Title Effects of the Cyclic-Humid Treatment on the Mechanical andSurface Properties of Wood-Based Panels( 本文(Fulltext) )
Author(s) SAHRIYANTI SAAD
Report No.(Doctoral Degree) 博士(農学) 甲第680号 Issue Date 2017-09-22 Type 博士論文 Version ETD URL http://hdl.handle.net/20.500.12099/73133 ※この資料の著作権は、各資料の著者・学協会・出版社等に帰属します。
Effects of the Cyclic-Humid Treatment on the
Mechanical and Surface Properties of
Wood-Based Panels
(
ᮌ
ᮌ㉁ࣃࢿࣝࡢᶵᲔⓗᛶ㉁࠾ࡼࡧ⾲㠃≉ᛶཬࡰࡍ
྾‵⧞㏉ࡋฎ⌮ࡢᙳ㡪
)
2017
The United Graduate School of Agricultural Science,
Gifu University
Science of Biological Resources
(Shizuoka University)
Effects of the Cyclic-Humid Treatment on the
Mechanical and Surface Properties of
Wood-Based Panels
(
ᮌ
ᮌ㉁ࣃࢿࣝࡢᶵᲔⓗᛶ㉁࠾ࡼࡧ⾲㠃≉ᛶཬࡰࡍ
྾‵⧞㏉ࡋฎ⌮ࡢᙳ㡪
)
TABLE OF CONTENTS Page 7,7/(3$*(««««««««««««««««««««««««««« i 7$%/(2)&217(176««««««««««««««««««««««. ii /,672)7$%/(6««««««««««««««««««««««««. iii /,672)),*85(6«««««««««««««««««««««««... iv CHAPTER 1 *(1(5$/,1752'8&7,21«««««««««««««. 1 1.1 %DFNJURXQG««««««««««««««««««««««.... 1 1.2 Objectives of the VWXG\««««««««««««««««««««« 3 1.3 Structure of the GLVVHUWDWLRQ«««««««««««««««««««« 4 CHAPTER 2 /,7(5$785(5(9,(:«««««««««««««««... 5 2.1 Durability performance concept««««««««««««««««.... 5 2.2 (YDOXDWLRQGXUDELOLW\EDVHGRQSURGXFWLRQWHFKQRORJ\«««««««««. 6 2.2.1 (IIHFWVRISURGXFWLRQYDULDEOHV««««««««««««««... 6 2.2.2 Evaluating potential material for new product«««««««««.. 9 2.3 $FFHOHUDWHGDJLQJWHVW«««««««««««««««««««««.. 10 2.3.1 Outdoor exposure test ««««««««««««««««««« 10 2.3.2 Laboratory exposure test««««««««««««««««««. 11 CHAPTER 3 EFFECTS OF CYCLIC AGING TREATMENT ON
PERFORMANCE OF WOOD-BASED PANELS «««««««««««« 12 3.1 ,QWURGXFWLRQ«««««««««««««««««««««««««« 12 3.2 Materials DQGPHWKRGV«««««««««««««««««««««.. 14 3.2.1 6SHFLPHQSUHSDUDWLRQ«««««««««««««««««««. 14 3.2.2 Mechanical properties evaluation ««««««««««««««.. 14 3.2.2 An accelerated aging treatment «««««««««««««««« 15 3.3 Results DQGGLVFXVVLRQ«««««««««««««««««««««.. 16
3.3.1 Intensity of the wet-dry cycle treatment condition «««««««« 16 3.3.2 Strength retention under the wet-dry cycle treatment «««««««. 18 3.3.3 Estimating internal bond strength «..«««««««««««««. 21 3.4 &RQFOXVLRQ««««««««««««««««««««««... 22
Table of contents (continued) Page
CHAPTER 4 MECHANICAL AND SURFACE DEGRADATION OF
WOOD-BASED PANELS AS THE EFFECT OF AGING TREATMENT ...««. 31 4.1 ,QWURGXFWLRQ««««««««««««««««««««««««... 31 4.2 0DWHULDOVDQGPHWKRGV«««««««««««««««««««««.. 33 4.2.1 Specimen preparation and accelerated aging «««««««««... 33 4.2.2 Measurement and testing «««««««««««««««««.. 33 4.3 ResXOWVDQGGLVFXVVLRQ«««««««««««««««««««««.. 35 4.3.1 Effect of the aging treatment on surface quality «««««««««. 35 4.3.2 Effect of aging treatment on mechanical properties «««««««« 38 4.3.3 Correlation of surface and mechanical degradation ..«««««««.. 39 4.4 &RQFOXVLRQV«««««««««««««««««««««««««.. 40 CHAPTER 5 GENERAL CONCLUSIONS ««««««««««««... 53 5()(5(1&(6««««««««««««««««««««««««««. 55 ACKNOWLEDGEMENTS «««««««««««««««««««««. 64 6800$5<«««««««««««««««««««««««««««.. 66 Summary in English««««««««««««««««««««««« 66 6XPPDU\LQ-DSDQHVH««««««««««««««««««««««. 70 LIST OF PUBLICATIONS CONCERNING THE DISSERTATION «««««. 72
LIST OF TABLES
Page Table 3.1 Characteristics of the specimens «««««««««««««. 24 Table 3.2 Coefficient values of the panels for the wet-GU\F\FOHWUHDWPHQW«« 25 Table 4.1 Initial properties of panel specimens in aging treatment ««««... 42 Table 5.1 Surface roughness parameters of the panels before and after the
LIST OF FIGURES
Page Figure 3.1 Panel types used in this study. See 7DEOHIRUDEEUHYLDWLRQV««. 25 Figure 3.2 Schematic diagram of a wet-dry cycle treatment. Weight change
(WC) occurs corresponding to the humidity changes under wet- and
dry-conditions «««««««««««««««««««««. 26 Figure 3.3 ǻWeight change (%) of panel products after a complete 80 cycles... 26 Figure 3.4 ǻ7hickness swelling (%) of panel products after a complete 80
cycles «««««««««««««««««««««««« 27 Figure 3.5 The changes of thickness swelling (%) for all panels during wet-dry
cycles treatment ««««««««««««««««««««.. 27 Figure 3.6 The fluctuation of thickness swelling (%) for PWa and OSBa over 80
cycles ««««««««««««««««««««««««. 28 Figure 3.7 Ed (GPa) changes during each state over 80 cycles «««« 28
Figure 3.8 Ed retention (%) of the panel products in the wet state ««««« 29
Figure 3.9 Ed retention and retention model (F(t)) of PBa and OSBa during the
wet-dry cycle treatment in the dry state«««««««««««« 29 Figure 3.10 Tan į of the panel products in relation to the wet state during the
wet-GU\F\FOHWUHDWPHQW«««««««««««««««««« 30 Figure 4.1 A stylus-type profilometer used in this study. a surface profile
measurement on the specimen, b tracing of the detector stylus .««.. 44 Figure 4.2 A non-destructive vibration method using sound level meter with fast
Fourier transform (FFT) analyzer «««««««««««««« 44 Figure 4.3 Typical vicinal surface profile before (left side) and after (right side)
the aging treatment; (a, b): PB(PF), (c, d): PB(MDI), (e, f):
MDF(MUF), and (g, h): MDF(MDI) ««««««««««««.. 45 Figure 4.4 Average roughness (Ra) changes during aging treatment; (a): PB(PF),
E 3% 0', F 0') 08) DQG G 0') 0', «««««.. 46 Figure 4.5 Fig. 4.3 Thickness swelling (TS) during aging treatment; (a):PB(PF),
E 3% 0', F 0') 08) DQG G 0') 0', «««««« 47 Figure 4.6 Micrographs (x100) of the surface of unaged MDF(MUF) (a), aged
0') 08) E XQDJHG0') 0', F DQGDJHG0') 0', G «. 48 Figure 4.7 Average Ed values of panels before and after the aging treatment«« 49
Figure 4.8 Ed retention during aging treatment; (a): PB(PF), (b): PB(MDI), (c):
MDF(MUF), and (d): MDF(MDI). White circles are values in dry
state and black circles are values in wet state ««««««««« 50 Figure 4.9 Measurements of tan įduring aging treatment; (a): PB(PF), (b):
PB(MDI), (c): MDF(MUF), and (d) MDF(MDI). White circles are
values in dry state and black circles are values in wet state «««« 51 Figure 4.10 Relationship between changes of surface roughness and mechanical
properties at dry condition. (a): Ra increment and Ed retention, and
(b) Ra incement and tan įincrement «««««««««««« 52
CHAPTER 1
GENERAL INTRODUCTION
1.1 Background
Wood-based panel is a group of products made of discrete wood elements combined with adhesive or resin under heat and pressure. It has long been introduced as substituted product for solid wood in response of diminishing supply of large diameter logs, decreasing quality of wood resources and increasing need for housing by expanding human population. Wood-based panel has been extensively developed for furniture industry and later continued to expand for structural purposes.
The most common wood-based panel products are plywood, oriented strand board (OSB), particleboard, and fiberboard. Nowadays, the demand of those panels still shows the uptrend especially for housing construction. Since the applications of wood products are in indoor and outdoor sphere, some alteration of the products will take place which are influenced by changing condition surrounding the products. They will lose the ability to sustain their performance in the real-use condition. Those matters bring about a thought that it is important to understand about durability performance of the wood-based materials in service life. Durability performance of wood-based materials is defined as SURGXFW¶V resistance to deteriorating factors. Many factors affect durability performance of wood-based materials such as moisture, temperature, UV light, biological attack and mechanical load. Nevertheless moisture and temperature are considered as mainly factors which influence longevity of wood-based panels.
Laboratory accelerated aging treatments have been developed as a short-term experiment which can be used to predict long-term durability performance of wood-based materials. It can assess the occurrence of products degradation by the changing properties of wood product. Many studies have been conducted to assess durability performance of wooden product with different treatment, such as water immersion, boiling, steaming, freezing and drying. Some of them were specified as standardized method. A method developed by ASTM (American Society for Testing and Materials, ASTM-D 1037), called ³ASTM 6-cycles test´ is a laboratory aging method which was widely used in many studies to predict the resistance of wood-based fiber and particle panel materials. Evaluating durability of wood-based panels using other accelerated tests have been performed, such as WKH :HVW &RDVW $GKHVLYHV 0DQXIDFWXUHU¶V $VVRFLDWLRQ :&$0$ WKH %ULWLsh Standard (BS) 5669, boil test (JIS-B), cyclic soak-dry test (APA D-1), the cyclic boil-dry test, Vacuum-Pressure-Soak-Dry (VPSD), V313 (European standard 321 1993) (Lehmann 1977; McNatt 1989; Kajita et al. 1991; Kojima et al. 2010, 2011). The common problem which faced in those methods is too severe to simulate the actual condition as the procedures involved extreme conditions which are unreal to happen in the reality either continually or intermittently.
The abovementioned accelerated aging tests are employed over a short period of time and set to rapidly approximate the effects of many years of outdoor exposure. These considered suitable for quality control purposes. However, they are inadequate to provide degradation mechanisms of the wood-based panel because generally, the changes in properties are evaluated only before and after the aging treatment. Some questions may be asked to assess our understanding of degradation mechanisms of the wood-based panel itself. At what extent the changes in properties may occur in the start, during, and in the end
of the aging treatment? What actually progressing during aging treatment before reaching failure? Are similar performance of inside and surface parts of the wood-based panels?
Another issue for evaluating performance indicators such as mechanical properties of the specimens under aging treatment is commonly using destructive methods. Researchers have recognized a key shortcoming of the destructive testing: once the specimen is broken, it cannot be used again, different specimens will be evaluated at each measured time and then many specimens are needed to conduct the test.
In this study, we designed and proposed milder conditions for laboratory aging test. We consider a humid condition as a mild treatment, and employed it in repetitive high-low humidity condition and involves heat with elongated cycle times. This condition is reasonably occur in natural environment. This aging treatment is assumed to only small losses in the panel SURGXFW¶VSURSHUWLHV7KXVWKLVPHWKRGPD\HIIHFWLYHO\GHJUDGHDSURGXFW in ways that emulate environmental conditions. Furthermore, using non-destructive test helps to gather information about the degradation behavior of the wood-based panels over the course of the treatment.
1.2 Objectives of the study
This study was to investigate the effects of the designed mild aging treatment for some commercial wood-based panel products. Employing mild aging condition intended to simulate main factors of natural exposure in an accelerated way, altogether with non-destructive methods, is to assess the degradation rate of the wood-based panel products and to distinguish degradation behaviors induced by the mild aging test. Various characteristic of the panel specimens were inspected in order to confirm the effect of test condition on
wood-EDVHG SDQHOV¶ SHUIRUPDQFH and to determine durability of panels by physical and mechanical property changes during and after accelerated aging treatment.
1.3 Structure of the thesis
This thesis consists of five chapters. Chapter 1 presented general introduction, background and objectives of the study.
Chapter 2 introduced the concepts of the durability performance studies. It also includes previous studies conducted which related to this research subject.
Chapter 3 described the effect of cyclic aging treatment on performance of wood based-panels. A mild aging treatment was proposed. Intensity of the wet-dry cycle treatment was examined by changing thickness swelling and weight changes. Strength degradation was investigated and strength loss behavior was predicted by non-linear least-squares regression analysis.
Chapter 4 discussed the mechanical and surface degradation of wood-based panels as the effect of aging treatment. Dynamic modulus of elasticity, loss tangent and surface roughness parameters of mat-formed panels were examined during the aging treatment. The changing behaviors inside and surface part of panels were evaluated and their correlation was characterized.
CHAPTER 2
LITERATURE REVIEW
2.1 Durability performance concept
Evaluating durability of panel product is purposed to describe product performance in which how well the panel will stand up for long-term use, in service-life. Durability study divided into two determination concept. First concept is how the panel degraded by certain exposure condition which corresponds to the production technology. This concept is from view point of producer. Panels are fabricated based on a standard manufacturing and then subjected to aging testing either natural weathering or laboratory exposure. For instances, the standards for producing panel product are ANSI A208.1; CS 236-66 as standards for mat-formed wood particleboard and PS-1-74 for construction and industrial plywood. Physical and mechanical properties reduction after the exposure is commonly reported as test results. That results is used to depict ability of the product in maintaining their required properties, in other words it is used as quality control of the evaluated product.
Second concept is durability based on performance requirement by particular standard. The concept is related to XVHUV¶ point of view. It was developed as approaching the wide-range application of panel product especially for structural uses. The performance-based standard evaluates performance of panel product regardless of their manufacturing variables or techniques. Durability performance in this context is to find out the performance and predict how long does the product last. To evaluate durability performance in this context, various commercial panel types are chosen and exposed in the same exposure method either laboratory or weathering exposure. The discussion would be the properties reductions over exposure period and if they still able to fulfill standard requirement of certain
building code then it could mean product can perform satisfactory in service life. Approaching this durability concept, in 2004, researchers in Japan through the Research Working Group on Wood-based Panels of the Japan Wood Research Society have been conducting the weathering exposure test at eight sites in Japan. The test evaluated several commercial wood-based panels for structural use and it was up to 7 years exposure test.
2.2 Evaluating durability based on production technology
2.2.1 Effects of production variables
As wood-based panel is a mat formed and hot pressed product, in consequence all variables involved in its manufacturing process has noticeable influence to the properties of resultant panel product. Since early period of wood-based panels had been introduced to the market, extensive studies has been conducted in regard to how manufacture variables affect properties of the wood-based panels and looked for optimizing processing parameters to reach acceptable panel properties as well. Pertaining diverse uses of wood-based panels either for interior and exterior application, dimensional stability is a property of panel product which is most questionable. Resistance of swelling, especially thickness swelling (TS) or linear expansion (LE) when panel takes on moisture is considered as the most required property among other properties. It is well known that panel swelling initiates strength reduction even failure for the worst effect. Thus the best performance will come about when thickness change occurs in minimum extent. Hence, many studies investigated moisture resistance of wood-based panels as effect of specific manufacturing variable and those kind studies were considered as durability properties evaluation.
Enormous number of manufacture variables either directly or indirectly affect dimensional stability, however board density, resin type, resin content level, element type,
and pressing condition are considered to have significant influence on dimensional stability properties in which of resultant panels (Kelly, 1972). Board density affects dimensional stability under laboratory dimensional stability and aging test. Board with high density was being less durable board in the term of dimensional change. Gatchell et al. (1966) reported that an increase in density of particleboard resulted in an increase in TS which was subjected to ASTM accelerated aging test and 1 year exposure. Halligan and Schniewind (1972) had revealed that TS of urea-bonded particleboard increased with increasing density which were exposed to various RH levels (ranging from 30 to 97 percent). According to the study by Hse (1975) increasing density from 0.633 g/cm3 to 0.793 g/cm3 when exposed to change
humidity exposure, 5-hour boiling, and vacuum pressure soak test, gave various effect for TS of particleboards from nine southern hardwood. Generally, TS increased as board density increased. Wong et al. (1999) investigated uniform vertical density profile and conventional U-shaped density profile of particleboard on TS under dry-wet cyclic conditioning. It was found also that conventional density profile possessed higher TS compared to the uniform density profile. Suzuki and Miyamoto (1998) found swelling in length known as linear expansion (LE) increased with the increasing board density of Japanese cedar particleboard with range of 0.4-0.8 g/cm3. Ganev et al. (2005) also reported TS and LE of medium density
fiberboard (MDF) increased with increasing board density from 540 kg/m3 to 800 kg/m3
with those effect was significantly stronger to the LE than the TS under 50-80% RH exposure.
Dimensional stability is affected by resin type and resin content. Gatchell et al. (1966) had reported that TS decreased by increasing amount of resin. Hann et al (1962) found that phenolic-resin type was more durable than urea or urea-melamine resin type however urea bonded flakeboard which containing more high amount of resin approached the properties of phenolic-resin type. Jokerst (1968) used various resin type and amount of
resin to produce flakeboard and exposed them to four kinds of humidity exposures, as well 8 years outdoor exposure. Results were in agreement with Hann et al. (1962) for comparing urea bonded flakeboard and phenol-bonded flakeboard. Lehmann (1970) found that increasing resin content from 2 % to 4 % greatly decreased TS particleboard, while from 4 % to 8 % resin content was leveling-RIILQ76DIWHUGD\VRI ၨ) ၨ& and 90 %RH exposure. Hiziroglu and Kamden (1995) found that increased resin content in the range of 0% to 2% reduced TS (small extent) of wet-process hardboard which was evaluated by water soak testing. Suzuki and Miyamoto (1998) showed little increase a LE of particleboard with increasing resin content from 6% to 12% and decreased the TS humidity test and immersion test.
3DUWLFOHVL]HDOVRDIIHFWVGLPHQVLRQDOVWDELOLW\RISDQHOSURGXFWV,Q/HKPDQQ¶VZRUN (1974), thin particle has been reported less effect in thickness swelling than thick particle under humidity and vacuum pressure soaking (VPS) exposure test but there was no significant effect of particle length on dimensional stability properties of the flakeboard. Study by Shuler and Kelly (1976) also showed that no significant difference between 1 and 3 inches of flake length in LE. Miyamoto et al. (2002) reported LE increased with decreasing particle size under combination temperature in 40 τC and 90% relative humidity (RH). Hiziroglu and Suchland (1993) also reported insignificant effect of geometry particles by screen mesh on LE of particleboard.
Other variables which were evaluated its effect on dimensional stability under water exposure and/or fluctuated humidity exposure are wax addition by Heebink and Hann (1959); Gatchell et al. (1966); Albrecht (1968), pressing condition by Lehmann (1974); Steven and Woodson (1977); Suchland et al. (1983); Wong et al (1998) and Geimer and Kwon (1999), Miyamoto et al. (2002), particle alignment variable by Geimer (1980); Avramidis and Smith (1989); McNatt et al. (1992); Wu (1999); Wu and Lee (2002), and
chemical modification by Nicholls et al. (1991). Some studies also examined effect of given treatment on durability properties. Acetylation treatment on flakeboard had improved aging capability based on resistance of TS and retained strength compared to unacetylated board (Vick et al., 1991). PF resin-impregnated particleboard was produced by Yung et al. (2007) and it was found had excellent performance after accelerated aging in hot water. Heat post-treatment (240 τC) demonstrated by Hsu et al. (1989) and durability of waferboard was improved and had a better inherent ability to withstand severe exposure conditions than the regular boards.
2.2.2 Evaluating potential material for new product
In response of growing demand of wood based panel, researchers considered to utilize other materials than wood resources either for finding new product or as alternative raw materials. Some of those studies including evaluation durability of resultant panels under certain aging exposures on experiment study. Lehmann and Geimer (1974)utilizing various proportions of sound wood, bark, dead wood, decayed wood, and branches; McNatt (1978); McNatt and Link (1978) utilized Douglas-fir forest residues from logging operation. Pugel et al. (1989) used fast grown tree for producing flakeboard, particleboard, and fiberboard as well Massijaya et al. 2005 which evaluate particleboard from small diameter fast growing species. Later on researchers began to explore potential use of agricultural resources as basically wood-based panels can also be made of any lignocellulosic materials. Singh et al. (2000) utilize jute fiber as reinforcement of composite while Kim and Seo (2006) used sisal fiber for composite and evaluate their durability. Okuda and Sato (2007, 2008) used kenaf core for binderless boards. Fiorelli et al. (2012) explore coconut fiber particleboard; Sassoni et al. (2015) promoted hemp-based composite. Sugarcane bagasse
composite was studied by Fiorelli et al. (2016) and reported that the waterproofed particleboards can be used in moist environments based on standardized aging test.
2.3 Accelerated aging tests
2.3.1 Outdoor exposure test
Application of wood-based panel as component of housing construction requires to perform satisfactory for long time periods after being put in intended place. In conjunction with this, testing procedure concerning long-term weathering exposure, condition which product would facing in service life, is essential to be evaluated. Outdoor or weathering exposure considers as a test which can provide basic information because it combines various influencing deterioration factors. Outdoor exposure tests have been carried out by many researchers and employed with various period of time in many sites all over the world. The investigation of outdoor exposure was embarked in United States. Hann and Blomquist (1962) exposed Douglas-fir flakeboard for 3 years exposure in Madison, Wisconsin. Carrol
et al. (1969) exposed poplar plywood for 5 years in Canada, North America. Ten years
exposure of softwood plywood was carried out by Black et al. (1976) in Madison and Gulfport. Biblis (2000) exposed untreated six species commercial plywood for 6 years in Alabama and it was found that Douglas-fir plywood was the best regarding the change of surface properties and the retained mechanical properties. Dinwoodie (1981) had exposed plywood to natural exposure in Europe. Deppe and Schmidst (1989) also did the same in Madison, North America.
2.3.2 Laboratory exposure test
Laboratory exposure test is simulated condition test which employed over a particular time frame and commonly used for short period tests. The test is intended to rapidly approximate
the effects of many years of outdoor exposure and avoid the time consuming of long-term test. Durability of particleboard as effect of temperature and humidity have been conducted Hann et al. (1963); Kajita et al. (1991) evaluated particleboard by compared four approval methods of aging test. Kojima and Suzuki (2011a) also have compared five accelerated aging conditions and found some correspondence between some methods. Suchland (1973), performed cyclic exposure of ten commercial particleboards. Deppe (1981) produced reliable prediction method of accelerated aging and found 24 weeks of XENOTEST (weathering apparatus) represents about 5 years outdoor exposure. Kojima and Suzuki (2011b) found 5 years outdoor exposure quite close severity to 5 cycle VPSD.
CHAPTER 3
EFFECTS OF CYCLIC AGING TEST ON PERFORMANCE OF WOOD-BASED PANELS
3.1 Introduction
Aging test is a term for a test where the panel products are exposed to condition in which it is expected to cause degradation. Accelerated aging test is a test where the degradation is deliberately increased, while the condition involved will never occur in actual service life. In determination of durability performance of panel products, accelerated aging tests are being used more for quality control purposes. As for it, the accelerated test methods have been designed to evaluate rapidly the effect of particular climate agents, since outdoor exposure is kind of time-consuming test. Intensity of the accelerated aging treatment would be a factor that approximate the effect of years of outdoor exposure test.
The simulated condition test is employed over a particular time frame and is commonly used for short period tests. The test is intended to rapidly approximate the effects of duration of outdoor exposure and avoid the time consuming of long-term test. A number of studies have used this kind of aging test to determine the durability performance of panel products by using either standardized or laboratory-developed methods. Kajita et al.(1991), Alexopoulos (1992), Kojima et al. (2010), and Kojima and Suzuki (2011a, 2011b) have worked with standardized methods such as the ASTM 6-cycle, V313, APA D-1, and JIS. Others such as McNatt and Link (1989) and McNatt and McDonald (1993) have used laboratory-scale methods, with some modified procedures. Some studies have also discussed WKLV PHWKRG¶V UHODWLRQVKLS ZLWK RXWGRRU H[SRVXUH 0RVW RI WKH UHVXOWV IURP WKHVH standardized methods tests have found much more severe damage than those from outdoor
exposure tests. As a result, simulated tests cannot be used to reliably estimate outdoor exposure, as reported by Hann et al.(1962) , Vick (1987), River (1984) , and Korai et al. (2014).
The severity of simulated aging tests is caused by the extreme conditions of the treatment, particularly those involving liquid water, such as boiling, soaking, spraying, and vacuum pressure soaking. Therefore, several studies have attempted to use vapor water conditions in their aging tests. Chiu and Biblis (1973) and Pu et al. (1993) have reported on humid conditions in aging tests over short periods, which involved several hours of cycles that were evaluated using destructive bending tests. However, researchers have recognized a key shortcoming of the destructive testing: once the specimen is broken, it cannot be used again and many specimens are needed to conduct the test. Non-destructive evaluation has been well investigated for wood products, and some (e.g. Zhenbo et al (2006), and Bos and Casagrande (2003)) have focused on wood-based panels. Nevertheless, there are few studies that focus on non-destructive evaluations under accelerated aging tests. However, Sun and Arima (1998) have performed these tests using particleboard and oriented strandboard as specimens. The specimens were treated with boiling water for 60 minutes and then evaluated for strength properties and dimensional stability over several time periods. We observed that elongated cycle timeVZLWKPLOGHUDJLQJWUHDWPHQWVOHGWRRQO\VPDOOORVVHVLQWKHSURGXFW¶V properties. Thus, this method may effectively degrade a product in ways that emulate environmental conditions. Furthermore, using non-destructive test helps to gather information about the degradation behavior of the wood panels over the course of the treatment.
Our research proposes milder conditions for the simulated aging test over longer time exposure. We consider a humid condition as a mild treatment, and this involves using vapor-based water to degrade panel products. Our objective in this chapter is to investigate the
effects of the proposed aging condition, to assess the degradation rate of the panel products, and to distinguish degradation behaviors induced by our mild aging test. Degradation behaviors can provide practical information on the durability of a panel product, which is especially relevant for commercial panel products in Japan.
3.2 Materials and methods
3.2.1 Specimen preparation
We used four groups of commercial panel products as specimens: plywood (PW), oriented strand board (OSB), particleboard (PB), and medium density fiberboard (MDF). Each group of panels consisted of two panel types that differed in adhesive type, panel thickness, or wood species. The characteristic for each panel type are shown in Table 3.1. All specimens were prepared to be 50 mm wide and 250 mm long with four replications for each measurement (Fig. 3.1)7KHVSHFLPHQVZHUHLQLWLDOO\FRQGLWLRQHGDWÛ&IRU48 hours, after which we measured the dimensions and weight of each specimen and specified as the initial dimensions and weight.
3.2.2 Mechanical properties evaluation
We conducted a dynamic bending test using transverse vibration. A microphone was placed at one end of the specimen and the specimen was hit using a small hammer at the opposite end. The vibration signal, which was obtained by a microphone, was converted into a power spectrum by a fast Fourier transform (FFT) analyzer. We calculated the elastic constant (Ed)
and the loss tangent (tan į) from the peak resonance frequency, which corresponded to the first vibration mode and the amplitude of the resonance curve, respectively, as shown in the following equations:
2 2 4 2 6 . 500 48 t f L EG S U GPa (Eq. 3-1) S O G tan (Eq. 3-2)
where ȡ is density (gcm-3), L is the length (mm), f is the frequency of the resonance peak
(Hz), t is the thickness (mm), and Ȝ is the logarithmic decrement of the resonance amplitude which calculated by Hilbert transformation.
3.2.3 An accelerated aging treatment
We used an accelerated aging treatment designed as a mild test. Figure 3.2 shows the schematic diagram of the treatment. The treatment included a wet-dry cycle where each cycle had wet and dry states; 40°C and 90% relative humidity (RH) for 120 hours was characterized as the wet state, while 40°C for 48 hours without humidity control was characterized as the dry state. We repeated this treatment for 80 complete cycles in a climate-controlled chamber.
At the end of every state, we measured the dimensions and weight of each specimen and determined Ed and tan į via the dynamic bending test. We calculated weight change (WC) and thickness swelling (TS) in every cycle on the basis of the initial weight and dimensions of the specimen, respectively. For conformity intensity of the condition test, we GHWHUPLQHGǻ:&DQGǻ76ZLWKWKHIROORZing formulas:
ǻWCi = WCwi - WCdi (Eq.3-3)
where, WCwi is WC of wet condition at i cycle and WCdi is WC of dry condition at i cycle.
ǻTSi = TSwi - TSdi (Eq.3-4)
3.3 Results and discussion
3.3.1 Intensity of the wet-dry cycle treatment condition
The effects of the aging treatment are shown in Figs. 3.3 and 3.4. Figure 3.3 shows the mean YDOXH DQG VWDQGDUG GHYLDWLRQ RI ǻ:& IRU F\FOHV :H IRXQG WKDW WKH ZHW-dry cycle treatment caused 8±ǻ:&LQWKHSDQHOV7KHUHZHUHDIHZGLIIHUHQFHVLQǻ:&IRUWKH mat-formed panel types (PB, MDF, and OSB) and in the veneer-EDVHGW\SH7KHǻ:&RI plywood was approximately 11.5%, whereas the mat-formed panels ranged from 8±10%. This suggests that plywood absorbs much more moisture than do mat-formed panels. The PHDQYDOXHDQGVWDQGDUGGHYLDWLRQRIǻ76IRUDOOSDQel types are given in Fig. 3.3. We found that the wet-dry cycle treatment caused a 2.5±ǻ767KHǻ76IRUWKHSO\ZRRGZDV and the mat-formed panels ranged from 5±6%.
Figures 3.3 and 3.4 show that the intensity of the wet-dry cycle treatment accelerated the mat-IRUPHGSDQHOV7KHFRQGLWLRQVRIWKLVWUHDWPHQW\LHOGHGDǻ:&RI approximately 10%. However, by the private note (Kojima et al., 2009), ǻ:&RI$670-cycle test has EHHQIRXQGWKDWFRXOGUHDFKDOPRVW)RUǻ76WKHYDOXHVUHDFKHGDSSUR[LPDWHO\IRU the wet-dry cycle treatment, while a study by Kojima et al. (2009) found a value of approximately 26%. Our wet-dry cycle treatment resulteGLQVPDOOHUǻ:&YDOXHVLQGLFDWLQJ that this treatment is milder than the standardized method. However, we consider that 10% of the moisture change in the wet-dry cycle treatments could be quite high.
Interestingly, the particleboards TS for each 1% change in moisture content was estimated using approach of weight change value and found to be about 0.4±0.6%. This was quite close to the thickness change in standard particleboard, which is approximately 0.3± 0.5% for each 1% change in moisture content (EWPAA, 2008). This confirmed that the conditions designed for this wet-dry cycle treatment yielded fairly small thickness changes
for the panels and was nearly ideal for evaluating moisture changes that would occur during alternating wet and dry conditions in the environment.
TS of all panels increased with increasing the number of cycle with the exception of veneer base panels (Fig. 3.5). Although the trend was similar but the increasing TS among the panels has differently proceeded during the wet-dry cycle treatment. To clarify the effects of the treatment on TS behavior, we observed OSB and PW for comparison as shown in Fig. 3.6. TS increased during the wet state because panels swell from moisture absorption and decreased during the dry state because panels shrink from moisture desorption. We dropped points and drew lines for the TS values in each state during the cycles. There was an obvious and repeated rise and fall over the 80 cycles. The thickness swelling of the OSB rapidly increased in the early cycles of the treatment, and then tended to level off after the first 20 cycles.
When panel products absorb moisture, either in vapor or liquid form, overall TS occurs both from the swelling of the particles themselves and the springback (Halligan, 1970; Moslemi, 1974). Particle swelling occurs because wood, from which the particles are made, is naturally hygroscopic. Furthermore, absorbed moisture could cause bond breakage and bring about particle separation. Springback is a result of releasing the stress of the board, which is caused by the compression of pressing. Nevertheless, when a board desorbs moisture and returns to a dry stage, there is part ofthe overall swelling never recovers to its original form. That is springback which generates permanent change of thickness and henceforth called irrecoverable TS.
The overall TS of the OSB after completed cycle was approximately 15% and about 9% of that did not recover which we defined as irrecoverable TS. When OSB was exposed to a wet state in the first cycle, TS was about 9% and when OSB continued to a dry state,
only 6% of its swelling recovered, which we defined as recoverable TS. Notably, 3% was irrecoverable TS. The TS from the 20th cycle to the end of the cycles was quite similar. This
indicates that the amount of irrecoverable TS increased from the first cycle up to 20 cycles but that the amount of recoverable TS in each subsequent cycle did not increase. Unlike OSB, the overall TS for the PW panel was about 3% and the PW panel almost retained its initial thickness when it was exposed to a dry state in subsequent cycle. Even at the end of cycle, the treatment resulted in almost 3% TS value for PW. The amplitude of recoverable TS is almost same with advancing cycles in PW. As a result, we found that a wet-dry cycle treatment is not adequate for assessing the degradation of PW.
Overall, our wet-dry cycle treatment can cause TS to reach 9% after completing cycle. Conversely, ASTM 6-cycles can cause TS to reach about 24% (Kojima, 2009). Thus, the wet-dry cycle treatment resulted in smaller thickness changes than did the ASTM 6-cycles. Therefore this condition was thought to be well designed to act as a mild accelerated aging test.
3.3.2 Strength retention under the wet-dry cycle treatment
We examined strength property changes during the treatment using Ed. Ed values during 80
cycles for all panel types are shown in Fig. 3.7. Ed values also show an alternating rise and
fall during the treatment, which corresponded to changes in TS and WC. These rise-fall curves illustrate Ed behaviors during the wet-dry cycle treatment. Generally, they decreased
gradually during the early cycles and then levelled off with later cycles.
There were some differences in Ed amplitude, which we calculated from dry to wet
state between the panel types. In other words, the effect of this treatment was different among panels. To explore this, we selected two of the eight panel types for comparison, i.e.
PBa and MDFa which had same 1.1 GPa initial amplitude. At the end of total cycle, it became 0.4 GPa for PBa and 0.8 GPa for MDFa. Thus, PBa experienced a larger loss of strength than did MDFa.
We tested the extent of Ed retention in the wet state of cycle treatment and defined
as follows:
Ed retention (%) = (Ed wet state/ Ed initial)100 (Eq. 3-5)
Ed retention of panel products is shown in Fig. 3.8. The wet-dry cycle treatment
resulted in decreasing Ed values over 80 cycles, with different changes for each panel.
Retention values were largest in PB, then followed by MDF, OSB, and PW. Ed retention in
the wet state were 70±90% in the first cycle, then declined to 50±80% at the end of the cycle. Furthermore, the Ed retention value at the end of the complete cycle for the PW panels was
higher than that for the mat-formed panels. That indicates that PW panels retained greater strength than did mat-formed panels. The rate of retention decreased fairly rapidly during the early cycles and then continued to decrease gradually until the final cycle. These retention values represent a loss of strength due to moisture deteriorating the structure of the panels during the adsorption process.
The Ed retention for all panels in Fig. 3.8 shows a clear decreasing trend, but in fact
retention behavior during the wet-dry cycle treatment was very different for the different panels. To further explore this phenomenon, we used a degradation model developed by Suzuki and Saito (1988) to elucidate the strength loss behavior for each of the panels over a prolonged treatment. Calculating Ed retention behavior for all the panels is based on the data
F(t) = A + (100 ± A) exp(t/B) (Eq.3-6)
Coefficient A is the saturation value of the Ed retention and refers to the durability of
the panel. Coefficient B is the decreasing rate and refers to how fast retention decreases. When panels are continuously subjected to the same treatment condition, we assumed that their strength would never reach zero. This means the panels would retain their strength after an infinite number of cycles. Using this model, we can predict the specific saturation value for strength retention after successive treatment cycles. Table 3.2 shows the coefficient values for all the panels, which are determined by Equation (3-6).
The coefficients in Table 3.2 were determined by non-linear least-squares regression analysis and show great variation. These coefficients also provided a useful overview of the different Ed retention behaviors among the panels over certain cycles. Except PWb, all
panels show that the values of A are higher in the dry state than in the wet state. Another finding, except MDFb, the PW types had higher values of A than did the mat-formed panel types. Comparing between mat-formed panels during the wet state, the values of A for PBa and OSBa were 52% and 63%, respectively. These results indicate that, by this calculation, PBa was less durable than OSBa during the wet-dry cycle treatment, even though thickness swelling of PBa was slightly less than that of OSBa.
Based on the values of B, some panels experienced a decrease in strength at the very beginning of the cycles, and others toward the end of the cycles. However, to better explain this difference, we compared PBa and OSBa during their dry states. PBa, where B is 13, has a higher coefficient than OSBa, where B is 8. Since the rate of decrease corresponds to the number of cycles, the strength retention of PBa needed more time to reach its saturation value than did that of OSBa.
To clearly identify the different behaviors of strength loss for both PBa and OSBa panels, we plotted all Ed retention data during the wet-dry cycle treatments and drew
regression lines calculated by non-linear least-squares regression. Plots show the values of
Ed retention rates, while the solid and dotted lines represent the predicted value of the
strength loss. Figure 3.8 shows the relationship between Ed retention, when panels were
tested at a dry state, and the number of cycles. As shown in those curves, the wet-dry cycle treatment degradation rates of the two panels differed significantly. The strength loss of PBa decreased continuously and moved more slowly toward its saturation value, whereas OSBa rapidly decreased within the first few cycles and then became relatively stable, as it had almost reached its saturation value.
The regression lines agreed well with the Ed retention values from our experimental
data as shown in Fig. 3.9. This means that Equation 3-3 can be used effectively to evaluate the strength loss of panels under wet-dry cycle treatments for an infinite number of cycles.
3.3.3 Estimating internal bond strength
The loss tangent (tan į) values of all the panels, which were computed by Equation 3-2, are plotted in Fig. 3.10 for each number of cycles. The tan į behavior of the panels during the wet-dry cycle treatment is shown in Fig. 3.10. Loss tangent is a rheological term that corresponds to internal loss when a material vibrates. This is likely related to element attribution inside a board. Logically, this could be used as an indicator of internal bond strength.
The values of tan į increased with each increase in the number of cycles. As shown by Obataya et al. (2003), a TS of wood that results in Ed reduction also results in increasing
for all panels during the wet-dry cycle treatment. The tan į values showed an overall increasing trend, growing for the first 20 cycles and subsequently levelling off. This trend resembled the TS trend. This is understandable because moisture changes occurred during the treatment by adsorption and desorption, which caused TS in the panels. TS reduces the number of bonding points inside the board and leads to the degradation of internal bond strength.
Figure 3.10 shows the changes of tan į during the wet-dry cycle treatment for 80 cycles. The tan į values differed among the panels. The largest tan į occurred in MDFa, which ranged from 0.03 at the first cycle to 0.045 at the end of cycles. The smallest change in tan į from the first cycle to the last occurred in PBa, which ranged from 0.025 to 0.03. There was no big difference for both PW types while OSB types and PBb were found to be quite same trend. However, under this aging condition, the changes of tan į were not consistent when associated with sequences of the changing TS values. We assumed that this was due to differences in wave propagation through the board, as element size, element direction, and adhesive type varied among panels. We thought that tan įcould be used as an index of mechanism degradation inside the panel. Therefore, relation between tan į and durability is important issue to be discussed furthermore.
3.4 Conclusion
We subjected four groups of commercial panel products to a wet-dry cycle treatment. The proposed aging conditions yielded an 8± ǻ:& DQG D ± ǻ76 2XU WHVW XVHG considerably milder conditions than those used in the ASTM 6-cycle standardized method. Our wet-dry cycle treatment caused a rise and fall of thickness swelling during the treatment and resulted in a significant irrecoverable thickness swelling for mat-formed panel types, though not for plywood types.
The wet-dry cycle treatment considerably reduced the Ed value over increasing
numbers of cycles. The rate of Ed retention decreased fairly rapid during early cycles in the
treatment and then tended to level off. The residual Ed retention rate of the panel products
in the wet state was between 50±80% after completing 80 cycles.
The degradation behavior of each panel product was caused by extended wet-dry cycle treatments. This was well traced using vibrational non-destructive test and was also obvious from the predictive model. The value of tan į increased with increasing numbers of cycles. However, these results suggest that this proposed aging condition was fairly well designed and may be effective for providing degradation information on wood-based panels. This could help predict the durability performance of mat-formed panel products.
24 T able 3.1 Ch ar ac te ri stic s of t he s pe ci me ns Pa ne l ty pe A bbre vi ation a A dhe si ve T hic kne ss (m m) D ens it y b (g cm -3 ) MOR c (M Pa ) MOE d (M Pa ) No te s Pl yw ood PWa PF e 12.2 0.60 49.3 ± 13.4 6.55 ± 0.84 5 plie s PWb PF 8.8 0.58 71.8 ± 13.1 8.78 ± 1.16 3 plie s O ri ente d s tra nd bo ar d OS B a PF 12.4 0.63 37.7 ± 8.9 4.90 ± 0.69 Asp en OS B b PF 11.8 0.65 36.0 ± 6.9 4.68 ± 0.62 Pi ne Pa rt ic le boa rd PB a PF 12.2 0.74 21.6 ± 3.5 3.44 ± 0.46 - PB b MD I f 12.1 0.77 29.7 ± 2.4 3.97 ± 0.19 - M edium de ns ity fi be rboa rd MD Fa MU F g 12.2 0.75 44.9 ± 3.0 4.07 ± 0.22 - MD Fb MD I 9.1 0.71 33.8 ± 1.4 3.10 ± 0.15 - a L ow er-ca se le tte rs a re s ta nding for di stin gu is hin g a m ong tw o pa ne ls in the s ame p ane l ty pe b D en sit y ar e g iv en as av er ag e va lu es c M odulus of rupture , g iv en a s a ve ra ge ± s ta nda rd de vi at ion d M odulus of e la stic ity , g iv en a s a ve ra ge ± s ta nda rd de vi ation e P he nol form al de hy de f M eth yl en e diphe ny l diis oc ya na te g M ela min e ure a form al de hy de
Table 3.2 Coefficient values of the degradation model of panels under the wet-dry cycle treatment
Panel
typea AbDry state Bc A Wet state B
PWa 87 15 79 13 PWb 82 9 85 3 OSBa 73 8 63 9 OSBb 78 12 67 9 PBa 60 13 52 23 PBb 74 15 61 17 MDFa 73 13 59 11 MDFb 89 19 69 35
a The abbreviations refer to Table 3.1 b The saturation value of the Ed retention. c The decreasing rate
Fig. 3.2 Schematic diagram of a wet-dry cycle treatment. Weight change (WC) occurs corresponding to the humidity changes under wet- and dry-conditions.
Fig. 3.4 ǻThickness swelling (%) of panel products after the complete 80 cycles
Fig. 3.6 The fluctuation of thickness swelling (%) for PWa and OSBa over 80 cycles
Fig. 3.8 Ed retention (%) of the panel products in the wet state
Fig. 3.9. Ed retention and retention model (F(t)) of PBa and OSBa during the wet-dry cycle
treatment in the dry state. F(t) = A + (100 ± A)exp(B/t), where t is number of treatment cycles.
Fig. 3.10 Tan į of the panel products in relation to the wet state during the wet-dry cycle treatment
CHAPTER 4
MECHANICAL AND SURFACE DEGRADATION OF WOOD-BASED PANEL AS THE EFFECT OF AGING TREATMENT
4.1 Introduction
Properties of wood-based panels change during their service life. Therefore, panel durability over application time is an important property that needs to be evaluated. Accelerated aging tests such as ASTM 6-cycle, EN321, APA D-1, and JIS-Wet bending have been used to determine the durability of wood-based panels. Dimensional stability (Pugel et al. 1990; Moya et al. 2009; Kojima et al. 2009), bending strength (Hann et al. 1963; Kojima and Suzuki, 2011; Korai et al. 2014), and internal bond (Kajita et al. 1991; Kojima and Suzuki, 2011; Korai et al. 2015) are parameters usually evaluated for determination of panel durability. In contrast to this, surface properties are less explored, and usually evaluated separately with those aforementioned parameters. The strength properties are important for meeting the requirements during use, whereas changes in surface properties should be evaluated in order to predict effective performance of panels during their service life.
Surface roughness, a surface property that is considered as latent, becomes notable when a panel is subjected to conditions which change its properties, such as humidity. Numerous studies have reported surface roughness of wood-based panel products (Hiziroglu and Suchland 1993; Hiziroglu et al. 2004; Hiziroglu and Kosonkom, 2006; Hiziroglu and Suzuki, 2007; Kilic et al. 2009; Tabarsa et al. 2011; Baharoglu et al. 2012). However, less information is available on surface roughness of panel as function of accelerated aging. A study exposing medium density fiberboard (MDF) to some level of relative humidity (RH)
found that roughness values increased as panels were exposed to higher humidity levels ranging from 65 to 85 % (Ozdemir et al. 2009). However, the RH exposure was used for once until the intended equilibrium moisture content of MDF was reached. A previous study on hardboard and MDF subjected to one cycle of 50-86-50 % RH exposure (Hiziroglu, 1996) and found roughness instability after re-exposure to lower RH. Those RH exposures mentioned above provide insufficient information on panel surface instability that might occur owing to natural swelling and shrinkage during service life.
Ostman (1983) studied surface roughness of painted and unpainted commercial particleboard (PB) and fiberboard that were subjected to different methods of accelerated aging. The methods used were not suitable for evaluating surface roughness, because those methods employed extreme conditions that resulted in drastic degradation. The relation between surface degradation and strength degradation has been discussed; however, strength degradation and surface degradation have been studied separately. More information that is comprehensive and based on simultaneous assessment of surface and strength property degradations owing to accelerated aging tests is required.
In addition, evaluation of strength degradation is usually carried out using destructive tests. This requires plenty of specimens, and different specimens are measured on each occasion. Therefore, a mild accelerated aging method using the non-destructive bending test and the stylus technique for evaluating mechanical properties and surface roughness evaluation, respectively, is required for obtaining precise information on degradation properties.
The objective of this chapter was to evaluate the changes in surface roughness and dynamic mechanical properties that occurred in MDFs and PBs, when subjected to
accelerated aging treatment. In addition, we investigated the possible correlation of surface degradation with mechanical degradation.
4.2 Materials and methods
4.2.1 Specimen preparation and accelerated aging
Panels used in this chapter were commercially manufactured PB and MDF. Two types of PB included PB bonded with phenol formaldehyde (PB(PF)) and methylene diphenyl diisocyanate (PB(MDI)). Two types of MDF included MDF bonded with melamine urea formaldehyde (MDF(MUF)) and methylene diphenyl diisocyanate (MDF(MDI)). Two panel boards with dimension of 300 mm × 300 mm were selected for each panel type. The two boards were cut into ten specimens of 50 mm × 300 mm and then randomly selected five specimens for each panel type. A total of twenty specimens were used for all panel types, and those were conditioned to 20 °C and 60 % RH for two weeks prior to the accelerated aging treatment. The initial properties of the specimens are shown in Table 4.1.
The accelerated aging condition (hereafter referred to as the aging treatment) consisted of fifteen cycles of exposure to high and low RH at a constant temperature of 60 °C. The cycle started with the wet state (90 % RH for 120 h), followed by the dry state (no humidity control for 48 h). Thus, each cycle lasted one week. This present study used water vapor instead of liquid water, as required by several standardized accelerated aging test.
4.2.2 Measurement and testing
Dimensions and weight of specimens, measured after conditioning at 20 °C and 60 % RH, was specified as initial state of the specimens. The same specimens were evaluated for thickness swelling (TS), surface quality, and a dynamic bending test was carried out at the
end of every state during each cycle of the aging treatment. For surface quality evaluation, surface profiles were measured with a stylus-type profilometer (SJ-301, Mitutoyo Surftest). The tracing length was 15 mm with a constant speed of 0.5 mm/s. Measurements were made perpendicular to direction of the board production (Fig. 4.1). Three points on the surface of the each specimen, one point was in the center surface and the other points were 20 mm distance from the both panel edges, were marked. Those are to ensure that the same point was measured on each occasion. Calibration of the device was performed before the measurement.
Three roughness parameters, average roughness (Ra), maximum height roughness
(Rz), and ten points mean roughness (RzJIS), which are commonly used to evaluate surface
characteristic of wood and wood-based panels, were calculated to determine the degradation indicated by surface roughness over the humidity change exposure. The definitions of these three parameters are available in JIS B 0601-2001. Increasing rate of those surface roughness parameters were determined by Eq. 4-1.
100 aging before value aging before value aging after value (%) rate increasing Roughness 㸫 u (Eq. 4-1)
In addition to surface roughness measurements, microstructure of the panel surfaces was examined using scanning electron microscope (JSM-6510LV, Joel). Unaged specimens and specimen after completion of aging treatment were inspected. Moreover, the term surface degradation in this study is limited in the increment of surface structure change compared to its initial surface condition.
Evaluation of mechanical properties was performed simultaneously with surface roughness measurements. Dynamic modulus of elasticity (Ed) values of (2,0) vibration mode
sound level meter (LA-1410, Ono Sokki) with fast Fourier transform (FFT) analyzer (CF-7200, Ono Sokki). A microphone was placed above the specimen end and a small hammer stroked on the opposite end. The vibration signal, which was obtained by a microphone, was converted into a power spectrum by FFT analyser (Fig.4.2). The Ed was calculated from the
peak resonance frequency by Eq. 4-2 and tan į was calculated from the amplitude of the resonance curve by Eq. 4-3.
2 2 4 2 6 . 500 48 t f L EG
S
U
(GPa) (Eq. 4-2) S O G tan (Eq. 4-3)where ȡ is density (g cm3), L is the length (mm), f is the frequency of the resonance peak
(Hz), t is the thickness (mm), and Ȝ is the logarithmic decrement of the resonance amplitude which calculated by Hilbert transformation. The previous study (Saad et al. 2016) determined the Ed based on the initial thickness and density values, whereas in this present
study, based on the thickness and density values at the test. Panel Ed degradation upon the
aging treatment was expressed by Ed residual value and calculated using Eq. 4-4.
100 (%) retention u LQLWLDO DW YDOXH DWWHVW YDOXH G G G E E E (Eq. 4-4)
4.3 Results and discussion
4.3.1 Effect of the aging treatment on surface quality
Surface profile was recorded before, during and after the aging treatment. Surface profile provides a prompt visual view of panel surface changes. Surface structure changed after the
aging treatment, which was evident from the longer raised and lowered strips than those visible before the aging treatment (Fig. 4.3). However, the figure seems quite difficult to be used for comparing the changes between different types of panels. Therefore, quantitatively surface roughness before and after the aging treatment was identified. Three roughness parameters, Ra, Rz, and RzJIS, reported as the mean of fifteen different profiles for each panel
type, increased with the aging treatment (Table 4.4). The increasing rate in Ra appeared to
be larger than that of Rz and RzJIS. MDF(MUF) showed the largest increasing rate with the
aging treatment (76 %), followed by PB(PF) (74 %), PB(MDI) (62 %), and MDF(MDI) (22 %). The order of increasing rate in Rz was the same as Ra, whereas the largest increasing rate
in RzJIS was found in PB(PF) (44 %), followed by MDF(MUF) (40 %), PB(MDI) (27 %),
and MDF(MDI) (9 %). Surface changes obtained were considered to be very small compared to surface changes under vacuum pressure-soak-dry (VPSD) aging reached 325 % for PB phenolic resin and 222 % for fiberboard phenolic resin (Ostman, 1983).
Even though PB(MDI) had the coarsest surface initially, it showed lower increasing rate in surface roughness parameters than the two types of amino-based panels, (i.e., MDF(MUF) and PB(PF)). In contrast, MDF(MUF) with the smoothest surface initially, became rough after the aging treatment, indicating heavy degradation. However, MDF(MDI) exhibited the least increasing rate in surface roughness parameters. In addition, the change in surface roughness of the panels depends on the adhesive type.
The surface changes occurred in panel specimens owing to moisture content variation by the aging treatment. A change during the aging treatment is valuable information for understanding the progress of degradation over time and predicting aging resistance. Since the panels did not degrade drastically by the aging treatment, the stylus technique could trace the surface changes during the treatment. Ra increased with the number
evident in the first cycle, followed by relatively gradual degradation. Degradation appeared to progress in PB(PF), PB(MDI), and MDF(MUF) panels, whereas Ra remained constant in
the MDF(MDI) panel.
Surface roughness degradation is associated with dimensional changes owing to swelling-shrinkage phase during aging treatments (Suchland, 1973). Each panel had different extent and course of change in TS following the aging treatment (Fig. 4.5). TS of PBs increased gradually at the dry state of each cycle, whereas MDFs showed little change in the dry states. Unlike TS change, which increased and decreased intermittently at wet and dry states, respectively, Ra did not show a similar trend in each cycle. Changes in surface
roughness owing to aging treatment were irreversible. Increasing surface roughness appeared consistent with increasing TS in PB panels, whereas it did not in MDF panels. Even though MDF(MUF) experienced little change TS, swelling-shrinkage could loosen and raised some individual fibers on the surface. It did not affect to overall TS but fiber-pop affected the scanning track of the stylus. Comparing the courses of surface degradation owing to the aging treatment revealed that MDF(MDI) had the greatest surface stability among the panel types compared. This might be attributed to the combination of MDI resin used and low density of the panel (lower compaction ratio) could produce high surface stability. High density particleboard surface are not as stable as low density when subjected to RH changes (Hiziroglu and Suchland, 1993). .
Micrographs of the unaged two type of MDF panel surfaces appeared almost similar (Fig. 4.6), consistent with the quantitative measures of surface roughness values. However, the surface structure became uneven after the aging treatment, with more cracks and raised fibers in MDF(MUF) than in MDF(MDI) panel surface.
4.3.2 Effect of aging treatment on mechanical properties
4.3.2.1 Dynamic modulus of elasticity (Ed)
The Ed value before the aging treatment ranged from approximately 4.0±5.3 GPa (Fig. 4.7).
In contrast, after the aging treatment, the Ed value ranged from 2.1±3.3 GPa and 2.7±4.0 GPa
in the wet and dry states, respectively. The results revealed that the aging treatment reduced
Ed values of all the panels. Among the panel types compared, PB(PF) had the lowest Ed
value after the aging treatment. Observing Ed changes using non-destructive test could
follow the progress degradation owing to the aging treatment. Ed retention of the panels,
which showed the progress during the treatment revealed that unlike surface degradation, the aging treatment had similar effects on Ed at each wet and dry state in every cycle (Fig.
4.8).
Degradation increased with increasing number of cycles. We observed that rapid degradation occurred at the first cycle, followed by slow decrease in rate of degradation at successive cycles. Ed retention of PB(PF), PB(MDI) and MDF(MUF) decreased
exponentially with increasing number of cycles, whereas that in MDF(MDI) hardly decreased. PB(PF) showed the least Ed retention (59 %) after the completion of the aging
treatment, followed by MDF(MUF) (75 %), PB(MDI) (75 %), and MDF(MDI) (93 %). Generally, amino-based panels, in which PF is known as a less durable resin than MDI, showed larger Ed loss than that in MDI-bonded panels. However, it was revealed that
PB(MDI) and MDF(MUF) had similar decreasing trend and equal values in Ed retention
(Fig. 4.8). It could be interpreted that both panels have similar resistance or performance under this aging treatment. Therefore, resin type was not major factor forEd degradation as
it is found in surface degradation. Interaction factor between resin type and panel constituent might contribute to the Ed degradation as well.
4.3.2.2 Loss tangent (tan į)
The tan į values after the aging treatment did not show marked changes compared to the initial state, except in MDF(MUF). Slight increase in tan į with increasing number of cycles was observed during the aging treatment in the wet state of PB(PF) and MDF(MUF), whereas PB(MDI) and MDF(MDI) showed hardly changed (Fig. 4.9). Moreover, the change in tan į between the dry and wet states of the two amino-based panels was larger than that in the two MDI-bound panels. This indicates that the change in tan į is relatively dependent of resin type and MDI-bonded panels were able to maintain their formation against humidity exposure. Furthermore, the change in tan įwas much less than that in surface degradation and Ed degradation.
4.3.3 Correlation of surface and mechanical degradation
In order to compare degradations more easily, a ranking was made by using value of the changes after treatment cycles, based on their respective initial values. The panel types studied could be arranged in decreasing order of surface degradation as follows: MDF(MUF), PB(PF), PB(MDI), and MDF(MDI). In contrast, Ed and tan į degradation were
similar in PB(PF), MDF(MUF), PB(MDI), and MDF(MDI). However, Ed loss of PB(PF)
was much more marked than surface degradation, in contrast to the observed trends in MDF(MUF). Nevertheless, the two panels bonded with MDI resin exhibited the same rank for those surface and mechanical degradations.
To evaluate the feasibility of predicting total degradation, correlation between surface degradation and mechanical degradations was determined. Changes of Ra increment,
Ed retention and tan į increment values at each dry state during the aging treatment were
their correlation coefficient values were higher than that between Ra increment and tan į
increment (Fig. 4.10). This relationship depicted that surface roughness increased and strongly caused decreasing of Ed but increasing surface roughness was less influence the tan
įThis is understandable because Ed is calculated by taking into account the dimensions of
the board. Swelling in the thickness apparently occurred by the aging treatment and surface become rough then this lead to affect resulted Ed. Meanwhile, tan į was less affected by the
change of surface because tan įrelates to the property of in-plane board. Inner part changes are considered more correlate to the bonding point loss. Ra increment and Ed retention of
PB(PF) and MDF(MUF) had high correlation (0.96 and 0.93, respectively). In case of correlation between Ra and tan įincrement, PB(PF) was the one which appeared to have
higher correlation than the other panels (0.82). These correlation means surface roughness change might be used to predict change in dynamic modulus of elasticity as the increasing of Ra simultaneous with the decreasing Ed at successive cycle. However, correlation between
change in tan į by the change in surface roughness might be valid only for PB(PF).
4.4 Conclusion
Surface roughness and dynamic mechanical properties of commercial PB panels and MDF panels subjected to the aging treatment of repetitive cycles of RH fluctuation were evaluated. The stylus technique and a non-destructive vibrational test were effective in following degradation during the aging treatment. The aging treatment increased surface roughness and tan įvalues, but decreased Ed values of the panels compared to their respective initial
values. The Ra increment and Ed retention observed were more marked than tan įincrement.
The degradation behavior during aging treatment appeared to be different among the panels. Surface roughness and tan į seemed to be relatively dependent on resin type, whereas Ed
dynamic mechanical properties. There was strong correlation between Ra increment and Ed
retention. Surface roughness increment might be used to predict dynamic modulus of elasticity, particularly in PB(PF) panel.
42 T able 4.1 I niti al prop er tie s of pa ne l s pe ci m en s for t he a ging tre atme nt Pa ne l ty pe A dhe si ve S ymbol T hic kne ss a (m m) D ens it y b (g c m 3 ) Ed c (G Pa ) Pa rt ic le boa rd (P B ) PF d PB (P F) 12.0 ± 0.03 0.75 ± 0.01 4.5 ± 0.23 MD I e PB (M D I) 12.0 ± 0.09 0.79 ± 0.02 5.3 ± 0.30 M edium de ns ity fi bre boa rd (M D F) MU F f M D F( MUF ) 12.1 ± 0.08 0.74 ± 0.01 5.1 ± 0.08 MD I M D F( MDI ) 9.0 ± 0.03 0.72 ± 0.02 4.0 ± 0.13 a T hic kn es s is g iv en a s a ve ra ge v alue ± s ta nda rd de vi ation b D ens it y is g iv en a s a ve ra ge v alue ± s ta nda rd de vi ation c D yna mic modulus of e la stic ity , g iv en a s a ve ra ge v alue ± s ta nda rd d ev ia tion d P he nol form al de hy de e M eth yl en e diphe ny l diis oc ya na te f M elamine ur ea for mald ehy de
