Growth of Carbon Nanotubes by Pyrolysis of Composite Film of Poly (Vinyl Alcohol) and Modified Fly Ash ()
1. Introduction
Carbon nanostructured materials were synthesized by pyrolysis of benzene at 1000˚C over Ni powder for 60 min [1]. Carbon nanocapsules with silicon carbide nanoparticles were reported by thermal decomposition of PVA at 500˚C in argon [2]. Catalytic graphitization was reported using PVA with an iron-containing catalyst for the formation of CNTs at 600˚C - 800˚C in nitrogen flow for 2 h. The catalyst particles were uniformly distributed on the surface and filled the hollow channel of CNTs. The CNT structures depended on the nature of the catalyst and carbon sources as well as the pyrolysis conditions [3].
Coal combustion in power stations generates huge amounts of FA as a by-product. The storage and proper disposal of this material are a significant global concern on both environmental and economic fronts [4]. FA is commonly processed by disposal as landfill in dams and lagoons. It typically consists of crystalline aluminosilicate, mullite and α-quartz (~96.5%) spherical particles between 1 μm ~ 100 μm in diameter, along with traces of metal oxides (~3.5%) (iron, calcium, magnesium, potassium, sodium, titanium), depending on the processing conndition and the chemical composition of the coal burned [5].
In 2007, less than 20% of 14.5 million tons produced in Australia was re-used [6]. The remaining FA was disposed in low land area. Successful results have been achieved in the use of FA of different researchers are in agriculture and soil management, the production of polymer composites [5] cement and concrete [7], metal composites [8] as catalytic support for hydrogen production, steam methane reforming [4] and iron-nitrate-impregnated support for the synthesis of multi-walled CNTs (MWCNTs) [6].
Research on the recycling of FA as a filler and catalyst support has been moving towards a green technology for value-added products and composite materials. We have recently reported the dual functions of fly ash as filler and catalyst for the synthesis of CNT materials [9]. We report here a significantly more cost-effective pyrolysis method than the expensive arc-discharge and CVD methods, using waste composite film of PVA and modified FA for growth and self-assembled CNT ropes, where FA has dual roles as filler in composite film and catalyst in pyrolysis condition.
2. Experimental Procedures
2.1. Materials
An industrial waste FA material was obtained from Swanbank Coal Fire Plant, Qld, Australia. The PVA (molecular weight, 125,000 g/mol; degree of hydrolysis ~89% with remaining 11% acetate) from S.D. Fine-Chemical Ltd., Chennai, India, and reagent grade sodium hydroxide from Laboratory Unilab Reagent, Sydney, were purchased and used as received.
2.2. Preparation of Composite Films
The detailed modification process of FA with sodium hydroxide at 85˚C for 8 h was reported in our earlier research. The PVA composite film was fabricated by a casting method from an aqueous suspension of PVA and modified FA. The PVA was dissolved in distilled water at 80˚C for 2 h to prepare the 1.2 wt% solution. The modified FA particles (5 wt%) were dispersed and sonicated for 5 min. The resulting mixture was cast in glass petri dishes (100 mm diameter) and bubbles were removed by shaking and air blowing. The petri dishes were kept at room temperature until completely dry. The films were peeled out and dried in an oven at 60˚C under vacuum for 6 h. The thickness of the films was 50 µm - 70 µm [5].
2.3. Pyrolysis Method
The pyrolysis of the samples was run at an isothermal condition in a horizontal quartz tube furnace (HTF 6035) under nitrogen flow. The hot zone of the furnace was preheated to 500˚C using thermocouple. The furnace connection scheme was as reported in our work [9]. The weighed amount of sample was kept in an alumina crucible and placed on a graphite feeder, which was inserted into the cold zone with simultaneous nitrogen purging at 2 L/min, held for 30 s, and passed through into the hot zone for 10 min. After pyrolysis, the sample was relocated from hot to cold zones via the graphite feeder for 30 s and removed from the alumina crucible at room temperature. The sample was stored in a small bottle, and used for calculation and analysis.
2.4. Analysis and Analytical Instruments
X-ray photoelectron spectroscopy (XPS) was conducted using an ESCALAB220i-XL (VG Scientific, Thermo Scientific, UK) to calculate the atomic composition of the CNTs. The standard conditions were used: >2 × 10–9 mbar, mono-chromated Al K alpha (energy 1486.6 eV), 200 W (20 mA and 10 kV), spot size ~1 mm in cross, photoelectron take-off angle of 90, pass energy of 100 eV for survey scans and 20 eV for region scans, and step-sizes of 1 eV for survey scans and 0.1 eV for region scans.
Raman spectroscopy was performed on an inVia Reinshaw Raman microscope using green laser excitation (514 nm) for the calculation of the degree of ordered CNT structures. Scan was recorded on an extended range (100 cm–1 - 4000 cm–1), with 10% power of 1.2 mW. The samples were prepared by pressing onto a glass slide.
A Hitachi 4500-II scanning electron microscope (SEM) was used to examine the morphology of the sample. The sample on carbon tape was coated with chromium in a chromium sputter unit to improve conductivity. Transmission electron microscopy (TEM) images were obtained using a Philips CM200 with 200 kV. The sample was dispersed in ethanol and a drop of diluted suspension was poured onto a carbon-coated copper grid, which was directly injected into a sample injection holder after airdrying for three days.
3. Results and Discussion
The pyrolysis of PVA composite film with 5 wt% modified FA left behind black solid residue on the alumina crucible. The yield was 53.8% depending on the FA contents in the composite films.
X-ray photoelectron spectroscopy (XPS) was conducted using an ESCALAB220i-XL to calculate the atomic carbon composition in the solid residue [10]. XPS spectrum of the material is shown in Figure 1. The peaks at 285.1 eV and 532.2 eV were generated by photoelectrons emitted from the carbon 1 s and oxygen 1 s core levels, respectively [10].
The broad peak was isolated by the Lorentzian method to divide and fit the curve by composed of five peaks [9]. The peak at 285.1 eV was generated by the excitation of C-C/C-H bonds in the graphite configuration with ~83.8%, including its shake-up structure. The peak was centred at 286.6 eV, which can be attributed to photoelectrons emitted from carbon atoms bonding with C-O-H group configuration with 6.8%. Three other small peaks were centred at 288.1, 289.2 and 290.7 eV, emitted from the carbon atoms belonging to C-O-C (4.1%), C=O (3.6%) and -COOH/-COOCH3 (1.6%) groups, respectively.
Therefore, it can be assumed that oxygen is a component of CNTs. The intensity of the oxygen 1 s core level
Figure 1. Wide-range XPS spectrum of CNT materials.
was much higher than those of the Si and Al 2 s core levels. The relative concentration of oxygen and carbon are nearly 18% and 74%, respectively; the remainder consisted of mixtures of Na, Si, Ca and Al in the CNT sample. However, the percentages of O, Si and Al were ~60.2%, 25% and 6%, respectively, including 4% unburned carbon in FA, which was of amorphous structure.
The graphitic configurations of CNT materials were analyzed by Raman spectroscopy for more reliable authentication of the carbon structure. The characteristic peaks were identified and shown in Figure 2 on excitation with 514 nm laser light.
The strongest Raman G band line, the characteristic peak for well-graphitised CNT, appeared at 1598 cm–1 with relatively narrower FWHM, 71 cm–1. Another peak at 1380 cm–1 with 112 cm–1 FWHM is the D-line band, indicating a typical sign for defective graphitic structures on CNT. The intensity of the D-line is lower than that of the G-line, evidence of the more ordered carbon structure in CNT. CNT materials consist mainly of crystalline graphitic carbons along with amorphous carbons. The intensity of isolated peaks was calculated and used to determine the characteristic features of the graphitic wall of CNT materials [6].
The intensity ratio (IG/ID) of the G band and D band was treated to determine the degree of carbon graphitization in CNT configurations. The higher IG/ID ratio indicated a greater degree of wall graphitization, suggesting superior CNT quality. The IG/ID ratio of the CNT materials grown on FA surfaces was 3.43, higher than the 1.17 for MWCNTs yielded for iron-catalyst-coated FA by the CVD method [6].
The line width and intensity ratio of (IG/ID) varied depending on the structure of the carbons and the conditions used for analysis, e.g., laser power and energy [11]. The well-known phenomenon, Raman band intensities and widths, increased with the decrease of wavelength of exciting laser, whereas these lines were shifted to lower wave number regions with an increase of the exciting laser power [12,13].
A Hitachi 4500-II scanning electron microscope (SEM) was used to examine the morphology of the sample. The morphology of the solid residue was visualized and analyzed by SEM and TEM. The morphologies of CNT ropes in SEM images are presented in Figures 3(a)-(b). Figure 3(a) clearly shows the active site where CNT ropes were initiated and grown randomly with curved and entangled structures on the FA surfaces.
Figure 3(b) shows several isolated, long, flexible CNT ropes captured with a 500 µm scale bar. The ropes were observed in highly entangled with tightly twisted structures. The self-assembly of two, three and complicated