A study on some physical and mechanical properties of molded thermal insulation materials produced from perlite and boric acid added forestry by-products

ABSTRACT
 The tree bark and cones are either burned or left in the forest. In both cases, it causes both environmental problems and a significant loss of economic potential. In this study, some physical and mechanical properties of molded thermal insulation materials produced from natural raw materials were investigated experimentally. In this context, 15 types of sandwich panel materials containing honeycomb-shaped core were produced from pine (Pinus brutia Ten.) bark and cones. The mean moisture content, density, compressive strength, tensile strength perpendicular to the surface, tensile strength parallel to the surface, and dimensional stability values of the materials were 10.6%, 269.717 kg/m3, 0.493.06 N/mm2, 0.011 N/mm2, 0.150 N/mm2 and −0.156%, −0.054%, 0.942%, respectively. According to the results, it was determined that the increase in particle size and perlite ratio in the materials produced from the bark decreased the density. Moreover, it was found that the particle size-moisture content relationship and the perlite ratio-moisture content relationship varied in the materials containing bark and cones, that the mechanical properties were higher in the materials containing cones, and that the dimensional stability did not show a regular change.


Introduction
The envelope of a building significantly influences the surrounding microclimate, creates a boundary between the indoor and outdoor environment, and plays an important role, as it affects the thermal comfort and energy losses of occupants during use. Interest in environmental and energy issues has grown exponentially in recent years, and many international and national policies have been adopted to provide the planet with a more sustainable future (Schiavoni et al. 2016). Natural resources are often described as "renewable" or "non-renewable." Renewable resources are the ones that can be regularly renewed or harvested, such as wood and agricultural products, and can be renewed to any desired level as long as production conditions are suitable. Non-renewable resources are the ones that can only be harvested once. Most of these are significantly limited; it is usually expressed as a reserve (Berge 2009). Glass wool, rock wool, polyurethane foam, expanded polystyrene foam and extruded polystyrene foam are the most widely used artificial thermal insulation materials. They are produced from non-renewable resources, and their production process is difficult and requires expensive investment. They also have a lot of negative environmental effects during and after use (Cetiner and Shea 2018). Concerns about the consumption of energy resources, raw materials and excessive pollution have increased in recent years. Much more emphasis is placed on recycling and the use of non-toxic materials in many studies (Blanchet et al. 2000;Korjenic et al. 2011). Environmental awareness cannot be limited to energy saving only; minimum energy and resource use and minimum production pollution during the manufacture of insulation materials are also the issues to be considered (Berge 2009). There is a growing trend in the world towards the importance of a livable environment. Parallel to this trend, interest in non-toxic, renewable, easy, cheap and abundant natural resources, such as bark (especially coniferous species), cones and lignocellulosic wastes, which are easy to produce, is increasing day by day (Efe 2022). The above-mentioned natural raw materials are still not industrially converted into high value-added products; they cause environmental pollution and are generally burned (Kain 2016). Forests produce tons of waste and by-products, such as bark and cones, every year. Of these, the bark is resistant to microorganisms, has low thermal conductivity, and is environmentally friendly (Kain et al. 2015), while cones, despite being a renewable fiber source, are also not used effectively (Buyuksari et al. 2010;Arrakhiz et al. 2013). According to the 2021 statistics of the Food and Agriculture Organization (FAO), it was reported that 1.398 billion cubic meters of coniferous wood and 2.572 billion cubic meters of nonconiferous wood were produced in 2019 (FAO 2021). Taking into account that the tree contains 10% bark on average (Xing et al. 2007;Hoong et al. 2011), it can be stated that the annual potential usable bark volume is 397 million cubic meters. Huge amounts of bark from forest products industry activities are either thrown away or burned each year (Barbu 2011;Pásztory et al. 2016). On the other hand, it is not known how many pine cones appear in the world each year, except for the stone pine. However, given the current tree inventory, it can be estimated that they are in very large quantities. It was determined that many studies were performed on bark and cones in the literature. Sirisha (Khuntia and Biswas 2020), birch (Réh et al. 2021), beech (Bekhta et al. 2021), cork (Pinto et al. 2018;Andrzejewski et al. 2019;Amaro et al. 2020;Mirski et al. 2020), pine (Zhang et al. 2020;Borysiuk et al. 2021), red cedar (Chen and Yan 2018), walnut, chestnut, fir and spruce (Aydın et al. 2010) species were investigated as a filler in the production of wood-based composites. One of the focus of attention in recent years is bark-based insulation materials. In this context, cork (La Rosa et al. 2014;Ferreira et al. 2016), pine (Kain et al. 2012;Tudor et al. 2020a), black locust (Pásztory et al. 2017), spruce (Tudor et al. 2020b), poplar (Tsalagkas et al. 2019;Busquets-Ferrer et al. 2021), larix (Kain et al. 2014(Kain et al. , 2015(Kain et al. , 2018, spruce, pine, birch (Holmberg et al. 2016), and eucalyptus (Wesolowski et al. 2014) were investigated. Pine cones, another non-wood forest product, are a good candidate raw material for various uses with their strong fibrous structure, high specific gravity, and chemical content different from wood and bark. Many studies were conducted on pine cones. It was determined that the use of cones in the production of chip material or medium density fiber material (MDF) adversely affected its mechanical properties (Ayrilmis et al. 2009;Buyuksari et al. 2010;Sahin and Arslan 2011;Niemz and Sandberg 2022). The effects of cone fiber or flour contained in some matrices, such as epoxy (Baştürk and Kürşat 2015;Kolář et al. 2019), polypropylene (Arrakhiz et al. 2012), high-density polyethylene (HDPE) (Guo et al. 2019), and polycaprolactone (PCL) (Jha et al. 2018), were investigated. On the other hand, many mineralbased additives used in nano-or micro-scales to improve composite properties were the subject of different studies. Of these, wollastonite (Taghiyari et al. 2020(Taghiyari et al. , 2021a(Taghiyari et al. , 2021b and perlite (Sengul et al. 2011;Sun and Wang 2015;Efe 2022) were used in a wide variety of fields. In summary, it is understood that studies on the performance of bark and cone as composite, filler, thermal insulation material and filter were carried out in the literature. Some researchers reported in their studies that the bark could be supplied in high amounts as a raw material compared to other biobased insulating materials (Busquets-Ferrer et al. 2021), that the relationships between particle size, material density and thermal conductivity would be a good topic for research (Kain et al. 2015), that the fire behavior and water absorption of insulation materials needed to be investigated in further studies (La Rosa et al. 2014;Pásztory et al. 2017;Tsalagkas et al. 2019), and that it was necessary to work on different chip geometries (Blanchet et al. 2000). In addition, some researchers' recommendation that the strength and mechanical properties of thermal insulation materials should also be investigated (La Rosa et al. 2014;Ferreira et al. 2016;Pásztory et al. 2016;Tsalagkas et al. 2019) emphasizes the importance of the subject. The materials produced in the literature research are either gap-free, completely filled or loose; so far, no research has been found on the production of barkand/or cone-based thermal insulation materials using honeycomb molds. The aim of this study is to determine some physical and mechanical properties of the molded composite material produced from pine cone and bark using a different technique and to analyze the relationship of these properties with the particle size and perlite ratio of the raw materials in the material content. The gap ratio of the produced materials is about 55%. In order to determine the physical and mechanical properties and to show the relationships of these properties with the production parameters, 15 types of materials were produced. Cones and barks were used in three particle sizes as fine, medium and coarse. Paraffin, boric acid and expanded perlite were used as additives in varying proportions. The data obtained in the study is promising for scientists doing research in this field, for the sector representatives producing commercial building sheathing materials, for the evaluation of the relevant people, and for the natural wastes to be the subject of long-term projects in parallel with national and/or international policies.

Materials and methods
The bark and cones of Calabrian Pine (Pinus brutia Ten.) were used as raw materials because Turkey's most widely distributed coniferous species (OGM 2013) has the thickest bark. The bark and cones were collected in Yenice district, Çanakkale province, Turkey. The raw materials were procured in July 2018 because in this month, the production of logs is common, the cones shed their seeds, and there is a fastnatural drying. In the production of materials, paraffin, boric acid, and expanded perlite were added to improve moisture and fire resistance, lightness and insulation properties. Ureaformaldehyde resin was used as glue. Perlite was obtained from Genper Expanded Perlite Industry Business Corporation in Kütahya, Turkey; boric acid was obtained from Forscher, Turkey; and paraffin and urea-formaldehyde resin were obtained from Kastamonu Integrated Wood Industry, and Trade Corporation Particleboard Plant, Balıkesir, Turkey. The technical properties of the urea-formaldehyde resin, boric acid, perlite and paraffin are given in Tables 1 and 2. A hammer mill with 6 and 9 mm aperture sieves, 24 crushing jaws and a 3 kW motor used in grinding raw materials, and an air-circulating drying oven that can be adjusted digitally up to 300°C were purchased from Doğangül Makine Company, Gaziantep, Turkey. Steel sieves with 0.5, 1, 2, 3, 4 and 5 mm openings were used manually in the classification of the raw materials. A honeycomb-shaped aluminum mold set, with 500 × 350 × 50 mm, was purchased from Alper Torna Company, Gaziantep, Turkey, to produce the core layers of all materials. The surface layers were produced in Kocayusuf Piton-KP-1 brand hydraulic press, which can reach 130°C and 200 bar. Some other equipment used in production were as follows; A curing oven adjustable up to 200°C ; digital humidity and temperature meters; 3 hydraulic jacks, each with a lifting capacity of 5 tons; and a digital balance with a 40 kg capacity. The tests were carried out in the Çanakkale Onsekiz Mart University Yenice Vocational School Forestry Department laboratory and Kahramanmaraş Sütçü İmam University Forestry Faculty laboratories.

Production of materials
The materials were produced in three stages: the production of 4 mm thick surface layers, the production of 32 mm thick core layers, and the assembly of the material. Material types and contents are shown in Table 3.
In the production of all materials, 10% glue, 1% boric acid and 0.5% paraffin were used by weight. Cones and barks having air dry (12%-15%) moisture content were first ground in a hammer mill and then sieved by hand, and particles smaller than 1 mm and larger than 4 mm were removed as unused raw materials. Finally, it was dried in the oven until it reached 1%-2% moisture. In the production of the surface layers, a frame made of medium density fiberboard (MDF) with 20 × 450 × 550 mm as a template was used. Cone/bark, perlite, glue, boric acid and paraffin in varying proportions for each material were mixed with a hand mixer and laid in the template. Thus, material mats were created. Surface layer mats were pre-compressed with an 18 mm thick MDF material using 4 joiner's clamps. Finally, 4 mm thick materials were produced in a hydraulic press using a compression ratio of 1:5. In the production of the surface layers, 200 bar pressure was applied at 130°C for 8 min. The mixture of the same content as the surface layers was used in the production of the core layers. Here, the mixture was laid in a honeycombshaped aluminum mold, the voids of the material were removed by vibration, and 2 kg/cm 2 pressure was applied with hydraulic jacks. The mold was fixed under pressure and cured at 160°C for 30 min, resulting in a 32 mm thick middle layer. The apparatus for the production of the honeycomb-shaped aluminum mold and the middle layer is shown in Figures 1 and 2, respectively.
By bonding the surface layers to both sides of the core layer, the target material in the form of a 40 mm thick sandwich was produced. In this final stage, the pressing parameters were 130°C, 20 bar and 8 min.
2.1.1. Tests 2.1.1.1. Moisture content. TS EN 322 (1999a) standard was used to determine the moisture content. Four samples of 50 × 50 × 40 mm from each material were weighed first, the initial weight was recorded, and then they were kept in an oven at 103 ± 2°C. This process was continued by measuring every 6 h until the difference between the last two weighing was 0.1%. The samples removed from the oven were kept in the desiccator until they reached room temperature. The moisture content was calculated according to: where H is the moisture content, m H is the moist weight, and m 0 is the dry weight.
where σ m is the CS (kPa), σ 10 is the CS corresponding to 10% strain, F m is the maximum force, A 0 is the initial cross-sectional area of the sample:   where F 10 is the force (N) corresponding to 10% strain, and A 0 is the initial cross-sectional area (mm 2 ) of the test specimen.   Z010 machine at a loading speed of 10 mm/min. Preloading was not applied to the samples. The IB was calculated by determining the maximum force at break (MPa). Five samples of 50 × 50 × 40 mm were used for each type of material. The IB was calculated with the help of: where σ m is the IB, F m is the maximum tensile force, A is the cross-sectional area of the sample, l,b is the length, and width of the sample.  (5). Figure 3 shows the test setup.
where σ T is TS, F m is the maximum tensile force, d is the thickness of the test area, and b is the width of the test area. . After the initial dimensions of the samples were measured with an accuracy of 0.1 mm, they were placed in the air-conditioning cabinet. It was removed after 48 h and kept at 65 ± 5% relative humidity and 23 ± 2°C for at least 3 h, and then the sample sizes (lt, bt, dt) were measured again. Final measurements were made for the initial msurement points (l t1 , l t2 , l t3 and b t1 , b t2 , b t3 and d t1 , d t2 , d t3 , d t4 , d t5 ). The size changes Δε l , Δε b and Δε d in the samples were calculated with the equations (6), (7), and (8).
where Δεl is the length change Δεb is the width change, Δεd is the thickness change (%), l 0 , b 0 , d 0 are the initial dimensions (mm), and l t , b t , d t are the final dimensions.

Moisture content
The moisture (MC) of the materials was in the range of 7.10%-15.96%. The mean MC of all materials was 10.60% with a standard deviation of 1.82%. As seen in Figure 4, the MC values changed in direct proportion to particle size in the 100% bark-based group A materials and inversely in the 100% cone-based group B materials. Although there seems to be a big difference between the MCs of the materials in Figure  4, it can be said that the MC values are close to each other numerically. The average MC of all materials was 10.90%. The fact that B1 had the highest MC can be explained by the better adhesion of the water in the glue to the cellulose molecules in the cone fiber. B1, which had more particle surface, retained more glue and its MC was higher than B2 and B3. The variation of the MC values in the perlite containing materials is shown in Figure 5. It was determined that the perlite ratio had an effect that increased the MC in the D Figure 3. Test setup of tensile strength parallel to the surface. Figure 4. Particle size-MC relationships. Values were adopted from Efe (2022).
group materials and decreased it in the E group materials. It was observed that perlite did not have the same effect in both groups due to the structural differences of the bark and the cone. The samples containing perlite were expected to have higher MC than the others. However, except for E1, the MC values of the D and E group samples were lower. This can be explained by the increased density of perlite under pressure and the decrease in hygroscopicity. Kain et al. (2012Kain et al. ( , 2013b produced materials with particle sizes between 0 and 45 mm from pine (Pinus sylvestris) bark, and found a mean MC of 12.2% with a standard deviation of 0.6%. The MC of 20 mm thick boards produced from Larch (Larix decidua Mill.) bark was reported to be 15.6% with a standard deviation of 0.7% (Kain et al. 2015). Tsalagkas et al. (2019) found the MC of the boards produced from poplar (Populus sp.) bark was in the range of 7.29%-9.12%. Materials produced from 19 to 25.4 mm thick larch and poplar bark were reported to have MC in the range of 3%-8.6% (Tudor et al. 2020b). On the other hand, the MC of 19-21 mm thick materials produced from bark of spruce (Picea abies L.) and larch (Larix decidua Mill.) was reported to be 8%-9% (Tudor et al. 2020a). In general, although the MC results obtained in this study were similar to some of the studies mentioned above, there were differences in terms of wood type, raw material particle size, pressing technique, board form, other materials added to the board, and board thickness.

Density
Density has a significant effect on many properties of the material, such as water absorption, bending strength, and thermal insulation performance (Xing et al. 2007;Gupta and Yan 2011). The lowest and highest densities of the materials were 244.48 kg/m 3 in the C3 group and 305.43 kg/m 3 in the B1 group, respectively. The mean density was 269.70 kg/m 3 with a standard deviation of 16.50. In this study, it was observed that the densities of the materials were affected by the structural properties such as raw material density, fiber, extractive substance, etc., and production parameters such as gluing, laying, additives, and compression ratio. In this context, although the expected results were obtained in some of the materials; some were not available.
It is thought that the density differences are due to the different resistance of bark and cones with different particle sizes under pressure. Figures 6 and 7 give clues about density-particle size and density-perlite ratio relationships. There was a regular variation depending on the bark/cone mixture in the group C samples and depending on the cone/perlite mixture in the group E samples. However, in the other groups, it was observed that the density results did not have a regular variation depending on the content. It was expected that there would be an inverse proportion between particle size and material density in the group A and B samples, but it was seen that there was no such relationship. The main reasons leading to this situation are considered as non-homogeneous internal adhesion, laying and pressing errors. The density of the cone particles was higher than that of the bark. Therefore, a higher density was measured in the B and E group materials than the others due to the cone content. The group D samples consisting of shell-perlite mixture were expected to have the lowest density, but it was determined that it was not so due to the possible errors mentioned above. Studies with similar raw materials were examined in the literature. For example, the densities of 20 mm thick materials produced from 6 to 10 mm thick larch (Larix decidua Mill.) bark were between 270 and 540 kg/m 3 (Kain et al. 2014), the densities of 20 mm thick boards produced from the bark and woods of   Picea abies, Pinus sylvestris, and Abies alba were between 350 and 500 kg/m 3 (Kain et al. 2013a), the densities of 10-20-30-40 mm thick boards produced from 1 to 8 mm, 8 to 13 mm and 13 to 45 mm black locust (Robinia pseudoacacia) barks were between 185.8 and 548.3 kg/m 3 (Pásztory et al. 2017), and the densities of 30 mm thick materials produced from the 3 to 6 mm and 10-30 mm particles of larch (Larix decidua Mill.) barks were found to be between 191 and 609 kg/m 3 (Kain et al. 2018). Moreover, the densities of 10 mm thick boards produced using stone pine (Pinus pinea L.) cones were between 730 and 760 kg/m 3 (Ayrilmis et al. 2009), and the densities of 20 mm thick materials produced using poplar (Populus sp.) bark were 336.80-413.07 kg/m 3 (Tsalagkas et al. 2019), and the mean densities of the boards produced with pine and larch bark were reported to be 960 kg/m 3 (Rudenko 2010). When the densities of the materials in the above studies were compared with the data in this study, it was seen that the mean density was 269.717 kg/m 3 and that the change interval of 244.485-305.430 kg/m 3 was considerably lower than them.

Mechanical properties
There is a need for thermal insulation materials with sufficient CS in horizontal or slightly inclined applications in buildings. If the thermal insulation material does not have sufficient CS, the material will be deformed against the forces acting on it from the external environment and will not be able to fulfill the task expected from it (Akıncı 2007). One of the important features of thermal insulation materials is the mechanical resistance they show against loads of variable duration. In thermal insulation materials, a decrease in the thickness of more than a certain value causes an unacceptable deterioration in the performance of the material. At this time, even if the material carries a load, it cannot fulfill its main task. For this reason, in thermal insulation materials, the CS at 10% deformation (that is when there is a 10% decrease in thickness) is taken as the basis. This value is called CS at 10% deformation, and is shown by σ 10 . The highest CS was measured with 0.891 N/mm 2 in the B3 group material, and the lowest CS was measured in the A1 group material with 0.266 N/mm 2 . The mean CS value was found to be 0.493 N/mm 2 with a standard deviation of 0.21. On the other hand, the highest values were recorded in the B and E group materials containing cones. It is considered that the main reasons for this variability in the data were the density differences of the raw materials (the average bulk density of the bark was 210-220 kg/m 3 and the average bulk density of the cone was 220-270 kg/m 3 ) and structural differences. In addition, the possible reason for the low CS of the perlite-containing groups may be that perlite and the other raw materials could not provide good internal bonding. On the other hand, the CS increased in direct proportion to the particle size in the group A materials and in direct proportion to the perlite ratio in the group D materials. It decreased inversely with the perlite ratio in the E group materials and inversely with the cone ratio in the C group materials. No regular change was observed in the B group materials. Figure 8 shows the particle size-density-compressive strength relationships. The general expectation is that the CS increases with the increase in density. However, the findings show that there was no such a relationship between density and CS; however, it showed that the particle size increase had a positive effect on the CS. It is thought that gluing and pressing errors had an effect on the results, especially in the production of core layers. Figure 9 gives an idea about the perlite ratio-density-compressive strength relationships. As the perlite ratio increased in the D group materials containing shell, the CS increased, while it decreased in the E group materials containing cones. It was determined that the CS changed inversely with the density of the group D materials and directly proportional to the density of the group E materials. Figure 10 shows the behavior of the group A materials in the CS test. According to the graph, the maximum deformation of the A1, A2 and A3 group materials occurred at the end of the application of force of 4.92, 7.53 and 9.32 N, respectively. These results showed that the CS increased as the particle size increased. Figure 11 shows the behavior of the group B materials in the CS test. According to the graph, the maximum deformation of the B1, B2 and B3 group materials occurred at the end of the application of force of 20.15, 17.61 and 24.83 N, respectively. These results showed that there was an irregular relationship between particle size and CS in the group B materials. Figure 12 shows the behavior of the  group C materials in the CS test. According to the graph, the maximum deformation of the C1, C2 and C3 group materials occurred at the end of the force application of 14.38, 10.01 and 8.60 N, respectively. These results showed that the CS of C group materials decreased as the shell ratio decreased. Figure 13 shows the behavior of the group D materials in the CS test. According to the graph, the maximum deformation of the D1, D2 and D3 group materials occurred at the end of the force application of 14. 87, 9.15 and 12.20 N, respectively. These results showed that although there was no linear relationship between the perlite content and the CS in the group D materials, the general trend was that the perlite ratio decreased the CS. Figure 14 shows the behavior of the group E materials in the CS test. According to the graph, the maximum deformation of the E1, E2 and E3 group materials occurred at the end of the application of force of 19.34, 19.58 and 14.90 N, respectively. These results showed that although there was no linear relationship between the perlite content and the CS in the group E materials, the general trend was that the perlite ratio increased the CS.
The force-deformation graphs showed that in general, all materials underwent a similar plastic deformation from the beginning of the test up to 2 N, and that after this point, a permanent deformation took place. In summary, the group A, C and D materials containing shells had lower CS than the others; the group B materials showed the highest CS; and the perlite content had a different effect depending on whether the raw material was bark or cone. It was determined that the structural properties and production parameters of the bark and cone also affected the CS. The CS of mineral wools at 10% deformation, which is one of the materials currently widely used in the insulation sector, is given between 0.5-500 kPa in TS EN 13162 + A1 (2015). The CS at 10% deformation for EPS is given as 30-500 kPa in TS EN 13163:2012+ A2 (2017, and the CS of XPS at 10% deformation is given as 100-1000 kPa in TS EN 13164 + A1 (2015). According to EN 13171:2012, the CS at 10% deformation required by wood fiber thermal insulation boards was reported between 5 and 500 kPa. The CS data at 10% deformation of all material groups produced in this study met the requirements in the standards given above. (Kain et al. 2012(Kain et al. , 2013b reported     that the panel density showed a highly significant (p < 0.001) positive correlation with CS and some other mechanical properties (Dost 1971;Kain et al. 2012). It was reported that the increase in density did not directly affect the mechanical properties (Gupta and Yan 2011), and that the mechanical properties of shell-based materials produced in different particle sizes weakened as the particle size increased (Kain et al. 2012). In addition to the particle size, which determines the density on mechanical properties, the glue ratio, gluing, and pressing conditions should not be ignored. Giannotas et al. (2022) measured the CSs of the composites produced from cement and Scots pine (Pinus sylvestris L.) and black pine (Pinus nigra L.) shells between 4.15 and 17.77 N/mm 2 , and they reported that the CS ratio decreased as the content of bark increased. In a study, the composites of different densities were produced using gypsum, cork (Quercus suber) and sawdust, and CSs were found between 2.27 and 6.58 N/mm 2 . It was reported that there was an improvement in CS values in parallel with the increase in density (Hernández-Olivares et al. 1999). IB and TS are the mechanical properties that are expected to meet the standards in thermal insulation materials. These properties give clues as to whether the components that make up the material are sufficiently bonded to each other, and thus whether the integrity of the material can be maintained for long periods of time during use. On the other hand, it is important that the material maintains its dimensional stability during storage, transportation, and in variable environmental conditions (temperature, humidity, wind, etc.) at the place of use. Otherwise, undesirable effects, such as surface blistering, fluctuations, and pulling back of fasteners, may occur.
According to Figures 15 and 16, the TS of the group A, B, D and E materials was higher than IB. The results are expected to be like this due to the hollow structure of the materials, the shape of the specimens in the IB and TS tests, and the test mechanisms. The IB and TS of the group B materials were higher than those of the group A materials. This result was given by the fibrous structure of the cones that made up the B group materials and thus a stronger bonding in gluing. The TS of the group A materials decreased as the particle size increased, whereas the IBs did not show a regular variation. While the TSs of the group B materials were expected to decrease in parallel with the increase in particle size, it was observed that there was no regular decrease. This result is thought to be caused by some deficiencies or errors in the production stages. The IBs of the B group materials were very close to each other. The TS of the D group materials was lower than that of the E group materials. It was found that as the perlite ratio in the material content increased, the TS values generally decreased and that the IB values were independent of these ratios. Taghiyari et al. (2020) reported that the use of wollastonite caused a 37-70% decrease in the IB value of the panels, as well as reductions in the MOR and MOE values. Similarly, in this study, a decrease was observed in mechanical values as the percentage of the perlite added to the material increased. Although this decrease was not regular in all materials, it was clear in the E group materials. The wollastonite used in previous studies (Taghiyari et al. 2020(Taghiyari et al. , 2021a(Taghiyari et al. , 2021b was in gel form and had micron-sized particles and a low material participation rate (Approximately 10%), and the panels produced were void-free and high density (670 kg/m 3 ). All of these affected the similarity rate of the results reached in this study. The IB values of the three-layer MDF with a density of 850 kg/m 3 produced using black spruce bark in the middle layer were measured between 0.37 and 0.58 N/mm 2 (Xing et al. 2007). Kain et al. (2015) determined the IBs of the composite materials produced from larch shells at densities of 500-450-400-350-300-250 kg/m 3 between 0.32 and 0.06 N/mm 2 . Particle boards were produced by using the bark of black spruce (Picea mariana (Mill.)) and trembling aspen (Populus tremuloides (Michx.)) in fine, medium and coarse grain sizes. It was determined that the mechanical properties weakened as the bark ratio increased, and that the materials with fine bark grain size showed higher IB (Yemele et al. 2008a(Yemele et al. , 2008b. In addition, the IB values of the particle boards produced by (Buyuksari et al. 2010) with pine (Pinus nigra Arnold var. pallasiana) and beech (Fagus orientalis Lipsky) wood and pine (Pinus pinea L.) cones were between 0.29 and 0.57 N/mm 2 , and it was stated that as the cone ratio of the boards increased, the IB values decreased. In some studies, the bending and breaking strengths of composite materials produced using Polycaprolactone (CAPA 6800) pine (Pinus pinea L.) cones were inversely proportional to the cone ratio (Jha et al. 2018), and the mechanical properties of composites produced from polypropylene and pine (Pinus pinea L.) cones improved with the increase in cone ratio (Arrakhiz et al. 2012). In this study,  the dimensional change results of the materials were found to be −0.156% (Δεl), −0.054% (Δεb) and 0.942% (Δεd). The maximum and minimum Δεl values were −0.007% and −0.393% in the E2 and A2 groups; the maximum and minimum Δεb values were 0.335% and −0.258% in the B3 and C1 groups, and the maximum and minimum Δεd values were 5.180% and −4.413% in the B2 and E2 groups, respectively. As can be seen in the findings, the dimensional changes of the materials occurred at reasonable rates, and the dimensional stability data in terms of width and length gave better results than thickness. As possible reasons for the differences, it can be suggested that the free-falling particles in the laying phase generally overlap in the horizontal position and provide better bonding, and that especially the pressure applied in the production of the middle layer is lower than that of the surface layers.

Conclusion
In this study, some physical and mechanical properties of biobased composite materials produced from tree bark and cones with added perlite and boric acid were investigated. The data obtained in this research is promising in terms of being remarkable for many people related to the subject, especially for scientists and sector representatives producing building thermal insulation materials, and in terms of natural wastes being the subject of long-term projects in parallel with national and/or international policies. In general, the following can be stated: . The fact that the materials in this study were 55% voit played an important role in reducing the weight and raw material consumption. . It was determined that the density tended to decrease with the increase in particle size in the bark-based group A materials and with the increase in the perlite ratio in the group D materials. . The particle size-MC relationship and the perlite ratio-MC relationship varied in the materials containing bark and cones. It can be thought that this is due to the structure of the bark and cone. . It was found that the mechanical properties of the conebased materials were better than the bark-based ones. It was found that the dimensional stability did not show a regular change in all groups and that each material was, however, more sensitive in the thickness direction compared to the width and length direction.
As a result, by changing the bark and cone form, they can be converted into composite materials with increased thermal insulation and mechanical properties. In further studies, the materials to be produced will be able to show better insulation and mechanical properties by thinning honeycomb walls and surface layers in the material design.