Effects of Nano-Clay on Physical and Mechanical Properties of Medium-Density Fiberboards Made from Wood and Chicken-Feather Fibers and Two Types of Resins

Medium-density fi berboards (MDF) were produced, using two different resins of urea-formaldehyde (UF) and phenol-formaldehyde (PF) at 10 % and 8 % contents, respectively. In order to fi nd new source of raw material to satisfy the increasing need of composite manufacturing industry, 10 % of chicken feather was added to the furnish. Moreover, nano-clay was added to investigate its potential improving effects on MDF panel properties. Results showed that panels with PF resin demonstrated signifi cantly lower water absorption and thickness swelling values in comparison to the panels with UF resin. However, mechanical properties of panels containing UF resin were generally higher; this was partially attributed to the higher resin content. It was concluded that panels with PF resin are recommended for applications where panels may be more exposed to water and vapor. In cases where the mechanical properties are of prime importance, panels with UF resin are more preferable. NC can be recommended in panels containing UF-resin to improve the properties, while it is not recommended in panels with PF-resin. Moreover, chicken feather can be included in MDF furnish to provide part of raw materials, though its addition can have diminishing effect on the properties to some extent.

Ključne riječi: vlakna od perja, mineralni materijali, nanokompoziti, nanovlakna gline, drvna vlakna Wood is a renewable material for many applications that helped mankind keep a sustainable development over centuries of civilization (Daly-Hassan et al., 2014; Fernandez et al., 2014 and 2017; Arce and Moya, 2015).Fast-growing species provide an opportunity to provide the raw materials necessary for wood industries and composite products (Behling et al., 2011;Gbetoho et al., 2017); however, the wood produced of these trees is usually of low quality and density and it is usually more susceptible to wood deteriorating bioagents (Schmidt, 2006;de Medeiros et al., 2016;Schmidt et al., 2016;Ayata et al. 2017).Therefore, many modifi cation techniques were used to improve its properties (Hill, 2006).However, it is mostly used in composite panels, engineered wood, and paper manufacturing industries (Candan and (Malanit et al. 2005;Valenzuela et al., 2012).However, for a steady production of composite panels, a continuous fl ow of raw materials is essential.Iran is short of woody materials and therefore, many other substitutes have been considered (Hosseinkhani, 2015).
Feather fi bers were reported to be used in some materials including concrete (Hamoush & El-Hawary, 1994;Koch, 2006;Acda, 2010).In an offi cial report by the Ministry of Jihad-e-Agriculture of Iran, the chicken feather produced in 2012 was about 80,000 metric tons in Iran.This amount of chicken feather would be enough to manufacture more than 20 million composite panels with standard size of 366 × 183 × 16 mm with 5 % feather content.Other countries also produce immense amount of chicken feather on a monthly basis.Chicken feathers from poultry farms used to be a valuable fi lling for blankets and matrices.However, synthetic materials substituted this natural substance and now chicken feathers are considered waste materials in Iran.Application of a percentage of chicken feathers in composite panels were studied before and the results were promising (Winandy et al., 2003 and2007;Taghiyari et al., 2014ab;Taghiyari and Sarvari, 2016).However, the cited authors (Winandy et al. 2003 and2007) only used phenol-formaldehyde (PF) resin in their study.Though PF is very practical in many applications, it is not popular in composite-manufacturing countries in Iran and some South-East Asian countries like Malaysia.Therefore, urea-formaldehyde (UF) resin was used here along with PF for comparison purposes.In addition, previous studies used quill-free feathers, that is, quills were removed from feathers.The process of removing quills is costly, adding to the fi nal cost of the composite panels produced.Therefore, in the present study, the whole feather was used in the production of composite panels.
Moreover, nanomaterials have been successful in improving properties in many materials (Majidi, Suganya et al., 2017).In wood science and technology, different metal and mineral nanomaterials were used to decrease hot-pressing in wood-composite panels, to improve biological resistance of solid wood and composites, and to ameliorate the negative effects of thermal modifi cation.In the present research project, nano-clay was also added as an additive to the resins to investigate its effects.

Specimen procurement 2.1. Nabava uzoraka
Composite panels were manufactured from a mixture of fi ve species of poplar, alder, hornbeam, maple and beech.The fi nal thickness of panels was 16 mm with a density of 0.67 g/cm 3 .Two types of resins were separately used for panels, namely urea-formaldehyde (UF) and phenol-formaldehyde (PF) resins.Specifi cations of the resins are given in Table 1.
Urea-formaldehyde (UF) and phenol-formaldehyde (PF) resins were procured from Dehghani Co. and Iran Composite Co. in Iran, respectively.The temperature and duration of hot press was 175 °C and 8 minutes, and 200 °C and 6 minutes for UF and PF resins, respectively.Once produced, panels were kept in conditioning chamber (25 °C, and 40±3 % relative humidity) for four weeks before being cut, ready for tests.The moisture content of the specimens were measured at the time of testing to be 7.4 % in all treatments, because wood has a thermo-hygromechanical behavior and its properties depend on the combined action of temperature, relative humidity, and mechanical load variations (Figueroa et al., 2012).Five replicate boards were produced for each treatment.From each board, two specimens were cut for each kind of test.
Chicken feathers were bought from a chicken farm located in Ghazvin city, Ghazvin Province.Only the feathers of the chicken body were used because the quills of the wing feathers were thick and infl exible.Moreover, the process of removing the quills of wing feathers is not commercially competitive for composite manufacturing factories.The chemical combinations of chicken feathers are given in Table 2.As can be observed, they are mostly comprised of serine (C 3 H 7 NO 3 ).

Nano-clay application 2.2. Nanošenje nanogline
Nano-clay (NC) powder was produced in cooperation with Mehrabadi Manufacturing Company in Tehran, Iran.The size range of nanoparticles was 30-110 nm.NC composition is given in Table 3. NC was mixed with the resins and sprayed on to the wood fi bers in a rotary drum-mixer before the hot press.A magnetic mixer stirred the resin-nanoclay mixture for 25-30 minutes for each production batch before being sprayed on the furnish.Consumption level of NC powder was 10 % based on the dry weight of the composite mat.Physical and mechanical tests, as well as number and location of the specimens, were carried out in accordance with the Iranian National Standard ISIRI 9044 PB Type P2 (compatible with ASTM D1037-99) specifi cations, using INSTRON 4486 test machine, with fi ve kN capacity.Figures 1 and 2 show thickness swelling and MOR specimens, ready for measurement.Statistical analysis was conducted using SAS software program, version 9.2 (2010).One-way ANO-VA was performed to discern signifi cant difference at the 95 % level of confi dence.Grouping was made between treatments, using the Duncan's multiple range test.Hierarchical cluster analysis, including dendrogram and using Ward methods with squared Euclidean distance intervals, was carried out by SPSS/18 (2010).Cluster analysis was performed to fi nd similarities and dissimilarities between treatments based on more than one property simultaneously.The scaled indicator in each cluster analysis shows how much treatments are similar or different; lower scale numbers show more similarities, while higher ones show dissimilarities (Taghiyari et al., 2014ab).Surface and contour plots were made by Minitab software, version 16.2.2(2010).

REZULTATI I RASPRAVA
Results showed that MOR values of panels produced with UF-resin were higher when the composite furnish was comprised only of wood fi bers (Figure 3).This was attributed to the higher UF-resin content (10 %) in comparison to PF-resin content (8 %).Addition of chicken feathers resulted in a decrease in MOR values in all treatments, though the decreases were not statistically signifi cant in some cases.The decrease was partially attributed to the incompatibility of resins with the chemical components of feather.Similar chemical incompatibility between UF-resin and chicken feather components were reported to decrease biological resistance of composite panels (Taghiyari et al., 2014ab), and to increase liquid permeation (Taghiyari & Sarvari, 2016).Addition of NC to the composite furnish reacted differently in each treatment.In panels with 100 % wood fi bers and UF resin, NC improved MOR values by 10.7 %.This was because NC acted as a fi ller in UF resin, improving the overall strength of the composite matrix.Similar improvement was reported by addition of other minerals.However, in panels produced with PF-resin, resin content was lower (8 %) and part of the resin was absorbed by NC particles, avoiding this portion to be actively involved in the process of sticking the fi bers together; ultimately MOR values decreased, though the amount of decrease was not signifi cant.No signifi cant difference was observed by addition of NC to panels with 10 % feather-content.It was concluded that NC can act as a fi ller and improve MOR values only in panels produced with UFresin.
Modulus of elasticity (MOE) values illustrated somehow different trends in comparison to MOR values.Addition of chicken feather to the furnish generally improved MOE values in the produced MDF panels (Figure 4).This was a result of higher fl exibility of feather fi bers in comparison to wood fi bers.Addition of NC to the furnish signifi cantly decreased MOE values only in panels with PF-resin and feather content, similar to the decrease that occurred in MOR value of this treatment.
Results of water absorption (WA) measurements showed signifi cant difference between panels produced with UF-resin and those produced with PF-resin (Figure 5).Panels with PF-resin generally showed lower WA values.This was due to the BWP (boil and water proof) nature of this resin.To be specifi c, lower WA values were achieved in panels produced with PF-resin although PF-resin content (8 %) was 2 % lower than UF-resin content (10 %), and this was because of water resistance nature of PF-resin.Addition of feather did not have statistically signifi cant effect on WA values.Thickness swelling (TS) values generally followed the same trend as WA values, though the improving effects of PF-resin were more conspicuous on TS values in comparison to WA values (Figure 6).Addition of NC generally did not affect TS values, though there were some fl uctuations.It was concluded that in MDF composite panel production, when water resistance is of prime importance, PF-resin is more preferable than UF-resin, though lower PF-resin provided an edge of commercial competition over UF-resin.
Results of the surface analysis showed clear inverse relationship between MOR values versus both WA and TS values in panels produced with PF-resin (Figure 7).In panels produced with UF-resin, no clear relationship was found between MOR versus WA and TS.The inverse relationship indicated that, in panels containing PF-resin, physical stability of MDF panels can easily be anticipated with regard to their mechanical property of MOR values.However, in panels produced with UF-resin, the interactions between different production factors of NC-content and feather-content were so high, varying from treatment to treatment, that ultimately it made no clear anticipation possible.Contour plots clearly demonstrated inverse relationship between MOE versus WA and TS values in panels produced with PF-resin (Figure 8A).Contour plot between MOR versus WA and TS demonstrated direct relationship between MOR versus WA values, but inverse relationship versus TS (Figure 8B).The in-verse relationship indicated that both MOR and TS values were in close agreement with the overall strength of the matrix; that is, the higher strength and integrity among wood fi bers can be translated into higher MOR values, and consequently lower TS values.However, WA values cannot necessarily be related to the strength of material, but it is closely related to other factors such as porous structure/system of the composite.
Cluster analysis based on the properties studied in the present project demonstrated remote clustering of panels produced with feather and PF-resin containing NC, that is, treatment WF-P-NC (Figure 9).Therefore, based on the properties discussed above and the cluster analysis, it was concluded that addition of feather and NC to panels produced with PF resin would not be recommended due to the unsatisfactory physical and mechanical properties.Close clustering of the two panel treatments produced with PF-resin and 100 % wood fi bers with or without NC content (W-P and W-P-NC treatments) indicated that in these panels, the improvement by NC was so low that extra expenses for addition of NC would not be commercially recommendable.The same decision was true about panels with feather content and produced with UF resin (WF-U and WF-U-NC treatments).

ZAKLJUČAK
Wood composite industry is in constant need for raw materials to keep up a continuous production line.Therefore, in the present research project, 10 % of chicken feathers was added to medium-density fi berboard (MDF) furnish to investigate if chicken feathers can be used as part of the mat.Two different resins of urea-formaldehyde (UF) and phenol-formaldehyde (PF) were used at 10 % and 8 % contents, respectively.Nanoclay was mixed with the above mentioned resins and sprayed on the furnish to investigate its potential improving effects on physical and mechanical properties.Panels containing PF-resin demonstrated better stability towards water.However, mechanical properties of panels containing UF resin were generally higher; this was partially attributed to the higher UF-resin content (10 %).It was concluded that NC can be recommended only in panels with UF-resin to improve the properties, while in panels with PF-resin, the improvements are not so high to compensate for the accompanying expenses.Moreover, chicken feathers to be added in MDF panels as part of the furnish demonstrated promising results, though they reduced the properties to some extent.