go over to other sugars, and there is no indication of this.
Instead one can generally see a proportional variation of all sugars between time points.
Therefore a likelier explanation is variations in sensitivity during the HPLC
measurements or errors in the dilution during the preparation of the samples. Since all of the samples that show this pattern were saccharified and analyzed together and each time point for all of these samples was analyzed in a group, it is difficult to decide which explanation that is correct, but the fact that a pipette was found to be broken and was replaced somewhere around the time when these samples were analyzed, in
combination with the fact that the 3 and 6 hour samples seem to be unreasonably high, may indicate that dilution errors might have caused the problem. A repeated analysis of these samples was performed, but unfortunately this coincided with further problems with the sensitivity of the PAD detector on the Dionex system and the results had to be discarded. At this point it was not possible to perform any further carbohydrate
measurements due to lack of time within this thesis project and the solution of this problem had to be postponed until a later stage.
Sugar content of the Accelerase enzyme solution
The Accelerase enzyme solution is known to contain sugar, which comes from the fungal culture solution. In order to determine how much sugar that was added to the
saccharification process together with the enzymes the sugar content of the Accelerase enzyme mixture was measured using the same method as for the saccharification samples. These measurements showed that there were detectable levels of galactose, glucose and mannose present (Figure 14), but they were not high enough to significantly influence final yields, since the sugar added with Accelerase would correspond to about 0.1 % of the total sugar content of most samples.
Tabulated saccharification rates and ethanol potentials
Calculated saccharification rates are tabulated in Appendix B. The rates are reported in g
sugar / g substrate for glucose, xylose and the combined values for the other sugars
(arabinose, galactose, mannose and cellobiose). Theoretical estimations of ethanol
production from the released sugars in each sample are also listed. These values are
0,0 5,0 10,0 15,0 20,0 25,0 30,0 35,0 -10
13 25 38 50 63 75 88
100 080630 #416 accelerase-2 ED_1
nC
min
1 - Arabinos - 9,075 2 - Galaktos - 10,325 3 - Glukos - 12,350 4 - Mannos - 16,600 6 - Cellobios - 28,500
