The Effect of Nano TiO 2 and Nano Boron Nitride on Mechanical , Morphological and Thermal Properties of WF / PP Composites

This study evaluated the effect of nano boron nitride (BN) and nano titanium dioxide (TiO2 ) on some physical, mechanical and thermal properties of WF/PP composites. Polyproplene as a polymer matrix and wood fl our obtained from particleboards were used as reinforcing fi llers to prepare the composites by using a single screw extruder. It was observed that density in all composites did not change signifi cantly with the increasing of wood fl our. It was found that BMOR and BMOE of the composites increased with the increasing of the wood fl our content and nanoparticle types, while the TMOR and TMOE decreased. According to the results of thermal properties (TGA), thermal degradation of all composites was found to be lower compared with pure PP.

Boron nitride (BN) has been widely used in the thermal management industry for years.BN is a good lubricant and abrasive, and it has a high thermal conductivity, high electrical resistance, and high temperature resistance.The familiar structures of BN are the hexagonal (hBN) and cubic (cBN) crystal structures.The structure of hBN is more stable than that of cBN (Meneghetti et al., 2008).BN is a low atomic numbered nonmetallic compound; its melting temperature (∼3000 °C) is too high to be used for thermal insulation.BN/polymer composites can decrease thermal expansion and increase thermal conductivity, while enhancing the electrical insulation properties (Zhou et al., 2007).Also, the addition of a small amount of BN in polymers might enhance their extrusion by increasing their fl owablity (Jung et al., 2010).Nano TiO 2 particle is one of the promising inorganic nano fi llers used in polymer matrix composites to enhance the mechanical properties (Nayak et al., 2016; Bardak et al., 2016).Among the investigated inorganic nano fi llers, TiO 2 nanopowder is being increasingly investigated because it is non-toxic, chemically inert, low cost, corrosion resistant and has a high refractive index, UV fi ltration capacity and high hardness (Mirabedini et al., 2008).
The goal of this research was to investigate the usability of wood fl our obtained from particleboards in wood plastic composites.Furthermore, the effects of nano TiO 2 and nano BN on the physical, mechanical and thermal properties of WF/PP composites have determined.

Materijali
Wood fl ours (WF) were supplied by Kastamonu Entegre Ağaç Sanayi A.Ş. Wood fl ours consist of 50 % Gymnospermae (Pinus nigra Arn. and Pinus sylvestris L.) wood and 50 % Angiospermae wood (Fagus orientalis L. and Populus alba L.).Oven-dry densities of WFs used in the study were 0.54 g•cm -3 , 0.48 g•cm -3 , 0.63 g•cm -3 and 0.40 g•cm -3 , respectively.All WF samples were of the same size made using 0.5 mm sieve.Polypropylene (EH241) was supplied by PETKIM Inc, in Turkey.The properties of the PP -EH241 are listed in Table 1.

INTRODUCTION
1. UVOD Wood-plastic composites (WPCs) are principally concerned with thermoplastic polymers reinforced by wood and wood derivatives such as fi bers or fl our (Soccalingame et al., 2016).Wood fl our (WF) actually represented a waste material, which had to be eliminated from sawmills in the past.Many waste utilization strategies have been introduced over the last century, such as bedding, composting, combustion, gas generation, use as feedstock for chemical industry, etc.Among them, the use of WF as raw material for making new solids is the most positive due to its convenience in application and low energy costs (Ashori, 2008;Okamoto, 2003).
In general, wood fl our is used as plastic fi ller, which tends to increase the stiffness of the composite but does not improve its strength.Natural fi bers can be used to reinforce plastics rather than fi ller rate, which increases strength as well as stiffness.Natural fi bers can be used to reinforce fi lled plastics, by increasing both strength and stiffness.Wood and other lignocellulosic fi bers typically have higher particle sizes than those of wood fl our (Osswald and Menges, 1995).WPCs combine the best properties of the pure components and can show superior performance in many application areas.Compared with potential traditional competitors, WPCs offer better thermal and acoustic isolation than aluminum, better durability and lower maintenance than wood, and often lower price than pure plastics (Garcia et al., 2009).
Lignocellulosic materials have made signifi cant contributions to the thermoplastic industry, which has led to the emergence of wood-plastic composites in the construction industry.Products such as decking, fencing, siding, window framing, and roof tiles are being introduced into the market.The use of WPCs is also increasing in construction, transportation, industrial and consumer industries.Growing interest in renewable resources-based products is due to social and environmental concerns.Commercial thermoplastics such as polyethylene (PE), polypropylene (PP), polyvinyl chloride (PVC), polystyrene (PS), and styrene maleic anhydride (SMA) are commonly used in the manufacture of plastic/wood fi ber composites (Sobczak et al., 2013;Bledzki and Gassan, 1999;Rowell et al., 1997;Zor et al., 2016).The use of technical and standard plastics has approached the application of natural fibers thanks to their low prices and steadily rising performance (Wittig, 1994).The composites reinforced with wood have shown a great growth due to many advantages.Their processing is easy, economic, and ecological.They have relatively high strength and stiffness, low cost, low density, low CO 2 emission, and they are biodegradable and renewable (Deka and Maji, 2011).However, these polymers have low thermal conductivities.Therefore, some researchers have worked to improve the thermal conductivity and electrical insulation of polymers by adding different fi llers and fi re retardants.
Hexagonal boron nitride (hBN) was supplied by BORTEK -BOR Technologies, Inc. Boron nitride has the formula BN, so it consists of boron and nitrogen elements.The hexagonal formation is stable and softest among the BN polymorphs.Boron nitrides cannot be naturally obtained, so they are chemically synthesized by reacting boron trioxide or boric acid with ammonia or urea (Rudolph 2000, Robert andChaitanya 1990).The h-BN has a specifi c gravity of 2.27 g•cm -3 and a melting range of 2700-3000 °C.It consists of thin plates that have an average diameter of about 200 nm and a thickness of 80 nm (Ayrilmis et al., 2014).

Metode
WF was oven dried at 103±2 °C to obtain moisture content less than 1 %.PP was used as matrix polymer, while WF was used as fi llers.Loading ratios of WF were 10 and 20 % wt.Nano materials (nano TiO 2 and nano boron nitride) loading level varied from 0.5 to 1 % by compounding weight (1000 gr).A general survey of PP/WF compound after mechanical mixing is shown in Figure 1.The production formulations are given in Table 2.
The materials used in the compounding were fi rst mixed to achieve better dispersion by mechanical mixer for 15 minute.The obtained samples were extruded at 50 rpm by a single screw extruder.During the extrusion, the zone temperatures ranged from 170 to 180 °C, the melting pressure of the extruder varied between 5 and 10 bars depending on material blends, the screw speed was 50 rpm, and the material output was 1 kg•h -1 .When exiting the extrusion, the obtained compounds, which were melted, were cooled and solidifi ed directly in a water-cooling system, while being pulled with end drive conveyors.Then, the solidifi ed materials were pelletized through a pelletizer.The pellets obtained were injection molded to obtain the test samples.All samples were conditioned at 20 °C and 65 % relative humidity prior to tests.First, the weights of samples (m) were measured by 0.001 g precision scales.Sample volumes (V) were calculated using dimensions and densities (D) of samples determined according to D=m/V equation.
The samples were air dried at 70 °C until a constant weight was reached prior to the immersion in a water bath.The specimens were periodically taken out of the water, wiped with tissue paper to remove surface water, reweighed and measured again and immediately put back into the water.The water absorption of the composites was determined according to ASTM D 1037-93.
Bending strength (BMOR), fl exure modulus (BMOE), tensile strength (TMOR) and tensile modulus (TMOE) were determined according to ASTM D 790-03 Test Method 1 and ASTM D 638-03 Type I, respectively.These tests were conducted using a Zwick tester with a 10-kN load cell capacity.Test speed was used at a rate of 0.2 inch•min -1 for all tests.The izod impact strength (IIS) tests were conducted according to ASTM D 256-06.The notches were provided with a NotchVIS machine (Ceast trademark) and tests were performed with a Resil 50 B impact tester.The morphological properties of the samples were observed with a scanning electron microscope (SEM) (TESCAN MAIA3) with an accelerating voltage of 15 kV under nitrogen.The fracture parts of all samples were sputter-coated with gold using a Denton sputter coater for enhanced conductivity.The thermal stability of all the composites was investigated using a TGA and DSC (Perkin Elmer, TA Instruments, USA).When using a TGA, the samples were heated from 25 ºC to 600 ºC with a heating rate of 10 ºC/min and a nitrogen fl ow of 100 mL•min -1 .The samples weighing about 10 mg were used for the tests.Degradation temperatures at 10 %  weight loss (T 10% ) and 50 % weight loss (T 50% ), maximum degradation temperature in the derivative thermogravimetric peaks (DTG max ), and mass loss of the samples in the TGA curves were measured and compared with the results obtained.
One-way analysis of variance (ANOVA) was performed to identify signifi cant differences at the 99 % confi dence level.The Duncan test was used to determine the difference between groups.The important differences between formulations were shown with letters, such as A, B, C, and D.

Physical and mechanical properties 3.1. Fizikalna i mehanička svojstva
The highest density was determined to be 0.94 g•cm -3 in pure PP samples.This was followed, in order, by PP+W1B1, PP+W1 and PP+W1T1 samples.The lowest density was obtained in the W2T1 sample.Low density of wood fl our is the factor with the highest impact on decreasing densities.Densities of WFs used in the study varied between 0.40 g•cm -3 (Pinus nigra Arn.) and 0.63 g•cm -3 (Fagus orientalis L.).There is a signifi cant difference in variations (F ratio = 4.706; P value = 0.000 < 0.05) according to one-way analysis of the variance (ANOVA).Homogeneity within variations was determined by using Duncan test.Figure 2 and Table 3 show densities and homogeneity groups of WPC, respectively.
Figure 2 and Figure 3 show that the addition of wood fl our increased BMOR and BMOE values.The highest value of BMOR was found to be 51.60 MPa in the PP+W1B1 samples.PP+W2B2 samples showed the highest BMOE with 2207 MPa strength.Control samples (Pure PP) showed the lowest BMOR value with 41.80 MPa and BMOE with 1256 MPa, respectively.Supporting WPC with wood fl our resulted in the increase of 23.4 % in the bending strength (PP+W1B1), 75 % in the modulus of elasticity (PP+W2B1).Bouafi f et al., (2009) state that lignocellulosic materials in-creased the bending and tension strength as well as elasticity modules in bending and tension tests.BMOR and BMOE values are shown in Figure 3 and Figure 4, respectively.
The multi-way ANOVA analysis was conducted to fi nd the effects of nanoparticle type, nanoparticle rates and fi ller rate on bending strength of WPC and the obtained data are given in Table 4.
The effect of nanoparticle rate (0.5 % and 1 %), nanoparticle type (TiO 2 and BN) and fi ller rate (10 % and 20 %) on the BMOR was found to be signifi cant according to the result of Duncan test as seen in Table 5.The effect of interaction of nanoparticle type and fi ller rate on the bending strength was statistically signifi cant.Duncan test was applied to determine the differences between groups.The effects of nanoparticle rate, nanoparticle type and fi ller material rate on the bending strength are given in Table 5. Increasing in nanoparticle rate increased proportionally both BMOR and BMOE values (Table 5.1).BN increased the BMOR strength, while TiO 2 contributed to the development of BMOE strength.It can be said that BN is more effective in bending resistance when compared to TiO 2 (Table 5.2).Adding wood fl our to pure PP increased the BMOR and BMOE, but the fi ller rate was not important in BMOR test according to Table 5     Figure 5 and Figure 6 show TMOR and TMOE values of the samples, respectively.TMOR of pure PP was determined as 29.7 MPa.Decreasing was observed in TMOR values of PP+W1 and PP+W2 samples.These decreases were calculated to be 4.94 % and 12.2 %, respectively.All samples containing nano boron nitride showed increased TMOR values when compared to control samples.TMOE of pure PP increased in all of the samples.The highest increase in TMOE of PP+W2B1 and PP+W2B0.5 was determined to be 83 % and 75.8 %, respectively.
According to the multi-way ANOVA analysis, it was found that the fi ller rate (A), nanoparticle type (B) and nanoparticle rate (C) were statistically signifi cant at 0.05.Table 6 shows the results of multi-way ANO-VA analysis.
It was found that the interaction between fi ller rate (1) and nanoparticle rate (3) was not signifi cant.In-   As seen in Figure 7, the highest izod impact strength (IIS) was obtained in PP+W1and PP+W2 samples, 2.17 kJ•m -2 and 2.07 kJ•m -2 , respectively.The lowest IIS was determined in W1samples, which contained 0.5 % of nano boron nitride.Generally, it can be said that nanoparticles decreased the izod impact strength of WF/PP composites.

Toplinska svojstva
Thermal stability of the samples was investigated using TGA/DTG analysis.As sees in Fig 8, it can be said that no signifi cant difference occured in TGA analysis with the addition of nanoparticle types and fi ller.The fastest mass losses were observed in samples containing PP+W2T1 and PP+W2NB1.On the other hand, the PP+W1T1 curve was the nearest to the curve of pure PP.Onset temperature of the composites was also determined to decrease with adding the nanoparticle types and fi ller.Similarly, as the fi ller loading increased, the thermal stability of the composites slightly decreased, whereas the fi nal ash content monotonically increased (Kiziltas et al., 2011a, b).
DTG curves in Figure 9 showed maximum degradation at 453 °C for W2NB1%, and the peaks of the composites were found to be between 350 °C and 470 °C.As seen in Table 8, DTA curves indicated the two peaks of melting peak (T m ) and decomposition peak (T d ).It is well known that PP, which is extremely hyg-roscopic in nature, is consumed at 426 °C without formation of any char residue (Baeza and Freer, 2001).
As seen in Table 8, the maximum value of T m was found in the wood composites with 1%NB (167.5 °C), whereas the minimum value of T m was determined in the wood composites with 1%NB.T d values, the maximum and minimum values, were found to be 1%NB and 0.5%T, respectively.The summary of the thermogravimetric analysis is presented in Table 8.

ZAKLJUČAK
Adding wood fl our has no signifi cant effect on the density of WPC composites.The BMOR and BMOE of the composites were increased both with the addition of wood fl our and nanoparticle rate.The TMOR, TMOE and izod impact strength of the composites were negatively affected by the increase of the rate of wood fl our.It was found that thermal stability of the composites (TGA) decreased with both nanoparticle types.It was concluded that wood fl our obtained from particleboards in WPCs has a signifi cant effect on the material characterization (mechanical properties, thermal properties, etc.).Based on the fi ndings obtained from the present study, the use of W1T1 can further increase mechanical performance of all composites.

Table 5
(2)ect of nanoparticle rate (A), nanoparticle type (B) and fi ller rate (C) on BMOR according to Duncan test Tablica 5. Utjecaj količine nanočestica (A), vrste nanočestica (B) i količine punila (C) na čvrstoću na savijanje istraživanih kompozitnih materijala prema Duncanovu testu the rate of wood fl our decreased the tensile strength of WPC when compared to control samples.However, adding nanoparticles to WF/PP compound led to a signifi cant increase.It can be said that BN provided a better fi t with wood fl our than TiO 2 .The effects of nanoparticle rate (1), nanoparticle type(2)and fi ller rate (3) on the TMOR are according to the results of Duncan test.The related test results are shown in Table7.

Table 7
Effect of nanoparticle rate (A), nanoparticle type (B) and fi ller rate (C) on TMOR according to Duncan test