THE EFFECT OF VARIABLE ENVIRONMENTAL CONDITIONS ON DIMENSIONAL CHANGES IN THIN WOOD-BASED MATERIALS. PART I. ABSORPTION CHANGES

Radosław Mirski
Department of Wood-Based Materials, Poznań University of Life Sciences, Poland

ABSTRACT

In this study dimensional changes caused by cyclical changes in relative humidity at constant air temperature were analyzed in thin wood-based materials, according to EN 318. Tests were conducted on MDF, HDF, PB and OSB. Boards used in the tests, except for MDF, were boards with an enhanced resistance to the action of air with elevated humidity. It was found that final relative changes in board dimensions significantly depend on changes in dimensions occurring at individual stages of conditioning. The smallest relative change in length after the last stage of conditioning was recorded in OSB; however, dimensional changes in these boards in relation to their initial length are much bigger than those of the other boards. In turn, the biggest dimensional stability was found for particleboard.

Key words: MDF, HDF, PB, OSB, dimensional stability.

INTRODUCTION

Major changes observed in used wood seem to be dimensional changes, caused by natural or accidental changed in relative humidity. Linear changes and changes in volume in many natural wood species have already been thoroughly investigated and presented in the tabular form [6]. However, products manufactured from natural wood due to their high price and the depleting material base of good quality timber are being replaced wherever possible by wood-based materials. Easy manufacture of products with a large usable area, made mainly from poor quality wood or small-size timber, unusable in minimally processed form, results in a situation when board production industry has been developing dynamically for many years, with wood-based materials finding new applications in an increasing range of areas [4].

The use of different degrees of fineness and a variety of binding agents on the one hand makes it possible to manufacture large, homogenous surfaces and on the other hand significantly improves properties of manufactured materials in relation to the original wood, including also its hydrophobicity. However, it results from an analysis of bibliography on the subject that exposure of wood-based materials to air with elevated or cyclically changing relative humidity results in a deterioration of their mechanical properties [2,3,7,10]. It seems that, apart from insight into changes in mechanical properties of boards caused by changes in relative humidity, it is also essential to know how under such conditions their dimensions are modified. On the one hand, this facilitates the determination of an optimum width of expansion gaps and on the other hand, cooperation in combination with other materials. For this reason in many research centres numerous research projects are being conducted to determine the level of dimensional changes, mainly length, taking place in wood-based materials under the influence of changing humidity [8,9,11,12].

The aim of this study was to investigate dimensional changes in selected wood-based boards caused by cyclical changes in relative humidity at a constant ambient temperature.

MATERIAL AND METHODS

Tests were conducted on the following types of boards, with thickness ranging from 6 to 8 mm:
– MDF – resinated with UF resin,
– HDF – resinated with MUF resin,
– PB (particleboards) – resinated with MUPF resin,
– OSB/3 – outer layers resinated with MUPF resin and the core – with PMDI resin,
– OSB/4 – outer layers resinated with MUPF resin and the core – with PMDI.

Boards tested in this study, except for MDF, were boards with an elevated water resistance, to be used as floor panels or siding elements, periodically exposed to elevated humidity. Prior to the onset of the tests their basic properties were determined and they are presented in Table 1. As it results from the given data, all boards used in the study were characterized by very good mechanical properties, especially in terms of bending strength and modulus of elasticity.

In order to determine the effect of humidity at a temperature of 20°C boards were subjected to the cyclical action of air at humidity of 30%, 65% and 85%. Adopted levels of relative humidity and the used testing machine complied with respective standard EN 318. This standard assumes only three conditioning degrees (1–3), whereas in this study this number was increased and the concept of a stage was introduced, referring to a period after which it was decided to determine dimensional changes of boards following formula 1.1. The complete conditioning cycle is presented in Table 2.

where:
x – the level of relative humidity, after which relative change in length or thickness is determined,
n – degree n = 1, 3, 5, 7, 9,
δl_65,x – relative change in length at a change in relative humidity from 65% to x, [mm/m],
δt_65,x – relative change in thickness at a change in relative humidity from 65% to x, [%].

The assumed system of changes in relative humidity for degrees 1, 2, 3 corresponds to absorption changes as defined in Standard EN-318, thus it was decided to define the adopted conditioning schedule as absorption schedule.

In order to determine changes in dimensions occurring in the tested material as a result of the entire conditioning period, also relative changes in length taking place between stage 10 and 0 were also determined (65₁₀ → 65₀).

RESULTS AND DISCUSSION

Tests concerning dimensional changes in selected wood-based materials, caused by cyclical changes in relative humidity at a constant ambient temperature, showed that after stage zero of conditioning relative change in length of fiberboards is comparable, amounting to -0.45 mm/m and -0.43 mm/m, respectively, for MDF and HDF boards (Tables 3 and 4). Relative changes in length of particleboards and OSB analyzed along the longer axis and recorded after this stage of conditioning were only slightly higher (Tables 5-7). In contrast, the biggest shrinkage was recorded for OSB/3 analyzed along the shorter axis (Table 8), being by approx. 40% bigger in comparison to the value determined for the longer axis. Much bigger changes in length for the shorter axis of OSB result most probably from the specific arrangement of chips in the board, as most of them are arranged along the board axis and changes in length are defined in this case in the radial or tangential direction of wood. Such assumptions seem justified, since it is commonly known that shrinkage of wood in pine (Pinus sylvestris L.) at a change in its moisture content by 5% results in the longitudinal direction in a change of its dimensions by 0.6 mm/m. Changes recorded for tested OSB for obvious reasons are smaller. In turn, for the other directions these changes may be even 10 times bigger.

However, this problem is much more complicated, since these are not free changes. Moreover, different resination rates and different types of adhesives are found in tested boards and individual layers have different densities. Moreover, different weight fractions may be observed for individual layers. Presented factors have a highly significant effect on the behaviour of tested wood-based material and as such require in-depth analyses. In the following stage of the study board samples were subjected to gradual wetting. In case of fiberboards it may be observed that after the third degree of conditioning they exhibit much lower moisture content than particleboards or OSB. Moisture content of these boards is around 8%, while for particleboards and OSB it is approx. 12 and 15%, respectively. However, despite such large differences in moisture contents for all variants, except for the shorter axis of OSB and longer axis of OSB/4, relative changes in length were similar, being 0.72–0.79 mm/m. In turn, the lowest value of elongation of 0.62 mm/m was recorded for OSB/4. The biggest relative changes in length were observed, similarly as after the zero stage, for the shorter axis of OSB, being by 70% bigger than for the other boards. However, results found for fiberboards after this stage of analysis were not consistent with the results reported by Ayrilmis [1]. In his study he showed that with an increase in density of MDF increasing values are also observed for their relative dimensional change (MDF/HDF – 720:1000 kg/m³), determined in accordance with the analogous standard DIN EN 318. In his study the author used boards with a thickness of 11 mm, which relative change in length after an analogous conditioning time (stage I) was approx. two times bigger, while that of thickness by approx. 40%. Such big differences in the recorded relative change in length of boards may be the effect of differences in the type of wood used in their manufacture or applied resination method. Similarly, testing results recorded in this study for particleboards are in divergence with the results reported by Suzuki and Miyamoto [9], who showed a linear correlation between elongation and density of particleboards. Those authors showed that single-layer boards made from cedar, 10 mm thick and with a density of 800 kg/m³, after conditioning at 40°C and relative humidity of 90% lengthened by 4.3 mm/m (0.43%), i.e. over 5 times more than the particleboard used in the study. This confirms the significant effect of many factors, such as the type of wood, degree of fineness and the number of chip layers, applied resins and pressing parameters or other technological factors, which prevents direct reference of results given in literature sources to wood-based materials, which are identical only in name.

Significant differences between boards were found as late as the second stage of conditioning. The very big shrinkage observed in case of fiberboards and particleboard, was comparable with the shrinkage of OSB/3 determined along the shorter axis. We need to stress here especially high values of relative change in length of fiberboards after this stage of analyses, although their moisture content decreased by only 3% and 2% for MDF and HDF, respectively. Such behaviour of fiberboards may indicate that after transfer from conditioning stage 3 to 4 these boards did not return to their original size and significant shrinkage occurred only after a reduction of relative humidity by a further 35%. As a consequence relative changes in length were so big. Such an assumption may also be confirmed by the fact that despite sample conditioning being repeated only twice at relative humidity of 85%, all boards used in the tests after conditioning degree 10 were characterized by a positive relative change in length, calculated in relation to the initial dimension (Fig. 1).

In successive conditioning stages (III, IV) for all variants of boards an increase was observed in relative change in length in relation to that recorded earlier, both after stage I for absorption changes and after stage II for desorption changes. Consistently bigger changes were found for analyzed fiberboards and particleboard. Relative change in length of MDF or HDF after conditioning stage IV was by almost 1 mm/m bigger than that recorded for OSB/3 or OSB/4 for the longer axis. Despite that fact, the biggest increment in length for OSB, primarily OSB/3, was observed along the smaller axis, caused by conditioning performed according to the assumed schedule (Fig. 1). This was so because a change in length along this axis for OSB/3 was approx. 0.8 mm/m and by over 50% bigger than that for OSB/4 tested in the same direction.

It results from values of relative change in thickness presented in Tables 3-7 for boards used in this study that after the zero conditioning stage the smallest change in thickness was found for HDF. In this case shrinkage was only 0.22%, in comparison to an almost two times bigger in MDF and almost 5 times bigger in OSB/4. This low value of shrinkage in HDF was probably caused by their high density and good hydrophobic protection. In turn, it is the structure of OSB that is responsible for considerable changes in thickness in these boards, observed also after the second stage of conditioning including sample wetting. High roughness of board surface resulting from the specific chip dimensions makes precise measurement of thickness very difficult. The measurement site (the place where the sensor is applied) may be either one chip with a large area or several adjacent chips. Irregular chip swelling or damage to the glue line in one of them may cause a local strong increase in board thickness. In contrast, particleboard used in this study was characterized by small changes in thickness. As it results from data presented in tab. 5, the action of neither air with elevated (85%) nor that with reduced (30%) relative humidity had any significant effect on changes in thickness of this board, since after the third stage of conditioning its swelling was only 0.64%, and shrinkage after the fourth stage was -1.63%. For particleboards also the smallest changes in thickness were recorded after the completion of analyses in comparison to the initial thickness and its final swelling in thickness was almost 8 times smaller than that of OSB/3 (Fig. 2). Such good hydrophobic properties of particleboards is probably the result of high resination rate and high pressing rate, which as a consequence facilitates the generation of high density not only in outer layers, but also in the core. Such a line of reasoning seems to be confirmed by values of basic properties of particleboards (Table 1). High values of the modulus of elasticity and internal bond for these boards result primarily from their high density and high resination rate.

CONCLUSIONS

As a result of conducted analyses it was found that after the zero conditioning stage the smallest relative change in length was recorded for fiberboards. Shrinkage for these boards was almost by ½ smaller than that recorded for OSB tested along the shorter axis. For successive desorption stages an increase in shrinkage was observed for all tested boards, which was more intensive in case of fiberboards. The smallest relative change in length after conditioning stage IV was found for fiberboard. Also relative changes in thickness determined for successive stages of conditioning in case of this board were relatively much smaller than those of the other boards. Such good hydrophobic properties of particleboards may probably ascribed to the high resination rate, applied hydrophobic agents and high pressing degree, which as a result makes it possible to obtain high density, almost uniformly distributed at the board cross-section.

REFERENCES

1. Ayrilmis N., 2007. Effect of panel density on dimensional stability of medium and high density fiberboards. J. Mat. Sci., 42, 8551-8557.

2. Czarnecki R., Derkowski A., Mirski R., 2003. Comparison of properties of properties of OSB/3 and OSB/4 boards subjected to action of humid conditions. Ann. Warsaw Agric. Univ. For. Wood Technol. 53, 28-31.

3. Derkowski A., Łęcka J., Mirski R., 2002. Properties of OSB/3 boards in dependence upon the environment humidity. Ann. Warsaw Agric. Univ. For. Wood Technol. Spec. No I 2002, 91-94.

4. Hikiert M. A., Oniśko W., 2006. Development strategy wood based materials industry in Poland to 2013. Kom. Gosp. Poznań, 2006.

5. Kangas K., Baudin A., 2003. Modelling and projections of forest products demand, supply and trade in Europe. A study prepared for the European Forest Sector Outlook Study (EFSOS). New York.

6. Kokociński W., 2004. Drewno [Wood measurement of physical and mechanical properties]. Wydawnictwo-Drukarnia Prodruk, Poznań [in Polish].

7. Mohammad M.A.H., Smith I., 1996. Effects of multi-phase moisture conditioning on stiffness of nailed OSB-to-lumber connections. For. Prod. J. 46(4), 78-83.

8. Pritchard J., Ansell M.P., Thompson J.H., Bonfield P.W., 2001. Effect of two relative humidity environments on the performance properties of MDF, OSB and chipboard. Wood Sci. Technol. 35(5), 395-403.

9. Suzuki S., Miyamoto K., 1998. Effect of manufacturing parameters on the linear expansion and density profile of particleboard. J. Wood Sci. 44, 444-450.

10. Wu Q., Lee J.N., 2002. Thicknes swelling of oriented strandboard under long-term cyclic humidity exposure condition. Wood Fiber Sci. 34(1), 125-139.

11. Wu Q., Suchsland O., 1996. Linear expansion and its relationship to moisture content change for commercial oriented strandboards. For. Prod. J., 46(11/12), 79-83.

12. Wu Q., Suchsland O., 1997. Effect of moisture on the flexural properties of commercial oriented strandboards. Wood Fiber Sci., 29(1), 47-57.

Accepted for print: 24.09.2009

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.