DIMENSIONAL STABILITY OF PARTICLEBOARD Sergej MEDVED1, Milanka ĐIPOROVIĆ-MOMČILOVIĆ2 Mlađan POPOVIĆ2, Alan ANTONOVIĆ3, Vladimir JAMBREKOVIĆ3 [email protected] 1

3

University of Ljubljana, Biotechnical Faculty, Department of Wood Science and Technology, Ljubljana Slovenia 2 University of Belgrade, Faculty of forestry, Belgrade, Serbia

University of Zagreb, Faculty of forestry, Department of Wood Technology, Zagreb, Croatia

Abst ract : At exp osur e to wat er or moist envir onment par ticleb oar ds t ends to s well and ex pand in all dir ections. T he most vis ib le is s welling in thickness t hat is higher t ha n hor izontal ex pans ion. T hickness s welling of par ticleb oar ds is inf lu enced b y ma ny fact or s. Among most imp or ta nt ar e r aw mat er ial us ed, typ e and s har e of a dhes ive used a nd dens it y of pr odu ced boar ds. T he pur pos e of this pap er is to pr es ent t he impact of wood sp ecies us ed, shar e of adhes ive and b oar d dens it y on t hickness s welling aft er 24 hour immer s ion in wat er . S ever al t hr ee-la yer and s ingle- la yer par ticleb oar ds bonded wit h ur ea - f or ma ldehyde a dhes ive wer e ma de in lab or ator y condit ions wher e wood sp ecies (t hr ee-la yer ma de fr om spr uce, b eech, oak a n p op lar par ticles), r es in cont ent (t hr ee- la yer , cor e layer : 6 t o 9 %; sur face la yer : 10 to 13% ) and dens it y of b oar d (s ingle- layer wit h dens it y b et ween 0,7 and 1,0 g/cm 3 ) wer e a lt er ed. At b oar ds wher e r es in cont ent a nd dens it y was alt er ed industr y ma de par ticles wer e us ed. T hicknes s swelling was det er mined wit h 24- hour immer s ion t est. It was det er min ed t hat highest swelling was det er min ed when b eech par ticles wer e us ed, and lowest when oak a nd spr uce par ticles wer e us ed. T he b iggest cha nges in s welling a nd pr essur e wer e obs er ved when t he r es in cont ent was changed in cor e lay er . T he s wellin g was a ls o influ enced b y dens it y of b oar ds. It was det er min ed t hat highest swelling was obs er ved at boar ds wit h highest dens it y.

Key words: particleboard, thickness swelling, wood species, resin content, density

1525

1

INTRODUCTION

Dimensional stability of particleboards is an interesting property. When we discus about the dimensional stability length or width wise it can be determined that it is very good, better compared to wood itself. The planar or linear expansion of particleboards is 0.2 to 1.0%1. Low expansion of particleboard is due to the alignment of constituents (particles). Cross lamination is achieved by alternating the particle direction in the panel. While in plane expansion of boards is low, the expansion in vertical direction (thickness swelling) is high, higher than that observed for solid wood. The thickness swelling of particleboard is between 5 and 30% and depends on several factors. But before we discus about this factors we must explain why swelling and expansion of the board in the presence of water/vapour occurs. Nageli (1854) reported that material would swell if it takes up a liquid (in our case water) and if three conditions are fulfilled:  The dimensions of material increase with an accompanying thermal change as a result of the taking up of another phase  In a microscopic sense the material retains its homogeneity  The cohesion of material is diminished but not eliminated. Considering wood and its relation towards water we see that wood is such a material. Due to its chemical composition and specific surface, caused by the macro (lumens) and micro (capillary) structure, wood easily adsorbs and desorbs water. Due to adsorption or desorption, the mass and dimensions are changed, while the structure remains homogenous but more soft and flexible. The sorption behaviour is caused by bound water, which is held within the cell walls in molecular form by physical forces of attraction. Residing in the cell wall, bound water affects wood's bulk. This in turn affects the mass and dimensions of the wood in proportion to the volume of water gained or lost (Mizi et al., 2009).

1

From 35% to 95% relative humidity at 20 °C

1526

Since particleboards are made from wood constituents and synthetic polymer– adhesive during pressing (high pressure, high temperature, presence of moisture), it can be determined that swelling in thickness is the result of:  Swelling of wood, also called reversible or recoverable swelling and  Swelling due to the pressing and hydrothermal treatment during pressing, also called irreversible or irrecoverable swelling. As we can wood particles are not only one responsible for swelling but also the conditions under which they are pressed into board. Due to high pressure (3 N/mm2 or more) severe compression crushing of particles occurs. During exposure to water or vapour, compressed and crushed, cell wall will absorb water what will lead to regaining the original shape; shape before pressing. This leads to release of compression stresses from pressing hence swelling occurs. In addition, there is vivid effect of swelling due to the regaining of the particle original shape, i.e. particle straightening. Regaining the original shape and straightness of particles in the presence of water/vapour can cause the breakage of the glue bond hence swelling occurs (Halligan, 1970; Mizi et al., 2009). According to Medved et al (2006) the interaction between moisture and the particleboard leads to the development of swelling stresses that cause the failure of the resin bond and the dislocation of the particles within the panel. According to Liiri (1960), Buschbeck et al. (1961a, 1961b), Kehr (1962), Caroll (1963), Kehr and Schilling (1965a, 1965b), Stegmann and Bismark (1968), Halligan (1970), Grigoriou (1981), Jossifov (1989), Niemz and Steinmetzler (1992), Sekino and Asakura (1993), Mantanis et al. (1994), Dix and Marutzky (1997a, 1997b), Xu and Suchsland (1998), Ikeda and Suzuki (1999), Medved (2005), Papadapolous (2006) are:  wood species used  resin type and share of resin  production process o wood processing o drying o pressing  board density and density of individual layer

1527

The influence of wood species used is related to the sorption properties and chemical composition of wood itself, to the size and shape of particles created during chipping, behaviour of particles during pressure and later board density (Liiri (1960), Buschbeck et al. (1961a, 1961b), Kehr (1962), Caroll (1963), Kehr and Schilling (1965a, 1965b), Stegmann and Bismark (1968), Suchsland (1972) Grigoriou (1981), Niemz (1982), Jossifov (1989), Panjković and Bruči (1991), Medved (2000), Brochmann et al. (2004). Thoemen (2006)). Important is also resin type used. The lowest swelling was observed when phenol– formaldehyde or pMDI adhesive were used. Low swelling was also observed when melamine or melamine–urea–formaldehyde adhesive was used. The highest swelling was observed when urea–formaldehyde adhesive was used (Suchsland (1972), Sekino and Asakura (1993), Ikeda and Suzuki (1999), Papadapolous (2006), No and Kim (2007), Hervillard et al. (2007)). During pressing, blended particles are exposed to the high temperature that can also cause some degradation. Temperature during pressing is needed for the adhesive to cure and particles to cross link and overlap. The stronger the bond between crosslinked and overlapped particles the better the sorption properties should be. On the other hand, during this hydrothermal treatment wood is crushed, compressed, causing fractures, micro fractures and collapsed cells occurs, thus creating the conditions for irreversible swelling (Mizi et al., 2009). Klauditz (1955), Halligan (1970), Xu and Suchsland (1991), Niemz and Steinmetzler (1992), Schwab et al., (1997), Irle (1999), Medved and Resnik (2001) determined that stresses created due to pressing at high temperature and pressure are not released. Usually these unreleased stresses are greater in boards with high density and hence boards with higher density have higher swelling and higher expansion. The aim of this paper is to present the impact of wood species used, share of adhesive and board density on thickness swelling after 24-hour immersion in water.

1528

2

MATERIALS AND METHO DS

For this investigation industry particles and particles made from beech (Fagus sp.), oak (Quercus sp.), poplar (Populus sp.) and spruce (Picea sp.) were used. At boards where resin content and density was altered industry made particles were used. Those particles were made from 75% of softwoods (mostly spruce) and 25% of hardwoods (mostly beech and oak). Moisture content was between 2 to 4%. Particles were made from fresh round wood in a laboratory chipper. The moisture of wood at chipping was between 60 and 70%. Particles were then dried to 2 to 3% moisture content approximately and separated into two different size classes that are characteristic for surface and core layer respectively. For core layer particles that remained on the sieve with mesh size 1.5 mm and above were used, while for surface layer particles that passed through the sieve with mesh size 1.5 mm were used. The particles were glued in laboratory conditions with a urea–formaldehyde resin. Single layer particleboards made from the core layer particles were blended with 9% adhesive. When three layer particleboard made from different wood species were made particles for core layer were blended with 7.5% and particles for surface layer 11.5% adhesive. The share of adhesive applied on particles for the determination of impact of resin share was 6 to 9% for core and 10 to 13% for surface layer (table 1) Table 1:

Resin content of individual layer with regard to the layer Resin content Type

Surface layer

Core layer

%

%

A – D (wood species)

11.5

7.5

B – H (single layer)

/

10

I

10.0

7.5

J

11.5

6.0

K

11.5

7.5

L

11.5

9.0

M

13.0

7.5

Wax emulsion (2% in both layers) and hardener (3% in core layer) were also added to the resin. As hardener ammonium nitrate (NH4NO3) was used.

1529

Target thickness was 16 mm and target density was 0.65 g/sm3, except for the boards were impact of the density was investigated. At the determination of the density impact on thickness swelling the density of the boards was 0.7, 0.8, 0.9 and 1.0 g/cm3. The increase in density was obtained with the increase in mass of particles and related material (adhesive, wax and hardener). Blended particles were pressed into 500500 mm board. Particle mat was pressed at 180 C, 3 N/mm2 for 3 minutes. Thickness swelling was determined according to SIST EN 317 by 24–hour immersion in water. 3

RESULTS AND DISCUS ION

We have determined that wood species used, resin content and density of the board influences the thickness swelling of particleboard (table 2). Table 2:

Thickness swelling with regard to the board composition Resin content

Board

Type

SL

CL

% A

Three layer

%

Thickness mm

Density

Thickness swelling

3

g/cm

%

11.5

7.5

16.04

0.694

17.1

11.5

7.5

16.08

0.705

21.3

11.5

7.5

16.12

0.701

11.7

11.5

7.5

16.09

0.699

13.2

– spruce B

Three layer – beech

C

Three layer – oak

D

Three layer – poplar

E

Single layer

/

10.0

16.11

0.712

11.8

F

Single layer

/

10.0

16.13

0.819

14.5

G

Single layer

/

10.0

16.15

0.921

16.4

H

Single layer

/

10.0

16.16

0.998

17.5

I

Three layer

10.0

7.5

16.21

0.711

12,32

J

Three layer

11.5

6.0

16.18

0,711

12,43

K

Three layer

11.5

7.5

16.20

0,709

10,24

L

Three layer

11.5

9.0

16.17

0,712

8,31

M

Three layer

13.0

7.5

16.18

0,715

9,62

1530

As we can see the lowest swelling can be observed at boards made from oak, single layer boards with low density and three layer board where core layer was blended with 9% adhesive. 3.1

WOOD SPECIES

As it was determined by Liiri (1960), Buschbeck et al. (1961a, 1961b), Kehr (1962), Caroll (1963), Kehr and Schilling (1965a, 1965b), Stegmann and Bismark (1968) Grigoriou (1981), Jossifov (1989), Panjković and Bruči (1991), Medved (2000) and Medved (2005) wood species is important when thickness swelling is in question (figure 1).

Figure 1:

Thickness swelling with regard to the wood species used

The impact of wood species on the thickness swelling is related to its sorption properties, anatomical and chemical composition, size of particles and its compressibility. Looking at chemical composition the thickness swelling is related to the content of lignin and cellulose. The higher the content of the lignin the lower the swelling. With regard to the chemical composition of individual wood species, the share of cellulose also plays an important role. If the sorption behaviour of chemical elements in wood is compared then the most hygroscopic is hemicelluloses, followed by cellulose and lignin (Gorišek, 2009). Consequently, it is logical that using a wood species with a high proportion of hemicelluloses would result in unstable particleboard. According

1531

to Mantanis et al. (1994) swelling is also influenced by extractives. They determined that removal of extractives caused an increase in the maximum swelling of wood in water. According to Moslemi (1974) less swelling is also expected when wood species with high density are used due to the lower degree of mat compaction. If we compare thickness swelling of boards made from oak compared to those made from spruce and poplar similar can be concluded. 3.2

DENSITY

The board density is important for thickness swelling (figure 2).

Figure 2:

Thickness swelling with regard to the particleboard density

The relation between density and thickness swelling is due to the higher densification of particles, and due to induced stresses during pressing. As we mentioned particles are, during pressing, crushed, compressed, also overlapped and in curved shape hence causing fractures, micro fractures and collapsed cells walls. Since particles, pressed to higher density are more deformed there is a greater potential for reversal of densification and particle straightening hence higher thickness swelling. One of the reason for higher selling of boards with higher density is also related the amount of

1532

the particles in the board. Since the goal was to produce the boards with same nominal thickness at same pressure, we had to increase the mass of wood particles. Since more wood was used higher swelling, at boards with higher density, was determined. 3.3

RESIN CONTENT

Swelling of particleboards is also affected by the resin content (figure 3).

Figure 3:

Thickness swelling with regard to the resin content

Thickness swelling of particleboards decreases with increasing resin content. Similar was observed by Schneider et al (1996). More adhesive added to particle means that greater area of particles is covered with adhesive. If greater area is covered by adhesive that also means that particles are more bonded together, hence less water can penetrate between particles and also less regaining of the particle original shape, i.e. particle straightening is observed. As we see in figure 3 the greatest decrease in swelling was determined when resin content was altered in core layer. Core layer is less dense and more porous than surface layer. Higher porosity, more free spaces between particles enables faster penetration of water between particles. Since the resin does not cover entire particle surface and its penetration is small, that water can easily penetrate into the particles. Easy penetration of water into micro voids and micro fractures results in greater thickness swelling and a higher swelling stress.

1533

4

CONCLUSIO NS

Dimensional stability, especially thickness swelling is important property of particleboards. In this paper we investigated the impact of wood species used for particles, resin content and board density on thickness swelling of particleboards. It was determined that lowest swelling was observed when boards were made from oak wood, and highest swelling when beech particles were used. When poplar particles were used swelling was higher compared to oak, but lower compared to spruce and oak. Decrease in density resulted in less swelling compared to the boards with higher density. Less swelling was also observed at higher resin content. The lowest swelling was observed when resin content was 9% for core layer and 11.5% for surface layer. Higher thickness swelling was observed at board with low resin content in core layer. It can be conclude that the condition of core layer (resin content) have higher impact on thickness swelling than changes of resin content in surface layer.

1534

5

REFERENCES

Brochmann, J., Edwardson, C. Shmulsky, R. (2004) Influence of resin type and flake thickness on properties of OSB. Forest products journal. 54, 3, 51-55. Buschbeck, L., Kehr, E. and Jensen, U. (1961a) Untersuchungen über die Eignung verschiedener Holzarten und sortimente zur Herstellung von Spanplatten – 1. Mitteilung: Rotbuche und Kiefer. Holztechnologie, 2, 2, s. 99–110 Buschbeck, L., Kehr, E. and Jensen, U. (1961b) Untersuchungen über die Eignung verschiedener Holzarten und sortimente zur Herstellung von Spanplatten – 2. Mitteilung: Kiefernreiserholz. Holztechnologie, 2, 3, s. 195–201 Carroll, M. (1963) Whole wood and mixed species as raw material for particleboard. Bulletin 274 Washington State University Pullman Dix, B. and Marutzky, R. (1997a) Nutzung von Holz aus Kurzumtriebsplantagen (I). Holz–Zentralblatt, 123, 9, s. 141–142 Dix, B. and Marutzky, R. (1997b) Nutzung von Holz aus Kurzumtriebsplantagen (II). Holz–Zentralblatt, 123, 10, s. 154–155 Gorišek, Ž. (2009) Les: zgradba in lastnosti: njegova variabilnost in heterogenost. Ljubljana: Biotehniška fakulteta, Oddelek za lesarstvo, 2009 Grigoriou, A. (1981) Der Einfluß verschiedener Holzarten auf die Eigenschaften dreischichtiger Spanplatten und deren Dechschichten. Holz als Roh– und Werkstoff, 39, 3, s. 97–105 Ikeda, M. and Suzuki, S. (1999) Evaluation of the Durability Performance of Woodbased Panels Subjected to Outdoor Exposure., Mokuzai Gaikaishi, 23, s. 25-36 Haligan A. F. (1970) A review of Thickness Swelling in Particleboard. Wood Science and Technology, 4, 4,: 301-312 Hervillard, T., Cao, Q. and Laborie, M.P.G. (2007) Improving waterresistance of wheat straw-based medium density fiberboards bonded with aminoplastic and phenolic resins. BioResources, 2, 2, 148-156

1535

Irle, M.A. (1999) An Investigation of the Influence of Wood Density on the Rheological Behaviour and Dimensional Stability of Hot-Pressed Particles. 3rd European Panel Products Symposium, 7-18 Jossifov, N. (1989) Wechselbeziehungen zwischen der Dichte und wesentlichen physikalisch–mechanischen Eigenschaften industriell hergestellter mehrschichtiger Spanplatten aus Hartlaubholz. Holztechnologie, 30, 4, s. 200–202 Kehr, E. (1962) Untersuchungen über die Eignung verschiedener Holzarten und sortimente zur Herstellung von Splanplatten. Holztechnologie, 3, 1: 22–28 Kehr, E. and Schilling, W. (1965a) Untersuchungen über die Eignung verschiedener Holzarten und –sortimente zur Herstellung von Spanplatten – 6. Mitteilung: Birke. Holztechnologie, 6, 225 - 232 Kehr, E. and Schilling, W. (1965b) Untersuchungen über die Eignung verschiedener Holzarten und –sortimente zur Herstellung von Spanplatten – 7. Mitteilung: Eiche, Aspe, Pappel, Hainbuche, Ulme, Lärche sowie als Vergleichsholzarten Fichte und Kiefer. Holztechnologie, 6, 225 - 232 Klauditz, W. (1955) Entwicklung, Stand und holzwirtschaftliche Bedeutung der Holzspanplattenherstellung. Holz als Roh– und Werkstoff. 16, 459 – 466 Liiri, O., Kivistö, A. and Saarinen, A. (1977) Der Einfluß von Holzarten, Spangröße und Bindemittel auf Festigkeit und die Quellung von Spanplatten mit höheren elastomechanischen Eigenschaften. Holzforschung und Holzverwertung, 29, 6, s. 117–122 Mantanis, G.I., Young, R.A. and Rowell, R.M. (1994) Swelling of wood – Part 1: Swelling in water. Wood Science and Technology, 28, 119-134 Medved, S. (2000) Vpliv zgradbe zunanjega sloja na sorpcijo in trdnost iverne plošče = Influence of structure of surface layer composition on sorption and strength of the particleboard. Les, 52 (1/2), 5 – 13 Medved, S. (2005) Advantages and disadvantages of mixed wood species utilization for wood based panels. Proceedings of the COST Action E44 Conference: June 14th -

1536

15th 2005, Vienna, Austria, (Lignovisionen, 9). Vienna: Boku Vienna, 2005, str. 167173 Medved, S. and Resnik, J. (2001) Thickness swelling of the individual layers of a three layer particleboard. Proceedings of the Fifth International Conference on the Development of Wood Science, Wood Technology and Forestry, ICWSF 2001, 5th 7th September 2001, Ljubljana, Slovenia. Ljubljana: Biotechnical Faculty, Department of Wood Science and Technology, 2001, 115 – 123 Medved, S., Šernek, M. and Šega, B. Thickness swelling and swelling pressure of wood-based panels curing. Wood resources and panel properties : conference proceedings: Cost Action E44-E49, Valencia, Spain, 12-13 June 2006. Valencia: AIDIMA, Furniture, wood and packaging technology institute, 2006, 123 – 129. Mizi, F. et al., 2009. Performance in use and new products of wood based composites. London: Brunel University press Moslemi, A. A. (1974) Particleboard – Volume 1: Materials. Amsterdam, London, Southern Illinois University Press, s. 7–19 Nageli, C.V. (1854) Die starkenkorner. Morphologische Physiologische, ChemischPhysikalische und Systematisch-Botanische Nomographie, Zurich, Switzerland Niemz, P. (1982) Untersuchungen zum Einfluß der Struktur auf die Eigenshaften von Spanplatten – Teil 1: Einfluß von Partikelformat, Rohdichte, Festharzanteil und Fastparaffinanteil. Holztechnologie, 23, 4, s. 206–213 Niemz, P. and Steinmetzler, J. (1992).

Untersuchungen zum Quelldruck bei

Feuchteänderung von MDF. Holz-Zentralblatt, 34, 6, 83. No, Y.B., Kim, M.G. (2007) Evaluation of melamine-modified urea-formaldehyde resins as particleboard binders. Journal of Applied Polymer Science, 106, 6, 41484156 Panjkovic, I. and Bruci, V. (1991) Utjecaj razlicitih vrsta drva na fizicko–mehanicka svojstava troslojnih iverica. Drvna industrija, 42, 3/4, s. 55–60

1537

Papadapolous, A.N. (2006) Property comparisons and bonding efficiency of UF and pMDI bonded particleboards as affected by key process variables. BioResources, 1, 2, 201-208 Schwab, E., Steffen, A. and Korte, C. (1997) Feuchtebedingte Längenänderungen von Holzwerstoffen in Plattenebene. Holz als Roh– und Werkstoff, 55, 227 – 233 Sekino, N. and Asakura, N. (1993) Humidity control efficiency of low-density particleboards for interior walls II. Measurement of EMC and calculation of moisture capacities. Mokuzai Gakkaishi = Journal of the Japan Wood Research Society, 39(10), 1146 – 1151 Stegmann, G. and Bismarck, C. (1968) Verarbeitung von Laubhölzern für die Spanplattenherstellung. Holzforschung und Holzverwertung, 20, 1 – 11 Suchsland, O. (1972) Linear hygroscopic expansion of selected commercial particleboards. Forest Products Journal. 22(11), 28 – 32 Thoemen, H. (2006) The Effects of the Wood Raw Material on Panel Properties: A Fundamental Approach. Wood resources and panel properties : conference proceedings: Cost Action E44-E49, Valencia, Spain, 12-13 June 2006. Valencia: AIDIMA, Furniture, wood and packaging technology institute, 2006 Turner, H.D. (1954) Effect of particle size and shape on strength and dimensional stability of resin bonded wood particle panels. Forest Products Journal, 4(5), 210 – 223 Xu, H. and Suchsland, O. (1991) The expansion potential: a new evaluator of the expansion behavior of wood composites. Forest Products Journal, 41(6), 39 – 42 Xu, W. and Suchsland, O. (1998) Variability of particleboard properties from single– and mixed–species process. Forest Products Journal, 48, 9, s. 68–74 EN 317. Particleboard and fibreboards – Determination of swelling in thickness after immersion in water. 1993

1538