Biomass Quality Improvements by Sieving and Fractioning
Harvested stumps and small trees from thinning.
Work ReportSubproject 3. Chemical and physical requirements of the forest biomass to biorefineries,
fractionation of biomass, and preparation of bio-chemicals
AuthorHåkan Örberg
Swedish University of Agricultural SciencesDepartment of Forest Biomaterials and Technology
30.6.2014
Contents
1. Abstract........................................................................................................................................................1
2. Background..................................................................................................................................................1
3. Introduction................................................................................................................................................1
4. Material and methods................................................................................................................................2
4.1 Biomass samples used in this study...............................................................................................2
4.1.1 Stumps... ........................................................................................................................2
4.1.2 Small trees.......................................................................................................................3
4.2 Factors that were studied...............................................................................................................3
4.2.1 Stumps.............................................................................................................................3
4.2.2 Small trees.......................................................................................................................3
4.3 Gravimetric fractioning...................................................................................................................3
4.4 Fractioning by sieving.....................................................................................................................4
4.5 Responses.......................................................................................................................................4
4.5.1 Cellulose..........................................................................................................................5
4.5.2 Hemicelluloses.................................................................................................................5
4.5.3 Lignin...............................................................................................................................5
4.5.4 Extractives.......................................................................................................................5
4.5.5 Ash..................................................................................................................................5
5. Experimental design....................................................................................................................................6
5.1 Design for stumps...........................................................................................................................6
5.2 Design for small trees.....................................................................................................................6
6. Results..........................................................................................................................................................6
6.1 Stumps............................................................................................................................................7
6.1.1 Ash content.....................................................................................................................7
6.1.2 Extractives.......................................................................................................................8
6.1.3 Statistical evaluation.......................................................................................................8
6.1.3.1 Stumps.............................................................................................................8
6.1.3.2 Small trees.......................................................................................................9
6.2 Small trees.....................................................................................................................................10
6.2.1 Ash content.....................................................................................................................10
6.2.2 Extractives......................................................................................................................11
6.3 Combined effects...........................................................................................................................12
7. Discussion...................................................................................................................................................12
KeywordsFractioning, ash, extractives, cellulose, stumps, small trees
1. AbstractIn this work two main methods for fractioning and sieving of forest biomass have been evaluated. A general need for enlarged feedstock supply to biorefining industries in the future is creating a need to look at mate-rials from forest harvest operations that are not generally used today. Therefore, this work has focused on stumps from clear cutting and small trees from thinning. The experiments have aimed to show whether the quality of these raw materials can be improved by fractioning and sieving to reach a quality level good enough for aiming at the so called “sugar platform”, i.e. fermentation. Evaluated parameters have been ash contents, extractives, cellulose, hemicellulos, lignin and sugars.
This work has shown that different fractioning methods can be used to increase the quality of forest biomass from spruce/pine stumps and small trees from thinning. This is especially obvious for ash contents in both stumps and small trees. Ash is mostly attached to small particles, and by sieving and gravimetric method a great share of these small particles is reduced. Concerning extractives, it is also shown that these methods can lower the amount of small bark particles that are richer in extractives and can therefore raise the value for fermentation. However the statistical evaluation shows that the standard deviation for extractives is too high to get statistical significance. Due to this more detailed studies should be completed.
2. BackgroundThere is a future need for increased feedstock of biomass for biochemical conversion in order to replace fossil raw materials. By looking at new forest resources that today are not much used, like stumps and small trees from thinning operations, a wider raw material base could be available.
In the fermentation process of biomass into ethanol, and in similar processes, some compounds in the raw material are considered to be process disturbing. This is especially so for ash and extractives. Ash is disturbing for the process both mechanically and chemically while extractives mostly lower the outcome of the process. By sieving and gravimetric fractioning the contents of these compounds can be lowered. In order to increase the quality in terms of reducing the contents of unwanted compounds from soil and minerals different frac-tioning methods can be used. Contamination from soil is very common for example in stumps for natural rea-sons since they are lifted from the ground. It is more or less impossible to avoid that gravel and soil particles are lifted together with stumps when they are harvested.
It is also interesting to determine the amount of cellulose and hemicelluloses for this new raw material in order to evaluate the share of fermentable sugars. The content of lignin should be as low as possible for the fermentation process.
3. Introduction The overall target of this subproject was to evaluate methods for improving biomass quality for chemical pro-cesses aiming at the refinement of different sugars, i.e. the sugar platform. In addition to the main product, i.e. lignocellulosic ethanol, a lot of other interesting products are generated from this system, e.g. certain sugar compounds and byproducts. Sieving and fractioning methods are used to lower process disturbing and unwanted substances connected to a certain fraction. The methods are also used to lower impurities that are only useless like ash forming minerals.
1
The purpose of this sub-project was to characterize the physical and chemical properties of appropriate bio-mass fractions from stumps (pine and spruce) and thinning materials from young stands (birch, pine and spruce). Characterized parameters were moisture content, ash content, extractive content, cellulose, hemi-celluloses, lignin, bulk density and particle size distribution. All sampling was performed according to the demands of the different characterization methods. The charts will specify sample sizes, increment numbers, and handling procedures.
Research and development of different fractioning tech-niques have focused on mechanical sorting technologies. Mechanical sieving of common forest biomass assortments in the Botnia-Atlantica region was performed in a Mogens-en Sizer Type E0 554 separator (10-1000 kg/h), using dif-ferent sieve combinations. The choice of materials for frac-tioning was decided in cooperation with project partners and actors on the market. The beneficiaries of this project will be receiving industries in the form of improved feed-stock qualities to lower overall cost, and also some machin-ery and process equipment manufacturers.
4. Material and methods
4.1 Biomass samples used in this study1. Stumps from spruce and pine 2. Small delimbed trees from spruce, pine and birch
4.1.1 Stumps Stumps were lifted shortly after clear cutting of a stand by an excavator that included a tool for lifting and splitting the stumps in the same operation.
All stumps were transported to the BTC plant in Umeå where they were crushed in a hammer crusher (Peters-son 4700 B). After crushing the biomass it was dried to about 15 % w.c. in low air temperature (30°C). Sample size after drying was about 500 kg. Before sieving and fractioning the samples were comminuted in a shred-der (Micromat 2000) with a sieve size of Ø 15 mm.
Figure 1. Excavator with a mounted tool for stump lifting.
Figure 2. Crusher in action with stumps.
Figure 3. Small trees of pine, spruce and birch. Figure 4. Harvester for thinning operation.
2
4.1.2 Small trees Small trees were collected from a stand in the coastland of Västerbotten. The average age of the trees was 30 years.
From each assortment of delimbed trees of pine, spruce and birch 10 m3 was transported to BTC plant in Umeå shortly after harvest. All assortments were chipped in a stationary chipper (Edsbyhuggen 250H Electric power 65A) with two chipping lengths (8,0 mm and 12,0 mm). After chipping all assortments were dried to about 15 % w.c. in low air temperature (30°C).
4.2 Factors that were studied
4.2.1 Stumps 1. Effects of sieving in seizer with one sieve Ø 1,0 mm. Sample size of about 250 kg of each pine and
spruce was sieved. For pine 12,5 % of the fine materials were sorted out and for spruce 7,3 % of the fine materials were sorted out from the start weight.
2. Effect of gravimetric fractioning. For spruce 11,5 % of the lightest and 0,7 % of the heaviest frac-tion (mostly rocks) was sorted out. For pine 9,9 % of the lightest and 0,7 % of the heaviest fraction (mostly rocks) was sorted out.
3. Effect of both sieving and gravimetric fractioning.
4.2.2 Small trees 1. Effects of sieving in screener with one sieve Ø 1, 9 mm and 14 mm for 8 mm chips and Ø 1, 9 mm
and 16 mm for 12 mm chips. Sample size of about 250 kg of each pine and spruce was sieved. For pine 12,5 % of the fine materials were sorted out and for spruce 7,3 % of the fine materials were sorted out from the start weight.
2. Effect of gravimetric fractioning 3. Effect of chipping, sieving and gravimetric fractioning
4.3 Gravimetric fractioning Gravimetric fractioning uses differences in specific particle weight in treated biomass. After the biomass has been dried and comminuted to an appropriate particle size by chipping or shredding it is fed into the gravi-metric separator in an even flow. This is done by an air tight rotary valve or screw. The biomass is then trans-ported in a thin layer by a vibrating table.
Figure 5. Gravimetric separator for commercial use. Figure 6. Principle set-up of a gravimetric separator.
3
An adjustable nozzle is placed at the end of the table and is lifting up lighter fractions by an air stream letting heavi-er fractions like rocks, gravel and metal pieces pass by. The equipment can also be used the other way round to reduce the amount of very light and unwanted fractions.
In this study the biomass was treated twice in order to sepa-rate and exclude both the finest material (light fraction) and heavy fractions. For spruce 5,6 % and for pine 6,1 % by weight was sorted out in the light fraction and for spruce 1,0% and pine 0,8% by weight of the heavy fraction was sorted out.
4.4 Fractioning by sieving Commonly fractioning in industries is done by sieving. Sieves can be made in different design with several active screening surfaces varying from 2-5 m2 for each flat. Sieve sizes vary from Ø 0,1-30 mm.
The screener is driven by a vibrator which gives it the typical elliptical motion pattern that counteracts blinding and loos-ens and stratifies the feed material. Screens are manufac-tured with up to 5 decks, suiting the screening duty.
By the screening operation of biomass two main fractions are mostly rejected, first too big particles that might make problems in downstream operation and secondly too small particles that usually are more ash rich.
Sieving = Fraction > 1.0 mm (10-12 % mass reduction) Gravimetric fractioning in two steps:
• Reduction of light particles (dust 8-10 % mass reduc-tion)
• Reduction of heavy particles (mostly gravel 0,8-1,0 % mass reduction)
When analyzing the contents of carbohydrates and lignin, py-rolysis combined with GS/MS is used.
4.5 Responses Main components of forest biomass are carbohydrates, lignin, extractives and ashes. The carbohydrates can be divided into cellulose and hemicelluloses. Cellulose and hemicelluloses can be hydrolysed to sugars and fermented to ethanol. The lignin part of the biomass cannot be used in fermentation processes but it can be used as a solid fuel or for additives in e.g. pellet production. It is not only important to know the ratio between cellulose, hemicelluloses, lignin and the rest but also to get information of the different sugars from the hemi-celluloses. When analyzing the contents of carbohydrates and lignin, pyrolysis combined with GS/MS is used.
Figure 7. Experimental set-up of sieving.
Figure 8. Principle set-uo of a screen separator for sieving.
Figure 9. Combined treatment of both sieving and gravimetric fractioning.
4
4.5.1 Cellulose Cellulose can be split into glucose but very strong H-linkages between glucose groups make cellulose very stable and difficult to split. Strong methods like acid hydrolysis in high temperature and pressure might de-compose cellulose. A problem is thereby that exposed glucose will quickly turn in to unwanted aromatic compounds.
Especially developed enzymes are capable of decomposing cellulose, and if combined with weak acid hydrol-ysis, will directly take care of the exposed glucose and together with the use of yeast ferment it to ethanol.
The total amount of sugars in celluloses and hemicelluloses is analyzed with GC/FID. All analyses are done with three replicates.
4.5.2 Hemicelluloses Hemicelluloses are built up of different five-carbon (Pentose’s) sugars as Xylose and Arabinose but also six-carbon sugars (Hexoses) as Glucose, Mannose and Galactos. Glucose and Xylose are the easiest sugars to ferment to ethanol. Pentoses are more difficult.
The amount of hemicelluloses is analyzed with GC/FID. All analyses are done with three replicates
4.5.3 Lignin Different to the other cell wall components of biomass, lignin is mostly insoluble in mineral acids. Therefore lignin can be analyzed gravimetrically after hydrolyzing the cellulose and hemicelluloses fractions with sulfuric acid.
The amount of lignin measured by this method is called Klason lignin and is used in this work. This is the amount of biomass that does not solute in sulfur acid (72 %, H2SO4, 30 ± 1oC, for 2 hours).
4.5.4 Extractives Extractives are a collective term for a lot of fairly light molecular compounds. The content of extractives in forest biomass is relatively low but some compounds might have a high value for medical and biological use. Most of the extractives, when industrially upgraded, are today used as liquid fuel like tall oil.
For the fermentation process of biomass to ethanol some extractives can be process disturbing.
To determine the content of extractives like fatty acids, resins, turpentine and alcohols the sample can be dissolved in ether (etoxyethan).
4.5.5 Ash The amount of ash in biomass is in some respect of lower interest when the concentration is low like in pure stem wood (0,4-0,5 %) but for some biomasses and fractions the ash contents can be significantly higher. In these cases ashes might be process disturbing. Ash compounds are normally not reactive at low temperatures but might cause other problems. The amount of ash gives a measure of purity/impurity to the biomass.
5
5. Experimental design
5.1 Design for stumps For stumps the experimental design is quadratic where the effect of sieving and gravimetric fractioning is studied.
5.2 Design for small trees
For small trees the design is cubic where the effect of chipping, sieving and gravimetric fractioning is studied.
6. Results In order to create a representative sample for an inhomogeneous biomaterial like chipped small trees, crushed stumps or logging residues the method for taking samples and how samples are treated is crucial. For the evaluation of the chemical composition in samples analysis the sample size must be reduced in a correct way. It is also extremely important that the methods for chemical analysis are adapted to the final sample size. This is because of the importance of reducing the standard deviation in the results. In this work a model for sample preparation and sample size reduction has been developed
Operation before sampling • In this “Forest Refine” work samples from the forest were taken from:
• Stumps: pine and spruce stumps starting with a weight of 500 kg/sample • Small trees from thinning: pine, spruce and birch with a weight of 500 kg/sample
• After first particle size reduction by chipping (small trees) or crushing (stumps) samples were dried in low temperature air to w.c. of ˜ 12 %. During the chipping and crushing samples for chemical analysis were taken continuously in the flowing stream. Sample size was ˜ 1,0 kg.
Sample size reduction • After gravimetric fractioning and sieving A+B samples were taken out with a weight of 1 kg/sample. • These samples were reduced and milled down to particle size of Ø 1 mm and sample size was now
down to 100 g/sample.• For the chemical analysis samples were reduced to 1 g/sample.• All sample size reductions were done after thorough mixing of the sample.
Figure 10. Schematic picture of the experimental model for stumps.
Figure 11. Schematic picture of the experimental model for small trees.
6
6.1 Stumps
6.1.1 Ash content Small fractions of pine and spruce stumps are rich in ash which leads to a possibility to reduce ash contents by siev-ing and reducing the finer fractions. A problem with sam-pling and analysis, especially in the case of stump material, is that the standard deviation for ash contents is high. This is a problem for the statistical evaluation.
In figure 15 ash content is shown in spruce and pine stumps after sieving and gravimetric treatment and by the combi-nation of both methods.
Figure 12. Schematic picture of sam-ple size reduction in stump and small tree samples.
Figure 13. Ash contents in spruce stumps after sieving.
Figure 14. Ash contents in pine stumps after sieving.
Figure 15. Ash contents in spruce and pine stumps after different fractioning treatments.
7
6.1.2 Extractives The standard deviation for extractives in stumps is much lower compared to ash content.
In figure 16 extractive content is shown in spruce and pine stumps after sieving and gravimetric treatment and by the combination of both methods.
6.1.3 Statistical evaluation
6.1.3.1 Stumps
Figure 17. Extractive contents in pine stumps after sieving.
Figure 16. Extractive contents in spruce stumps after sieving.
Figure 18. Extractive contents in spruce and pine stumps after different fractioning treatments.
Figure 19. Response coefficients for ash contents in spruce stumps for sieving and gravimetric treatment.
Figure 20. Response coefficients for extractives in spruce stumps for sieving and gravimetric treatment.
8
6.1.3.2 Small trees
Figure 21. Response coefficients for ash contents in spruce stumps for sieving and gravimetric treatment.
Figure 22. Response coefficients for extractives in pine stumps for sieving and gravimetric treatment.
Figure 23. Response coefficients for ash contents in small spruce trees for sieving and gravimetric treatment.
Figure 24. Response coefficients for extractives in small spruce trees for sieving and gravimetric treatment.
9
Figure 25. Response coefficients for ash contents in small pine trees stumps for sieving and gravimetric treatment.
Figure 26. Response coefficients for extractives in small pine trees for sieving and gravimetric treatment.
6.2 Small trees
6.2.1 Ash content The ash content in small spruce, pine and birch trees is on a low level, but as in the case of stumps, the small particles < 1,9 mm are rich in ash. This leads to the possibility of reducing the ash content in small trees by sieving and reducing the finer fractions in the same way as was done with the stumps.
Figure 30 shows that bark parts of the small trees are much richer in ash compared to the stem wood. This is extremely obvious for spruce bark.
Figure 27. Ash contents in small spruce trees after sieving.
Figure 28. Ash contents in small pine trees after sieving.
10
6.2.2 Extractives The content of extractives in the bark fraction of spruce and pine trees is in general much higher when com-pared to stem wood parts of the trees. As shown in figure 31 extractives in pine bark is 5-6 % of DM and in birch bark up to 20 % of DM. In blank samples of pine and spruce, which are chipped samples of small tree logs including bark, the extractive content is 2-3 %. Pure stem wood has even lower values.
Figure 29. Ash contents in small birch trees after sieving.
Figure 30. Ash contents in stem wood / bark in birch, pine and spruce.
Figure 31. Extractive contents in bark of pine, spruce and birch.
Figure 32. Extractive contents in small spruce trees after sieving.
Figure 33. Extractive contents in small pine trees after sieving.
Figure 34. Extractive contents in small birch trees after sieving.
11
6.3 Combined effects When both methods of fractioning were used on the material, reduction of ash was obvious. In general the ash content was lower for all pine samples compared to those of spruce.
7. Discussion In these experiments it has been shown that sieving and fractioning of small trees can lower the content of extractives. This depends a lot on the ability to reduce the amount of small bark pieces during the sieving and fractioning treat-ment.
Looking at figure 37 it shows that the bark frac-tions are high in extractives (5-20 % ) compared to the stem wood of spruce and pine (2-3 %). It is obvious that an optimized process of combined sieving and fractioning can lower the amount of extractives in the chipped material.
The experiments show that the mean value of both ash content and extractive content has been lowered by fractioning but the statistical evaluation shows that the standard deviation is too high to get statistical signifi-cance. Based on this more detailed studies should be completed.
Figure 35. Ash content in spruce and pine stumps after gravimetric and sieving treatment.
Figure 36. Ash content in small pine and spruce trees after gravimetric and sieving treatment.
Figure 37. Extractive contents in pine, spruce and birch bark.
12
