*Article* **Management of Forest Residues as a Raw Material for the Production of Particleboards**

**Marta P ˛edzik 1,2 , Karol Tomczak 1,3 , Dominika Janiszewska-Latterini 1 , Arkadiusz Tomczak 3 and Tomasz Rogozi ´nski 2, \***


**Abstract:** Expanding the base of raw materials for use in the production of wood-based materials, researchers and panel manufacturers around the world are increasingly trying to produce panel prototypes from raw materials available in a given area and climate, or by managing waste from wood industry processing. The aim of the study was therefore to test the hypothesis that forest residues de-rived from Scots pine roundwood harvesting have the same suitability for the production of three-layer particleboard as the wood of the most valuable part of the Scots pine stem, by comparing selected properties of raw wood material and final product—particleboard. Characterization of both the raw material and the physical-mechanical and hygienic properties of the produced panels was carried out. For these panels from the tree trunk, MOR was 14.6 N/mm<sup>2</sup> , MOE 1960 N/mm<sup>2</sup> and IB 0.46 N/mm<sup>2</sup> . The MOR and IB values turned out to be higher for the panel from the branch and are 16.5 and 0.72 N/mm<sup>2</sup> , respectively. Excessive swelling of the panels resulted in all manufactured particleboards meeting the standardized performance requirements of EN 312 for interior furnishing panels (including furniture) for use in dry conditions (type P2).

**Keywords:** wood-based materials properties; formaldehyde; alternative raw materials; forest residues

#### **1. Introduction**

According to the Food and Agricultural Organization of the United Nations report [1], global roundwood production in 2020 (including wood fuel (WF) and industrial roundwood (IR)) was estimated at 3966 million m<sup>3</sup> (WF—1945 million m<sup>3</sup> and IR 2021 million m<sup>3</sup> ). Compared to 2000 [2], global timber production has increased by about 24%. Available models and calculations show that if the world's human population reaches 10 billion, the demand for wood will be higher than the world's supply of this raw material, which may lead to increased wood prices and uncontrolled deforestation of protected forest areas for illegal wood trading. The importance of forests and the need to protect its resources is the reason to take steps toward environmental sustainability as one of the main parameters when selecting raw materials for industrial purposes. Therefore, to maintain stability in the production of wood and protect the most valuable wood resources, measures should be taken to increase the production of wood-based materials that can replace raw wood.

Forests residues (FR) represent a raw wood material which occurs during logging operations in both mature stands and thinning interventions [3,4]. Forest residues that are eligible for further use include branches, with needles and leaves, and tree tops, as well as undersized or damaged stem parts [5]. FR is usually left to decompose naturally or for the local population, for domestic heating purposes [5–7]. However, FR has the potential to be a suitable renewable source of energy also at the industrial scale [8]. While searching for

**Citation:** P ˛edzik, M.; Tomczak, K.; Janiszewska-Latterini, D.; Tomczak, A.; Rogozi ´nski, T. Management of Forest Residues as a Raw Material for the Production of Particleboards. *Forests* **2022**, *13*, 1933. https:// doi.org/10.3390/f13111933

Academic Editor: Nadir Ayrilmis

Received: 27 October 2022 Accepted: 14 November 2022 Published: 16 November 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

alternative uses of a given raw material, it is necessary to take into account the three pillars of sustainable development, which are environmental, social, and economic aspects [9]. According to estimates, the amount of forest residues available for further use in Polish forests ranges from a few to several million cubic meters [10,11]. The average share of FR in the total mass of harvested wood is about 11% and depends mainly on the fertility of the habitat [6,12]. The average wood volume of forest residues (mainly pine wood residues) in the total harvest was estimated at 37.7 m3/ha [6].

Besides utilizing FR for energy purposes [5,13–15], one of the possible alternative use of forest residues is to exploit them as material for wood-based boards, i.e., particleboards [16,17]. Currently, forest wood, lignocellulosic biomass [18,19], recycled wood [20,21], and industrial wood (residues in the form of cuttings and sawdust) [22–24] are processed for the production of wood-based materials including particleboards. Combinations of lignocellulosic raw materials, recycled wood and industrial particles [25], and even wood bark [26,27] can also be used. By expanding the base of raw materials for use in the production of wood materials, researchers and panels producers from all around the world are more and more often trying to produce prototypes of boards from raw materials available in a given area and climate, or using waste from the processing of the wood industry for added-value applications. The search for new alternative raw materials is one of the key issues in line with the bioeconomy approach, especially for particleboard production, which is in high demand [28].

In addition, it is possible to produce boards from raw materials of other origins, such as annual plant residues, e.g., straws and grasses [28,29], crop residues [30,31], postproduction residues from the food industry, e.g., seed husks [32,33], or residues from garden tree pruning [34,35], sugarcane bagasse [29] and other lignocellulosic wastes [18,19]. When cereal straw is used, to obtain a satisfactory board quality, binders other than those used commonly are often employed to improve the adhesion of straw particles [36,37]. When using alternative raw materials other than wood, it is most advantageous to use mixtures of wood and non-wood raw materials, as this allows to control of the strength parameters of the manufactured boards and reduces the radicality of adjusting technological parameters. At the same time, the use of alternative raw materials additives contributes to the reduction of significant amounts of waste (of various origins) and also reduces the costs of purchasing raw materials, which has positive environmental and economic aspects.

Several studies have focused on the evaluation of FR for particleboard production. Iwakiri et al. 2017 reported that it is possible to produce particleboards based on 50% of forest residues from branches, tree tops, stumps, and roots in a mixture with industrial pine particles highlighting that the obtained panels meet the requirements of EN 312 standard [38]. Panels produced from a mixture of industrial particles and forest residues in a 50/50% ratio were considered the maximum possible, as they showed a statistically equal average to panels produced with a predominance of industrial particles and tree tops and branches (in a 75/25% ratio) and also a mixture of three other types of materials. Wronka and Kowaluk (2022) proved that there is a possibility to use different content (up to 50%) of Scots pine (*Pinus sylvestris* L.) branches to produce three-layer particleboard. Their results indicate that the higher bulk density of branch particles than the industrial material, along with an increase in the content of branch particles, results in a significant increase in internal bonding. On the other hand, the values of flexural strength and modulus of elasticity decreased with an increase in the content of branch particles up to 100%. Therefore, a 50% content in the panel of branch particles characterized by a maximum diameter of 40 mm was also considered maximum [39]. Nurek et al. 2019 showed that the mean density of unfractionated forest residues was app. 800 kg/m<sup>3</sup> [8]. They also noticed the correlation between density and practicle size—the smaller the fraction the greater the density. Rahman et al. (2013) showed that a practical board from stem wood has better properties than particleboard made from branches and twigs of surian tree (*Toona sinensis* Roem) [40]. In addition, wood from garden resources (citrus branches) and beech twigs (*Fagus orientalis* Lipsky) and poplar wood trunks (*Populus alba* L.) can be

successfully used to produce particleboards with good mechanical properties [41]. On the other hand, Nemli et al. 2004 identified the presence of branch wood as a parameter which is negatively influencing the mechanical properties of particleboards produced from black locust biomass [42]. The same phenomenon was observed by Kowaluk et al. (2011). Therefore, further research is needed to test the possibility of applying FR in the sustainable production of particleboards and converting them into value-added products [43]. Most of the published studies are focusing on the properties of wood-based boards produced from mixing wood and branches wood [30,39] or other materials [19,33,35,41]. Moreover, in particular, knowledge about the properties of particleboards produced from FR derived from logging interventions in Scots pine (*Pinus sylvestris* L.) stands is still limited.

The aim of the study was therefore to test the hypothesis that forest residues derived from Scots pine roundwood harvesting have the same suitability for the production of three-layer particleboard as the wood of the most valuable part of the Scots pine stem, by comparing selected properties of raw wood material and the final product—particleboard.

#### **2. Materials and Methods**

#### *2.1. Raw Wood Material*

The wood used in the study was harvested as part of a clearcut in a 94-year-old Scots pine stand in the State Forests in Poland. Wood raw material from 3 trees was taken and debarked for laboratory analyses and particular board preparation. Figure 1 presents graphically the elements of the raw material obtained at individual stages.

**Figure 1.** The process of obtaining individual elements of raw materials.

A 1.0 m long log was taken from each tree, measured from the base of the trunk, and pieces of branches were cut from each crown. Immediately after the tree was cut, approximately 4 cm thick discs were cut at the breast height from each tree for green (GD) and basic (BD) density estimation. Respectively the same was done with branches. Each disc was weighed with an accuracy of 0.001 g using a Steinberg laboratory scale (Steinberg Systems SBS-LW-200A, Berlin, Germany). After weighing, the thickness of each sample was measured to an accuracy of 0.01 mm using a certified Vogel calliper in five different places, and the diameter of each disc was measured by using the cross-over method according to the minimum and maximum size of the sample (Vogel Germany GmbH & Co. KG, Kevelaer, Germany). Then, the volume of each disc was calculated by Equation (1) according to Pérez-Harguindeguy et al. dimensional methodology [44]:

$$\mathbf{V} = \boldsymbol{\pi} \times \left(\mathbf{0.5r}\right)^2 \times \mathbf{h} \tag{\text{m}^3} \tag{1}$$

where V is the volume of disc, r is the average diameter of the disc, and h is the average height of each sample.

In the next stage, discs were transported to the laboratory where were dried for 48 h at 105 ◦C in a laboratory oven. After drying the samples were placed in a desiccator until cooled. Next, each sample was weighed and measured by the same procedure that had been used for samples in fresh condition. The green density was calculated using Equation (2), while basic density was calculated using Equation (3):

$$\text{GD} = \frac{m\_f}{v\_m} \quad (\text{kg/m}^3) \tag{2}$$

$$\text{BD} = \frac{m\_s}{v\_m} \quad (\text{kg/m}^3) \tag{3}$$

where *m<sup>f</sup>* is the mass of fresh felled wood, *m<sup>s</sup>* is the mass of an oven-dried sample and *v<sup>m</sup>* is the volume of the fresh sample.

The remaining parts of both raw materials were first cut into smaller elements on a format saw (Felder, Hall in Tirol, Austria) and then shredded in a cutting mill Condux (Mankato, MN, United States). Obtained particles were divided into fractions on a vibrating sorter (Allgaier, Uhingen, Germany) equipped with sieves with mesh sizes: 8.0, 2.0, 1.0, and 0.5 mm. The core layer fraction consisted of particles retained on a 2.0 mm sieve, and the surface layers were particles from a sieve 1.0 mm in diameter and smaller. The fractional composition of the tree stem and branches stem intended for core layers are determined on a laboratory vibrating screen AS 200 tap (Fritsch, Idar-Oberstein, Germany) with the following sets of mesh sieves: 8.0, 4.0, 2.0, 1.0, 0.50, 0.25, and <0.25 mm. Their bulk density (ρ) was also determined according to Equation (4):

$$
\rho = \frac{m\_c - m\_n}{V} \quad (\text{kg/m}^3) \tag{4}
$$

where *m<sup>c</sup>* is the weight of the measuring vessel with the raw material (kg), *m<sup>n</sup>* is the mass of the measuring vessel (kg) and *V* is the capacity measuring vessel (m<sup>3</sup> ).

#### *2.2. Adhesives*

A melamine-urea-formaldehyde (MUF) adhesive (Swiss Krono Sp. z o.o., Zary, Poland) ˙ was used for the production of particleboards. Table 1 presents the selected properties of the adhesive. The resination was 11% for the surface layers and 9% for the core layer particles of dry resin content referred to dry particles (*w/w*). The hardener is a 40% water solution of NH4NO<sup>3</sup> which was used in an amount of 2 wt.% for a core layer and 3 wt.% for the surface layers. A paraffin emulsion (0.8 wt.% for the core layer and 0.2 wt.% for the surface layers) was also added to the resin to protect the manufactured boards from exposure to water.

**Table 1.** Properties of MUF adhesive.


#### *2.3. Particleboard Production*

Two three-layer particleboards with a nominal density of 670 kg/m<sup>3</sup> and dimensions 500 × 700 × 16 mm were produced the share of surface layers in the panels was 35 wt.%. The raw materials were dried at 100 ◦C to a moisture content of 2–3% in a chamber dryer

with forced air circulation (Ashaki Kagaku Co., Ltd., Tokyo, Japan) and then glued in a rotary laboratory sealer with pneumatic spraying (Lodige, Paderborn, Germany).

The boards were pressed on a hydraulic single-level press (Simpelkamp, Krefeld, Germany) using the pressing parameters: unit pressure 2.5 MPa, temperature 200 ◦C, pressing factor 7 s per one mm of nominal board thickness After manufacturing, respectively, the boards were conditioned in an air-conditioning chamber at a relative humidity of 65 ± 5% and a temperature of 20 ± 2 ◦C. Then parameters were determined, such as density, modulus of elasticity in bending (MOE), modulus of rupture (MOR), internal bond (IB), thickness swelling (TS) after 24 h water immersion, and water absorption (WA), formaldehyde content by perforator method and formaldehyde emission by the chamber method according to procedures defined in the European standards: EN 120 [45], EN 310 [46], EN 317 [47], EN 319 [48], EN 323 [49], EN 717-1 [47].

#### *2.4. Statistical Analysis*

In the first step, the Shapiro–Wilk test was performed to verify the normal distribution of data. The result of the test rejected the normal distribution hypothesis. To compare data between samples the non-parametric Mann–Whitney U test was performed. Statistical inference was performed at significance level α = 0.05. The program Statistica 13.1 (TIBCO Software Inc., Palo Alto, CA, USA) and RStudio and the R package (R Core Team 2022) were used for the calculations.

#### **3. Results**

#### *3.1. Characteristics of Raw Material*

Wood discs collected from branches were characterized by a greater green density of wood than wood collected from tree stems. The difference between these two wood locations was app. 11 kg/m<sup>3</sup> . The maximum GD of branches was over 1000 kg/m<sup>3</sup> and was higher than the maximum stem wood GD by 61.83 kg/m<sup>3</sup> (Table 2).

**Table 2.** Basic statistics of green density (kg/m<sup>3</sup> ) of raw wood material collected from tree stems and branches of wood.


\* Ns—no significant differences.

When in the case of basic density wood from branches was characterized by lower density. The differences between obtained results from the stems and branches were app. 160 kg/m<sup>3</sup> and was statistically significant (Table 3).

**Table 3.** Basic statistics of basic density (kg/m<sup>3</sup> ) of raw wood material collected from tree stems and branches wood.


The bulk density of the material is closely related to the density of the raw material from which it was obtained. Already shredded to particle form, the material from the tree trunk showed a bulk density of about 90 kg/m<sup>3</sup> , while that from tree branches was about 70 kg/m<sup>3</sup> .

Figure 2 presents the fractional composition of particles from tree stems and branches. The analysis shows that the fractions with the size of 4.0, 2.0, and 1.0 mm represent the largest mass share of particles obtained from the tree stem, respectively 30, 36, and 28%. In turn, for the particles from the branches, the largest fraction came from the 2.0 and 1.0 mm sieves—a total of 72% of the entire particle mixture. Very small amounts of fine particles from the sieve with a mesh smaller than 1.0 mm were observed for the trunk particles it is approx. 7% and for the branches particles approx. 4%.

**Figure 2.** Fractional composition of particles from tree stem and branches.

#### *3.2. Properties of Particleboards Depending on Raw Wood Material*

Physical-mechanical properties of the investigated particleboards are reported in Table 4 and the hygienic properties are given in Table 5. No statistically significant differences were found regarding MOR. All the mechanical properties of the manufactured boards have met the requirements of P3 type 16 mm thick boards—non-load-bearing boards for use in humid conditions. For these panels, MOR 14.6 N/mm<sup>2</sup> , MOE 1960 N/mm2, and IB 0.46 N/mm<sup>2</sup> were obtained. On the other hand, the MOR and IB values turned out to be higher for the branch panel and are respectively 16.5 and 0.72 N/mm<sup>2</sup> . Such high values made it possible to qualify the board to the P5 type—load-bearing boards for use in humid conditions. Comparing the achieved results of internal bond tests to the EN 312 standard requirements it should be noted, that the IB values of tree branches panels exceed the minimal requirements for P7 type—heavy duty load-bearing boards for use in humid conditions, which is 0.70 N/mm<sup>2</sup> . In the case of TS and WA, the stem panels showed better performance, but were still insufficient to achieve adequate water resistance.

**Table 4.** The mean value and standard deviation of examined properties of 3-layer particleboards from tree stem and branches wood.


\* Ns—no significant differences.


**Table 5.** Hygienic properties of 3-layer particleboards from tree stems and branches of wood.

The formaldehyde content was determined using the EN 120 perforator method, and it was found that boards manufactured with formaldehyde-containing resin achieved very low formaldehyde content. Many factors can affect the formaldehyde content of particleboard, and certainly, these are the pressing parameters, the raw material, the type and amount of hardener, and the properties of the resin itself: the type of resin and the molar ratio (of urea and formaldehyde) and the formaldehyde content in the resin [28]. A lower formaldehyde content was observed in panels made from the stem portion of 1.2 mg/100 g oven dry board. The obtained formaldehyde emission level is also satisfactorily low. For the E1 emission class, acceptable values for raw wood-based boards are ≤8mg/100 g oven dry board content and release ≤0.124 mg/m<sup>3</sup> air.

#### **4. Discussion**

The main aim of this study was to estimate the suitability of forest residues (pine branches) for the production of three-layer particleboard. Moreover, particleboards from the stems of trees from which the forest residues were collected, were produced for comparison.

Raw wood material collected from pine stems was characterized by a mean green density of around 982 kg/m<sup>3</sup> when the basic density of the same material was 555 kg/m<sup>3</sup> . Obtained green mean values results were similar to results noticed on freshly felled pine logs by Tomczak et al. (2020) and Tomczak et al. (2016) [50,51]. The wood of branches is characterized by a different anatomical structure and structure than stem wood [52]. Mainly due to the high occurrence of reaction wood, which, among other things, has an impact on wood density [53,54]. The mean value of wood branches green density was 993 kg/m<sup>3</sup> when the basic density was 395 kg/m<sup>3</sup> . Findings about branches basic density concerning mostly tropical species [55–57]. When studies about wood branches properties of European species are very limited. The basic density of branches of wood was significantly lower than the density of samples collected from the steam. This result is not comparable with a study carried out by Dibdiakova and Vadla (2012) [58] and Gryc et al. (2011) [59], who compared the basic density of branches and wood basic density of the steam of spruce from different sites. In both presented studies obtained branch wood density was higher than in our study.

Currently, alternative lignocellulosic raw materials are high in high demand [19] which will be able to compete with other industries due to the lack of wood, such as the paper industry, or for energy purposes. Forest residues also could be a good alternative for high quality as a raw material for board production. Alamsyah et al. (2020) produced OSB and particleboard from branches and twigs of surian tree (*Toona sinensis* Roem) [60], Rahman et al. (2013) manufactured boards from branches and stem wood in three types: only from stem wood, only from branch wood and stem-branch mix of bahdi (*Lannea coromandelica* Merr.) [40]. According to the comparison they presented, stem particles were characterized by the highest mechanical properties, then branch-stem particles, and the lowest quality were obtained in branch particles. In the study by Jahan-Latibari and Roohnia (2010), two types of forest residues were used to make particleboard, they were poplar branches, small-diameter poplar wood (3–8 cm), and beech wood [61]. Their results showed that the characteristics of particleboard made from poplar branches and smalldiameter wood were comparable to that made from mature beech wood. In our study panels made of branch particles were characterized by significantly higher quality than stem ones. Which led to classified branch panels as P7 type—heavy-duty load-bearing boards for use in humid conditions according to EN 312. Such differences between our and

presented studies can be explained by species' wood properties. At the stage of production of wood-based materials, it is important to increase the durability of products, which extends their life cycle, thus influencing the extension of the period of carbon sequestration in the wood contained in them, which is a beneficial environmental aspect. The use of a board with such good mechanical parameters for the manufacture of furniture from them can allow reducing the amount of waste wood-based materials created by slower consumption of products and replacement with new ones. Due to the non-emissivity of wood-based panels, the control of free formaldehyde content and emissions is important. Sustainable production of low-emission materials raises ecological issues that are extremely important for the environment and subsequent use, without polluting the air and disposing of products.

In our study, the bulk density of raw material obtained from stem wood was higher than the bulk density of the wood from branches. The research of Kowaluk et al. 2019 used a mixture of industrial wood chips only with an admixture of coniferous wood (mainly *Pinus sylvestris* L.), thanks to which it obtained a bulk density of approx. 164 kg/m<sup>3</sup> [62]. In the same study, applewood particles (classified as moderately heavy species) were used, the density of which was approx. 245 kg/m<sup>3</sup> . The higher the bulk density of the lignocellulosic material, the smaller the thickness of the mat prepared to be pressed to form a particleboard. Knowledge of the material density is especially important when preparing mats from alternative lignocellulosic raw materials, which are characterized by a much lower density than pine wood particles. An example is straw particles obtained from alternative lignocellulosic raw materials, such as rapeseed—approx. 42 kg/m<sup>3</sup> and rye approx. 25 kg/m<sup>3</sup> [36]. The lower bulk density of materials may cause some technical limitations in the industrial-scale manufacturing process. In addition, there may be some logistical problems in developing an optimal and sustainable transportation mechanism [28]. Due to the high volume of logged timber—around 40 mln m<sup>3</sup> per year, State Forests (SF) in Poland are a huge source of forest residues [63]. The exact amount of FR production is difficult to estimate. Due to many environmental factors that affect the amount of waste [6,12]. Gendek et al. [6], in reference to estimates in a paper by Zaj ˛aczkowski (2013) [64], estimated that forest residues in Poland in 2031 could amount to 2.34 million m3, an increase of 12.5% over the 2021 estimate. In recent years SF in Poland offer to sell FR for energy purposes as M2E assortments [65]. The starting price of the auction starts around 4–5 EUR/m<sup>3</sup> (accessed on 1 September 2022). According to the results of our study, this raw material is also relevant and sustainable for high-value particleboard production. Due to the relatively low price, high accessibility, and quality of the final product. Unfortunately, the destabilization of the timber market in Poland, caused among others energy problems and war in Ukraine (stopped wood import). Moreover, EU sanctions have stopped the import of forest residues from Belarus. Therefore currently the final auction price of wood biomass sells by State Forest significantly increased even to over 100 EUR/m<sup>3</sup> .

From an environmental and economic point of view, particleboard production could be based on the idea of using less valuable wood waste and residues. For example, those generated in the woodworking process like particles and residues from the production of wood-based panels, constitute a significant waste [66]. This would create a value-added product that contributes to improved waste management. Moreover, additions of other materials such as waste from the leather industry [67,68], tetrapak packaging [69,70], and construction materials [71,72] can be used. When material innovations are used to make panels, the hygienic properties of the panels made from them must be controlled, the standards of which must be strictly observed. When using wood material, the likelihood of high emission values and formaldehyde is much lower than for alternative raw materials. There are many additives to amine resins that are designed to reduce formaldehyde content in the board, but often their mechanical and physical properties are adversely affected. An example of the use of formaldehyde (FS) Scavenger Solution described in Basbog 2022, which was synthesized using a mixture of monoethanolamine (MEA), ammonium chloride (AC), and distilled water (DW) [73]. Adhesives that are considered biobased on a mixture

of citric acid and glycerin [74], waste melamine impregnated paper [75], protein-based [76], and cornstarch and tannin-based wood adhesives [42,77] can be used.

#### **5. Conclusions**

One possible alternative use of forest residues is their use as an alternative material for wood-based panels, namely particleboard. Based on the research, it can be concluded that forest biomass in the form of pine branches can replace roundwood. The results obtained clearly show that the use of wood material in the form of forest biomass residues improves MOR, which is worth noting, and significantly increases the IB value from 0.46 N/mm<sup>2</sup> for boards made of stem wood particles to 0.72N/mm<sup>2</sup> . Mechanically, the obtained panels meet the minimum requirements for P5-type boards, but nevertheless, the TS value is insufficient. All of the particleboards produced met standardized performance requirements, making them suitable for use as a board for interior fitments (including furniture) for use in dry conditions (type P2).

Future research should be directed toward improving the quality of particleboard produced from raw materials that are alternatives to roundwood. Consideration should be given to raw materials with lower market prices and the use of additives in the form of alternative adhesives, including those that increase water resistance. In addition, further research is needed to reduce competition for raw materials in the energy sector, which is currently a major obstacle to the cultivation of many industrial crops.

**Author Contributions:** Conceptualization, M.P. and K.T.; methodology, M.P., K.T. and A.T.; formal analysis, D.J.-L., A.T. and T.R.; investigation, M.P., K.T., D.J.-L.; data curation, M.P. and K.T.; writing original draft preparation, M.P. and K.T.; writing—review and editing, D.J.-L., T.R.; visualization, M.P. and K.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


## *Article* **Ignition of Wood-Based Boards by Radiant Heat**

**Iveta Marková 1, \*, Martina Ivaniˇcová 2 , Linda Makovická Osvaldová 1 , Jozef Harangózo <sup>2</sup> and Ivana Tureková 2**


**Abstract:** Particleboards (PB) and oriented strand boards (OSB) are commonly used materials in building structures or building interiors. The surface of boards may hence become directly exposed to fire or radiant heat. The aim of this paper is to evaluate the behaviour of uncoated particleboards and OSB exposed to radiant heat. The following ignition parameters were used to observe the process of particleboard and OSB ignition: heat flux intensity (from 43 to 50 kW.m-2 ) and ignition temperature. The time-to-ignition and mass loss of particleboards and OSB with thicknesses of 12, 15 and 18 mm were monitored and compared. The experiments were conducted on a modified device in accordance with ISO 5657: 1997. Results confirmed thermal degradation of samples. Heat flux had a significant effect on mass loss (burning rate) and time-to-ignition. OSB had higher ignition time than particleboards and the thermal degradation of OSB started later, i.e., at a higher temperature than that of particleboards, but OSB also had higher mass loss than particleboards. The samples yielded the same results above 47 kW.m−<sup>2</sup> . Thermal analysis also confirmed a higher thermal decomposition temperature of OSB (179 ◦C) compared to particleboards (146 ◦C). The difference in mass loss in both stages did not exceed 1%.

**Keywords:** particleboard; OSB; heat release; time-to-ignition; mass loss

#### **1. Introduction**

The production of wood-based boards encompasses the utilization of wood of lower quality classes [1–4] and obtaining suitable materials with improved physical and mechanical properties [5–10]. Properties of particleboards (PB) are described in detail in the work of [11,12]. The oriented strand boards (OSB) belong to this product group, but they are also considered an input material in the furniture and construction industries [9,13–15]. A description of OSB in terms of their preparation and properties is defined in the work of [16,17]. These materials are also analysed within the scope of insulation materials [18–21]. They are a part of sandwich panels in low-energy houses [22–25]. They are typically used as an interior sheathing material [26] or furniture [27–30]. Research on the fire resistance of the mentioned materials is also rich [31–35].

Large-size wood-based materials form the largest percentage of wood material in timber houses [36–38]. These materials can be directly exposed to fire [39–41] or the effect of radiant heat [42,43]. Thermal degradation and potentially even ignition of wood-based boards are caused by the action of the ignition source [41–48]. These processes are affected by both the combustible material and the environment in which it is located [49,50]. The ignition process cannot be characterized by a single property [51]. Rantuch et al. [52] used ignition parameters to define the term ignition. Two of these ignition parameters (critical heat flux and ignition temperature) are used here to compare OSB and PB with thicknesses of 12, 15 and 18 mm. This article presents the differences in the results of the research between PB and OSB due to the influence of external heat flux.

Ignition is the ability of a sample to ignite under the action of an external thermal initiator and under defined test conditions, according to [53]. According to ISO 3261 [54],

**Citation:** Marková, I.; Ivaniˇcová, M.; Osvaldová, L.M.; Harangózo, J.; Tureková, I. Ignition of Wood-Based Boards by Radiant Heat. *Forests* **2022**, *13*, 1738. https://doi.org/ 10.3390/f13101738

Academic Editors: Réh Roman, Petar Antov, L'uboš Krišt'ák, Muhammad Adly Rahandi Lubis and Seng Hua Lee

Received: 23 September 2022 Accepted: 18 October 2022 Published: 20 October 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

it is the ability of a material to ignite. The process of ignition is characterized by the timeto-ignition of a sample, which depends on the ignition temperature, thermal properties of materials, sample conditions (size, humidity, orientation) and critical heat flux [55]. Definition of "ignition temperature" can be interpreted as the minimum temperature to which the air must be heated so that the sample placed in the heated air environment ignites, or the surface temperature of the sample just before the ignition point [56–58].

Separate attention is paid to the issue of simulating the ignition of wood under external heat flux from calculations of ignition parameters [59,60]. A prediction model presented in Chen et al.'s paper [61] studies the pyrolysis and ignition time of wood under external heat flux. The solution of the model provides the temperature at each point of the solid and the local solid conversion, and the time-to-ignition of the wood is predicted with the solution of surface temperature [62]. Chen et al. [61] obtained good agreement between experimental and theoretical results.

The aim of this paper is to evaluate the behaviour of uncoated particleboards and OSB exposed to radiant heat. The significant influence of board density and thickness on time-to-ignition and mass loss of PB and OSB samples is monitored and observed. At the same time, the difference in the thermal degradation of PB and OSB samples is sought by comparing the results between time-to ignition and mass loss of PB and OSB samples.

#### **2. Materials and Methods**

#### *2.1. Experimental Samples*

Particleboards (PB) and OSB with thicknesses of 12, 15 and 18 mm (Figure 1a) were used as samples. Selected thicknesses correspond to those typically used in the construction and insulation of houses, in the construction of ceilings, soffits, partitions, etc. The samples were sourced from the company BUCINA DDD, Zvolen, under the product name ˇ Particleboard raw unsanded (Table 1). These particleboards contain softwood strands, mainly spruce, and a urea–formaldehyde adhesive mixture [63]. Č

The samples of oriented strand boards were obtained from the company Kronospan-Jihlava, under product name OSB/3 SUPERFINISH ECO (Figure 1b), without surface treatment. These OSB are multi-layered boards made of flat wood chips of a specific shape and thickness. The chips in the outer layers are oriented parallel to the length or width of the board, the chips in the middle layers may be oriented randomly or generally perpendicular to the lamellae of the outer layers. They are bonded with melamine formaldehyde resin and PMDI, and they are flat-pressed. The boards contain mainly a mixture of different softwood species [64].

**Figure 1.** Example of experimental samples prepared in accordance with ISO 5657 [65]: (**a**) particleboard (PB); (**b**) OSB.

The samples of OSB were cut to specific dimensions (165 × 165 mm) according to ISO 5657: 1997 [65]. Selected sheet board materials were stored at a specific temperature (23 ◦C ± 2 ◦C) and relative humidity (50 ± 5%).

There were tested air-conditioned samples, because the change in moisture will affect the thermal parameters of the samples and subsequently the thermal processes [16,66]. The density of samples (Table 1) was determined according to EN 323: 1996 [67]. The remaining parameters were obtained from the safety data sheets (Table 2).

**Table 1.** The density of PB and OSB samples according to EN 323: 1996 [67].


**Table 2.** Physical and chemical properties and fire-technical characteristics of particleboards and OSB with thicknesses of 10–18 mm.


#### *2.2. Methodology*

−

−

2.2.1. Determination of Mass Loss and Time-to-Ignition

The measuring instrument was calibrated, and heat flux values used for selected samples were logged in Tureková et al. [68,69].

Time-to-ignition and mass loss were determined for the selected level of heat flux density and thickness of the sheet board materials according to a modified procedure based on ISO 5657: 1997 [65]. This modification included a change of the igniter. The ignition was caused only by heat flux, without the use of an open flame (Figure 2).

− **Figure 2.** Scheme of the equipment for determination of flammability of materials at radiant heat flux of 10–50 kW.m−<sup>2</sup> according to ISO 5657: 1997 [65]. (**a**) Real test equipment and equipment scheme. (**b**) Scheme of the used equipment with description of components: 1—heating cone, 2—board for sample, 3—movable arm, 4—connection point for recording experimental data. (**c**) Detailed look at the burning of the particleboard sample with 18 mm thickness in 100 s.

The samples were placed horizontally and exposed to a heat flux of 43 to 50 kW.m−<sup>2</sup> by an electrically heated cone calorimeter. Orientation experiments determined the minimum heat flux required to maintain flame combustion. Time-to-ignition and mass loss were

− −

− −

monitored in the interval of 43 to 50 kW.m−<sup>2</sup> at each thickness of the sheet board material in a series of five repetitions.

The factors which affect time-to-ignition and mass loss are type of sample, board thickness and heat flux density. The obtained results of the ignition and mass loss temperatures were statistically evaluated by statistics. The following factors were used: mixture samples, board thickness (12, 15 and 18 mm) and heat flux density (from 43 to 50 kW.m−<sup>2</sup> ).

#### 2.2.2. Thermal Analysis (Thermogravimetry TGA) of PB and OSB

This analytical method was chosen as the weight of the analysed samples in milligrams. These methods are used in observations and comparison of thermal decomposition of samples, and in the research of the changes and conditions of the chemical reaction course. Thermogravimetry (TGA) studies the course of both thermolysis and polymer burning and records the changes in the weight of the heated sample. The sample was stabilized for 24 h under standard conditions; the test was conducted on a Mettler TA 3000 with a TC 10A processor and TG 50 thermogravimetric weights module in the air and flow rate of 200 mL.min−<sup>1</sup> . The heat increased at a rate of 10 ◦C.min−<sup>1</sup> . The test was carried out up to a temperature of 700 ◦C. The samples for TGA were specifically prepared by disintegration, and the weight of the OSB sample was 10.225 mg and PB was 10.543 mg.

#### **3. Results and Discussion**

The minimum value of radiant heat flux for particleboards and OSB was approximately 43 kW.m−<sup>2</sup> . This value represented the critical heat flux for the selected samples. The maximum value of the radiant heat flux, to which the selected sheet board materials were exposed, was 50 kW.m−<sup>2</sup> . The heat flux was gradually increased by 1 kW.m−<sup>2</sup> (Table 3).



The sample was placed horizontally under the cone calorimeter and exposed to selected heat fluxes which led to gradual thermal degradation and generation of flammable gases. Thermal degradation (Figure 3) is manifested by mass loss (Table 3). Ignition occurs when the critical temperature is reached [69]. Time-to-ignition was recorded, while considering only the permanent ignition of the surface of the analysed sample when exposed to a selected level of heat flux density. The carbonized residue (Figure 4) remained on the surface which has been exposed to radiant heat [70–73], which proves the thermal

− − **Figure 3.** (**a**) Burning process of particleboard samples with 15 mm thickness after their ignition by radiant heat at 48 kW.m−<sup>2</sup> in 80 s; (**b**) OSB sample with 15 mm thickness during experiment in 80 s at 48 kW.m−<sup>2</sup> . − −

**Figure 4.** Cooled samples 10 min after the end of the experiment, sample thickness of 18 mm: (**a**) PB; (**b**) OSB.

− − Time-to-ignition of particleboard and OSB samples of the same thickness (Figure 4) differed in experiments with lower heat flux values, i.e., at 43 to 46 kW.m−<sup>2</sup> . Particleboards and OSB with thicknesses of 12 and 15 mm had the same time-to-ignition values starting from 47 kW.m−<sup>2</sup> (Figure 4 and Table 1). Samples of particleboard and OSB with a thickness of 18 mm showed the same time stamps starting from 48 kW.m−<sup>2</sup> (Figure 5).

**−**

**−**

In comparison with OSB, particleboards generally showed lower time-to-ignition values. The cause can be found in the board structure. OSB consist of larger wood chips compared to particleboards.

The box plot graph for time-to-ignition OSB and PB samples shows the dispersion of the obtained data (Figure 5c). PB samples, in all thicknesses, have comparable results (in Figure 5c), marked with the numbers 2 as PB 12, 4 as PB 15 and 6 as PB 18. The above matrix presents the data obtained from heat flux 43 to 50 kW.m−<sup>2</sup> . It confirms the fact that the thickness of the sample does not have a significant influence on time-to-ignition for PB samples. OSB samples show a significant dispersion of the obtained data and confirm the ratio with increasing heat flux; the ignition time is shortened (see also in Figure 5a).

**Figure 5.** *Cont*.

− − − − − **Figure 5.** Comparison of time-to-ignition and mass loss values of particleboards and OSB depending on the heat flux values. (**a**) Comparison of time-to-ignition results with box graph, where X Axis is heat flux 43–50 kW.m−<sup>2</sup> and Y Axis is time-to-ignition for PB and OSB samples. (**b**) Comparison of mass loss results with box graph, where X Axis is heat flux 43–50 kW.m−<sup>2</sup> and Y Axis is mass loss for PB and OSB samples. Legends: PB 12—PB samples with 12 mm thickness, PB 15—PB samples with 15 mm thickness, PB 18—PB samples with 18 mm thickness, OSB 12—OSB samples with 12 mm thickness, OSB 15—OSB samples with 15 mm thickness, OSB 18—OSB samples with 18 mm thickness. Box graphs have X Axis marks as 2—43 kW.m−<sup>2</sup> ; 3—44 kW.m−<sup>2</sup> ; 4—45 kW.m−<sup>2</sup> ; 5—46 kW.m−<sup>2</sup> ; 6—47 kW.m−<sup>2</sup> ; 7—48 kW.m−<sup>2</sup> ; 8—49 kW.m−<sup>2</sup> ; and 9—50 kW.m−<sup>2</sup> . Confidential interval 95%.

The values of time-to-ignition and mass loss of OSB have a greater dispersion of results, as evidenced by the created box graphs (Figure 5). The variability results from

−

− − − − −

the nature of the board, which is composed of large-area wood particles from pressed flat chips that are pressed under the influence of high pressure and temperature (Figure 6). The binder is a formaldehyde-based resin [74]. Osvald et al. [75] do not assume the influence of the bonding material (glue as well as other additives) on the thermal degradation of the OSB surface.

**Figure 6.** *Cont*.

**Figure 6.** *Cont*.

**Figure 6.** *Cont*.

**Figure 6.** *Cont*.

**Figure 6.** Graphical representation of time-to-ignition and mass loss dependence on board thickness and heat flux with box plots. Legends: blue colour is marked for OSB samples, PB is marked by red, lineárny OSB is linear OSB curve and Lineárny PB is linear curve of PB. Confidential interval 95%.

− − − When comparing the mass loss of particleboards and OSB, lower mass loss values are observed in particleboards of all thicknesses. This difference decreases with increasing sample thickness. Mass loss values of particleboard and OSB samples with a thickness of 18 mm are the same (Figure 5b). A detailed analysis of time-to-ignition and mass loss results for individual sample thicknesses exposed to selected heat flux values is shown in Figure 6. The comparison of time-to-ignition values of particleboards and OSB showed interesting results, apart from the results with the heat flux of 43 kW.m−<sup>2</sup> (Figure 6a). Figure 6 shows the linear dependences of time-to-ignition increase on the sample thickness. At the same time, the graphs are supplemented with quantitative analysis through box graphs. The presented graphs confirm the description of the behaviour of OSB and PB due to the action of radiant heat. Particleboards record lower time-to-ignition values than OSB up to the heat flux of 47 kW.m−<sup>2</sup> (Figure 6b–e). Subsequently, the particleboard and OSB time stamps become identical (Figure 6f–h). All linear dependences maintain an increasing tendency (Figure 6a–h), i.e., the time-to-ignition increases with increasing sample thickness. The given increasing tendency was, however, no longer found at heat flux of 49 and 50 kW.m−<sup>2</sup> (Figure 6g,h).

− Naturally, mass loss (∆*m*) results show the opposite tendency: ∆*m* decreases with increasing sample thickness (Figure 6i–p), while the ∆*m* of OSB is generally greater than the ∆*m* of particleboards. Interesting results can be seen at the heat fluxes of 43 (Figure 6i), 44 (Figure 6j) and 46 (Figure 6l) kW.m−<sup>2</sup> , where there is a change in ∆*m* occurring in samples with a thickness of 18 mm. These cases show higher ∆*m* values of particleboard samples compared to OSB.

The results confirm relatively similar behaviour of particleboard and OSB samples. OSB have generally higher time-to-ignition values, i.e., they withstand the effect of radiant heat longer than particleboards. On the other hand, OSB have a higher ∆*m* value compared to particleboards during thermal degradation and subsequent combustion.

Our results show that as the thickness of samples increases, the differences in the behaviour of the samples disappear under action radiant heat, which can be seen in Figure 6. Practice should take into account the importance of thickness when applying these materials in building structures or elements.

For the purpose of this analysis, another parameter evaluating the behaviour of solids in the event of a fire was calculated, namely the burning rate of OSB (Figure 7a) and particleboards (Figure 7b). The process of thermal degradation of wood-based materials is associated with the charring of the surface, hence some authors [49] call this parameter the charring rate. Once again, dependence between the increase in the rate of burning and the increase in heat flux was confirmed. The burning rate (g.m−<sup>2</sup> .s−<sup>1</sup> ) is calculated as the ratio of mass loss ∆*m* to the time of thermal degradation. The results show a decrease in

the rate of burning with increasing thickness of the sample (Figure 8), which is also stated by Richter et al. [49]. This fact confirms that particleboards act as thermal insulators.

− −

− −

− − − − − − − − **Figure 7.** Graphical dependence of burning rate of OSB and PB samples depending on thermal stress. Legend: 12, 15, 18 are values of thickness. Box plots, have X Axis marks as 2—43 kW.m−<sup>2</sup> ; 3—44 kW.m−<sup>2</sup> ; 4—45 kW.m−<sup>2</sup> ; 5—46 kW.m−<sup>2</sup> ; 6—47 kW.m−<sup>2</sup> ; 7—48 kW.m−<sup>2</sup> ; 8—49 kW.m−<sup>2</sup> ; and 9—50 kW.m−<sup>2</sup> . Confidential interval 95%. − − − − − − − −

− − **Figure 8.** Comparison of thermogravimetric records showing the decomposition of selected board materials at a heating rate of 10 ◦C.min−<sup>1</sup> in an atmosphere of air.

The box plots added to Figure 7 show the same tendency for the burning rate to increase. The values of 43,44, 45 and 46 kW.m−<sup>2</sup> have exactly the same burning rate values, and significant changes occur at heat flows of 48-50 kW.m−<sup>2</sup> .

Despite the previous linear dependences, it is not possible to draw a clear conclusion. This fact is also confirmed in Figure 7. The results show a relationship between the thickness of the samples and the burning rate, which is again linear, but the lines differ (Figure 7).

Richter et al. [49] addressed the effect of oxygen concentration and heat flux on the ignition and burning of particleboards. The experiments were performed on samples of particleboards with different oxygen concentrations (0%–21%), heat fluxes (10–70 kW.m−<sup>2</sup> ), sample densities (600–800 kg.m−<sup>2</sup> ) and sample thicknesses (6–25 mm). The results of Richter et al. [49] showed the effect of heat flux and oxygen concentration on the rate of burning, ignition time and combustion type (pyrolysis, smouldering, combustion).

Maciulaitis et al. [70] watched, among other things, the influence of 30, 35, 40, 45 and 50 kW.m−<sup>2</sup> heat flows in accordance with LST ISO 5657: 1999 [65] with 6 mm, 10 mm, 15 mm and 18 mm thick oriented strand boards (OSBs).

#### **Statistical evaluation of measurement data**

The assessment of the impact of the kind samples (PB, OSB) and the impact of thickness (12,15 and 18 mm) on time-to-ignition and mass loss was carried out by statistical analysis. We used the multifactor analysis of variance (ANOVA) using LSD (95% level of provability) of the test (software Statistica 10).

Table 4 confirms significant differences for thickness. The OSB 18 mm has the highest time-to-ignition value.


**Table 4.** The impact of samples (PB, OSB) and impact of thickness (12.15 and 18 mm) on the time-toignition through the 1-factor analysis of variance (ANOVA) (α= 0.05).

ANOVA–LSD test (α = 0.5): a, b, c, d, e—statistically significant difference.

The mass loss for all samples was 15% of the original weight of the samples. The obtained statistical data did not confirm the significance of the influence of the kind of sample and its thickness on mass loss (Table 5).

**Table 5.** The impact of samples (PB, OSB) and impact of thickness (12.15 and 18 mm) on the time-toignition through the 1-factor analysis of variance (ANOVA) (α = 0.05).


ANOVA–LSD test (α = 0.5): a, b, c—statistically significant difference.

Thermal analysis is another method which uses constant heating to analyse the sample. The results confirm thermal decomposition of samples in two stages [49], as is the case with other cellulosic materials (Table 6). Individual stages of thermal decomposition of particleboard and OSB samples were defined with the use of thermogravimetric analysis in an atmosphere of air.


**Table 6.** Thermogravimetric analysis of OSB and particleboard samples.

Thermal decomposition of the OSB sample (Figure 8) took place in two stages. The first stage of thermal decomposition, the main decomposition of the sample, occurred at a temperature of 179 ◦C. The highest mass loss (65.07%) was recorded at 325.7 ◦C within the first stage of decomposition, which ranged between the temperature of 179 ◦C and 381 ◦C. The second stage of thermal decomposition began at 381 ◦C. At this stage, the second maximum rate of mass loss was recorded at 443 ◦C, with a mass loss of 39.34% and a resistant residue of 0.61% after decomposition.

A similar course of thermal degradation was observed in particleboards. The main decomposition of the particleboard sample occurred at a temperature of 146 ◦C within the temperature range of up to 378 ◦C. At the same time, the highest mass loss of 64.65% was recorded at the temperature of 320.3 ◦C. In the second stage of thermal decomposition, which took place at the temperature range of 378 ◦C to 525 ◦C, the second maximum rate of mass loss was recorded at 445.7 ◦C. At this stage, there was a mass loss of 29.53% and the resistant residue after decomposition amounted to 0.64%.

Given values show the behaviour of boards subjected to thermal stress, where the OSB with a thickness of 12 mm begins to thermally degrade at 179 ◦C and its ignition time is 107 s at a heat flux of 43 kW.m−<sup>2</sup> .

Particleboard with the thickness of 12 mm begins to degrade at 146 ◦C and its ignition time is 89 s. The reported results are consistent in all sample thicknesses and heat flux values.

Sinha et al. [76] studied the effect of exposure time on the flexural strength of OSB and plywood at elevated temperatures. They reached a critical temperature of 190 ◦C at which the strength decreased and thermal degradation occurred. Very interesting research on time-to-ignition on Ancient Wood was conducted by Wang et al. [77].

#### **4. Conclusions**

Based on the performed experiments, it is possible to draw the following conclusions:


**Author Contributions:** Conceptualization, I.T. and M.I.; methodology, I.T.; software, I.M.; validation, I.T., M.I. and J.H.; formal analysis, I.T. and L.M.O.; investigation, I.T. and M.I.; resources, I.T., J.H. and M.I.; data curation, M.I.; writing—original draft preparation, I.T.; writing—review and editing, I.M. and I.T.; project administration, L.M.O.; funding acquisition, I.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** This article was supported by Institute Grant of University of Žilina No. 12716 and the Cultural and Educational Grant Agency of the Ministry of Education, Science, Research and Sport of the Slovak Republic on the basis of the project KEGA 0014UKF-4/2020 Innovative Learning e-modules for Safety in Dual education.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Not applicable for studies not involving humans or animals.

**Acknowledgments:** This article was supported by the Institute Grant of University of Žilina No. 12716 and Project KEGA 0014UKF-4/2020 Innovative Learning e-modules for Safety in Dual Education.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

