Optimering av lämpliga enzymer för enzymatisk flisförbehandling för energieffektiv TMP tillverkning

Transcription

1 Optimering av lämpliga enzymer för enzymatisk flisförbehandling för energieffektiv TMP tillverkning ÅForsk ref.nr Silvia Viforr, Tomas Mårtensson, Lennart Salmén October 2012

2 Acknowledgements Holmen Paper AB, Braviken mill, is acknowledged for the supply of the Impressafiner chips, as part of their engagement in the CRUW (Centre for Research on Ultrastructure of Wood fibres) program at SLU. The supply of enzymes by Novozymes A/S and the technical advice regarding their use by Kasper Klausen of Novozymes is greatly appreciated by the authors. Aalto University and Prof. Herbert Sixta is acknowledged for their contribution regarding the use of the Wing-refining facilities at the Department of Forest Products Technology. Mr Timo Ylönen of the department is also greatly acknowledged for his technical advice and help in running the refining experiments. M.Sc. Lars Widell of ÅF AB is acknowledged for technical advice.

3 Table of contents Page 1 Summary Introduction Background Wood structure Enzymatic pre-treatments Enzymes Materials and methods Materials Methods Results Laboratory enzyme treatments Pilot scale impregnation Refining trials Pulp properties Conclusions References Innventia Database information... 27

4 1 1 Summary In order to evaluate the possibilities of using enzymes for pre-treating softwood chips in order to lower the energy demand in mechanical pulping impregnation and refining trials were performed using a set of different enzymes. In small scale impregnation trials it was shown that the enzymatic activity highly depended on the size, i.e. the amount of destruction of the chips with more disintegrated wood chips rendering a higher release of sugars. Of the enzymes tested Pectinase and Xylanase resulted in the highest amount of released sugars while Mannanase and Pectin lyase gave considerably lower sugar releases. In larger scale impregnation trials the treatment indicated that most of the sugar released occurred in the first 60 minutes and that activities thereafter seemed to level off. Refining trials using a small Wing refiner on chips treated with Pectinase, Xylanase and Mannanase did not show any energy savings to a given freeness level. Comparing strength properties the differences were small considering tensile index and light scattering where the Mannanase treated chips even showed a lower strength development as a function of energy input. In conclusion it was apparent that wood chips, even after an Impressafiner treatment is too coarse for allowing enzymes to sufficiently break up the fibre-fibre bonding allowing for lowering the energy demands in mechanical pulping.

5 2 2 Introduction Thermomechanical pulping (TMP) is a highly energy demanding pulping process with an electrical energy consumption of usually more than 2,000 kwh/t. Due to the increasing energy costs as well as environmental considerations there is a high demand for ways of making the process more energy efficient. One approach is to modify the raw material, i.e. wood chips, prior to the refining, by mechanical chemical or enzymatic means. In this study pre-treatment using different types of enzymes has been evaluated. Enzymes are complex proteins that have specificity for targeting particular molecular structures, and are effective catalysts in biochemical processes even in low dosages. It is well known that a weakening of the outer cell wall structures, i.e. of the primary cell wall, by chemical means results in improved fibre separation and lower energy demand in refining. Thus enzymes that could specifically attack the structures of the primary cell wall would be of particular interest for evaluation. Here treatments of different pectinases of xylanase as well as of a mannanase have been evaluated.

6 3 3 Background 3.1 Wood structure For producing a good thermomechanical pulp at reasonable energy demands it is essential that a high degree of intact fibres may be separated, fibres that can be easily further refined. Thus the fibre fracture should preferably take part in the outer fibre wall layers (Franzén 1986)in the primary wall, S 1 region. Here the properties of the primary wall are of particular interest as its composition varies considerably from that of the other cell wall layers. The primary wall being a reminiscence from the living cell wall is constituted, apart from an disordered cellulose, of xyloglucans, pectins, proteins and lignin, with multiple interactions between the components, as illustrated in Figure 3.1 (Stevanic Srndovic 2011). Attacking the primary cell wall specifically would therefore be of advantage. It has earlier been demonstrated that with the use of low chemical dosages, specifically with bisulphite, a fracture in the primary wall occurs leading to lower energy demand in the subsequent refining (Axelson and Simonson 1982a, b, Axelson and Simonson 1983a, b, Westermark et al. 1987). Lignin Cellulose Pectin + Protein Xyloglucan Figure 3.1 Schematic picture of the interactions between the polymer components of the primary cell wall (Stevanic Srndovic 2011). Treating wood sections with specific enzymes, like proteases, have also been demonstrated to results in a fracture pattern in the primary wall with a fibrillated structure similar to that achieved with bisulphite treatments (Salmén and Pettersson 1995), see Figure 3.2. Thus enzymatic treatments using such enzymes that can break the connections between the primary wall components could be of specific interest; i.e. proteinases, pectinases or xylanases. The main problem attempting to use enzymes for pre-treatment of wood chips is the accessibility of the enzymes for the desired structure of modification, i.e. the primary cell wall. In general the size of enzymes are too large to access the cell wall of intact chips and access may only be gained via ray cells and boarded pits into the middle lamella/primary wall region. Thus a mechanical destructuring of the chips, as in an Impressafiner is highly recommended for increasing the possibility of an improved accessibility.

7 4 Figure 3.2 Microscopic images of fracture pattern in wood chips after a mild sulphonation, left and after an enzymatic treatment with proteases removing protein from the primary cell wall, right (Salmén and Pettersson 1995). 3.2 Enzymatic pre-treatments It was not until rather recently that the introduction of enzymes as a pre-treatment of wood chips was studied, for example see (Richardson et al. 1998, Peng et al. 2003). Since then, tests in both laboratory and mill scale have been performed but relatively few reports have been published on pre-treatment of wood chips and results from experiments suggest a non-conformity regarding the possibility of introducing reductions in energy consumption during refining. Table 1.1 summarises studies regarding the implementation of enzyme pre-treatments of wood chips for TMP. In many cases energy reductions were documented at a specific freeness level. However when comparing pulps at similar tensile strengths the situation is mixed and often no energy savings were reported. Pre-treatments using cellulases have been shown to results in large energy savings using Wing-refiners (Pere et al. 2005). However, such energy savings have so far not been possible to reproduce in conventional refiners other than as en energy savings to a specific freeness level (Salmén 2008). Xylanases have been shown to give large energy savings when treating poplar; up to 25% (Petit-Conil 2005). In the case of treatment of spruce chips no significant impact on the energy-property relationship has though been noted (Petit-Conil 2005). Treatments using pectinases have shown a 10% energy reduction for spruce samples above 100 ml freeness, but only marginal effects when it comes to comparisons at a given strength level (Peng et al. 2003). A similar case is observed for proteinases with the breaking point in freeness being 250 ml (Mansfield et al. 1999).

8 Table 3.1. Summary of enzymatic pre-treatments of chips for TMP reported in literature. 6 Enzyme Dosage A Wood Treatm. (h) Freenes (CSF) ΔEnergy ΔTensile index ΔTear index Reference Xylanase 20 U g -1 5 U g -1 Poplar Norwegian Spruce % -27% +20.6% -10% +23.5% -20% (Chen et al. 2009) (Meyer et al. 2009) Pectinase 500 nkat g -1 Mixture B % 0 0 (Peng et al. 2003) Pectinase 100 U g -1 Scots Pine 6 Ca % - - (Maijala et al. 2008) Pectinase 10 U g -1 Norwegian Spruce % -5% -10% (Meyer et al. 2009) Pectinex3XL C 720 g/t wood Black Spruce E 2.5 const.80-2% (-9%) +6% (+39%) +5% (+25%) (Sabourin et al. 2009) Novozym 863 D 830 g/t wood Black Spruce E 2.5 const.80-3% (- 10%) -2% (+28%) +1% (+20%) MnP mixed with Glucose oxidase 100 U g U g -1 Scots Pine Norwegian Spruce % - + ca. 10% - const. - (Maijala et al. 2008) Proteinases 0.6 mnpu F (Mansfield et al. 1999) Cellulase 5 U g -1 Norwegian Spruce % -10% -13% (Meyer et al. 2009) Cellulase 0.63 mg g -1 Spruce Sapwood % +4% -5% (Pere et al. 2005) Viscozyme TM L 3 ml enz. sol. Scots Pine 6 Ca % - - (Maijala et al. 2008) MnP = Manganese peroxidase. A Enzymatic activity is often denoted either (U) or (kat). [1 U] is defined as 1 µmol min -1 (µmol of substrate transformed per minute under defined conditions - usually optimal). [1 kat], or katal for catalytic activity, is defined as 1 mol s U = 1 µmol min kat (Dybkaer 2001) B Mixture of white spruce, red and black spruce, balsam fir. C polygalacturonase. D polygalacturonase containing other pectolytic activities and hemicellulytic activities. E Reference was fiberized and impregnated with water; percentage without fiberizing in brackets when supplied. F Neutral Proteinase unit/kg

9 7 3.3 Enzymes In the studies here performed it was decided to examine the following enzymes based on the ability to attack the primary wall, availability and thermal stability (the latter property of importance for industrial processing of steam treated wood); pectinases, xylanases, mannanases Pectinases Pectinases can be divided into three main categories; protopectinases, esterases and depolymerases. The former degrade protopectin by adding water; esterases remove methoxyesters (-R-COOCH3 -R-COOH); and the latter cleave α-1,4-bonds in galaturonic acid either by a hydrolytic mechanism, where a water molecule is introduced across the oxygen bridge, or by trans-elimination lysis, where no water molecule is needed to break the glycoside bond. For a complete classification of pectinases the reader is referred to Jayani et al. (Jayani et al. 2005). According to Jakób et al. (2009) pectin lyases by themselves can degrade pectin, whereas pectinesterases and polygalacturonases must co-work to degrade pectin completely. Reactions are illustrated in Figure 3.3 (a) (b) Xylanases Due to the complexity of the substrate, the xylan, with different side groups there are no definite reactions for this class of enzymes. A general reaction can be seen in Figure 3.3 (c)) Manannases Mannanases break down the connection between mannan units in glucomannans. A general reaction can be seen in Figure 3.3 (d).

10 8 a) b) c) d) Figure 3.3 Structures of wood substrates and reactions of some enzymes (a)-(b) pectin (R = CH3 or H) (pectin lyase and pectin methylesterase), (c) xylan (R and R1 = Arabinose, other polysaccharides, etc.) (endoxylanase), and (d) mannan (endomannanase). (Jayani et al. 2005; Ek et al. 2009).

11 9 4 Materials and methods 4.1 Materials Wood The wood material used was Norwegian spruce chips supplied by Braviken Mill, Holmen Paper. The chips had at Braviken been treated in an Impressafiner in order to break up the structure, Figure 4.1. In this treatment chips are pre-steamed and thereafter compressed and sheared through the Impressafiner. The treated chips were delivered to INNVENTIA and frozen for storage. Prior to the trials the chips were thawed (two weeks before trials). Wood shavings of Norwegian spruce chips were used for comparison. The materials are illustrated in Figure 4.2. Figure 4.1 RT Impressafiner (Andritz AG). Figure 4.2 Visualization of the used wood material of Norwegian spruce. a = Impressafiner chips, b = wood shavings.

12 Enzymes The enzymes used were; pectin lyase in combination with a pectin methylesterase (PME) multipectinase (NS81272) xylanase (NS51115) mannanase (NS51054) all supplied by Novozymes. The recommended dosages were 1 kg of enzyme/tonne DMC (dry matter content) wood, except for the pectin methylesterase which was recommended at 0.1 kg/tonne DMC (dry matter content) wood. The optimal conditions for these emzymes were 50 C and ph 5.0 for all pectinases, 55 C and ph 5.5 for xylanase, and 80 C ph 5.0 for mannanase. Xylanase was reported to have a minor cellulase background activity and in major an xylanase endo-1,4 activity. For adjusting to the desired ph values buffers of sodium acetate were prepared by addition of glacial acetic acid mixed with sodium acetate trihydrate and diluted with dh 2 O to ph 5.0 (Pectinase and the mannanase), 5.1 (Pectin lyase/pectin methylesterase), and 5.5 (xylanase) (0.1 M). 4.2 Methods Laboratory enzyme treatments In order to be able to impregnate the wood material with enzymes as well as to perform the enzymatic treatments in laboratory scale an autoclave was used allowing for compression of steam treated chips, Figure 4.3. Steam and enzymatic solutions may be applied from the top connections, while sampling of solution is made from the bottom tap. The chips inside the autoclave may be manually compressed by a top screw. The spruce chips (64 g DMC of Impressafiner chips, 46g DMC of spruce shavings) and steamed for 5 minutes (1 bar) where after they were compressed to 50 % of the height of the filled container. 30 ml sodium acetate buffer was added (ph 5.5 (0.1 M) for xylanase, ph 5.1 (0.1 M) for pectin lyase/pectin methylesterase and ph 5.0 (0.1 M) for pectinase and mannanase). The apparatus was moved to a water bath, maintained at approximately 18 C, for 10 minutes during which 200 ml buffer was added. When mannanse was tested, the apparatus was cooled at room temperature for 5 minutes, reaching a temperature of 79 C. The apparatus was emptied of liquid and moved to a water bath (80 C for mannanase, 50 C for all pectinases, and 55 C for xylanase) for 5 minutes. 30 ml of enzyme solution was added, the mechanical press was released, the cap was substituted to the foot of an E-flask, the wood chips were loosened with a spoon, and the rest of the enzyme solution was poured onto the wood. Time was taken from when all the enzyme solution had been poured. During the incubation, samples of the solution were taken every 10 minutes. After an incubation of 90 minutes the solution was drained, collected. The wood chips were then steamed for 5 minutes with the bottom tap open, during which the drained solution was collected. The wood chips were then again compressed (approximately 50 %) to remove some of the remaining liquid. The procedure is outlined in Figure 4.4.

13 11 Figure 4.3 Autoclave used for laboratory impregnation and testing of enzyme activity on wood chips. Figure 4.4 Visualization of wood chip pre-treatment in laboratory trials. The perforated container in the lower left corner was covered in a 200 MESH (74μm) net.

14 Pilot scale impregnation Pilot scale impregnation of wood chips were made using a PREX screw connected to a larger storage tank at Innventia, Figure 4.5. The tank of approx. 400 l was filled with 200 l dh 2 O adjusted to a temperature of C. Enzyme solutions (1.5; 2.0; 2.0 kg enzyme product/tonne dry wood of pectinase, mannanase, and xylanase respectively) were added to the water and stirred. 10 kg of Impressafiner wood chips was first steamed for 10 minutes where after it was fed into the tank by a rotating screw. The suspension in the tank was stirred half way through the feeding. When all wood chips had been fed into the tank the solution was again stirred. The incubation of the wood chips was continued for 60 minutes where after half of the chips were taken out (excessive water was put back in the tank before the wood chips were transferred to another container). The collected wood chips were steamed for 10 minutes to stop any enzymatic activity, and the wood material was then put in a refrigerator. The remaining chips were incubated for another 60 minutes, then repeating the above handling of the wood. 64 g DMC of wood chips from the different pilot scale impregnations after 1 hour were pressed to 50% as described earlier to collect liquid remaining within the wood structure. chips enzyme solution steam PREX screw Figure 4.5 Schematic illustration of the PREX screw arrangement at Innventia Refining Refining was made using a Wing-refiner at Aalto University in Helsinki, see Figure 4.6. The procedure followed a method earlier developed for ranking of effects of enzymatic treatment of wood chips for mechanical refining (Viforr 2008). The impregnated wood chips described earlier were portioned into 125 g DMC batches and refined in a lowintensity wing refiner. The wing defibrator chamber consisted of a 20 counter blade cylinder with a distance of 1 mm from 4 wing-like rotating blades to the counter blades. Three empty runs of the steam heat treated refiner were used as a blank. Chips were steam treated at a temperature of 124 C ± 0.6 C for 5 minutes, during which the 4 wing-like blades were rotated 90 every 1.25 minutes to heat the chips evenly. After 2

15 13 minutes of steaming the condensate was let out during 10 seconds. After 4 minutes and 50 seconds of steaming the valve was closed, the pulse-meter zeroed and the run started after the 5 minutes of steaming. Refining was performed for 2, 4, 6, and 8 minutes respectively. The pressure in the chamber was during refining bars with the temperature rising from about 124 C to 136 C; the temperature depending on how long the experiment was continued. All enzymatic trials were run in singles, as well as trials at 2, 6, and 8 minutes for the reference. 4 minutes of refining for the reference was run 4 times. Unrefined/less refined wood chips/pulp between the wing refiner body and the wing refiner cap was removed. Small pieces of unrefined wood chips were observed inside the wing refiner. The pulps were centrifuged and freeness and DMC were recorded where after the pulps were transported to Innventia and stored in a fridge until pulp testing. Figure 4.6 Wing refiner at Aalto University in Helsinki Tests Enzyme activity The effects of the enzymatic activity in the experiments was determined by using the Nelson-Somogyi method which measures the amount of reducing sugars, i.e. sugars that contain aldehyde groups that are oxidised to carboxylic acids. The analysis is based on that Copper (Cu) (II) ions are reduced to Cu (I) ions by a saccharide molecule, e.g. D-glucose, D-galactose, maltose, etc., which in this case is released after enzymatic hydrolysis of the saccharide chain. The Cu (I) ions are thereafter oxidised back to Cu (II) by a colourless arseno(poly)molybdate complex, which in a reduced form is blue. The absorption at 500 or 520 nm is observed with a spectrophotometer (Farnet 2010, Sadasivam 2007, Wrolstad 2001)

16 14 The Nelson-Reagent or arsenomolybdate reagent, as well as the low-alkalinity copper reagent was prepared according to Wrolstad (2001). 25 g ammonium molybdate was solved in 450 ml dh 2 O at 37 C under continuous stirring. 21 ml concentrated sulphuric acid was added as well as 3 g of disodium hydrogen arsenate heptahydrate solved in 25 ml dh 2 O. The stirring was continued for 24 hours, where after the mixture was transferred to a 1 l glass-stoppered volumetric flask covered in aluminium foil and was stored at room temperature. The low-alkalinity copper reagent was prepared by adding 12 g of sodium potassium tartrate (Rochelle salt) and 24 g of anhydrous sodium carbonate in 250 ml dh 2 O. 4 g of copper sulphate pentahydrate and 16 g of sodium hydrogen carbonate were solved in 200 ml dh 2 O. The solutions were pooled. To the latter mix was added a solution of 180 g of anhydrous sodium sulphate solved in 500 ml boiling dh 2 O. The final mixture was diluted to 1 l and stored at room temperature in a 1 l glass-stoppered volumetric flask. A standard curve was prepared using g of D-glucose diluted in 5 ml dh 2 O. 1 ml low-alkalinity copper reagent was added to all standards, containing a volume ranging from 5 to 600 µl of the prepared glucose solution. The solutions were heated in boiling water for 40 minutes. To standards with less or equal to 100 µl glucose, 1 ml of arsenomolybdate was added, whereas 2 ml were added to standards with more than 100 µl glucose. The standard solutions were cooled for 15 minutes in room temperature. Samples were diluted to a total volume of 5 ml with dh 2 O. The absorbance at the wavelengths 500 and 520 nm were measured with a spectrophotometer (Hewett-Packard Agilent 8453) and trend lines were fitted to these absorbances (R-value 0.99) due to a Samples were measured against the standard curve and a mean value of glucose equivalents was calculated from the trend lines Pulp tests Dry matter content chips (DMC) SCAN-CM 39:94 Dry matter content pulp (DMC) SCAN-CM 3:78 Freeness SCAN-M 4: Paper tests Hot desintegration ISO :2004 Laboratory sheets ISO :2005 Testing of lab. sheets ISO 5270:1999 Basis weight ISO 536:1995 Structural density SCAN-P 88:01 Tensile index ISO1924-3:2005 Light scattering ISO 9416:2009 Fibre length L&WFiber-Tester

17 15 5 Results 5.1 Laboratory enzyme treatments The wood material had a dry matter content of 36.5% for the Impressafiner wood chips and 59.5% for the finer wood chips. In the laboratory autoclave the compression after the initial steam treatment increased the dry matter content to 39.5% for the Impressafiner wood chips while after absorption of the enzyme solution the dry matter content was reduced to 34.5%. The enzymatic pre-trials performed are summarized in table 5.1. Table 5.1. Summary of enzymatic laboratory pre-treatments of different chips. Enzyme Dosage (kg/tonne dry wood) Wood (Reducing Sugars) (µg/g dry wood) Sugars Enzyme / Reference Pectin lyase Impressafiner PME Pectin lyase + PME Impressafiner Pectinase 1 Impressafiner Reference Pect. Impressafiner 300 impr. a) Average 365 Reference Pect. Impressafiner 430 g/g imp. b) Pectin lyase Fine chips PME Pectinase 10 Fine chips Reference Pect. Fine chips 450 fine a) Reference Pect. Fine chips 420 Average 435 g/g fine b) Mannanase 1.5 Impressafiner Mannanase 5 Impressafiner Reference Man. Impressafiner 570 Xylanase 1,5 Impressafiner Reference Xyl. Impressafiner 440

18 16 It was clearly evident that the structure of the wood chips had a large impact on the accessibility of the enzymes to the wood polymers. Thus much higher amounts of released sugars were obtained from the finely disintegrated wood chips than from the impressafiner treated chips, although these latter had a rather disrupted structure. As seen in Figure 5.1 the amount of released sugars from the finer chips was nearly doubled in the case of the enzymes Pectin lyase/pme as compared to the case with impressafiner chips. For Pectin lyase/pme the use of 1.5 times, the recommended amount of enzyme product, only showed a slightly higher degree of degraded (reducing) sugars compared to the reference. No improvement was seen when increasing the dosage to kg/ tonne of dry wood, Table 5.1. Of the Impressafiner wood chips, Pectinase and Xylanase resulted in the highest amount of released sugars. For Mannanase, the use of the recommended dose of 1.5 kg/tonne dry wood did not result in much increase in reduced sugar release and increasing the amount to 5 kg/tonne dry wood only slightly increased the amount of released sugars. Based on these measurements it was decided to perform larger scale experiments using Pectinase and Xylanase. Also an additional test using Mannanase was included despite the low amount of sugars released but based on the high temperature tolerance, 80 o C, which from a process consideration is of interest. 250 Released reducing sugars, g/g dry wood Impressafiner chips Fine chips Figure 5.1 Amount of released reducing sugars from different types of chips using a mixture of enzymes, Pectin lyase/pme in a quantity of 1.5/0.5 kg/tonne.

19 Pilot scale impregnation The pilot scale impregnation was performed using PREX equipment. For the enzymes Pectinase and Xylanase samples were prepared both with 60 min and 120 min of incubation while for Mannanase only incubation for 60 minutes was performed. As evident from Table 5.2 larger amounts of sugars were in general released from the chips in these pilot trials, even for the reference treatment, as compared to the laboratory trials. This could be an effect of the additional mechanical disintegration of the chips in the PREX- equipment. The impregnations resulted though in a markedly reduced amount of degraded (reducing) sugars in the liquid for the enzyme Pectinase compared to the reference materials, whereas the amount of released sugars for the enzymes Mannanase and Xylanase was about the same as in the pre-trials. When pressing out the liquid from the treated chips it is clear that much more of the sugars were retained within the chips than had had the possibility of diffusing out from them, Figure 5.2. It is here noticeable that a clear enzymatic activity was apparent for the pecitinase, almost as high as for the xylanase, which was in better agreement with the pre-trials. Also the mannanases seem to have resulted in noticeable sugar degradation although being a bit lower than that of the other two enzymes. The fact that so much of the released sugars still remains in the liquid held by the chips makes it difficult to follow the development of the sugar degradation based on the concentration in the surrounding liquid. This is apparent in Figure 5.3 showing the amount of degraded sugars over time of incubation for the different treatments. Table 5.2. Summary of enzymatic pilot scale impregnations. Enzyme Dosage (kg/tonne dry wood) Time (min) (Reducing Sugars) (µg/g dry wood) Sugars Enzyme / Reference Pectinase Xylanase Mannanase Reference 1150 Pectinase Xylanase Reference

20 18 Figure 5.2 Amount of released reducing sugars from different enzymatic treatments of Impressafiner chips (total amount based on concentration in pressed-out liquid) Figure 5.3 Amount of released reducing sugars versus time for different enzymatic treatments of Impressafiner chips (amount in treatment liquor). As seen in Figure 5.3 the amount of released sugar increased with time up to 60 minutes. For the Pectinase no difference was seen as compared to the reference wood while for Xylanase and Mannanases a clear increase in the amount of released reducing sugars was seen. This probably means that the degraded sugars from the pectinase treatment had more difficulty of diffusing out of the wood chips as compared to the degradation products of the other two enzymatic treatments. At 60 minutes half of the

21 19 chips were removed from the reaction vessel. When taking out the chips, the liquid was allowed to drain by gravitation into the vessel. In this process it seems as if some of the released sugars have been absorbing onto the removed chips thus diluting the liquid remaining with the chips still subjected to the enzymatic treatment. Anyhow, as the amount of released sugars in the liquid did not seem to increase after this time it may be assumed that the degradation process is coming into a halt. One could however not also exclude that some degradation of released sugars has taking place decreasing the amount of reduced sugars with time. 5.3 Refining trials Refining experiments were carried out on the enzymatic treatments performed for 60 minutes in a Wing-refiner at Aalto University in Helsinki, Finland. As evident from Figure 5.4 the refining trials were highly reproducible at least from the power consumption versus time. Figure 5.4 Power consumption versus time for Wing-refining of batches of Impressafiner chips pre-treated with the enzyme Pectinase for 60 minutes. In Figure 5.5 the freeness development for the different trials are summarized. Evidently no significant effect of the different enzymatic treatments could be noticed with regard to the freeness level.

22 20 Figure 5.5 Energy consumption versus time for Wing-refining of batches of Impressafiner chips pre-treated with different enzymes for 60 minutes. 5.4 Pulp properties Pulp properties were evaluated for the different pulps having a freeness level between 200 and 85 in CSF as refined. Prior to paper making shives were removed in a Metso screen from these pulps also resulting in a decreased freeness of the screened pulps. As seen in Figure 5.6 none of the enzymatic pre-treatments resulted in improved strength properties with respect to the energy consumption. In the case of mannanase treatment the continued refining to a low freeness pulp seemed to have resulted in a less advantageous property development. A similar result is apparent in viewing the tensile freeness relationship as in Figure 5.7. The mannanse treated pulp showed a less favourable property development as compared to the other pulps while the xylanase treatment could indicate slight improvements. When considering optical properties it is apparent that for all treatments the light scattering coefficient followed a similar improvement when plotted against tensile index, Figure 5.8; no apparent difference between the treatments could be discerned. The effects of the refinings were also similar with respect to the fibre length which was not too adversely affected at the freeness levels reached, Figure 5.9. As a whole it could be stated that none of the enzymatic treatments seems to have resulted in improved energy savings at a given property target.

23 21 Figure 5.6 Tensile index as a function of energy consumption in Wing-refining of batches of Impressafiner chips pre-treated with different enzymes for 60 minutes. Figure 5.7 Tensile index as a function of freeness in Wing-refining of batches of Impressafiner chips pre-treated with different enzymes for 60 minutes.

24 22 Figure 5.8 Light scattering coefficient as a function of tensile index for Wing-refined pulps of Impressafiner chips pre-treated with different enzymes for 60 minutes. Figure 5.9 Fibre length as a function of energy consumption in Wing-refining of batches of Impressafiner chips pre-treated with different enzymes for 60 minutes.

25 23 6 Conclusions Impregnation of Impressafiner treated chips with enzyme solutions have been shown to result in increased degradation of the fibre material as indicated by an increased dissolution of degraded carbohydrates. In refining of such impregnated chips no energy savings as a function of freeness level was seen for the tested cases of pectinase, xylanase and mannanase. The property development was similar to that of reference pulps I the case of pectinase and xylanase while for chips treated with mannanase a less favourable development of the tensile index was noted. Considering the much higher enzymatic activity reached when the initial fibre material was even further disintegrated it could well be that the possibilities for enzymes to attack on desired structures of the fibre wall may have been too few even in the case of the Impressafiner treated material.

26 24 7 References Axelson P and Simonson R (1982a) Thermomechanical pulping with low addition of sulphite. Influence of the preheating temperature at mild sulphite treatment of spruce chips prior to defibration. Paperi ja Puu 64(11), Axelson, P. and Simonson R (1982b) Thermomechaniocal pulping with low addition of sulfite. Part 1. Effects of mild sulfite treatment of spruce chips prior to defibration. Sv. Papperstidning 85(15), R Axelson P and Simonson R (1983a) Thermomechanical pulping with low addition of sulfite. Part 3. Mild sulfite treatment of a mixture of spruce and pine chips. Sv. Papperstidning 86(3), R Axelson P and Simonson R (1983b) Thermomechanical pulping with low addition of sulphite. Part 4 A mill scale trial. Sv. Papperstidning 86(15), R Chen F, Zhang S, Wang G and Wang S (2009) Study of xylanase pretreatment mechanical pulping of poplar International Mechanical pulping Conference, Sundsvall, Sweden. Dybkaer R (2001) Unit "katal" for catalytic activity (IUPAC technical report). Pure and Applied Chemistry 73(6), Ek M, Gellerstedt G and Henriksson G, Eds. (2009) Wood chemistry and wood biotechnology. (Pulp and paper chemistry and technology). Berlin, Walter de Gruyter. Farnet A M, Qasemian L, Guiral D and Ferré E (2010) A modified method based on arsenomolybdate complex to quantify cellulase activities: Application to litters.." Pedobiologia 53(2), Franzén R (1986) General and selective upgrading of mechanical pulps. Nordic Pulp Paper Res. J. 1(3), Jakób A, Bryjak J and Polakovič M (2009) Selection of a method for determination of activity of pectinolytic enzymes in berry fruit materials. Chemical Papers 63(6),

27 25 Jayani R S, Saxena S and Gupta R (2005) Microbial pectinolytic enzymes: A review. Process Biochemistry 40(9), Maijala P, Kleen M, Westin C, Poppius-Levlin K, Herranen K, Lehto J H, Reponen P M, Mettälä O A and Hatakka A (2008) Biomechanical pulping of softwood with enzymes and white-rot fungus physisporinus rivulosus. Enzyme and Microbial Technology 43(2), Mansfield S D, Won K K Y and Richardson J D (1999) Improvments in mechanical pulp processing with proteinase treatments. Appita J. 52(6), Meyer V, Lecourt M and Petit-Conil M (2009) Bio-TMP process to save energy: Comparison of enzymes efficiency International Mechanical Pulping Conference, Sundsvall, Sweden. Peng F, Ferritsius R and Ängsås U (2003) Mechanical pulping with pectinase pretreatment of wood chips International Mechanical Pulping Conference, Quebec. Pere J, Ellmen J and Viikari L (2005) Process for preparing mechanical pulp. Finland WO Petit-Conil M, Hoddenbagh J, Marc A, Tolan J S (2005) Method for mechanical pulp production. WO 2005/ A1. Richardson J D, Wong K K Y and Clark T A (1998) Modification of mechanical pulp using carbohydrate-degrading enzymes. J. Pulp Paper Sci. 24(4), Sabourin M, Mäentausta O, Räsänen K and Braeuer P (2009) Targeting enzyme formulations at selective wood components for optimized thermomechanical pulping of spruce International Mechanical Pulping Conference Sundsvall, Sweden. Sadasivam S and Manickam A (2007) Biochemical methods. New Age International (P) Ltd. Salmén L (2008) Ecotarget. EU-project Personal comunication

28 26 Salmén L and Pettersson B (1995) The primary wall; important for fibre separation in mechanical pulping. Cellulose Chem. Technol. 29(3), Stevanic Srndovic J (2011) Interaction between wood polymers in wood cell walls and cellulose/hemicellulose biocomposites. Ph.D. thesis Chemical and Bioological Engineering, Chalmers University of technology. Gothenburg, Sweden Westermark U, Samuelsson B, Simonson R and Pihl R (1987) Investigation of a selective sulfonation of wood chips. Part 5. Thermomechanical pulping with low addition of sulfite. Nordic Pulp Pap. Res. J. 2(4), Viforr S (2010) Enzymatic pre-treatment of wood chips for energy reductions in TMP production; a method for ranking of enzymes. Report S09-218, Värmeforsk, Stockholm, Sweden, ISSN Wrolstad R E (2001) Current protocols in food analytical chemistry. New York, Wiley.

29 27 8 Innventia Database information Title Optimering av lämpliga enzymer för enzymatisk flisförbehandling för energieffektiv TMP tillverkning Author Silvia Viforr, Tomas Mårtensson, Lennart Salmén Abstract In order to evaluate the possibilities of using enzymes for pre-treating softwood chips in order to lower the energy demand in mechanical pulping impregnation and refining trials were performed using a set of different enzymes. In small scale impregnation trials it was shown that the enzymatic activity highly depended on the size, i.e. the amount of destruction of the chips with more disintegrated wood chips rendering a higher release of sugars. Of the enzymes tested Pectinase and Xylanase resulted in the highest amount of released sugars while Mannanase and Pectin lyase gave considerably lower sugar releases. In larger scale impregnation trials the treatment indicated that most of the sugar released occurred in the first 60 minutes and that activities thereafter seemed to level off. Refining trials using a small Wing refiner on chips treated with Pectinase, Xylanase and Mannanase did not show any energy savings to a given freeness level. Comparing strength properties the differences were small considering tensile index and light scattering where the Mannanase treated chips even showed a lower strength development as a function of energy input. In conclusion it was apparent that wood chips, even after an Impressafiner treatment is too coarse for allowing enzymes to sufficiently break up the fibre-fibre bonding allowing for lowering the energy demands in mechanical pulping. Keywords energy consumption, enzymatic treatment, impregnation, mechanical pulping, strength properties Classification 1151 Type of publication Innventia report Report number 361 Publication year October 2012 Language English

30 INNVENTIA AB is a world leader in research and development relating to pulp, paper, graphic media, packaging and biorefining. Our unique ability to translate research into innovative products and processes generates enhanced value for our industry partners. We call our approach boosting business with science. Innventia is based in Stockholm, Bäckhammar and in Norway and the U.K. through our subsidiaries PFI and Edge respectively. is a trade and industry group within INNVENTIA AB Tel Drottning Kristinas väg 61, BOX 5604 Fax info@innventia.com SE Stockholm Sweden VATno