NC State
BioResources
Khan, T. A., Gupta, A., Jamari, S. S., Jose, R., Nasir, M., and Kumar, A. (2013). "Synthesis and characterization of carbon fibers and their application in wood composites," BioRes. 8(3), 4171-4184.

Abstract

Carbon fibers were synthesized using a low-cost, economical method. Fresh rubber wood fibers (Hevea brasiliensis) were burned using a furnace in an inert condition at 350 to 450 oC for 2-4 hours, and after that the fibers were ground at 18000 rpm for 20 to 40 seconds. The effect of carbon fibers as a reinforcement agent on mechanical, physical, and morphological properties was investigated. In the composite preparation, carbon fiber dosages (0, 0.1, 0.25, and 0.5 wt.%) were used as variable factors, along with a urea formaldehyde content of 10%. The morphology of the specimens was characterized using X-ray diffraction (XRD), Thermogravimetric analysis (TGA), and Field Emission Scanning Electron Microscopy (FESEM). The mechanical tests indicated that when carbon fibers were added, the modulus of rupture (MOR) and internal bonding strength (IB) improved significantly. From the TGA graph it was observed that the thermal stability of the composites based on carbon fiber was higher than composites without it. The thermocouple readings showed that at a higher loading of carbon, the core temperature of the board increased faster than for the control board.


Download PDF

Full Article

Synthesis and Characterization of Carbon Fibers and their Application in Wood Composites

Tanveer Ahmed Khan,a Arun Gupta,a,* Saidatul Shima Jamari,a Rajan Jose,Mohammed Nasir,a and Anuj Kumar a

Carbon fibers were synthesized using a low-cost, economical method. Fresh rubber wood fibers (Hevea brasiliensis) were burned using a furnace in an inert condition at 350 to 450 oC for 2-4 hours, and after that the fibers were ground at 18000 rpm for 20 to 40 seconds. The effect of carbon fibers as a reinforcement agent on mechanical, physical, and morphological properties was investigated. In the composite preparation, carbon fiber dosages (0, 0.1, 0.25, and 0.5 wt.%) were used as variable factors, along with a urea formaldehyde content of 10%. The morphology of the specimens was characterized using X-ray diffraction (XRD), Thermogravimetric analysis (TGA), and Field Emission Scanning Electron Microscopy (FESEM). The mechanical tests indicated that when carbon fibers were added, the modulus of rupture (MOR) and internal bonding strength (IB) improved significantly. From the TGA graph it was observed that the thermal stability of the composites based on carbon fiber was higher than composites without it. The thermocouple readings showed that at a higher loading of carbon, the core temperature of the board increased faster than for the control board.

Keywords: Urea-formaldehyde; MDF; Wood composite; Carbon fiber

Contact information: a: Faculty of Chemical and Natural Resource Engineering, University Malaysia Pahang; b: Faculty of Industrial Science and Technology, University Malaysia Pahang, 26300, Kuantan, Pahang, Malaysia; *E-mail: arun@ump.edu.my

INTRODUCTION

Medium-density fiber board (MDF) is an engineered wood product composed of fine ligno-cellulosic fibers combined with a synthetic resin and subjected to heat and pressure to form panels (Irle and Barbu 2010). It is made up of wood fibers and can be used as a building material similar in application to plywood. It is stronger and much denser than normal particleboard.

The low thermal conductivity of wood fiber limits the production rates in existing production lines. The carbon fibers can be used as a filler to improve the thermal and mechanical properties of wood composite. Carbon fibers are widely employed in various fields because of their unique properties including adsorptive capacity, chemical stability, thermal conductivity, and electrical conductivity (Youssef et al. 2008; Yun et al. 2008; Kim et al. 2008; Roh et al. 2008).

Principles of composites demonstrate that for a given reinforcement and matrix, the properties of resultant composites are mainly dependent on interfacial adhesion, because a good interfacial adhesion guarantees effective transition of stress, and thus the reinforcement and matrix can take full action. However, a large number of studies have shown that the surfaces of carbon fibers exhibit inertness, and the interfacial adhesion between carbon fibers and organic resin matrix is generally very weak (Meng et al. 2009; Xu et al. 2007; Lu et al. 2007). Hence, how to improve the interfacial adhesion between carbon fibers and resin matrix has been one of the most important topics of developing advanced composites.

Many theories or models have been proposed to explain or forecast the effect of surface modification of carbon fibers on the interfacial adhesion of resultant composites. To date, it is generally believed by many researchers that the chemical interaction between carbon fibers and polymeric matrix is necessary to improve the interfacial adhesion; however, some scholars have recently stated that good interfacial adhesion between carbon fibers and matrix can be also obtained by the formation of physical interaction. For example, Lu et al. (2007) employed air plasma to modify carbon fibers, and they concluded that mechanical interaction has a dominant effect on the interfacial adhesion of composites. The application of nano-based materials opens new aspects in the field of wood science, such as their use in solid wood, wood-based panels, etc. (Cai et al. 2007a, 2007b, 2008, and 2010). Mixing of carbon fibers has improved the thermal and mechanical properties of the wood composite boards.

This work was aimed at analyzing the influence of using carbon fibers as a reinforcing agent on the physico-mechanical properties of wood composites. Experiments were conducted to estimate the curing time at different weight concentrations of carbon fibers when mixed with resin and wood fibers. In the present work, the surface morphology and carbon fibers distribution in the composites have also been studied. The novelty of this work is the use of carbon fibers in wood composite for the first time.

MATERIALS AND METHODS

Materials

Wood Fibers

The lignocellulosic material used for this study was fresh rubber wood (Hevea brasiliensis) fibers obtained from Robin Resources Pvt. Ltd.

Urea formaldehyde

Urea-formaldehyde (UF) liquid resin used for this study was obtained from Dynea Malaysia Sdn. Bhd. The viscosity of the UF resin at 300 oC was 178 centipoises, pH 8.79, density 1.286 kg/m3, and the gel time at 100 oC was 36 s.

Methods

Preparation of carbon fibers

The fresh rubber wood fibers (Hevea brasiliensis) were burned using a furnace in an inert condition at 350 to 450 oC for 2 to 4 h. After that the fibers were ground with a Retsch Chemical Grinder ZM 200 at 18000 rpm for 20 to 40 seconds. The purity of the carbon fibers was found to be 74.09% using an elemental analysis system (CHNS analyzer).

Mixing of carbon fiber with UF resin

A bench-top overhead mixer IKA®-WERKE, model RW 20 DZM, with a speed range of 72 to 2000 rpm was used for pre-mixing carbon fiber with the UF resin at 1800 rpm for 30 min. The weight fraction of carbon fiber and UF resin were based on the weight fraction of the oven dry wood fibers. After mixing the carbon fibers and UF resin in a bench-top mixer, the resin mixture samples were subjected to ultrasonic treatment with NANO-LAB Ultrasonic probe dispersion QS1 system for 1 h for uniform mixing of nanoparticles in UF resin. This procedure was used to prepare three samples with carbon fiber concentrations of 0.1%, 0.25%, and 0.5% by weight in the UF resin.

Preparation of MDF panels

The standard laboratory method was followed to manufacture 300 mm x 300 mm x 15 mm MDF boards of 800 kg/m3. Rubber wood fiber in an amount of 750 g at 12% moisture content was used to prepare MDF. A rotary drum blender was used to blend the wood fibers, carbon fibers, and UF resin uniformly. The drum consists of steel dowels arranged in a zigzag pattern, which were intended to facilitate tumbling and mixing with a rotational speed of 18 rpm. The carbon fiber of various concentrations (0%, 0.1%, 0.25%, and 0.5%) mixed into 10% resin (all percentages are calculated from weight of wood fiber) was sprayed with a spray gun on the wood fibers. A puff mat of 300 x 300 mm was prepared from the resin added wood fiber, which was then pre-pressed at 1.5 MPa pressure. Finally, the pre-pressed panel was hot-pressed at 180 oC for 450 s with a target thickness of 15 mm; Fig. 1 shows the images of MDF samples. Except for the carbon fiber concentration, all the experimental parameters such as fiber moisture, resin loading, pressing time, and platen temperature were maintained the same in all experiments.

Fig. 1. Images of MDF samples (a) the 300mm x 300mm x 15mm MDF board, (b) for internal bonding, (c) for modulus of rupture

The MDF board was fabricated into two different sets as presented in Fig. 2. The first set of boards were used to find the temperature profile inside the core of mat during hot pressing and second set of boards were used to find the physical and mechanical properties. For the core temperature analysis, three replicates of each treatment, i.e. a total of 12 boards, were prepared. For the board performance test, four replicates of each treatment, i.e. 16 boards, were prepared. Ten samples were prepared from each treatment to analyze MOR, IB, and TS.

Fig. 2. Brief detail of the experiments performed

Measurements

Measurement of core temperature

Each wood fiber mat was pre-pressed to half-thickness of the loose mat. Two K-type thermocouple wires were inserted at the core of the mat from two opposite sides. The two thermocouple wires were 50 mm apart from each other. The minor deviations in the values from the two thermocouples were due to a slight difference in elevation of thermocouples location while hot pressing. The averages of the two values were taken as the mid-plane temperature of the mat. Three test specimens were prepared for each, by “board with 0.0%, 0.1%, and 0.5% weight concentration”.

Mechanical properties measurements

The hardness for both treated and untreated samples was measured. The Modulus of Rupture (MOR) and Internal Bonding (IB), which are commonly used to evaluate MDF, for both treated and untreated samples, were measured according to standard testing method ASTM D 1037. Mechanical testing of the samples was done on a Shimadzu UTM AG-X plug series, and results were analyzed on Trapezium X-software.

A total of 10 samples for MOR and 10 samples for IB were tested for the final result analysis. IB, a tensile strength, is tested perpendicular to the plane of the boards with a cross head speed 1 mm/min. Flexure testing was done by a three-point static bending test to determine the MOR with a cross head speed 10 mm/min. Equations (1) and (2) were used to calculate the MOR and IB, respectively.

MOR (N/mm2) = 3 × P × L/ 2 × b × d2 (1)

where P is the breaking load, L is the distance between knife edges on which the sample was supported, b is the average specimen breadth, and d is the average specimen depth.

 (2)

Dimensional stability tests

The thickness swelling (TS) tests were conducted in accordance with ASTM D 1037. Before testing, the weight and dimensions, i.e. length, width, and thickness of each specimen were measured. Conditioned samples of each type were soaked in water at room temperature for 24 h. Samples were removed from the water, patted dry, and then measured again. Each value obtained represented the average of sample.

The values of the thickness swelling (TS) in percentages were calculated using the following equation,

TS (%) = (T2T1)*100/T1 (3)

where T1 is the initial thickness of sample and T2 is the thickness of wetted sample.

Morphological study

The extents of intercalation and exfoliation of carbon fibers inclusions in the matrix system were monitored by X-ray diffraction (XRD) and Thermogravimetric Analysis (TGA).

The TGA method measures the amount and rate of change in the weight of a material as a function of temperature or time in a controlled atmosphere. Measurements are used primarily to determine the composition of materials and to predict their thermal stability at temperatures up to 1000 °C. The technique can characterize materials that exhibit weight loss or gain due to decomposition, oxidation, or dehydration. The standard practice for calibration of temperature scale for thermogravimetry follows the ASTM 1582 method.

Thermal stability was investigated using non-isothermal thermogravimetry (TG, DTA) using a TA Instrument. Samples (6 + 0.2 mg) were placed in alumina crucibles. An empty alumina crucible was used as a reference. The samples were heated from 30 to 600 °C in a 20 cm3/min flow of nitrogen with a heating rate of 10 °C/min.

X-ray Diffraction (XRD) measurements of boards containing carbon fibers and without carbon fibers were studied. The X-ray diffraction (XRD) was performed in a XRD analyzer. The samples were scanned over the 2θ range of 3 to 80° at a rate of 1 deg/min. The generator was operated at Cu/30 kV/15 mA. The inter layer spacing (d002) of carbon fiber was calculated in accordance with the Bragg equation: 2d sinθ λ.

The morphological structure of the composites was investigated by JEOL JSM-7500F Field Emission Scanning Electron Microscopy (FESEM), which provides narrower probing beams at low as well as high electron energy, resulting in both improved spatial resolution and minimized sample charging and damage.

RESULTS AND DISCUSSION

Core Temperature Progression

Figure 3 depicts the progression of core temperatures for the boards with carbon fibers of weight concentrations of 0.1, 0.25, and 0.5% in comparison to the control board (CB). The rapid rise in the core temperature can be attributed to the steep vapor pressure gradient (Gupta 2007; Bolton et al. 1989a) that developed during the pressing period of 15 s to 70 s. The 100 °C temperature reached the core in 72 s in 0.5 wt.% of carbon fiber compared to the control where the temperature reached the core in 84 s, indicating that conductivity can be increased by using carbon fibers. In pressing time phase from 112 s to 172 s, a constant temperature in the central plane was observed; this could be ascribed to phase change occurring in the board. The vapor formed was observed to exit from the edges of the board due to higher vapour pressure formed at the core. From the time period of 172 s, a gradual rise in the central plane was observed, which is mainly due to heat conduction in the board. It could be observed that the conductivity of boards prepared with carbon fibers was increased.

Fig. 3. Core temperature profile of mat during hot-pressing of MDF panels with different weight fraction % of carbon fibers

Enhanced Properties of the MDF

Internal Bonding (IB)

Figure 4 depicts the average internal bonding strength of composites made with various weight percentages of carbon fibers and 10% urea formaldehyde loadings. The bonding strength results show that the composites containing 0.1 wt% of carbon exhibited more internal bonding strength compared to samples without it. This could be possibly due to better adhesion between matrix (carbon fibers and urea formaldehyde) and wood fibers, which cause tension concentration in medium density fiberboards. It is seen from Fig. 4 that unlike 0.1 wt% of carbon fibers, internal bonding strength decreased with addition of carbon fibers, as can be seen in 0.25 wt% and 0.5 wt% of carbon fibers. Reduction of bonding strength with increasing carbon fibers can be related to the increase of the probability for agglomeration that creates regions of stress concentrations that require less energy to elongate the crack propagation (Eitan et al. 2003). This is consistent with the results reported by most authors studying the subject (Ashori et al. 2012; Tavasoli Farsheh et al. 2011; Ziaei Tabari et al. 2011). In general, the results for internal bonding strength test showed that medium density fiberboards, which contain 0.1 wt%, had the highest bonding strength value.

Fig. 4. Internal bonding results of MDF based on different ratios of carbon fibers and control board (without carbon fiber)

Modulus of Rupture (MOR)

The average values of MOR properties that were calculated and compared with the control specimen are shown in Fig. 5.