! ! ! " #$
% & ' ( ) ( ) ( ( * + , " " + , - + !. !( , + /012
!
! " # $ ! % &
! "
# $ !% ! & $ ' ' ( $ ' # % % ) %* %' $
' + " %
& ' ! #
$, ( $ - $$$ . /"/"#012" )*3- +) * !4 !& 5!6%55787& % 2 9:;567!%2 ?9:;567!; 0 @ ! " A/((%B A /((%
%7;=5 * C & D '()*+,-.-/0*-112,1-/ " 3'4%& A 2 ! "A ( % 2
% % % % $ $ $A A
AA - $$$ /"/"#012" )*3- +) * !4 !& 5!6%55787& %) A 9:;567!%E?9:;567!; 0 @ ! B & " B + AC
D '()*+,-.-/0*-112,1-/ - A F8>78A /"/"#012" )*3- +)
"
& 8>78
ACKNOWLEDGEMENTS
I gratefully acknowledge the finalcial support for my study and this final work to the Gobernment of Yugoslavia. I am especially indebted to Prf. Dr. Joåe Leniþ for his assistence, advice, criticism and observations in course of my study as well as for my final work. I am very thankful to the Director of the Research Institute of the Factory "LESONIT", Chemical Engineer Mrs. Ana âtemberger and her co²workers. Thanks are also due to Chemical Engineer Mrs. Marijanka Prosen for her collaboration in the determination of the mechanical and physical properties of the hardboards tested. I wish to acknowledge the participation of the Chemical Engineer Mrs. Ana Prestor²Knez, in the prehydrolysis of beech wood and determination of furfural. I deeply appreciate the kindness of Mr. Boãtjan Anko M.Sc. for his collaboration in the correction of the text of this work.
CONTENTS
1. INTRODUCTION
2. PURPOSE OF WORK
3. GENERAL EXPERIMENTAL
5. CONCLUSIONS 6. LITERATURE
1. INTRODUCTION The origen of wood fiberboard goes back to Japan where as early as in the 6th.century B.C. heavy papers were used for the construction of walls for small houses. In Europe a patent was granted to the British inventor Clay in 1772 for the application of "papier mache" not only for use in dwellings, furniture, doors, but also for carriages. The idea of using the new material for big stiff building elements was evident. Since the middle of the 19th.century the proporsals for use of fiberboards have greatly increased. More than 2OO patents were issued in this field between 1858 and l928, more than 600 patents until 1957. In spite of these early developments the actual fiberboard industry started near the begining of the 20th.century in England and in the USA. The development up to 1926 was rather sporadic and without a remarkable increase in capacity. In 1926 by the foundation of the Masonite Co. with a plant in Laurel, Miss., USA Mason improved the invention by Lyman (1858) to separate wood fibers by the expansion of hot water, steam or compressed air. Steam explosion converted wood chips into fibers, which were hot-pressed to hardboard of high quality without addition of artificial bonding agents. In 1931 invention of Asplund, Sweden, to continuously defibrate wood chips under pressure and steam (170 to 175oC.). Simultaneous experiences conducted in Europe and in USA to convert conventionally produced wet process insulation board by pressure into hardboard, in 1958. In 1943 a development followed by Weyerhaeuser Timber Co. and in 1945 Plywood Research Foundation on an industrial I scale of transport of fibers and mat-formation by air. In the USA, the ideas for dry and semi²dry methods came from Heritage, Evans and Neiler. The trend to install in the future more mills producing hardboards by the semi²dry and dry process is evident. Lehotsky and Nagy (1963) reported that until the end of the year 1962 about 17 plants operating on a semi-dry or a dry method were established, approximately half of them in the USA. (43) 1.2. DEFINITION AND CLASSIFICATION One fundamental definition generally is recognized (FAO, 1958/1959, p.4): ³Fiberboard is a board generic term encompassing sheet materials of widely
1
varying densities manufactured from refined or partially refined wood fibres or others vegetable fibers. Bonding agents and others material may be incorporated in the manufacture of the board to increase strength, resistance to moisture, fire or decay or WRLPSURYHVRPHRWKHUSURSHUW\´ In the technical sense the ISO-definition is more precise: ³Sheet material generally exceeding 1.5 mm in thickness, manufactured from ligno-cellulosic fibers with the primary bond from the felting of the fibers and their inherent adhesive properties. Bonding matHULDOVDQGRUDGGLWLYHVPD\EHDGGHG´. Perhaps, the best factor for classifying fiberboard is the density. This is internationally recognized. There is a rather simple difference between pressed and not pressed sheets but the range of qualities is wide and there is an overlapping. The general term "fiberboard" is not only adopted for use in the publications of FAO and ECE, but is generally under- stood in the literature and in the industry. According to their density ranges, fiberboards may be classified into five type as shown in the table 1.1. TABLE 1.1 CLASSIFICATION OF FIBERBOARD ACCORDING TO DENSITY (43). FIBERBOARD DENSITY (g/cm3) NON²COMPRESSED Semi-rigid insulation board 0.02 0.15 Rigid insulation board 0.15 0.40 COMPRESSED Intermediate or medium density (half-hard) Hardboard Special densified hardboard
DENSITY 0.40 0.80 1.20
(g/cm3) 0.80 1.20 1.45
2. PURPOSE OF WORK In pulping of broadleaf species, abundant in pentosans (birch, beech), it was attempted by kraft process using prehydrolysis, to utilise distillates for the production of furfural and hydrolysis residue for the production of high grade alpha pulp. (38,58) By analogy there exists the possibility of utilising hydrolysis residue to obtain mechanical fiber. From several investigations we can say that a much higher yield of wood substance may be anticipated than by the sulphate process.
2
In this study an attempt was made to analyse the influence of steam hydrolysis on the beech (Fagus silvatica), wood fibres, in particular on some physical properties and a theoretical yield of furfural at different prehydrolysis times. The research is focused on the possibility of utilizing y this fibre in the state it is in after prehydrolysis. Special consideration is given to the character of prehydrolysis of wood chips and its role from the viewpoints of the physical and mechanical variations of the fiber and furfural. The furfural and organic acid content have been determined. Hardboard represents a construction in which the fiber is studied from this hydrolysis residue. Pentosans determination (60) before and after prehydrolysis carried out on beechwood chips gave the possible yield of furfural. 3. GENERAL 3.1. BASIC CHARACTERISTICS OF WOOD FIBRES The crystalline nature of the cellulose in wood has been known for several decades, based on evidence provided by studies with X-ray diffraction and polarization microscopy. The term fibril had a wide use and was applied to wall components that could be seen with the light microscope. Elementary fibrils are presumably the cellulosic strands of smallest possible diameter (42). Muhlethaler (51) applies this term to fibrils with a diameter of approximately 35o. An elementary fibril of this cross² sectional dimension could contain about forty cellulose chains (42). Aggregates of elementary fibrils are classed as microfibrils and occur in nature in a broad spectrum of sizes, depending on the source of cellulose. Microfibrils aggregate into larger units which may be called macrofibrils and these are joined into lamellae that are organized into cell wall layers. Within the elementary fibrils there are zones in which the cellulose chains are oriented to such a high order that X²ray diffraction patterns can be produced as with true crystals. These regions are termed crystallalites or micelles. The crystalline regions are interrupted by amorphous zones alog the elementary fibril. In these portions the cellulose moleculeare simply not as perfectly aligned (42). Surrounding the fiber, and heavily lignified and stiff is the middle lamella (M), shared with adjacent fibers.
3
The outermost layer of the fiber is called the primary wall (P), which was formed during the maturation of the fiber, is not homogeneous but subdivided into the outer secondary wall or transition lamella (S1), the main secondary wall (S2), and the inner secondary wall (S3), sometimes called the tertiary wall or lamella (58), See fig. 3.1
Fig3.1 Cell ±wall architecture, schematically. M - middle lamella, P primary wall. S1 - transition Lamella, S2 ± main layer of secondary wall. S3 - tertiary wall or tertiary lamella of secondary wallMeier. (Adapted from Panshin and De Zeeuw) The middle lamella is intimatelly connected to the primary walls of two adjacent fibers and the concept of ³comSRXQGPLGGOHODPHOOD´ includes the primary walls. The true middle lamella is about l-2 microns thick (45) varying with species and growing conditions. Its high lignin content makes the middle lamella a hard and hydrophobic sheathing around the fibers. In the cambial zone it is mainly made up of pectic material,but becomes heavily lignified during the phase of fiber maturation. The primary wall forms the original wall of the fibers and merily gorws in area, in contrast to the secondary wall, which grows in thickness. The primary wall is very thin, corresponding to the width of three elementary cellulose fibrils (18). The secondary wall, is not a morphologically homogeneous layer and should be considered as a collective term for all fiber layers formed in the secondary process of fiber maturation. The outer secondary wall or transition lamella has a thickness of about 0.15 microns (18) and consists of two counter-rotating helices of cellulose microfibrils. The main secondary wall forms the bulk of the fiber, with a thickness of 1²10 microns, varying with species and growing conditions. The inner secondary wall, also called the tertiary wall, is the structure limiting the fiber wall toward the lumen, and is hence of particular interest from the pulping point of view. It consists of one single lamella, 0.07²0.08 microns thick.
4
It is remarkably resistant to chemicals, which may be a result of its dense morphological structure. 3.1.1 MORPHOLOGY OF WOOD CELLS 3.1.1.1 BASIC ELEMENTS OF WOOD STRUCTURE AND THEIR DIMENSIONS When thin sections of wood are examined with a ligth microscope, cellular composition can be readily observed. The cells are held together by intercellular substances but should this lamella be dissolved through chemical or mechanical treatment, the cellular composite is separated into individual elements. This is, in fact, the commercial process known as pulping which is employed to produce fibers. Microscopic examination of pulp samples of hardwood and softwood reveals that there are significant differences in the size and shapes of cells from these two sources. The main part of the tree which is used for pulping is the trunk. The trunk is composed histologically of three parts, the xylem or wood, the cambium and the bark. See fig. 3.2
Fig- 3·2· Schematic section of four-year-old pine trunk. (Adapted from Hनgglund after Strasburger)
5
Cambium is a thin, green layer of growing cells between the bark and the wood. The bark is composed of a white, inner bar, called phloem or the bast zone, and a darker, outer bark, or cork. Apart from the cambial zone there is another zone of growth on the tree, the apical zone, located at the tips of the stems and roots. There are four main elements, parenchyma oells, fibers, tracheids and vessel elements. There are several forms of gradations between libriform fibres, which have only mechanical functions, and the tracheids, which have well developed conductive functions. The intermediate elements are called fiber tracheids. Although the term fiber in its strict, botanical sense is confined to dells with mechanical functions only, in pulp technology the word is used for all sorts of structural elements in wood and in other pulping raw material. Tracheids and libriform fibers, after being split off from the cambium, extend longitudinally and become stretched to a length of approximately fifty to a hunderd times their diameter(42). The length of softwood tracheids varies from 1-11 mm, and is usually 2-5 mm in commercially important species. The hardwood libriform fibers are shorter, about 1-2 mm. The vessel elements are very often wide and short, with thin walls. Their diameter may be as small as 0.02 mm and sometimes as large as 0.5 mm and their length shows similar variations (58). Similarly, it is clearly seen that average length of softwood tracheids is always much greater than that of hardwood libriform fibers. It is obvious, that with the variations mentioned, fiber length has a fairly wide distribution curve even within a single trunk. In softwood, with their fairly homogeneous composition of structural elements, a sample with an average fiber length of 3 mm has a standard deviation of about 0.5 mm, i.e. two-thirds of the fibers fall in the range of 2.5-3.5 mm and 90% within 2.04.0 mm. Hardwood vessels with an average length of 0.5 mm have a standard deviation of 0.l mm and hardwood libriform fiber and fiber tracheids with an average length of 1.2 mm show a standard deviation of around 0.2 mm. The thickness of the cell walls varies within the ranges 2-8 microns for most tracheids, 1.5-3.0 microns for vessels, 3-7 microns for libriform fibers and 2-5 microns for parenchyma cells (58). TABLE 3.1 VOLUME PERCENT OF STRUCTURAL ELEMENTS IN SOME EUROPEAN SPECIES (58)
6
Species Abies alba Pinus silvestris Betula verrucosa Eagus silvatica Populus tremula Fraxinus excelsior
Tracheids and fibres 90 93 (98)+ 65 (86) 37 61 62
Parenchyma cells 10 7 (2) 10 (5) 32 13 26
Vessel elements 25 (9) 31 26 12
+ Weight per cent in brackets TABLE 3.2 FIBER DIMENSIONS OF VARIOUS WOOD SPECIES (58)
Species Picea abies Pseudosuga taxifolia Pinus silvestris Betula verrucosa Eagus silvatica Populus tremula Quercus robur
Length, mm Range Av.
Av.
Wall ness, Range
thick mic
Av.
24-59
36
1.3-13
6
4.00
44
1.0-10
7
3.50
38
Width, mic Range 3.50
3.006.00 0.562.00 0.402.30 0.501.35 0.401.90
1.25
10-29
18
2.4-7.2
3.7
1.30
25-35
29
2.5-15
5.2
0.95
13-37
21
1.3-5.3
4.3
1.10
10-35
21
2.5-10
6
3.1.1.2 GENESIS OF WOOD CELL The primary organic compounds, which eventually build up the wood substance, are produced in the leaves by photosynthesis out of carbon dioxide and water. The photosynthesis in addition to energy also supplies building blocks for the formation of cellulose and hemicelluloses, as well as of lignin and minor components involved in the growth. There are two principal types of growth, apical or bud growth, and cambial or growth of the trunk circunference.
7
In dormant condition this consists of 1-4 cell tiers, the outer of which is the initiating layer and the remainder called xylem mother cells, which have been formed by successive cell divisions of the initiating layer. The xylem mother cells are in turn capable of division and redivision to produce wood destined cells (13), as schematically illustrated in figure 3.3
3.3 Diagrams illustrating sequence of cell divisions in the cambium IC initial cell, XMC xylem mother cell, PMC, pholoem mother cell. (Adapted from Bannan). This happens when the cells are activated in the spring. There is first a radial expansion because of considerable water uptake. Then follow division of chromosomes and formation of a longitudinal partitioning wall. In the beginning, the generation of new cells is a fairly slow process, the interval between the first and second divisions being up to two weeks, but the tempo increases and there may be as many as 10-30 cell layers taking an active part in the generative process of the xylem upon formation of springwood (58). Occasionally there is also a formation of fhloem mother cells from the initiating layer. After midsummer, cell division proceeds at a decreasing rate in the initiating layer, whereas the xylem mother cells eventually cease to divide, thereby diminishing the
8
active cambium zone to a few cell rows, and consequently the rate of wood production. The bark production, on the other hand, continues at an unchanged rate. As the diameter of the trunk grows, an increase in the cell number along the periphery, multiplicative division, is necessary. This initially occurs in the cambium layer, for a few species by simple radio longitudinal cell division, but for most species by a semitransverse division. A high frequency of multiplicative division will tend to bring about a decrease in fiber length, whereas the selection of the longest cambial cells for survival acts in the opposite direction. The former mechanism is likely to cause the differences in average fiber length between the juvenile wood of the core and the mature wood of alder trees. The xylem cells formed by longitudinal division of xylem mother cells have very thin primary walls, consisting of cellulose and some pectic substances, and a largely pectic middle lamella. As the growth proceeds, they eventually become increasingly remote from the cambium layer and lose their capacity for division. Then they enlarge and mature as tracheids por libriform fibers. Length growth is not more than about 10% in softwood and up to 500% in hardwood, but expansion in radial direction may be as high as fourfold. Towards the end of this expansion there is a thickening of the cell walls by deposition of a secondary wall, containing cellulose and hemicelluloses. During the final phases of secondary wall formation, lignification completes the maturation of the xylem cell. 3.1.2 INTERFIBER BONDS OF WOOD One of the major objectives of all pulping process is to liberate the fibers of the wood. Surprisingly little has been published on the nature of the interfiber bonds. The fibers in the wood are usually joined in a fairly cohesive structure. lntercellular spaces occur in reaction wood only, except for a few species such as Juniperus spp., where they may occur in normal wood. Because of the special way in which a tree trunk grows, the fibers are formed in fairly regular radial rows but the tangential order within each annual growth ring is less regular, as seen in a transverse section. A separation of the wood fibers after physical or chemical softening of the interfiber bonds therefore often starts along radial planes. As cambial cells divide, a cell plate is formed as a wall common to two adjacent fibers, whose protoplasts eventually develop their own thin primary walls.
9
After the cessation of surface growth the transition lamella, the secondary wall and the tertiary wall are formed. The middle lamella contains mostly lignin, about 70% in spruce and probably still more in birch, together with carbohydrates, mainly hemicelluloses with an amorphous structure: the primary wall is somewhat less lignified and contains cellulose microfibrils oriented in all directions and probably with an in²terwoven structure. The main part of the cell wall is carbohydrates with cellulose in microfibrils, which have a helical structure, and relatively little lignin. In the transition lamella this helical arrangement is rather flat and forms a crossed network at an angle of approximately 60° to the fiber direction, while the secondary wall has a helical structure with an angle that is almost parallel to the fiber direction. The inner layer of the fiber wall, borders the lumen and is fairly thin. Finally, the cell cavity, the lumen, occupies a space of 20-40 microns in diameter. The fiber structure is illustrated in the figure 3.1. Liberation of the fibers obviously requires a separation along the middle lamella or between the middle lamella and the primary wall of one of the fibers. However, as the secondary wall forms the major part of the fiber wall, the primary wall being very thin and the transition lamella relatively thin, a failure between the primary wall and the transition lamella, or between the transition lamella and the main part of the secondary wall does not lead to complete destruction of the fiber wall (58). Ag the lignin is mainly concentred in the middle lamella and carbohydrate in the inner parts of the fiber wall, the position of the failure will be dependent on the relative strength of the lignin and the carbohydrate material (45). Garland (22) says that microscopical examination after mechanical strength tests on untrated wood showed that failure seldom occurs in the middle lamella in tensile and comprenssion strength tests. Failures takes place instead either across the fiber walls or between the transition lamella and the main part of the secondary wall. On delignification of wood it is to be expected that the middle lamella, where the bulk of the lignin is located, will be weakened, and that failure will occur to an increasing extent in this layer. This is especially so in strength tests on moist wood, whereas dry delignified wood shows strengths the same as or higher than those of untrated dry wood (36). Obviously, in this ease new cohesive forces through hydrogen bonding are formed between the fibers, which now consist entirely of carbohydrates. Failures across the fiber wall are also probable. The wet strength of these delignified samples is very small, around 10% of the wet untreated wood. Strength values for wood will therefore indicate either the strength of the individual fiber, the strength of the bond between different layers in the fiber wall, or the
10
strength of the bonds between the fiber walls (interfiber bonds), depending on the pretreatment and tests conditions (58). An interesting observation is the decrease in individual fiber strength upon the successive extraction of the hemicelluloses which may indicate that the hemicellulose; content of a fiber is important not only for its paperbonding ability but also for its internal strength. 3.1.3. BONDS AMONG THE RETNTEGRATED FIBERS 3.1.3.1 BONDS AMONG THE FIBERBOARD FIBERS One special method for obtention of fibers is the Asplund process, developed in the early 1930's which used mechanical forces for fiberizing chips that had been softened with tha aid of water and high temperature. During this process a deacetylation and hydrolysis of the glucosidic bonds, occur, specially in hemicelluloses. However, the tensile strength of the individual fibers is close to that of wood. For an examination of the bonding mechanism the properties of fibre building boards or paper sheets should be compared at the same sheet densities. A fiber to fiber bonding area can be formed only where fibres are in very close contact with each other. Two types of forces bring the surfaces together in this case. One is the external pressure applied during the drying operation. The other is the surface tension force of residual water, which reaches a maximum value at approximately 30% solids content (52). Hydrogen bonding is known to take place gradually during drying, starting already at approximately 20 to 30% dry content. During the pressing physical and chemical reactions give a bond between the fibers, which take place in the dry lignocellulosic material situated in the press. The velocity of these reactions increases with temperature. The most important reaction is an auto-crosslinking, which improves most board properties, while the chain breakage and related decomposition of the lignocellulosic material is generally a disadvantage. The volatilization and redistribution of the natural resin also taking place, mainly affect the water resistance. The heat treatment of the boards then increases the stability of the fibers and the strength properties of the boards (59). It is important to notice that a prolonged time of heat reatment will decrease the bending strength due to the depolymerization of the cellulose (52). It was shown that for a mechanical pulp the pre-oxidation with periodate before sheet formation increased the velocity of crosslinking during heat treatment, while
11
the prereduction of the pulp with sodium borohydride had a retarding effect compared with a hardboard sheet produced from a non-treated pulp. See figure 3.4 The explanation of the mechanism of the curing of the boards can be based on three theories. The first one is the formation of water resistent bridges, may be of the type of hemiacetal and etheric inside and between the fibers. The second is the elimination of the hydrolyticand oxidative covalent linkages. The third one is the decomposition and reaction of the resins, oleics, and accessory components of the wood (47). The possible mechanism of crosslinking by hemiacetal bonds, after periodate oxidation can be seen in the figure 3.4. The mechanism is supposed to be generally valid for heat treatment in the presence of oxygen.
Fig. 3.4. Possible mechanism of creating the crossbonding by hemiacetal bonds. 3.2. PRE-HYDROLYSIS OF WOOD 3.2.1. FROM THE STANDPOINT OF CHEMICAL CHANGES IN WOOD It is difficult to judge where physical action of water ends and chemical hydrolysis become of importance for the ease of fiberizing. Water definitely has a softening effect on the interfiber bonds already at room temperature, as proved by mechanical strength determination on moist and dry wood.
12
This must be due to the splitting of the hydrogen bonds and subsequent swelling, predominantly in the carbohydrate regions (36). Hydrolysis occurs at elevated temperature at the acetyl groups of the hemicelluloses, at the glycosidic bonds of all carbohydrates, especially the easilly accessible parts, at the lignin-carbohydrates bonds which are probably also of glycosidic nature, and finally at intralignin bonds, mainly benzyl alkyl ether bonds. Due to the organic acids formed, these reactions are acompanied by an increase in acidity, as well as by a dissolution of carbohydrates and lignin. At the same time, condensation reactions of the lignin accur (38). 3.2.1.1 HEMICELLULOSES The more affected during the prehydrolysis are the hemicelluloses but the pentosans are the most unstable which are transformed topentoses and then to furfural. The clasical reaction in the obtaining of furfural as final product is shown in the figure 3.5.
Fig- 3.5 Scheme of the reaction yielding furfural from xylan Hexosans have a analogous structure as pentosans with oxymethil furfural, which is decomposed to furfural and formaldehyde (44). See figure 3.6
13
According to research by Kurschener and Melcerva (44a), in Germany, the decomposition of pentosans in beech wood heated 28 days at 80°C was greater than that of hexosans.
Fig- 3.6 - Decomposition of oxymethyl furfural. Klaudiz and Stegmann(35) have demostrated that in poplar wood when subjected to 200oC. during one hour, the decomposition of cellulose was 0.3%, lignin 1.4% and pentosans 4%. 3.2.1.2 CELLULOSE Husemann, et.al. (25), have shown that hydrolysis of cellulose runs 1500 times more slowly than the hydrolysis of pentosans. When wood is hydrolyzed with steam for the production of furfural, the effect of hydrolysis is not restricted to the hemicellulose portion of the wood. The cellulose and lignin are also attacked to a degree depending on the severity of the hydrolysis conditions. The increase in the reaction temperature and/or time increase the loss of cellulose and causes extended condensation of lignin (56). The loss of cellulose means a reduced yield of pulp or glucose whether the hydrolysis residue is intented for the production of pulp or saccharification. Pure cellulose, when heated with water, is only slightly attacked even at 200oC., but the rate of hydrolysis is increased as the pH decreases. As a-part of the hemicelluloses is hydrolyzed only with difficulty, also cellulose is attacked in the conditions required to complete conversion of the pentoses into furfural. To guarantee a good yield of both furfural and cellulose, the prehydrolysis with steam is usually carried out at 170²l85o C. (24).The critical temperature for the degradation of cellulose being apparently about 180oC. (24, 56).
14
3.2.1.3 LIGNIN Lignin, the encrusting substance in wood, is a three²dimensional polymer of phenylpropane units, linked together by C-O-C and C-C linkages (42). In softwood, each unit carries one phenolic oxygen and one metholxyl group, while in hardwoods approximately half of the units contains an additional methoxyl group. Softwood lignin has a methoxyl content of 15 - 16% while that of hardwood has 21% (42). See figure 3.7
Fig ± 3.7 ± Phenolic oxygensand methoxyl groups in the phenylpropane unit sof the lignin molecule. a ± Softwood; b ± Hardwood. Like all hardwood, beechwocd yields on degradation with nitrobenzene and alkaly, more syringaldehyde than vanillin (21). Beech lignin contains almost 22% methoxyl, or about oneand-a-half times as much methoxyl as spruce lignin. The average composition of two milled wood lignin preparations from beech is C9H6.43O2(H2O)0.53(OCH3)1.3. The composition of the mixture of the three cinnamyl alcohols from which beech lignin is produced has tentatively the molar proportions 55% coniferyl alcohol, 4% p-coumaryl alcohol and 43% sinapyl alcohol. The composition of such mixture is C9H1O2(OCH3)1.39: beech lignin contains 2.2 atoms less and 0.53 mole of water more (21).
15
Lignin is resistant to hydrolysis. The model lignin structure shows that structural units are probably linked through ether bonds for the most part, since C-C bonds, as was stated above, which are recognized to be resistant to hydrolysis. There may be one exception to this, namely, that p-hydroxybenzyl ether bonds, exhibited by structural units I (K), may be rather susceptible to hydrolysis, particulary acid hydrolysis as shown in the figure 3.8 (R=lignin), (55).
Fig- 3.8 - Structural units showing the P-hydroxyybenzyl ether bond hydrolysis. The fact that lignin is resistant even to hydrolysis by strong mineral acids serves as the basis for the Klason method for determining lignin content in wood, since carbohydrates components of wood are readily hydrolyzed by such tretment and are rendered water-soluble. At elevated temperatures, however, profound changes in the lignin structure occur: formic acid, methanol acetic acid, acetone, vanillin, and other products may be produced and a portion of the lignin may become condensed. (55) The dissolution rate of lignin has been stated to be between that of pentosans and cellulose, when wood is boiled with water (58). In hydrolysis with water at elevated temperature under pressure, a considerable part of the lignin in wood may be dissolved, and still more is extractable from the hydrolysis residue by organic solvents. There is evidence of chemical linkages between lignin and carbohydrates in wood, mainly xylan or to polyuronides, and they are stated to be stronger than the glycosidic bonds in xylan (44a). In steam hydrolysis, the condensation reactions may overshadow the hydrolytic degradation of lignin. Lignin condensation is also catalyzed by the acids liberated from wood (37). It is known that the further lignin condensation proceeds, the less is the aldehyde yield in nitrobenzene oxidation while the yield of benzene polycarboxylic acids in permanganate oxidation increases.
16
Klemola and Nyman (38) have demostrated that the lignin condensation in steam hydrolysis is shown most clearly by the acetone extracts from the hydrolysis residues. They advise the selection of proper catalysts that are capable to crack the carbon-carbon bonds in the condensed lignin. Especially Ayroud(6) has given attention to the possibilities of the dehydrogenation²hydrogenation catalysts used in the oil industry for the cracking of lignin to low molecular products in good yields. 3.2.2. FROM THE STANDPOINT OF OBTAINING THE WOOD FIBRES 3.2.2.1 MECHANICAL PULPING There are two different ways for the production of mechanical pulp. One is the grinding of debarked logs, the other one is the refining of wood chips. In both methods a mechanical treatment leads to the defibration of the wood. Though recently only about 7% of the world production of mechanical pulp derives from the refiner defibration, this proportion will increase in the future as hardwoods and wastewoods can also be converted in refiners. Logs of softwoods and low density hardwoods such as poplar can be defibrated only in grinders. In recent machines the logs undergo the wet abrasive action of the circular surface of a grindstone. The stones vary from 1.4 to 2.3 m in diameter. The average pulp production per day amounts to 20 to 50 metric tons in 24 h. The pressure with which the logs are pressed against the grindstone, the rotatory velocity of the grindstone, and the surface of the grindstone are important factors in determining the quality of the pulp. High temperatures soften the lignin and the polyoses in the middle lamella (intercellular substance) between the wood fibers and thus favors the separation of the fibers. Studies of the grinding process and the thermal behavior of the cell wall components showed that temperatures of 170_to 190oC. as formerly assumed are not reached in the grinding zone and are not necessary for the thermal softening. The maximum temperature in the log under grinding does not exceed l00oC. Better defibration, lower content of sheaves, and better strength properties of the pulp with water at 85oC than with cold water has been experimented by Birkeland (13b). 3.2.2.2 THERMOMECHANICAL PULPING The steam pressurized refining of chips was developed by Defibrator. The pressure was usually around 6 to 10 atm. gauge pressure, corresponding to a temperature of 165o to 185°C. (3). In this temperature range the lignin is softened so that the fibers are almost completely separated in undamaged condition at a low
17
consumption of energy. If, on the other hand, the chips are refined at a lower temperature in the range of 120-130o C., most of the lignin, though softened, is in a glassy state. The factures occur predominantly in the outer layers of the secondary fiber wall, and the wall fractures render the fibers accessible to fibrillation. Another method of obtaining the fibers is the Mason process which principle is the high temperature defibration by steam expansion. As this process utilizes higher temperatures than that of Asplund, a higher loss in pulp yield occurs through hydrolysis, and it is open to question whether the Mason process should be assigned to mechanical or semichemical pulping. In this process, wood chips are fed into small, digesters originally of around 0.1 t of wood capacity, the socalled Masonite guns. The temperature is rapidly brought up to around 200°C., where it is maintained for approximately l5 seconds. Then the softened chips are further heated to 280-285oC and the temperature kept at this level for 4-5 seconds. The content of the digester is then blown to the atmosphere in an explosion with an estimated velocity of 1.200 m/sec. The pulp yield is lower than in Asplund process whose yield is between 90-95%. In addition to deacetylation, which also occurs in the Asplund process, hydrolysis of glycosidic bonds takes place, especially in the hemicellulose part of the wood. The ambition of the Defibrator method of developing a field of wood waste utilization has only been partly realized. 3.2.2.3 CHEMICAL PULPING In the chemical pulping, substances of the middle lamella are chemically dissolved to an extent that makes fiberizing possible without mechanical treatment. The chief drawback of the chemical pulping methods is of course the comparatively high wood consumption, with yields ranging approximately from 35-55% of wood. For some cases chemical pulps are unnecessarily high²grade and expensive, whereas mechanical pulps may not fulfil the quality requirements. The dominating pulping process today is the kraft process. The main reasons for this are the comparative simplicity nad rapidity of the process, its insensivity to variations in wood species, as well as the valuable properties of the pulp produced. Kraft cooking involves reactions which may only influenced to some exten within the limits of the reaction conditions. For instance, this makes possible with either process to make a pulp low in hemicellulose without excessive degradation of the cellulose. In the kraft process, the dissolution of carbohydrates is still less selective.
18
The alkaline peeling reaction from the aldehyde end groups of the carbohydrate molecules preferentially dissolves short-chain material of both hemicelluloses and cellulose, but stabilizing reactions, including adsorption of xylan on the cellulose microfibrils, will cause more hemicellulose to remain in the pulp than in proportion to their initial molecular size relative to that of cellulose. A high cooking temperature with water only, a considerable part of the hemicelluloses and fairly little cellulose are hydrolyzed to shorter chains. The hemicelluloses to a considerable extent dissolve in polymer, although degraded, form and are further hydrolyzed to manoses in solution. The latter are partly decomposed to furfural, hydroximethyl furfural, levulinic and formic acids(58).The remaining hemicelluloses, degraded to a DP od about 30% of the original, are more easily dissolved in the kraft cook, as the increased number of end groups make the peeling reaction more efficient. .XUEHJRYLü(44a) has found in some research on prehydrolysis of beechwood, that good results are obtained in prehydrolysis at 170oC during two hours, in order to obtain alpha cellulose. 3.2.3 FROM THE STANDPOINT OF OBTAINING THE FURFURAL 3.2.3.1 GENERAL CONSIDERATIONS The industry of furfural is known from 1832 when the production reached a great scale. In USA the Quaker Oats Co., has started a production of furfural in l922 from residues oats (53). Furfural is obtained from material rich in pentosans, from wastes of cereals, as bran, oats, maize. Zoch and others (66), have demostrated that the processing of wasted liquors containing several percent xylose cligomers, as spent sodium-base acid sulfite pulping liquor, can render very favorable and economic yield of furfural. The basic reaction in the obtainment of furfural from pentosans takes place in two phases. The first one, is the hydrolysis of pentosans to pentoses and the second one, is the convertion of pentoses to furfural through several intermediate reactions, and dehydration of the pentoses. All this reactions are carried out in the presence of water, high temperature and organic acids as catalytic medium (62). To obtain a good yield of furfural from pentosan-containing material it is profitable to remove furfural from the reaction mixture as soon as it is formed because of its reactivity (38). The yield of furfural is far from quantitative in steam hydrolysis, the losses being due to the high reactivity and its precursors in the dehydration reactions of pentoses.
19
Furfural along polymerizes in presence of acidic catalysts, splitting off formic acids and giving resinous products. As the loss of furfural in 0.1 N acetic acid at 187.5oC is 7% in two hours, it is evident that the polymerization of furfural plays a minor role in steam hydrolysis compared with the condensation reactions involving furfural, its precursors and lignin (38). Zoch and others (66), say that the maximum yield of furfural from pure xylose is 64%, reacted at 240oC with sulfuric acid, of the theoretically attainable. Biüev (13a), says that the yield of furfural, from prehydrolyzed beechwood with sulfate process is 75% from the theoretically possible. Dunlop (17) has demostrated the mechanism of the reaction, in the decomposition of xylose to furfural and the condensed products by the following scheme:
K1and k2 are constants of speed of formation which depends of the decomposition of furfural. In the production of furfural, an increase in the reaction temperature results in an increase in maximum furfural yield and a decrease in the time required to reach maximum furfural yield. The term "maximum furfural yield" designates the maximum point on the time²yield curve. This point, of course, is a function of the reaction conditions. It is usually to add to the reaction mixture small amounts of mineral acids to act as catalysts. In steam hydralysis, however the organic acids liberated from wood act as catalysts, and possible improvement in the furfural yield by the addition of mineral acids is largely outweighted by the increase in the costs of the equipment. There are two technological processes to obtain furfural. Both processes can be continuous or discontinuous. Today the continuous process is more used. 4. EXPERIMENTAL 4.1 RAW MATERIAL Normal beechwood chips were taken from the store-place.
20
A selection of the chips with the size of approx. 20 x 20 x 2 mm was done. A certain quantity of this material was taken in order to determine the density and the pentosans content (60) before hydrolysis. To determine the density four parallel runs were made, and two parallel runs for pentosans content. 4.2 STEAM PREHYDROLYSIS. Chips with 44.17% of moisture content were subjected to a steam prehydrolysis at 185+/- 2oC. and for different prehydrolysis times, 40, 50, 55, 60 minutes. The prehydrolysis was carried out in a hydrolysis vessel with 216 liters volume. Steam was led into the hydrolysis vessel at the bottom and destillates were taken out at the top. The average rate of steam to wood was ca. 2 kg of steam to l kg of ovendry wood. The distillates were analyzed for: - Content of furfural (by titration) - Content of organic acids, as acetic acid,(by titration) The reactor plant used for the steam hydrolysis is shown in the figure 4.1. The content of these products of hydrolysis, depending upon the prehydrolysis time, is shown in the table 4.1 and graphically in the figure 4.2. As catalytic acted the organic acids formed in the wood substance due to the hydrolysis. FlG.4.1
Scheme of the reactor plant Capacity: 250 liters Cylinder volume: 216 liters Condensator: 2 pieces with cooling surface of 2,3 m2 each R- Rotameter: 0-100 liters/h H- Furfural condensate tank: 225 liters C- Ligno-cellulosic material container: 370 liters M- Manometer: Wika 0-16 kp/cm2
21
4.3 TESTING OF DISTILLATES 4.3.1 GENERAL CONSIDERATIONS ABOUT DISTILLATES The beechwood distillates are soluble in water which are composed of a complex mixture of substances and which can be formed in two groups: furfural and its derivatives, acetic acid, formic acid, methanol, methylacetat, monosaccharides and oligosaccharides, resin components. The analysis of this complex mixture is complicated. The furfural presents in the distillates can be determinate by several methods as: titration, colorimetric, polarigraphic, cromatographic, spectrofotometric and by precipitation.
22
4.3.2 DETERMINING OF FURFURAL AND SOME OTHER CHARACTERISCS OF DISTILLATES The distillates obtained were subjected to analysis in order to determine: - The furfural content by titration - The organic acids, calculated, as acetic acid, by titration - the quantity of distillates From the collected data during the prehydrolysis time several tables and figures were done in order to see the yield of furfural depending upon the prehydrolysis time and the concentration of furfural at different hydrolysis times. The table 4.1 shows the composition of distillates. The figure 4.2 shows the formation of furfural (upper curve) and of acids (calculated as acetic acid, lower curve). 4.3.3 RESULTS AND DISCUSSION The formation of furfural follows closely that of acids (figure 4.2).Interesting data from the commercial and technologic aim can be find in the table 4.1. TABLE 4.1 COMPOSITION OF DISTILLATES, PREHYDROLYSIS TIME, AT 185+/- 2oC
DEPENDING
ON
THE
PREHYDROLYSIS TIME, MIN SUBSTANCE TO BE DETERMINATE 40 50 55 60 1. Prehydrolyzatea 1.09 1.49 1.50 1.57 (Distillate) 2. Furfural in the 2.23 2.20 2.31 2.34 prehydrolyzateb 3. Total content of furfuralc 2.60 3.47 3.93 4.11 4. Organic acids in the 1.48 1.26 1.74 1.74 prehydrolyzateb 5. Total content of organic 1.69 2.28 1.74 5.18 acidsc a. Calculated in liters per kg of original wood substance b. Calculated as % of the prehydrolyzate c. Calculated as % of original wood substance oven dry Since the method employed differs from the others used, in steam hydrolysis, this results look some low (57, 58, 48), even so, positive results were obtained, by this procedure. The figure 4.2 shows as the formation of furfural increases with the time of hydrolysis. It is clear that a major presencie of acids give a mayor production of furfural.
23
Another interesting deduction can be done looking the figure 4.3 which shows the concentration of furfural depending upon prehydrolysis time. The major concentration of furfural was reached at the 60 minutes of hydrolysis, with a value of 2.34%. We can say that there is not a great difference with the value reached at the 55 minutes of steam hydrolysis. This can be explained by the same values obtained in concentration of organic acids at 55 and 60 minutes of hydrolysis.
Fig. 4.2. Production of furfural and of organic acids in %.
24
Fig. 4.3. Concentration of furfural in distillate
The yield of furfural obtained from the decomposed pentosans is shown in the table 4.1A, where we can see the quantity of pentosans decomposed during the steam hydrolysis at different times. The theoretical yield of furfural was calculated as a 75% of furfural obtained from the hydrolyzed xylose (66). The yield from the theoretic possible was calculated in %, dividing the actually obtained furfural by the theoretical yield. The highest results were obtained at 60 minute of prehydrolysis time, since major catalyst concentration also was reached at the same time and a greatest quantity of pentosans were decomposed. From the data obtained during the 60 minute of steam hydrolysis a graphic was done in order to see the weight of furfural in %, contained in the distillates. Also the absolute weight of furfural in grames, was determined and plotted. (See figure 4.4) Looking the curve we can observe that the best result was obtained at 55 minute of steam hydrolysis with a value of 3.75 g/100 g· At the same time a concentration of 2.52 g/100 g of organic acid, calculated as acetic acid, was also reached. Equally, it is clear that the decreases of the curve is less pronunced after this point due to the ideal reaction temperature, better catalyst
25
concentration, and major decomposed pentcsans. When a temperature of 185OC. is reached, an intense production of furfural is observed. The variation of furfural, in grames, with the variation of volume of distillate was also observed. The minimum quantity was obtained at l0 minute of prehydrolysis time with a value of 39 g/100 g and the maximum at 60 minute with a value of 279.5 g/100 g. The pH values remained constant during 28 and 50 minutes of steam prehydrolysis (2.55), increasing after to a value of 2.5.
TABLE 4.1A PENTOSANS DECOMPOSED DURING PREHYDROLYSIS AND THEIR YIELD AS FURFURAL, PREFERRING TO THE ORIGINAL WOOD SUSTANCE.
PREHYDROLYSYS TIME
CONTENT OF PENTONSANS BEFORE HYDROLYSIS %
CONTENT OF PENTOSANS AFTER HYDROLYSIS %
PENTOSANS DESCOMPOSED %
THEORETICAL YIELD1 %
0 26.3 26.3 40 26.3 6.2 20.1 14.7 50 26.3 5.8 20.5 15.0 55 26.3 4.6 21.7 15.8 60 26.3 4.0 22.3 16.3 1/73% Furfural of the quantity of decomposed pentosans
ACTUALLY OBTAINED FURFURAL %
YIELD FROM THE THEORETIC POSSIB. %
2.6 3.5 3.9 4.1
17.7 22.0 24.7 26.4
Fig. 4.4. Weight in % of furfural in prehydrolyz ate and absolute quantity in grames.
26
4.4 TESTING OF HYDROLYSIS RESIDUE 4.4.1 APPLICABILITY OF HYDROLYSIS RESIDUE 4.4.1.1 DEFIBRATION OF REMAINING HYDROLYSIS RESIDUE Hydrolysis residue in form of chips, prehydrelyzed as was described under point 4.2. were subjected to the thermomechanical defibering process (2). Defibering was effected in the Asplund laboratory type D, "Defibrator", Stockholm, under the following conditions: - Moisture content of the hydrolysis residue: 45-50% - Time of presteam: 2 minutes - Time of defibration: 2 minutes - Pressure of presteam and defibering: ca. 11 kp/cm2 (185oC) Refining of the pulp was carried out in the refiner laboratory type D, from the firm "Defibrator", Stockholm, with two disks. The distance between disks was 0.25 mm for the prehydrolyzed pulp and 0.08 mm for the normal pulp. The variation of those values is due to the high values obatined in dewatering of the hydrolyzed pulp. Consistency of the pulp was 3%. The degree of beating is shown in the table 4.2. In this table big differences can be seen between values obtained by normal pulp and prehydrolyzed pulp. This can be explained with the softening of the fibers during the prehydrolysis time. So, a more fine defibering was reached with hydrolyzed pulps. The classification of the fibers made by the Bauer-McNett fractionator can be seen in the table 4.3. Clears differences can also be seen between hydrolyzed pulp and normal pulp. 4.4.1.2 MANUFACTURE OF TEST HARDBOARDS From the pulps so prepared, hardboards in various variants and subvariants were manufactured according to the table 4.4. The letter X was taken in order to represent the zero specimens. The letters A to D represent the specimens, depending on the prehydrolysis time. The numbers (1 or 2) represent the subvariants and quantity of pulp employed for the manufacture of hardboards.
27
TABLE 4.2. FREENESS OF PULPS USED FOR THE TEST HARDBOARDS TYPE OF PULP AND FREENESS 1 MEASURING UNITS 1
UNITS
DEFIBRATED PULP REFINED PULP
DS CSF DS CSF
RAW MATERIAL HYDROLYSIS RESIDUE FROM NORMAL BEECH HYDROLYSIS TIME, MINUTES BEECH FIR 40 50 55 60 38.5 40 40 50 15 15 850 830 830 820 800 830 70 80 80 80 25 25 780 760 760 750 730 765
1/Measured in Defibrator Seconds (DS), but expressed also in Canadian Standard Freeness (CSF). TABLE 4.3 CLASSIFICATION OF THE FIBERS VALUES ACHIEVED IN THE DEFIBRATOR FRACTIONATOR DEPENDING ON PREHYDROLYSIS TIME. INDIVIDUAL FRACTIONS IN %
NORMAL BEECH FIR
PREHYDROLYSIS 40 50
TIME 55
60
REMAINED ON THE SCREEN:
1.0 mm 0.5 mm 0.2 mm 0.15 mm
4.15 9.10 15.66 5.66
3.25 24.32 9.33 22.43
3.25 5.21 18.32 31.43
3.17 4.74 17.32 5.77
0.85 8.31 23.38 8.48
4.67 21.61 9.60 12.19
21.08
18.33
30.43
15.09
18.57
49.55
PASSED THROUGTH THE SCREEN:
240
Thus e.g. the test hardboard of the variant X2 contains 70% of normal beech pulp and 30% of normal fir pulp. Hardboards of the variants D2 contain 70% of hydrolysis residue of 60 minutes prehydrolysis and 30% normal fir pulp. So the variants A, B, C, D, represents tests hardboards with 40, 50, 55, 60 minutes of prehydrolyzed residue. Each variant consists of a series of 5 same hardboards wich can be considerer as parallels. Consequently, the manufacture of test hardboards comprised 5 variants or 10 subvariants or 50 individual hardboards. Hardboards of the variant X can be considered as control specimen, or zero specimen, as they do not contain any hydrolysis residue. Test hardboards were made by means of laboratory equipment delivered by the firm "Defibrator", Stockholm, in the size of 30 x 30 cm and nominal thickness of 3.2mm.
28
The diverse phases in the manufacturing of the test hardboards can be enumerated as follow: 1. Manufacturing of the boards in the apparatus for formation of the wet²mat, 2. Cold²pressing in the press of laboratory press ³Defibrator´ (10 kp/cm2 during 30 seconds), 3. Hot-pressing in the press of laboratory press ³Defibrator´, 4. Heat-treatment in the apparatus laboratory ³Defibrator´, 5. Climatizacion in the climate room. No additives were used, except sulfuric acid in order ton obtain pH 4,5, only in the cases, in wich hydrolysis residue did not contain enough own organic acids. The conditions of pressing were following: - In the first period: 45 kp/cm2 during 15 seconds, - In the second period: 10 kp/cm2 during 3 minutes, - In the third period: 45 kp/cm2 during 2 minutes, 30 seconds. - Pressing temperature: l85+/- 2oC.
TABLE 4.4 VARIANTS AND SUBVARIANTS OF TEST HARDBOARDS X
VARIANTS SUBVARIANTS DIFFERENT PULP COMPOSITION
1 2
A B C D PREHYDOLYSIS TIME FOR BEECH, MIN 40 50 55 60 100% B 100% B 100% B 100% B 100% B 70% B 70% B 70% B 70% B 70% B 30% F 30% F 30% F 30% F 30% F
X. Zero specimen, with normal beech and fir pulp B. Beech pulp F. Normal fir pulp The heat treatment was carried out during 4 hours at 163+/- 2° C. Before determination of mechanical and physical properties, all test hardboards were exposed to 65+/- 3% of relative humidity and to a temperature of 20+/- 1oC. until the equilibrium moisture content was achieved.
29
4.4.1.3 MECHANICAL PROPERTIES OF TEST HARDBOARDS Mechanical properties of test hardboards of each subvariant are given as mean values obtained as aritmetical mean from 5 testtubes and for the specific breaking enrgy from 6 testtubes. The following properties were investigated: - Bending strength (30), - Modulus of elasticity (16), - Bending stiffness (11), - Tensile strength parallel to the plane (9,34), - Tensile strength perpendicular to the plane (27), - Specific breaking energy by Charry (40). The results of all mechanical properties investigated are summarized in the table 4.5. All the tests enumerated above, were carried out in the apparatus Alwetron T-2000 from the firm Lorentzen & Wettres, Stockholm, except specific breaking energy which was calculated from the results obtained in the apparat Wclpert eclatometer, 0-40 kpcm from the firm Otto Wolpert-Werke, Ludwigshafen and bending stiffness which was calculated by the following formula: SEI= Eb x I Where Eb is the modulus of elasticty, (MOE), in bending and I the momentum of inertia. TABLE 4.5 MECHANICAL PROPERTIES OF TEST HARDBOARDS X1 921 337
X2 936 355
A1 1087 337
VARIANTS AND SUBVARIANTS B C A2 B1 B2 C1 C2 1079 1079 1080 1073 1015 354 354 485 351 451
29840
36429
32954
34036
34036
45217
29591
41599
28504
35617
96.7
115.6
92.7
102.3
102.3
179.8
96.9
96.9
90.5
93.0
230.2
308.1
263.1
251.9
251.9
351.5
216.3
270.5
201.4
238.3
12.0
12.4
19.4
17.7
17.7
20.4
11.5
14.6
10.0
13.0
9.8
10.8
5.9
6.5
6.5
8.1
5.7
6.0
5.5
5.8
MEASURED PROPERTIES 3 Volume Weith Kg/m Bending Strenght 2 kp/cm Modulus of elasticity 2 kp/cm Bending Stiffness Kp.cm Tensile Strength 2 kp/cm Tensile Strength 2 kp/cm Specific Break 2 energy kpc/cm
X
A
+ Parallel to the plane ++ Perpendicular to the plane
30
D D1 1038 338
D2 1043 380
The modulus of elasticity was calculated from the graph given by the apparatus and according to the following formula: Eb= .
F x L3 4 x b x h3 x f
(kp/cm2)
Where: F= is the load applied in kp/cm2 L= is the support distance in cm b= is the sample wide in cm h= is the sample thickness in cm f= is deflection at load F in cm. So, the moment of inertia is proportional to the cube of the thickness of the hardboard. The board thickness thus has a dominant influence on the bending stiffness (11). 4.4.1.4 RESULTS AND DISCUSSION The bending strength is one of the basic mechanical properties in hardboards. The figure 4.5 graphically shows how this property varies under the influence of hydrolysis. It can be clearly seen in the same figure, that the addition of fir pulp is very significant for the improvement of bending strength. These fibers improve the interfiber bonds of fiber surfaces. The improvement of this propertie is also due to the addition of longest fibers as fir pulp fibers. Looking the table 4.5 we can see that there are not difference between the results obtained with the control specimen and 40 minutes prehydrolyzed pulp for the subvariants X 1 and A1. The best results were obtained with 50 minutes prehydrolyzed pulp. After this time of steam hydrolysis the value of bending strength falls even with the addition of the normal fir pulp. All results obtained with hydrolysis residue were better than the obtained with the control specimen.
31
Fig- 4.5 - Bending strength of test 5 hardboards. Looking the figure it can be see tha a maximum value is reached between 30 and 40 minutes of prehydrolysis time decreasing after this maximum. It can be assumed that the forces, acting on the fiber surface became so weak because of reduced share of hemicelluloses, especially pentosans, which are bearers of mechanical strength partly as such and partly in form of new substances originating from them. As the second important mechanical property of the hardboard can be considered the tensile strength parallel to the board surface. This strength, depending upon the prehydrolysis time, is-graphically shown in the fig. 4.6. By prolonged prehydrolysis to 50 minutes, the others values obtained with 55 and 60 minutes are reduced by approx. 20%. This property also is improved by the addition of normal fir pulp.
32
Fig. 4.6. Tensile strength parallel to the plain. Fig. 4.7 Tensile strength perpendicular to the plan
33
Fig. 4.8 Modulus od elasticity of test hardboards. The z-strength of the test hardboards is graphically shown in the figure 4.7. The influence of the prehydrolysis time, for this property, is more pronunced after 30 minutes, as can be seen in the graphibut even so better results were obtained with 40 and 50 minutes of prehydrolysis tip me than the obtained with the normal pulp. There is a similarity between the curves for tensile strength parallel to the plane and tensile strength perpendicular to plane, but the improvement of the property is more useful in the tensile strength parallel to the plane, by the addition of fir pulp, than in the tensile strength perpendicular to the plane.
34
Fig. 4.9 Bending stiffness. The modulus of elasticity decreases as a result of hydrolysis to the greatest extent in the range between 30 and 40 minutes prehydrolysis (see figure 4.8). A much higher effect is achieved,if 50% of normal fir fibers is added, whereby the modulus of elasticity improves by 25-35% in relation to the hardboards made from pure prehydrolysis residue. In this work the best results were obtained by the utilisation of 50 minutes prehydrolysis residue. In the figure 4.8 the influence of prehydrolysis can be seen. This property is very important when the production of fiberboards have the aim the manufacturing of panels forbuilding. Mechanical solidity and rigidity, expresed as bending stiffness, is considered to beethe great importance. Fig 4.9.
35
Fig- 4.10 Specific breaking energy.
Another important property, is impact-breaking energy, which results were plotted in the figure 4.10 and from this graph we can see the influence of steam hydrolysis. In this case the best results were achieved for the boards made with normal pulp. The best result was obtained by the utilization of 50 minutes prehydrolysis residue. Finally, we can say that the best results, in general, were achieved by the utilisation of 50 minutes prehydrolysis residue and the addition of 30% normal fir pulp. In general, the mechanical properties, measured in the test hardboards, are within the limits of properties of commercial hardboards under condition, that the prehydrolysis time does not exceed 50 minutes,
36
Fig. 4.11. Walter absorption Fig. 4.12 Swelling of test hardboards.
37
The addition of normal fibers to the prehydrolyzed fibers acts on the property of absorption and swelling of test hardboards as it can be expected. Since they contain hemicellulose in a much greater extent than prehydrolyzed fibers, it results in a higher absorption and swelling, especially, if normal beech fibers are added. The prehydrolysis acts similarly as on the water absorption also on the reversible variation of moisture content (weight) so that the quantity of humidity absorbed diminished. See tables 4.6 and 4.7. TABLE 4.6 WATER ABSORPTION AND SWELLING OE TEST BOARDS DUE TO 24 HOURS SUBMERSION IN WATER AT 20+/- 2oC.
VARIANTS AND SUBVARIANTS X A B C D
X1 X2 A1 A2 B1 B2 C1 C2 D1 D2
ABSORPTION % 60 46 25 31 20 28 19 31 17 30
SWELLING % 38 29 13 16 11 16 9 16 8 13
We can say, in this case, that the prehydrolysis time does not have a significative influence on the weight by area of the test hardboards. Because of that no change can be seen, at different prehydrolysis times. Equally a small weights were observed in the control specimens. Looking the volume weight, higher values were obtained in hardboards made out of prehydrolysis residues. See table 4.7. TABLE 4.7 SOME PHYSICAL PROPERTIES OF THE TEST HARDBOARDS VARIANTS AND SUBVARIANTS X A
WEIGHT BY AREA 2 kg/m X1 X2 A1 A2
2.99 3.00 3.5 3.5
38
MOISTURE CONTENT % 7.2 7.6 5.2 6.1
B C D
B1 B2 C1 C2 D1 D2
3.5 3.9 3.5 3.5 3.5 3.6
5.2 5.7 5.0 6.0 4.7 5.5
5. CONCLUSIONS The basic aim of the investigation work has been to determine the changes of the physical and mechanical properties of fibers, decisive for the estimation of their technical applicability without an additional chemical treatment. Beside the investigation of hydrolysis residue, also the prehydrolysis distillates were tested, whereby the interest was focused on the quantitative determination of furfural and organic acids. In general, the results of the investigation confirm the technical possibility of utilisation of prehydrolyzed beech for the production of fibers possessing specific properties which can serve for the production of hardboards. It was demostrated that the best results, given for the test hardboards, in the mechanical and physical properties tested, were obtained with fibers subjected to 50 minute prehydrolysis time. Finally, the results obtained give the possibility of a safer economic calculation, on which, further investigations of beech wood as a raw material in an industrial scale could be based.
39
6. LITERATURE 1. Adler, E., and Hernestam, S. Acta Chem. Scand. 9 (1955) 319. 2. Asplund, A., Brit. Pat. 393 159 (1933). 3. Asplund, A., and Bystedt, I. Defibrator AB. International Mechanical Pulping Conference. Stockholm,1973 pp 4-5. 4. Asplund, G. Svensk Papperstid. 55 (1952) 505. 5. Assunmaa,S. in Treiber,F., ed. Die Chemie der Pflanzenzellwand, Springer, Berlin, 1957 p.181. 6. Ayroud, A.M. Tech. Report. No.132 Pulp and paper research Institute of Canada. 1959, Seeboth: H. and Rieche, A. Brenstoff - Chemie 46 (1965) 361. 7. Back, E. Pulp and Paper Magazine of Canada. 35(1967) 8. Back, E. et.al. Textile Research Journal 37 (1967) 432 9. Back, E. and Ake K. Isacsson. Svensk Papperstid. 71(1968)544 10. Back, E. och Leif Klinga. Svensk Papperstid. 66 (1963) 11. Back, E. et.al. Forest Products Journal 24 (1974) 48 12. Bailey. A.J. Ind. Chem. Anal. Ed. 52(1936)8 13, Bannan, M.W. Tappi 40 (1957) 220 13a Biƙev, R., Himija u Industriji Sofija 38 (1966) 60p 13b Birkeland, S. Das papier 23 (1969) 705 14. Clavin, S., and Ernst Back. Svensk Papperstid. 71 (1968) 883 15. Defibrator AB. Determination of dimensional stability. Unplublished results. Defibrator AB. Stockholm. 16. DIN 52352 17. Dunlop,A.P. Ind. Eng. Chem. 4C(l948)204 18. Emerton, H.W. and V. Goldsmith. Holzforschung.*10(1956)lO8 19. FAO/ECE. Board Cons. Paper 5.13. 1.26 20. FAO/ECE. Board Cons. Paper 5.14. 1.12 21. Freudenberg, K., Neish, A.C. Constitution and Biosynthesis of lignin. Spring Verlag, Berling. Heidelberg. New York 1968 pp 110-114 22. Garland,H., Ann. Missouri Botan Garden 26 (1952) 70 23. Gierer, J. Svensk Papperstid. 73 (1970) 571 24. Hosaka,H.,et.al. Report of the Hokkaido Forest Products Research Institute 15 (1959) 1-41 25. Husemann, E., Schulz G.V. et.al. Phys. Chem 52 B (1942) 23. 26. ISO Definition and Classification 1975 27. ISO/TC 89/SC L/WG 9 14E. 28. ISO/R ± 766 1968 E. 29. ISO 767 1975 E. 30. ISO/R - 768 1968 E. 31. ISO/R - 769 1968 E. 32. ISO 819 1975 E.
40
33. Johanson, F., and Back, E. Svensk Papperstid. 69 (1966) 199 34. JUS DA 1 086/57 35. Klauditz,W, Stegmann, G. Holfzooschung 5(1951) 386. 36. " Holzforschung 6 (1952) 70 37. Klemola, A. Investigation of birchwood (Betula pubescens) lignin degraded by steam hydrolysis. Doctoral Thesis. Univ. of Helsinki. Otaniemi,1968 58. KIemola. A,and Nyman, G.A. Paperi ja Puu 48 (1966) 595. 59. Klinga, L. and Back, E. Forest Products J. Sep; (1964) 425 40. Klinga, L., Back, E. Svensk Papperstid. 24 (1965) 870 41. Klinga, L. Meddelande fran wallboardindustrins centrallaboratorium No 28 B Stockholm 1962 42. Kollmann, F., Cote, W. Principles of wood sciencies and Technology. Vol 1. Solid Wood. Spring-Verlag. Berlin 1968 pp: 18-19, 21-22, 40-4l, 51-55, 66-45. Kollmann, F. et.al. Principles 0f Wood Sciencies and Technology. Vol II. Wood Based Materials. Spring-Verlag,1975 pp: 551-555 44. Kurvegovié. M. StudijIoptimalnih uslova dobivanja celuloze za HWM viskozna vlakna iz bukovog drveta sulfatnim postupkom vodenom prehidrolizom. Disertacija. TechnoORdžN\)DFXOtet, Zagreb 1967 45. Lange, P.W. Svensk Papperstid 57 (1954) 525 46. Lampert, H. Faserplatten. VEB Fachbuchverlag, Leipzip.1967 47. Leniþ. J. Kemijske promjene bukovine, smrekovine pri izradi tvrdih plóca vlaknatica. Master Thesis. Faculty of Forestry. Zagreb,1966. 48. Leniþ, J. Utjecaj prehidrolize bukovine na kemijska i fizikalna svojstva vlakanaca. Doctoral Thesis. Faculty of Forestry. Zagreb, 1971. 49. Lowgren,V. Paper Trade J. 113 No.11.39 (1941) 50. Merenwether,J.W,T. Holzforschung 11 (1957) 65 51. Muhletlater,K. Die Feinstruktur der Zellulosemikrofibrillen. Beih. Zeit. Schweiz. Forstv. 30 (1960) 55 52. Norberg,K.G. Back, E. Svensk Papperstid. 71 (1968) 774 53. Opáci´c, I. Kemijska prerada drva. Izdavacki odjel sveúcili´cta u Zagrebu. 1968 pp: 272-282 54. Oshima, M. Wood Chemistry process engineering aspects. Chemical Process Monograph No. 11 Noyes Development Corp. New York, 1965 pp: 31 and 95 55. Pulp and Paper Science and Technology. Vol. I Pulp. McGraw Hill 1962 pp: 9394 56. Richter, G.A. Tappi 39(1956)193 57. Root,D.F., et.el. Forest Prod. J. 9 (1959) 158 58. Rydholm, S.A. Pulping Processes. lnterscience Publishers. New York, London, Sydney. 1967 pp: 49-51, 55-70, 286-294 370, 395, 397, 399, 403, 404, 410412, 652, 655-657, 826, 1145-1152, 1161-1164. 59. SCAN M-6:69
41
60. SCAN 39.43 61. Schwalbe, C.G. and Robinoff, M.Z. Angew. Chem. 24(1911) 526 âtemberger, A. Unpublished results from the Research Institute of the Factory Lesonit, II. Bistrica, 1970 65. Teppi 227 os-68 64. The Chemistry of Wood. (Ed. Browning B.L.) Interscience Publishers,New York, 1963 p 62 65. Van Buijtenen, J.P., et.al. Tappi 44 (1961) 141 66. Zoch, L.L., et.al. Tappi 52(1969) 468
42
Buy your books fast and straightforward online - at one of world’s fastest growing online book stores! Environmentally sound due to Print-on-Demand technologies.
Buy your books online at
www.get-morebooks.com Kaufen Sie Ihre Bücher schnell und unkompliziert online – auf einer der am schnellsten wachsenden Buchhandelsplattformen weltweit! Dank Print-On-Demand umwelt- und ressourcenschonend produziert.
Bücher schneller online kaufen
www.morebooks.de VDM Verlagsservicegesellschaft mbH Heinrich-Böcking-Str. 6-8 D - 66121 Saarbrücken
Telefon: +49 681 3720 174 Telefax: +49 681 3720 1749
[email protected] www.vdm-vsg.de
