Influence of Fibre Length on Properties of Glass Fibre Reinforced Bark Particleboards

• This study was carried out to assess the feasibility of glass ﬁ bres of various lengths (12 mm, 18 mm, 24 mm and 30 mm) as reinforcement on the mechanical performance of bark particleboards intended for thermal insulation. To evaluate their ef ﬁ ciency, the results of ﬁ bre reinforced particleboards at mass of 3wt% con-centration were compared with plain bark based boards. Thermal, physical and mechanical properties (modulus of rupture, modulus of elasticity and internal bond) were determined on unreinforced and reinforced specimens. In general, the results of the thermal conductivity measurements indicated that the bark panels could potentially be used as feedstock for thermal insulation panels. However, the glass ﬁ bres lengths had a direct adverse effect on the mechanical behaviour of the bark particleboard, instead of providing synergistic reinforcement. Furthermore, the static bending properties, mainly the modulus of rupture, gradually decreased with increasing lengths of glass ﬁ bre.


INTRODUCTION 1. UVOD
According to the European Union, the building sector is responsible for consuming 40 % of the total energy in Europe. Therefore, ways of improving thermal effi ciency and decreasing greenhouse gases emissions, such as thermal insulation of building envelopes, is one of the most crucial methods to maximize the energy savings of a building during winter heating and summer air-conditioning (Schiavoni et al., 2016).
Bark (as a lignocellulosic residue), is mainly burned as a fuel for heat energy in wood mills. However, recent research efforts consider the use of bark as raw material in the production of particleboard (Yemele et al., 2008), wood-plastic composites (Yemele et al., 2010), and as thermal insulation material in the form of polyurethane foam (D'Souza et al., 2016) or rigid panels (Kain et al., 2016;Pásztory et al., 2017).
One of the essential limitations on the exploitation of bark particles in wood industry is the resulting weakness of the mechanical properties of the manufactured wood-based panels or composites. A potential solution to overcome this issue could be the reinforcement of bark particle boards with common synthetic fi bres such as glass, carbon, basalt and aramid. Glass fi bres were examined as reinforcing fi laments in cement and concrete composites (Kizilkanat et al., 2015;Arslan, 2016) and wood-plastic composites (Zolfagari et al., 2015). The fl exural strength and the compressive strength of the concrete can be improved by adding glass fi bre to the matrix (Majumdar and Nurse, 1974;İskender and Karasu, 2018). The fl exural properties of solid and laminated wood beams, plywood and particleboards were also investigated by overlaying glass fi bre on the surface (Smulski and Ifju, 1987).
The objectives of this research were to i) examine the effect of the length of glass fi bres on the thermal, physical and mechanical performance of glass fi bres reinforced bark particleboards, ii) determine the most suitable critical length of glass fi bres reinforcement and iii) investigate their market potential as thermal insulation panels.

Materijali
The whole bark samples (inner and outer bark) used in this research were directly collected from the debarking units processing harvested poplar (Populus euramericana cv. Pannónia) logs with a wide diameter range, which were stored in a local sawmill in the area of Sopron, Hungary. The E-glass fi bre roving used for this study was supplied by PD Tatneft-Alabuga Fiberglass LLC (Yelabuga, Russia) ( Table 1). The lengths of 12 mm (GF_12), 18 mm (GF_18), 24 mm (GF_24) and 30 mm (GF_30) were manually cut from the fi breglass roving cylindrical packages ( Figure 1) similarly as Tsalagkas et al. (2019). The main properties of the glass fi bres and their lengths used in this work are given in Table 1 and Figure 1, respectively. The commercial UF resin and hardener used in this work was purchased from DUKOL Ostrava s.r.o. The resin properties at 20 °C were: solid content 67.65 %; pH value 8.8; dynamic viscosity 500 mPa·s; and Ford cup viscosity 84 seconds.

Particleboard manufacturing 2.2. Proizvodnja iverice
Initially, bark slices of various thicknesses (15-30 mm), consisting of inner and outer bark, were collected and dried below 20 % in a chamber. Consecutively, the inner and outer bark were cut into small pieces and  chipped into particles using a hammer mill equipped with 8 mm screening holes. The granulated bark particles ranged from 0.5 mm to 8 mm fractions and were used as raw material for the manufacturing of bark panels. The moisture content of the bark particles was adjusted to 6 to 9 % before further processing. The randomly oriented, chopped glass fi bres of the prepared length were placed in a laboratory blender and homogenized with the bark particles for fi ve minutes, before pressing. A mass of 4 % urea formaldehyde (UF) adhesive was sprayed on and was mixed to the mixture of the bark particles and glass fi bres.The adhesive contains 35 % of aqueous solution of ammonium sulphate (NH 4 ) 2 SO 4 as catalyst. The glued mixtures were immediately manually layered and formed in a wooden frame into a mat. Thereafter, the frame was removed, and the mats were pre-pressed by hand to compact the materials without heat transfer. Next, the mats were transferred to a single-opening hydraulic hot press machine (Siempelkamp). The pressing temperature was set at 180 °C with a pressing time of 18 seconds per millimetre of thickness. The initial press pressure was 2.86 MPa, which was reduced after 120 seconds to 2 MPa, and after 240 seconds to 1.15 MPa to decrease the risk of panel damage by vapour. Single-layered boards with dimensions of 500 mm × 500 mm × 20 mm with a target density of 350 kg/m 3 were produced.

Karakterizacija
All the composite boards were kept at 20 °C and 65 % relative humidity, until equilibrium moisture content (EMC) was achieved, prior to experimental measurements. All these boards were cut and trimmed into various test specimens.
Apparent density ρ at 9.5-10.2 % moisture content was measured on the same samples used for the mechanical tests, as the average of at least fi fteen specimens. The mechanical bending test was done according to the standard (EN 323:1993). Dimensional stability of the specimens from thickness swelling (TS) and water absorption (WA) were calculated according to European Standard EN 317:1993.
Thermal conductivity was measured across the thickness of the composite boards (500 mm × 500 mm × 20 mm) with a heat fl ow meter using a guarded hotplate method. The thermal conductivity measuring equipment was designed and assembled at the Technical University of Budapest. The measurement started when the steady state was achieved, i.e. when the fl uctuation of the last fi fty per minute measurements was under 0.002 W/mK.
The bending strength (MOR) and modulus of elasticity (MOE) of bark composite boards were measured using a universal testing machine Instron 5506 (static three-point bending), in compliance with the applicable European Standard EN 310:1993 at a crosshead speed of 8 mm/min. The tensile strength perpendicular to the surface (internal bond, IB) was determined by using 50 mm × 50 mm specimens from each panel according to EN 319:1993, at a speed of 0.8 mm/min.

Statistical analysis 2.4. Statistička analiza
The analysis of variance (ANOVA) was applied using Statistica13 software (TIBCO Software Inc., USA) to statistically evaluate the infl uence of the length of glass fi bres on the thermal, mechanical and dimensional stability of the glass fi bre reinforced bark boards produced in the current study. All data were checked for normality (Shapiro-Wilk test) and homogeneity of variance (Levene's test), at a 5 % significance level. Extreme values that did not fi t the homogeneity or normality were excluded from subsequent analyses. Post hoc tests were conducted with Tukey's HSD test method at 5 % signifi cance level. Table 2 presents the mean values of the physical, thermal and mechanical properties of the bark particleboards reinforced with various glass fi bre lengths ranging from 12 mm to 30 mm with a 6 mm interval. The mean density of the unreinforced and glass fi bre reinforced bark particleboards fl uctuated from 373.21 to 387.57 kg/cm 3 , while the moisture content was estimated between 9.58 % and 10.18 %. The dimensional stability (TS % and WA %) was found not to be statistically different among the boards, indicating that glass fi bres had no infl uence on these properties. In addition, the thermal conductivity of the boards was shown to be similar, in the area of 0.074-0.078 W/m·K, among all the panels made of bark-glass fi bre composite boards, whilst the thermal conductivity of the bark particleboards was calculated as 0.079 W/m·K. Compared with other bio-based natural insulation materials, the obtained values were higher or contiguous to other wood-based panels (Pásztory et al., 2017), yet in a similar range. It is evident that the addition of glass fibres, with their lengths varying from 12 mm up to 30 mm, had a negative effect instead of a synergistic one on the static bending properties of bark particleboards. The control boards had the highest MOR and MOE values compared to the reinforced bark boards. Further, the MOR and MOE decreased with increasing the fi bre length from 12 mm to 30 mm. The boards reinforced with a glass fi bre length of 12 mm indicated the best mechanical performance among the fi bre lengths tested. For specimens with a 12 mm glass fi bre length, the MOR of the panels decreased by 28.17 % compared to control specimens. Furthermore, the MOR of 18 mm length panels additionally decreased by 17.65 % (related to 12 mm), the MOR of 24 mm fi bre length panels further decreased by 10.71 % (in relation to 18 mm) and the MOR of 30 mm glass fi bres reinforced again decreased by 12.00 % compared to 24 mm long fi bres. Accordingly, the MOE value gradually decreased from 0.24 GPa to 0.19 GPa with fi bres 12 mm long and then remained constant with longer fi bre lengths.

REZULTATI I RASPRAVA
However, it has to be noted that diffi culties in workability or fl occulation of fi bres during mixing were observed with increasing fi bre lengths, at the suggested 3wt% concentration. A possible explanation for the decreasing fl exural rupture could be a combination of several factors, such as the shear stresses distribution, the air voids in the panel material and the overlapping degree of the reinforced bark particleboards with longer glass fi bre lengths. Furthermore, statistically, the IB values were not signifi cantly affected by the reinforcement, demonstrating that the glass fi bres had no interaction with the urea-formaldehyde bonded bark particles, therefore indicating no cohesion between the glass fi bres and bark particles. This assumption was made because the glass fi bres area on the breaking surface appears in higher ratio than the ratio of glass fi bres in the volume. Consequently, the connection between the fi bres and the urea formaldehyde glued bark particles was weaker than the bonding between glued bark particles. The bonding connection between urea formaldehyde resin and glass fi bre was investigated by Sharma et al. (2015) from different mechanical aspects, but the positive effect of the glass fi bre was not apparent.

ZAKLJUČAK
Glass fi bre reinforced bark particleboards were evaluated as thermal insulation panels. The thermal conductivity investigation demonstrated the viability of the proposed boards as insulation panels. However, test results showed a negative infl uence especially on the bending strength (MOR) of the reinforced bark particleboards using the various glass fi bre lengths. As the glass fi bre length became longer, the fl exural strength of reinforced bark particleboards reduced signifi cantly. Therefore, the glass fi bre 12 mm long demonstrated the best perfor-mance among the tested specimens. In addition, the internal bond strength of the reinforced specimens was not statistically different compared to control bark particleboards, indicating no cohesion between the glass fi bres and the bark particles bonded with urea formaldehyde adhesive. Therefore, this study will be focused on i) the use of shorter glass fi bre lengths such as 6 mm and 3mm and ii) the improvement of interfacial interaction between the bark particles and glass fi bres through surface modifi cation methods, or by examining other adhesives.