MASTER'S THESIS
Biogas Production from Enzymatically Pretreated Agricultural and Forest Raw
Materials
Christine Enberg 2014
Master of Science in Engineering Technology Sustainable Process Engineering
Luleå University of Technology
Department of Civil, Environmental and Natural Resources Engineering
I
Summary
Biogas is a methane and carbon dioxide containing gas, produced by biomass degraded by microorganisms in anaerobic environment. Forest and agricultural residues contains higher proportions of cellulose and hemicellulose than substrates used in biogas production today, which makes them highly attractive for biogas production. However, the biogas yield of these materials is low due to the complex crystalline structure of the cell wall matrix. By separating the cell wall matrix and reduce the carbohydrate chain length by performing enzymatic pretreatments, such as partial enzymatic hydrolysis. The biogas yield is expected to be high since improvement in biogas production has been shown when commercial enzymes are used in the pretreatment of cellulose rich, complex, raw materials. The goal of this project is to evaluate the efficiency of enzymatic hydrolysis and compare the biogas yield of the following untreated and pretreated materials: wheat straw, spruce, pine and birch.
In order to investigate the ability of enzymes to release sugars, enzymatic hydrolysis was performed. Since some of the materials to be investigated were pretreated as slurries, it was crucial to investigate how the slurry state of the material is affecting the overall hydrolysis of the material as well. Also, the materials were prehydrolyzed prior to the reactor experiments in order to study the effect on the biogas yield. In the reactor experiments, the ability of each pretreated materials to produce biogas was investigated. In addition, detoxifying agents was added to the reactor experiments in order to investigate if detoxification was necessary in order to improve the biogas production.
The results showed that hydrolysis of pretreated wheat straw released the highest amount of sugars, whereas hydrolysis of pretreated birch improved the amount of released sugars most, compared to untreated birch. Pretreated birch produced the highest amount of methane during the reactor experiments, whereas pretreated pine improved the production of biogas, compared to untreated, the most. Detoxification only improves the methane potential when laccase is added to toughtreated spruce and pine. Prehydrolysis did not improve the biogas production except for birch, for which the methane potential was increased by 6-7%.
Birch is the most efficient pretreated material to be hydrolyzed, compared to the untreated material, and the methane potential of pretreated birch is the highest value achieved by all the materials used.
The efficiency might be explained by the favorable composition of hardwoods which makes it easier to degrade than spruce and pine. The large improvement of pine during the reactor experiments is probably time dependent since the enzymes might require a lot of time in order to efficiently hydrolyze pine.
Sammanfattning
Biogas är en gas som innehåller metan och koldioxid och produceras då biomassa bryts ned av mikroorganismer i anaerobisk omgivning. Restprodukter från skog och jordbruk innehåller större mängder cellulosa och hemicellulosa än de substrat som används till biogasproduktion idag, vilket gör dem mycket eftertraktade för biogasproduktion. Men, utbytet av biogas från dessa material är lågt på grund av den komplexa, kristallina, strukturen i cellväggarna. Cellväggen kan brytas upp och kolhydratkedjornas längd kan reduceras genom att enzymatisk förbehandla materialet, som vid partiell enzymatisk hydrolys. Utbytet av biogas förväntas då bli högt eftersom förbättringar i produktionen av biogas har visats när kommersiella enzymer används till förbehandling av komplexa råmaterial rika på cellulosa. Målet med studien är att utvärdera effektiviteten av enzymatisk hydrolys och jämföra utbytet av biogas för följande obehandlade och förbehandlade material: vetehalm, gran, tall och björk.
Genom enzymatisk hydrolys undersöktes enzymernas förmåga att frigöra socker ut materialen.
Eftersom vissa av de förbehandlade materialen bestod av en blandning av fast material och vätska var det även viktigt att undersöka hur vätskan påverkade hydrolysen av materialen. Materialen förhydrolyserades även inför reaktorexperimenten för att undersöka hur denna förbehandling påverkar utbytet av biogas. Under reaktorexperimenten undersöktes varje materials förmåga att producera biogas. Utöver detta tillsattes avgiftningsmedel för att undersöka om eventuellt behov av avgiftning skulle förbättre produktionen av biogas under reaktorexperimenten.
Resultaten tyder på att hydrolys av förbehandlad vetehalm frigjorde högst andel socker, medan hydrolys av förbehandlad björk förbättrade mängden frigjord socker mest, jämfört med obehandlad björk. Förbehandlad björk producerade mest metan under reaktorexperimenten, medan förbehandlad tall förbättrade produktionen av biogas mest, jämfört med obehandlad tall.
Avgiftningen förbättrade endast produktion av metan då lackas tillsattes till gran som genomgått hård förbehandling samt tall. Förhydrolysen förbättrade inte produktionen av biogas bortsett för björk då produktionen av metan ökade med 6-7%.
Björk är det mest effektiva förbehandlade materialet under hydrolys, jämfört med de obehandlade materialen, och förbehandlad björk uppnår högsta mängden av producerad metan av alla materialen.
Effektiviteten kan förklaras av att lövträd har en mer gynnsam sammansättning som gör materialet enklare att bryta ned än barrträd. Den höga förbättringen för tall under reaktorexperimenten är antagligen tidsberoende eftersom enzymerna troligtvis behöver lång tid på sig för effektiv hydrolys av tall.
Table of contents
1 Introduction ... 1
1.1 Object ... 1
1.2 Biogas ... 2
1.3 Lignocellulosic biomass ... 2
1.4 Enzymatic petreatment ... 3
1.5 Biogas production ... 4
1.6 Previous studies ... 5
2 Biogas production from enzymatically pretreated agricultural and forest raw material ... 7
2.1 Hydrolysis ... 8
2.2 Prehydrolysis ... 9
2.3 Reactor experiments ... 9
2.4 Results ... 11
2.5 Analysis ... 19
3 Discussion ... 21
3.1 Conclusion ... 23
4 References ... 24
5 Appendix ... 25
5.1 Hydrolysis results of wheat straw ... 25
5.2 Hydrolysis results of spruce ... 27
5.3 Hydrolysis results of pine ... 29
5.4 Hydrolysis results of birch ... 31
Introduction
1
In Sweden, 2012, the amount of biogas produced was about 1,6 TWh from a total of 242 plants. The usage of the biogas produced was for production of heat and, by upgrading the biogas, as fuel and the substrates used where mainly sewage sludge, food and feed waste from municipals and industries.1
In the future, the biogas is believed to be produced from forest and agricultural sources. These feedstocks contain higher proportions of cellulose and hemicellulose, which are the main building blocks of plant cell walls, than the substrates for biogas production used today. They also have the advantage that they do not compete with any other productions because of the residual character of the material. However, less biogas is produced from feedstocks rich in cellulose due to the protective cell wall matrix. One possible way to produce biogas from forest and agricultural feedstocks may be to increase the biogas yield by performing enzymatic pretreatments. By using commercial available cellulases a partial enzymatic hydrolysis, also called liquefaction, is performed allowing a separation of the cell wall matrix and gradual reduction of the cellulose and hemicellulose chains.
1.1 Object
In this study, an enzymatic liquefaction step is included to biogas production by fermentation. The feedstock, substrate, is common cellulose rich forest and agricultural raw materials in Sweden such as wheat straw, spruce, pine and birch. By using commercial available cellulases, the pretreatment of the substrates allows to reduce the viscosity of the high-solid content substrates by releasing the cell wall matrix and reducing the chain lengths of cellulose and hemicellulose. Also, better mixing for the inoculation of the fermenting organisms in enabled due to the pretreatment.
The biogas produced from these feedstocks is expected to be high yielded according to previous studies, for which significant improvement in biogas production has been shown when commercial enzymes are used in the pretreatment of such cellulose rich, complex, raw materials.
The main goal of this study is to use wheat straw, spruce, pine and birch for the production of biogas in order to evaluate and improve the biogas yield, i.e. the production of biogas per gram of material, mainly by:
evaluating the efficiency of the enzymatic hydrolysis of each material
comparing the biogas yield of each material at different conditions
Since there are some previous studies on commercial enzymes used as pretreatment of cellulose rich materials, the main approach will be based on some of the results of these studies in order to save some time during optimizing of the approach.
1 Biogasportalen.se
1.2 Biogas
Biogas is a methane and carbon dioxide containing gas, produced when biomass, i.e organic material, is degraded by microorganisms in anaerobic environment. Anaerobic digestion is a biological process in which organic carbon is degraded, in the absence of oxygen, to the most reduced state (methane) and the most oxidizing state (carbon dioxide). Hydrogen sulphide, nitrogen, ammonia and hydrogen are trace gases that may be formed during the process.2 The main part of the biogas, methane, is very rich in energy, and thus, biogas is suitable as heat production. The fact that biogas is renewable, and does not contribute to any net emissions of carbon dioxide, makes it suitable as a fuel.3 Another environmental advantage is that the solid residue may be used as a good fertilizer.4
In the production of biogas, the biogas potential is the definition of the volume biogas produced per gram volatile solid (VS) of substrate. In theoretical calculations, the methane potential is often higher than the measured produced methane. The degradation of raw material into biogas may be limited by the biodegradability and eventual inhibitors produced in the process.5 The amount of sulphur and nitrogen in the feedstock is affecting the production of trace gases, but since these elements are nutrients required by the microorganisms used, they cannot completely be eliminated from the feedstock.6
1.3 Lignocellulosic biomass
The biogas process is limited by fractions of the raw material with low degradability. Thus, less biogas is produced using these low degradability containing raw materials and the efficiency of the process is therefore lowered. Such a low degradability fraction is found in lignocellulosic biomass such as hardwood, softwood, grasses and industrial, as well as, agricultural residues.
Lignocellulosic biomass mainly contains cellulose, hemicellulose and lignin, for which the compact structure and the physical shielding of cellulose and hemicellulose by the lignin, makes these materials more resistance to enzymatic impact, compared to conventional biomasses used.7 The low degradability fractions of lignocellulosic due to the complex structure of high crystallinity and lignin content makes the material hard to digest during anaerobic digestion. This result in low or incomplete digestion due to the difficulties to degrade the material, and thus, the biogas yield will be low.8 Because of the high interest of lignocellulosic biomass, it is of great importance to find a way to change the compact structure in order to access cellulose and hemicellulose for the enzymes to break down and convert cellulose and hemicellulose into fermentable sugars.9 Thus, a pretreatment is necessary due to the resistant and crystalline structure of the lignocellulose, in order to enable enzymatic impact of the material. The access of enzymes to the material is dependent on the covering matrix of lignin.10
2 Biogas production from lignocelluloses
3 Biogasportalen.se
4 Chemical Process Technology, by Moulijn, Makkee and van Diepen
5 Biogas production from lignocelluloses
6 Current Anaerobic Digestion technologies used for treatment of municipal organic solid waste
7 Biogas from lignocellulosic biomass
8 Biogas production from lignocelluloses
9 Biogas from lignocellulosic biomass
10 Alkali pretratment of softwood spruce and hardwood birch by NaOH/thiourea, NaOH/urea, NaOH/urea/thiourea, and NaOH/PEG to improve ethanol and biogas production
The main components of lignocelluloses; cellulose, hemicellulose and lignin, have different functions and chemistry in different materials. Cellulose is the main component of higher plants consisting of a polymer chain of glucose units, which in turn form tight aggregates of 30-60 parallel cellulose chains in the form of a three dimensional microfibril, which in turn assembles into cellulose fibres.11 The cellulose chains are stabilized by hydrogen and van der Waal bonds, each unit linked by two intra-chain hydrogen bonds and two to three inter-chain bonds, making the chains tightly packed and stable. Hemicelluloses are different types of branched, polysaccharides in the cell wall, found in the matrix phase, attached to cellulose fibres by hydrogen bonds. In hardwood and agricultural plants, xylan is the dominant hemicellulose, whereas for softwoods, it is glucomannan. Lignins are complex, amorphous, branched polymers constructed of different phenylpropane units which are keeping the lignocellulosic structure together by inter-linkages between lignin and hemicelluloses as well as lignin and cellulose. The linkages can be ester, ether or glycosidic bonds, which are causing the strong linkage that makes lignin extremely resistant to biodegradability.
Regarding lignocellulosic structure, for higher plants the cell walls consists of a primary wall (the outermost wall, usually not lignified) and a secondary wall. The secondary wall is thicker and stronger than the primary wall, providing strength and flexibility during the cell growth, and contains a large amount of lignin. The secondary wall is often further divided into three layers, in which cellulose are sorted in different directions in each layer. In the molecular structure of the cell walls, crosslinking between different polymers contributes to the complexibility. In the microfibrils, the cellulose is ordered in a crystalline manner. The microfibrils are further surrounded by a matrix of different polymers, such as lignin and hemicelluloses, connected to each other by covalent and non-covalent bonds forming a three dimensional structure. These are further connected to the matrix polymers, and the different polymers in the gel matrix are connected to each other. This complex lignocellulosic structure is protecting the cell walls from microbial and enzymatic degradation.
Even though the lignocellulosic feedstocks have the protective lignin matrix in common, there are some differences in structure that may influence the degradability of the materials. Agricultural residues have the advantage that in most cases they are easier to degrade in comparison with forest residues. This is due to the lower lignin content, as well as the dimension of the straw being relatively small, which results in a material that is more easily accessible for the microbial enzymes.12 Also, there is usually more difficult to hydrolyse softwoods than hardwoods because of the higher lignin content of softwoods than that of hardwoods.13
1.4 Enzymatic petreatment
The complex structure of the protective lignin in the cell wall of the ligncellulosic material is a way to prevent microbial and enzymatic degradation of the plant by physically shield cellulose and hemicellulose. The access of substrate towards enzymes is not only dependent on the interaction bonds within the cell wall but also on the lignin content and the cellulosic crystallinity of the
11 Strategies to enhance conversion of lignocellulosic biomass to fermentable sugars and to enhance anaerobic digestion of algal biomass for biogas production
12 Biogas production from lignocelluloses
13 Alkali pretreatment of softwood spruce and hardwood birch by NaOH/thiourea, NaOH/urea, NaOH/urea/thiourea, and NaOH/PEG to improve ethanol and biogas production
material. Also, the surface area of the material is of importance since a higher accessible surface area contributes to more available substrate towards enzymes.
In order to make the material more accessible to enzymes, the compact structure of lignin should be broken down and the crystallinity should be reduced, I.e the long-chained cellulose and hemicellulose would be converted into fermentable sugars. In this way, the porosity of the material is increased, causing the carbohydrate to be more accessible to enzymes and increasing the solubilisation of the substrate. Thus, the pretreatment would result in improved biogas yield during the subsequent fermentation step of the process. Although, it is important that valuable fractions are preserved without degrading, no losses of organic matter occur and that the formation of inhibitors is limited during pretreatment step.
There are different pretreatment methods that can be used. The pretreatments of lignocelluloses prior to biogas production have the same objectives compared to pretreatments prior to ethanol production. However, there are some exceptions; the pretreatments can be less extensive for the anaerobic biogas production since the organisms are, to some extent, able to degrade both the crystallinity as well as the chain length of cellulose and hemicelluloses by themselves, which are two of the main functions of enzymatically pretreatments. Even though the pretreatment may fulfil the requirements for biogas production, from a broad point of view, the requirements of enzymatic pretreatments are to; improve the degradability by access cellulose and hemicelluloses to the enzymes, avoid degradation or losses of organic matter, avoid production of potential inhibitors during the process, be cost and energy efficient and contribute to as low impact as possible on the environment.
Several pretreatment techniques have been studied with respect to the degradability of lignocellulosic substrates, each method having their advantages and disadvantages. The different types of pretreatments may be divided into mechanical, thermal, chemical and biological treatment.
A combination of these methods is possible as well as a method called co-digestion, in which the reactor is loaded with an advantageous blend of different types of organic substrates, making the bioprocess to balance itself at a low cost. Both mechanical, thermal and chemical pretreatments requires high energy input for improved conversion of biomass and often requires expensive instruments and chemicals during the process. By using microorganisms in order to enhance the degradability of biomass, the advantages offered are low-capital cost and low energy demand as well as the environmental advantage. However, this pretreatment requires long resident time due to the usually low rate of the biological hydrolysis.1415
1.5 Biogas production
In order to convert the organic material into methane and carbon dioxide, I.e biogas, the material has to be degraded by microorganisms in anaerobic environment, usually called fermentation. The substrate to be fermented is added to the reactor along with inoculum, in which the necessary microorganisms is found. In order to ensure maintenance of the microorganisms and ensure the proceeding production of biogas, an appropriate nutrient solution is needed which includes energy sources such as proteins, fats and carbohydrates for the activity of the microorganisms. The
14 Biogas from lignocellulosic biomass
15 Biogas production from lignocelluloses
nutrients needed is divided into macronutrients; carbon, nitrogen, hydrogen, phosphorus, potassium and sulphur, micronutrients; cobalt, copper, iron, molybdenum, nickel, selenium, tungsten and zinc, and vitamins. Some metal ions is important for the function of the microorganisms, so called trace metals, which might be added either to the reactor or added through the composition of the loading of the reactor.
The substrate determines the biogas yield due to the composition of the material, the total volatile solids (VS) of the material determines the fraction of organic material that is available for digestion.16The carbon/nitrogen ratio, C/N, is also an important factor of the process, if the ratio is too low there is a risk of ammonia inhibition and if it is too high it might lead to lower methane yield due to the deficiency of nitrogen available for the microorganisms.
Anaerobic digestion is a biological process in which organic carbon is degraded, in the absence of oxygen, to the most reduced state (methane) and the most oxidizing state (carbon dioxide). The over-all process can be divided into four phases: hydrolysis, acidogenesis, acetogenesis and methanogenesis.17 In the hydrolysis phase, the material is degraded into monomers performed by extracellular hydrolytic enzymes, which uses water to cut the bonds between the polymers. The hydrolysis of lignocelluloses, and other complex structured compounds, requires time and the degradation might not be completed. Acidogenesis is the phase where the monomers are further degraded into short-chain organic acids of one to five carbons, alcohols, hydrogen, ammonia and carbon dioxide. Some of the further degraded products can be used directly by methanogenes, whereas longer carbon compounds are degraded into acetic acid, hydrogen and carbon dioxide in the acetogenic phase. The last phase is the methanogenesis in which methane is produced from acetate, carbon dioxide, hydrogen, methylamines, alcohols and formate.
The degradation of raw material into biogas may be limited by the biodegradability and eventual inhibitors produced in the process.18 For this reason, it might be necessary to add some detoxifying agents to the process in order to reduce any inhibitors that might affect the production of biogas.
The amount of inhibitors that might appear during the production of biogas is dependent on the raw material used and on the conditions of pretreatment and fermentation.19 The amount of sulphur and nitrogen in the feedstock is affecting the production of trace gases, such as the semi-harmful contaminants hydrogen sulphide and ammonia as well as other trace gases such as nitrogen and hydrogen, but since these elements are nutrients required by the bacteria, they can't completely be eliminated from the feedstock.20
1.6 Previous studies
In a previous study, alkali pretreated, as well as untreated, spruce and birch was hydrolyzed using 5% w/v dry matter of an enzyme solution containing 50 mmol/l sodium citrate buffer at pH 4,8 with 20 fpu Celluclast/Novozyme (Celluclast 1.5L, Novozyme, Denmark and Novozyme 188, Novozyme, Denmark) per gram of dry wood. The hydrolysis was carried out at 45°C at 120 rpm for
16 Biogas production from lignocelluloses
17 Strategies to enhance conversion of lignocellulosic biomass to fermentable sugars and to enhance anaerobic digestion of algal biomass for biogas production
18 Biogas production from lignocelluloses
19 Comparison of strategies to overcome the inhibitory effects in high-gravity fermentation of lignocellulosic hydrolysates
20 Current Anaerobic Digestion technologies used for treatment of municipal organic solid waste
72 h with good results. Pretreated birch showed a large improvement of the saccharification, 83% to the theoretical yield. The corresponding improvement of pretreated spruce was 57%. Regarding the biogas yield for pretreated birch and spruce, the yield was improved by 56% and 600%, respectively.21 The same approach as the method described above was used when the production of ethanol and biogas from NMMO pretreated birch was investigated in another study, the main difference being that 0,1 g/l of surfactant was added to each flask and the hydrolysis was performed during 96 h. Results for untreated and pretreated birch showed that the total crystallinity of birch decreased by 18,2% and 49,7%, respectively. Also, the hydrolysis of birch showed significant improvement when pretreated with NMMO. The corresponding methane yield of pretreated birch was increased to about 80% of the theoretical yield.22
In a study concerning inhibitory effects in fermentation of lignocellulosic hydrolysates, the influence on ethanol yield was investigated by a separate hydrolysis and fermentation as well as a simultaneous saccharification and fermentation. Spruce slurry, prepared from spruce wood chips pretreated with diluted acid (SO2 in water) at SEKAB E-Technology AB (Örnsköldsvik, Sweden), was used as substrate during the trials. The reducing agent sodium dithionite (Na2S2O4) was added to the process as a detoxification method. There was a clear beneficial effect on the fermentability at an amount of 0.01 mol/dm3 Na2S2O4 (as the final concentration in the medium), whereas a higher amount of Na2S2O4 gave no significant improvement to the process. The lower concentration of dithionite contributed to an ethanol yield of 57% of the maximum theoretical yield of dilute acid- pretreated spruce. The study also indicated that if detoxification was performed, the need for nutrients was less crucial (except for addition of an inorganic nitrogen source).23
In the same study, laccase was stated to be useful for the removal of inhibitors from hydrolysates.
Another study stated that laccase, together with steam explosion, had a positive effect on the anaerobic digestion of digested biofibres. However, there was no improvement when the biofibres were only pretreated with laccase prior to anaerobic digestion.24
21 Alkali pretreatment of softwood spruce and hardwood birch by NaOH/thiourea, NaOH/urea, NaOH/urea/thiourea, and NaOH/PEG to improve ethanol and biogas production
22 Ethanol and biogas production from birch by NMMO pretreatement
23 Comparison of strategies to overcome the inhibitory effects in high-gravity fermentation of lignocellulosic hydrolysates
24 Biogas production from lignocelluloses
Biogas production from enzymatically pretreated
2
agricultural and forest raw material
The materials to be used during the experiments were: wheat straw, spruce, pine and birch. In addition, each of these four materials can be divided into untreated and pretreated, since the each material is pretreated in either one or two different kinds of pretreatments. Spruce, pine and birch was received as pretreated slurries of low pH from SEKAB. Each of these materials was hydrothermally pretreated with sulphuric acid, specifications according to Table 2.1.
Hydrothermally pretreated wheat straw was provided as solids by Biotechnology lab, School of chemical engineering, National Technical University of Athens, Greece, specifications given in Table 2.1.
Table 2.1 Pretreatment specifications of the materials used.
Hydrothermal pretreatment with sulphuric acid
Spruce, toughtreated 212°C for 4-8 minutes and pH 1,6-1,8 Spruce, mildtreated 200°C for 4-8 minutes and pH 1,8-2,0
Pine 210-215°C for 5 minutes and pH 1,5-1,7 Birch 190°C for 4-6 minutes and pH 1,8-2,0
Hydrothermal pretreatment
Wheat straw, toughtreated 195°C for 17 minutes Wheat straw, mildtreated 185°C for 12 minutes
The inoculum used during the project was collected from the biogas plant in Boden which uses food wastes and municipal waste waters. The inoculum works under thermophilic conditions (55°C).
Analysis of total solids (TS) and volatile solids (VS) was performed for each material and each new batch of inoculum collected in which the total solids was measured after drying at 100°C, whereupon the material was heated gradually to 550-600°C for 2-3 hours. The ash of the material was then weighed and volatile solids were calculated according to:
Following equipment was used during the experiments:
Photospectrometer: light source shines through a monochromator and an output wavelength is beamed at the sample. A fraction of the monochromatic light is transmitted through the sample and to the detector. The photospectrometer measures the quantity of light transmitted through the substance relative to a reference substance.
Thermomixer: combines mixing with temperature control of the samples.
Automatic Methane Potential Test System (AMPTS): an analytical device that measure bio- methane flows produced during anaerobic digestion at laboratory scale. The gas volume measurements and data logging is automatic and is made on-line. The device consist of a sample incubation unit, in which reactor bottles is temperature controlled, a carbon dioxide trap unit, in which 3 M NaOH with pH indicator traps the carbon dioxide formed during the process, and a flow cell array unit, in which liquid displacement and buoyancy of water is measured as the methane gas enters the unit.
𝑉𝑆 = 𝑇𝑆 − 𝑎𝑠ℎ
2.1 Hydrolysis
In order to investigate the reducing sugar content of each material, i.e. the ability of the enzyme treatment to break the complex shield and release the sugars, hydrolysis was performed. The sugars released may be fermented into biogas once released. Since some of the materials to be investigated were pretreated as slurries, it is crucial to investigate how the slurry state of the material is affecting the overall hydrolysis of the material. In order to do this investigation, each of the materials was used as washed and unwashed. The unwashed material is the initial slurry state of the materials, which was dried at 80°C in order to get a solid, powdery, material. By washing the slurry with destillated water, the material was rinsed from pretreating chemicals that might have been inhibiting the hydrolysis and allowed the pH of the sample to increase to about 7 (which indicated that the material was properly washed). Also, by washing the materials from the slurry, any dissolved xylan in the liquid fraction was removed. Thus, preventing any xylan from being hydrolysed along with the solid material for which the release of sugars an important investigation. The water was filtrated off the material, resulting in thick slurry which was dried at 80°C. Due to these pretreatments of the already pretreated slurry materials, a total amount of 14 materials was investigated by hydrolysis, see Table 2.2.
Table 2.2 Materials used during hydrolysis.
Materials
Wheat straw Toughtreated Mildtreated
Untreated
Spruce
Toughtreated, slurry removed Toughtreated, slurry
Mildtreated, slurry removed Mildtreated, slurry
Untreated
Pine Pretreated, slurry removed Pretreated, slurry
Untreated
Birch Pretreated, slurry removed Pretreated, slurry
Untreated
Each of the 14 materials where enzymatically hydrolyzed by CMAX and Celluclast/Novozyme 5:1, using 3% w/v dry matter of material in a 100 mmol/l di-sodium phosphate-citric acid buffer solution of pH 5 (also containing 0,01% sodium azide in order to protect the enzymes from harmful microorganisms).
Table 2.3 Specification of the enzymes used during hydrolysis.
Enzyme specifications
CMAX AlternaFuel CMAX, provided by DYADIC 86,22 fpu/ml works at 60°C Celluclast/Novozyme Celluclast 1.5l, provided by Novozyme, Denmark
Novozyme 188, provided by Novozyme Denmark
83 fpu/ml
(5:1 v/v) works at 50°C The hydrolysis was performed at different activities of each the enzymes, i.e. at different values of fpu (filter paper unit) per gram of substrate, in order to optimize the hydrolysis. The materials were continuously hydrolyzed for 24 hours at 1000 rpm and 60°C for CMAX and 50°C for
Celluclast/Novozyme whereupon the samples were analyzed for reducing sugars.
When preparing for the analysis of the hydrolysis, a DNS calibration curve was set up. An initial solution of DNS (3,5-dinitrosalicylic acid), solution specifications found in Table 2.4, in order to investigate how the absorbance of DNS changes as the concentration of the DNS solution changes by dilution. The measurements were achieved in a photospectrometer at 540 nm, giving a calibration curve of the DNS.
Table 2.4 Specification of the DNS-solution used during hydrolysis.
DNS-solution
200 ml NaOH 8% w/v (16 g of NaOH pellets) 500 ml water
10 g DNS
402,7 g potassium sodium tarte
Finally, the volume was fixed with water to 1000 ml.
To each hydrolysed sample, diluted DNS was added to a sample of the liquid phase of the sample, i.e. the supernatant, and was put into a 100°C waterbath for 5 minutes for the reaction of the DNS to occur. After the reaction, the samples were further diluted with water. The absorbance of the diluted samples was thereafter measured at 540 nm in a photospectrometer. The DNS calibration curve obtained during the preparation of the hydrolysis was used along with the absorbance of the samples in order to calculate the total amount of reducing sugars in the sample after hydrolysis was performed.
2.2 Prehydrolysis
In addition to the already pretreated slurries of the materials, a pretreatment step was added to all of the materials, including the already pretreated and untreated materials, in order to link the results of the hydrolysis step to the reactor experiments to follow. This pretreatment step was a prehydrolysis step prior to the reactor experiment, for which the prehydrolyzed material was used in the reactor experiments. Enzyme solution was added to a solid concentration of 10% w/w material. The amount of enzyme to be added to the material was calculated for a certain value of fpu per gram volatile solids of the material, according to the results from the hydrolysis step. The material, along with the enzyme solution added, was incubated for 12 hours at the specific working temperature for each of the enzymes. After 12 hours, the prehydrolyzed material was allowed to cool down to about room temperature before it was used in the preparation of the reactors.
2.3 Reactor experiments
In the reactor experiments, the ability of the different materials to produce biogas was investigated.
A total amount of 10 different materials was used during the reactor experiments, see Table 2.5. The reactor was loaded with substrates and inoculum according to the volatile solid (VS) inoculum/substrate ratio:
The methane potential, ml CH4/g volatile solid (VS), were calculated according to the amount of biogas produced relative to the volatile solids of the substrate:
Table 2.5 Materials used during reactor experiments.
Reactor materials
Wheat straw Toughtreated Mildtreated Untreated
Spruce Toughtreated Mildtreated Untreated Pine Pretreated
Untreated Birch Pretreated
Untreated
Fresh inoculum from the biogas plant in Boden was used during every reactor experiment. Also, the ability of the inoculum itself to produce biogas was investigated in order to be able to deduct any biogas that might have been produced by the inoculum itself during the reactor experiments. For the same reason, enzymes were added to inoculum alone to investigate how much biogas the enzymes is producing.
The ability of each material to produce biogas, in inoculum, was investigated along with the ability to produce biogas when adding enzymes to the reactor and when the material to be used was already prehydrolyzed (as described in the previous section; 1.4 Enzymatic pretreatment). Also, according to section 1.6 Previous studies, laccase (added to the material for 12 h incubation before preparation of the reactors) and dithionite (1 mM and 10 mM) was added to investigate whether any detoxification was necessary in order to increase the biogas yield. Laccase was also added to the bottles of spruce and pine along with the enzymes. Additionally, three different salt solutions were added in order to feed the microorganisms with nutrients for their maintenance during the reactor experiments.
The biogas producing experiments were continued until the accumulated gas production ceased, I.e.
when the daily methane production became approximately less than 1% of the total accumulated amount of methane.
𝑚𝑙 𝐶𝐻4
⁄𝑔 𝑉𝑆= 𝑚𝑙 𝐶𝐻4 (𝑠𝑢𝑏𝑠𝑡𝑟𝑎𝑡𝑒)− 𝑚𝑙 𝐶𝐻4 (𝑖𝑛𝑜𝑐𝑢𝑙𝑢𝑚)×𝑉𝑖𝑛𝑜𝑐𝑢𝑙𝑢𝑚
𝑉𝑡𝑜𝑡𝑎𝑙 − 𝑚𝑙𝑒𝑛𝑧𝑦𝑚𝑒𝑠 𝑉𝑆𝑠𝑢𝑏𝑠𝑡𝑟𝑎𝑡𝑒
2.4 Results
According to the calibration curve in Figure 2.1 the equation obtained in order to calculate the reducing sugar content during the hydrolysis experiments is:
Where y represents the concentration, g/l, of reducing sugars of the sample and x is the measured absorption of the sample.
Figure 2.1 Calibration curve of the DNS-solution used during the hydrolysis experiments, representing how the concentration changes with the measured absorption of the sample.
Hydrolysis
During the hydrolysis trials, there was a distinct indication that CMAX was not suitable for hydrolysis of spruce, pine and birch. Thus, the hydrolysis of these materials was performed, as triplicates for accuracy, with Celluclast/Novozyme at 15 fpu. However, for wheat straw CMAX appeared to be suitable, which is why the following methane potential trials for wheat straw was performed with both CMAX and Celluclast/Novozyme at 10 fpu respectively. These results are to be found in Appendix 5.1.
For the untreated materials, the highest net reducing sugar content after 12 hour hydrolysis was obtained by wheat straw, followed by spruce, pine and birch. Even as pretreated, wheat straw obtained the highest net reducing sugar content of all of the materials (mildtreated wheat straw being the most effective pretreatment), see Figure 2.2. However, the greatest increase of net reducing sugars when comparing untreated and pretreated materials is obtained by birch, which contributes to an increase of about 11 times the content when birch is pretreated and the slurry is removed than when untreated, see Figure 2.5. Pretreatment of pine is least effective, only contributing to a net sugar content of 3,8 times the net sugar content compared to untreated pine, Figure 2.4.
y = 1,8462x R² = 0,9756
0 0,5 1 1,5 2 2,5
0 0,2 0,4 0,6 0,8 1 1,2
Concentration on DNS (g/L)
Absorption
DNS calibration curve
Absorption Linjär (Absorption)
𝑦 = 1,8426𝑥
Figure 2.2 Net total reducing sugar content after 12 h hydrolysis of wheat straw materials.
For spruce alone, Figure 2.3, toughtreated spruce resulted in higher net reducing sugar content than mildtreated. However, the results for spruce are not consistent due to the net reducing sugar content when the slurry is removed and not. In this case, the toughtreated is most effective when slurry is removed whereas mildtreated is most effective when the mildtreated spruce slurry is hydrolyzed.
Figure 2.3 Net total reducing sugar content after 12 h hydrolysis of spruce materials.
Pretreatment of pine, Figure 2.4, has been shown by comaprison with the hydrolysis results of the other materials to be the least effective approach concerning the hydrolysis part of the project.
Pretreated birch, Figure 2.5, has been shown to be the most effective material to be hydrolysed by Celluclast/Novozyme at 15 fpu. However, both pine and birch show consistant hydrolysis results concerning the removal of slurry from each of the pretreated materials to be the most effective approach for these materials.
11,26
12,50
1,80
0,00 2,00 4,00 6,00 8,00 10,00 12,00 14,00
Net total reducing sugars (g/L)
Wheat straw
Wheat straw, toughtreated Wheat straw, mildtreated Wheat straw untreated
7,12
4,02
2,55
3,83
1,21
0,00 2,00 4,00 6,00 8,00
Net total reducing sugars (g/L)
Spruce
Spruce, toughtreated slurry removed Spruce, toughtreated slurry Spruce, mildtreated slurry removed Spruce, mildtreated slurry Spruce, untreated
Figure 2.4 Net total reducing sugar content after 12 h hydrolysis of pine materials.
Figure 2.5 Net total reducing sugar content after 12 h hydrolysis of birch materials.
Methane potential
At first, the methane potential of the different untreated and pretreated materials, with inoculum alone and with inoculum along with enzymes, was studied as duplicates. As in the case with the hydrolysis results, wheat straw contributed to the highest methane potential for each of the untreated materials, I.e. being the most effective material as untreated according to the methane potential obtained, see Figure 2.6. Regarding the pretreated materials, birch is the most effective material contributing to a methane potential of 1,5 times higher methane potential than the second most effective pretreated material, mildtreated wheat straw, see Figure 2.9. In Figure 2.8 it is clear that the biggest improvement of pretreating the material according to the method of SEKAB is made by pine using Celluclast/Novozyme at 15 fpu, giving an increase of 26,5 times higher methane potential compared to the methane potential obtained by using untreated pine.
The methane potential for wheat straw was slightly improved when pretreated wheat straw was used instead of untreated compared to the results of spruce, pine and birch. The mildtreated wheat straw
2,76
2,47
0,73
0,00 0,50 1,00 1,50 2,00 2,50 3,00
Net total reducing sugars (g/L)
Pine
Pine, slurry removed Pine, slurry Pine, untreated
5,57
4,75
0,50 0,00
1,00 2,00 3,00 4,00 5,00 6,00
Net total reducing sugars (g/L)
Birch
Birch, slurry removed Birch, slurry Birch, untreated
was marginally more effective than toughtreated wheat straw and CMAX contributed to a slightly higher improvement than Celluclast/Novozyme, Figure 2.6.
Figure 2.6 Methane potential of wheat straw, with and without enzymes added to the bottle. The activity is 10 fpu for both CMAX and Celluclast/Novozyme.
For spruce, the methane potential was increased as the activity of enzymes was increased, Figure 2.7. However, the improvement of methane potential for pretreated spruce compared to untreated spruce is about 22 times higher for both Celluclast/Novozyme at 15 fpu and at 30 fpu.
Figure 2.7 Methane potential of spruce, with and without enzymes added to the bottle. Celluclast/Novozyme was used at 15 and 30 fpu.
159,9 167,74 173,06 174,78
197,96
175,54
90,98
129,6
103,65
0 50 100 150 200 250
Wheat straw CMAX Celluclast/Novozyme
Methane potential (mL CH4/g substrat)
Wheat straw
Wheat straw, toughtreated Wheat straw, mildtreated Wheat straw untreated
146,37
210,2
276,28
95,36
232,08
259,41
9,47 10,33 12,03
0 50 100 150 200 250 300
Spruce Spruce 15 fpu Spruce 30 fpu
Methane potential (mL CH4/g substrat)
Spruce
Spruce, toughtreated Spruce, mildtreated Spruce untreated
Figure 2.8 Methane potential of pine, with and without enzymes added to the bottle. The activity is 10 fpu for both CMAX and Celluclast/Novozyme.
Even though pine is most effective regarding methane potential when the material is pretreated compared to untreated, pretreated birch contributes to the highest methane potential of all the materials. For pretreated birch, the fermentation is equally effective when Celluclast/Novozyme is used both at 15 fpu and 30 fpu, see Figure 2.9.
Figure 2.9 Methane potential of birch, with and without enzymes added to the bottle. The activity is 10 fpu for both CMAX and Celluclast/Novozyme.
143,32
179,88
165,81
5,41 7,41 15,77
0 50 100 150 200
Pine Pine 15 fpu Pine 30 fpu
Methane potential (mL CH4/g substrat)
Pine
Pine slurry Pine untreated
254,07
284,50 284,26
17,49 32,26 42,03
0 50 100 150 200 250 300
Birch Birch 15 fpu Birch 30 fpu
Methane potential (mL CH4/g substrat)
Birch
Birch slurry Birch untreated
In the detoxification trial laccase and dithionite was added to the bottles in order to investigate if detoxification is necessary to improve the methane potential of the pretreated materials even further.
The results of the detoxification is compared to the results of the initial trials for methane potential.
For wheat straw, comparing Figure 2.6 and Figure 2.10, there is generally no improvement of the methane potential when detoxificiation is applied. From Figure 2.10 it is clear that the higher concentration of dithionite, 10 mM, causes a severe decrease in methane potenital for wheat straw.
Figure 2.10 Methane potential of wheat straw, when detoxification is performed during the trial.
Comparing the results of spruce in Figure 2.7 and Figure 2.11, improvement of methane potential is shown for toughtreated spruce when laccase is used as detoxifying agent. As in the case of wheat straw, there is a severe decrease in methane potential when the higher concentration of dithionite is added.
Figure 2.11 Methane potential of spruce, when detoxification is performed during the trial.
Comparing the results of pine, Figure 2.8, and birch, Figure 2.9, to the results of the detoxification trial in Figure 2.12 there is a slight increase in methane potential for pine when laccase is used as
143,86
30,03
101,22 152,39
29,15
109,21
0 20 40 60 80 100 120 140 160 180
Laccase 10 mM dithionite 1 mM dithionite
Methane potential (mL CH4/g substrat)
Wheat straw, detoxification
Wheat straw, toughtreated Wheat straw, mildtreated
167,26
34,05
105,08 90,38
14,68
47,24
0 20 40 60 80 100 120 140 160 180
Laccase 10 mM dithionite 1 mM dithionite Methane potential (mL CH4/g substrat)
Spruce, detoxification
Spruce, toughtreated Spruce, mildtreated
the detoxifying agent. However, there is no improvement for birch either when laccase or dithionite is used.
Figure 2.12 Methane potential of pine and birch, when detoxification is performed during the trial.
According to the results of the initial hydrolysis part of the project, Celluclast/Novozyme showed to be more effective and is therefore used as enzyme for the prehydrolysis of spruce, pine and birch.
Since CMAX showed some advantages during the hydrolysis of wheat straw, both CMAX and Celluclast/Novozyme is used during the prehydrolysis of wheat straw at 10 fpu respectively.
Comparing the methane potential for wheat straw when enzymes was added to the bottles in Figure 2.6 to the prehydrolysis of wheat straw at the same activity of 10 fpu in Figure 2.13, prehydrolysis resulted in a quite lower methane potential for CMAX and Celluclast/Novozyme respectively.
Figure 2.13 Methane potential of prehydrolyzed wheat straw, using CMAX and Celluclast/Novozyme at 10 fpu, respectively.
Comparison of the methane potential for prehydrolysis of spruce, Figure 2.7 and Figure 2.14, indicates, as in the case of wheat straw, on a decreased methane potential when spruce is
157,43
99,23 250,85
209,33
0 50 100 150 200 250 300
Laccase 1 mM dithionite
Methane potential (mL CH4/g substrat)
Pine and birch, detoxification
Pine slurry Birch slurry
162,85
166,3
169,62 170,47
158 160 162 164 166 168 170 172
CMAX Celluclast/Novozyme
Methane potential (mL CH4/g substrat)
Wheat straw, prehydrolysis
Wheat straw, toughtreated Wheat straw, mildtreated
prehydrolyzed with Celluclast/Novozyme at both 15 fpu and 30 fpu.
Figure 2.14 Methane potential of prehydrolyzed spruce, using Celluclast/Novozyme at 15 fpu and 30 fpu.
Prehydrolysis of pine, Figure 2.15, contributed to a slight increase in methane potential when pine is prehydrolyzed with Celluclast/Novozyme at 30 fpu compared to the trial where Celluclast/Novozyme at the same activity was added to the reactor bottles, Figure 2.8. For 15 fpu, no improvement of methane potential was showed.
Comparison of the methane potential for prehydrolysis of birch, however, Figure 2.7 and Figure 2.14, shows that the methane potential is improved at about 6-7% for prehydrolyzed birch at both 15 fpu and 30 fpu of Celluclast/Novozyme.
Figure 2.15 Methane potential of prehydrolyzed pine and birch, using Celluclast/Novozyme at 15 fpu and 30 fpu.
205,13
247,67
136,02
170,35
0 50 100 150 200 250 300
15 fpu 30 fpu
Methane potential (mL CH4/g substrat)
Spruce, prehydrolysis
Spruce, toughtreated Spruce, mildtreated
156,98 176,17
304,80 305,49
0 50 100 150 200 250 300 350
15 fpu 30 fpu
Methane potential (mL CH4/g substrat)
Pine and birch, prehydrolysis
Pine, slurry Birch, slurry
2.5 Analysis
During the initial studies of hydrolysis, there was a distinct indication that CMAX was not suitable for hydrolysis of spruce, pine and birch. Thus, the hydrolysis of these materials was performed with Celluclast/Novozyme at 15 fpu. For wheat straw CMAX appeared to be suitable, thus, the following methane potential trials for wheat straw was performed with both CMAX and Celluclast/Novozyme at 10 fpu respectively.
Hydrolysis
After 12 hour hydrolysis with Celluclast/Novozyme at 15 fpu, wheat straw obtained the highest net reducing sugar content of all the materials used, both as untreated as well as pretreated. However, the greatest ratio of net reducing sugars comparing the untreated and pretreated materials was obtained by birch. When the slurry was removed from the pretreated birch the hydrolysis resulted in about 11 times higher net reducing sugar content than hydrolysis of untreated birch. Pretreated pine was the least effective material hydrolysed, only resulting in a net reducing sugar content of 3,8 times higher than when untreated. Hydrolysis of toughtreated spruce resulted in higher net reducing sugar content than of mildtreated. However, the results are not consistent since the results of removal of slurry of the two pretreated materials did not correspond. In the case of pine and birch, the results are consistent concerning the removal of slurry since in both cases the removal of slurry contributed to higher net reducing sugar content.
Methane potential
As in the case of the hydrolysis, wheat straw obtained the highest methane potential of the untreated materials. For the pretreated materials, pretreated birch obtained the highest methane potential, about 1,5 times higher for mildtreated birch than for the second most effective pretreated material, mildtreated wheat straw. However, the biggest improvement of pretreated material compared to untreated material is made by pine, contributing to an increase of 26,5 times higher methane potential compared to the methane potential obtained for untreated pine using Celluclast/Novozyme at 15 fpu. The least effective approach was the pretreatment of wheat straw, for which mildtreated wheat straw (the second most effective pretreated material) only contributed to a methane potential ratio of about 1,8 times the methane potential obtained by untreated wheat straw. Mildtreated wheat straw was marginally more effective than toughtreated wheat straw and CMAX contributed to a slightly higher improvement in methane potential than with Celluclast/Novozyme. The methane potential of spruce was increased as the activity of Celluclast/Novozyme was increased, but the improvement of methane potential for pretreated spruce compared to untreated spruce is about 22 times higher for both Celluclast/Novozyme at 15 fpu and at 30 fpu.
In the detoxification trial, there was generally no improvement in methane potential for wheat straw when detoxification was applied. Toughtreated spruce showed some improvement of methane potential when laccase was used as detoxifying agent. In the case with both wheat straw and spruce, the higher concentration of dithionite, 10 mM, was causing a severe decrease in methane potential during detoxification. As in the case of spruce, the methane potential of pine was also improved when laccase was used as the detoxifying agent. Birch, however, showed no improvement in methane potential during detoxification, either with laccase or with dithionite.
Prehydrolysis of the materials showed no improvement for wheat straw, rather an overall decrease in methane potential for prehydrolyzed wheat straw. As well as spruce indicated on an overall decrease in methane potential when spruce was prehydrolyzed compared to the trial where enzymes was added to the reactor bottles. Prehydrolysis of pine resulted in a slight increase in methane potential when pine is prehydrolyzed with Celluclast/Novozyme at 30 fpu, whereas prehydrolysis of birch contributed to an increase in methane potential of about 6-7% at both 15 fpu and 30 fpu of Celluclast/Novozyme.
Discussion 3
The initial studies of hydrolysis indicated that CMAX was less suitable for hydrolysis of spruce, pine and birch, whereas CMAX showed indications on being suitable for hydrolysis of wheat straw.
These results might depend on the effectiveness of the enzymes, Celluclast/Novozyme being more effective for enzymatic hydrolysis of more complex lignocellulosic materials such as spruce, pine and birch, see section 1.3 Lignocellulosic biomass.
The fact that wheat straw obtained the highest net reducing sugar content during hydrolysis, both as pretreated as well as untreated, is an expected outcome due to section 1.3 Lignocellulosic biomass where it is stated that agricultural raw materials are usually more easily hydrolyzed than forest raw materials. As for the investigation of the methane potential of each of the materials, wheat straw obtained the highest methane potential of the untreated materials, which is consistent with the results of the hydrolysis trials and is proving the theory of wheat straw being easier to hydrolyze.
However, during the methane potential trials, mildtreated wheat straw was the least effective material of all the materials. One possible explanation might be that wheat straw is already easily hydrolyzed and thus most of the lignocelluloses of the material are hydrolyzed. When pretreatment is applied there are just some inaccessible areas of the material that is even further revealed thus resulting in the low effectiveness of pretreating wheat straw in order to enhance the process. In other words, wheat straw is already efficiently hydrolyzed without any pretreatment. Of course, there might be some economical and practical advantages since there are in fact some improvement when wheat straw is pretreated, especially when mildtreated. However, it is not as improved as in the case of spruce, pine and birch.
Hydrolysis of birch resulted in the greatest ratio of net reducing sugars comparing the results of untreated and pretreated materials hydrolyzed. In the trials for methane potential of each material, pretreated birch obtained the highest methane potential of all the materials. This might be explained by the favorable composition of hardwoods, stated in section 1.3 Lignocellulosic biomass. Thus, sulfuric acid is allowed to more easily diffuse into birch and consequently improve the ability of the enzymes to break through the complex structure of the lignocellulosic material and reveal the valuable sugars for a successful fermentation to proceed.
During the methane potential trials, pine was the most effective material since it contributed to the biggest improvement of methane potential comparing the methane potential of pretreated and untreated material. However, during the hydrolysis trials, the least effective material according to the net reducing sugar content after 12 hour hydrolysis was the pretreated pine material. The most effective methane potential results was obtained when Celluclast/Novozyme was used at 15 fpu, which was the same enzyme and activity used when the least effective hydrolysis results was obtained. The results of hydrolysis and methane potential for pine is not consistent, but one possible explanation to this outcome might be that pine requires more than 12 hours of hydrolysis in order for the enzyme to effectively break through the material. Thus, during the reactor experiments for methane potential, which lasted for several days, the enzymes had the time to sufficiently release more of valuable the sugars of pine from which methane was produced during fermentation.
The results of spruce did not distinguish from the other materials during the trials, neither was it the most efficient material used nor was it the least efficient material used. During the reactor experiments, the methane potential of spruce was increased as the activities of the enzymes were increased. However, the ratio of the methane potential of pretreated spruce to untreated spruce was
