Fabrication of a reusable magnetic multi-walled carbon nanotube–TiO₂ nanocomposite by electrostatic adsorption: enhanced photodegradation of malachite green

Abstract

A simple, quick and efficient method for the fabrication of a magnetic multi-walled carbon nanotube–TiO₂ (MMWCNT–TiO₂) nanocomposite through electrostatic attraction was proposed as a novel method. First of all, TiO₂ nanoparticles were synthesized by an atmospheric pressure chemical vapor synthesis (APCVS) method and characterized by scanning electron microscopy (SEM), X-ray diffraction (XRD) and UV-Vis diffuse reflectance spectroscopy (DRS). Next, the magnetic multi-walled carbon nanotube (MMWCNT) nanocomposite was prepared. Finally, TiO₂ nanoparticles were coated onto the surface of the MMWCNT through electrostatic attraction in ethanol solution. The morphology and structure of the final composite were investigated by SEM, XRD, Energy dispersive X-ray spectrometry (EDS) and Fourier transform infrared spectrophotometry (FT-IR). Both TiO₂ and MMWCNT–TiO₂ were characterized by BET surface area/pore volume analysis. The photocatalytic function of the TiO₂ and MMWCNT–TiO₂ composite was validated for malachite green (MG) degradation under irradiation with ultra violet (UV) light. MMWCNT–TiO₂, adsorbing MG, was easily separated from the aqueous solutions with the help of an external magnet; so, no filtration or centrifugation was necessary. The effects of pH, irradiation time, catalyst concentration, MG concentration, etc. on the photocatalytic activity were studied. The optimal conditions were an initial MG concentration of 20 mg L⁻¹ at pH 5.0 with a catalyst concentration of 0.2 g L⁻¹ under UV irradiation for 240 min with good recyclisation of the MMWCNT–TiO₂ catalyst.

---

Introduction

Several chemical, physical and biological processes are presently utilized for wastewater treatment such as flocculation, ultrafiltration, adsorption, ozonation and chlorination.^1,2 These procedures are not effective because they come out in solid waste, thus creating other environmental problems requiring further treatment. Therefore, it is necessary to find an effective method of wastewater treatment in order to remove hazardous dyes and organics from industry effluents.³ One of the methods of wastewater treatment containing dyes is their photocatalytic degradation in solutions illuminated with ultra violet (UV) irradiation, which contains a suitable photocatalyst. Among all the catalysts, titanium dioxide (TiO₂) is found to be the most efficient catalyst for the photocatalytic degradation of a pollutant due to the faster electron transfer of molecular oxygen.⁴ TiO₂ is considered as the most promising materials due to its relatively high activity, non-toxicity, chemical stability and low cost. However, the practical applications are still retarded by several limitations:^5,6 (1) wide-band gap, limiting the photoactivation only in the ultraviolet light region (3.0 and 3.2 eV for the rutile and anatase phases, respectively); (2) high carrier-recombination probability; (3) poor adsorbing affinity toward organic molecules and (4) is the separation of micro- or nanosized photocatalysts from wastewater, which is an often a difficult and energy-intensive process. To overcome this shortcoming of TiO₂, many researchers have tried to improve the photocatalytic activity of TiO₂ under visible light. There has been considerable advancement in the production of novel functional materials by coupling TiO₂ with other organic or inorganic materials.⁷ One of the most commonly employed methods for improving the photocatalytic activity is the combination of multi-wall carbon nanotubes (MWCNTs) with TiO₂.^8–11

MWCNTs have been extensively studied because they have many useful properties such as good electrical conductivity, nanosize absolute black, excellent mechanical properties, large surface area and high adsorption capacity.¹² In addition, MWCNTs have a large electricity-storage capacity and, therefore, it may accept photon-excited electrons in mixtures or composites with titania, thus retarding or hindering recombination.⁷ Hence, the combination of MWCNTs with TiO₂ can reduce charge recombination, enhance reactivity and photocatalytic ability of TiO₂.^13,14 On the other hand, a dispersion of TiO₂ on the MWCNTs surface could create many active sites for the photocatalytic degradation. It was suggested that the photo generated charge carriers could transfer from the TiO₂ to the CNTs and increase the photocatalytic activity of TiO₂ because the excited electron in a conduction band of TiO₂ might migrate into the CNTs and the recombination of electron–hole pairs decreased.^15,16 To overcome problems of separation, photocatalysts with easily controllable magnetic properties that exhibit valuable advantages in environmental and biomedical applications have been developed.¹⁷ A number of binary magnetic photocatalysts consisting of TiO₂-coated Fe₃O₄,^18,19 TiO₂–Fe₃O₄ hollow spheres,²⁰ mesoporous mixed γ-Fe₂O₃–TiO₂,²¹ and γ-Fe₂O₃–TiO₂ Janus hollow bowls²² have been developed to realize photocatalyst recovery by taking advantage of the magnetic properties of γ-Fe₂O₃ and Fe₃O₄. Thus, TiO₂ (the semiconductor photocatalyst) degrades organic contaminants; Fe₃O₄ provides magnetic properties for separation and recovery; and MWCNT provides an electron pathway to suppress charge recombination and enhance photocatalytic activity. These functions of TiO₂–Fe₃O₄–MWCNT result in enhanced photocatalytic activity and decreased photocatalyst loss.

For the TiO₂ preparation, various methods such as microemulsion,²³ solvothermal,²⁴ sol–gel,²⁵ precipitation,²⁶ combustion synthesis,²⁷ electrochemical synthesis,²⁸ chemical vapor synthesis (CVS),²⁹ inert gas condensation (IGC),³⁰ chemical vapor deposition (CVD),³¹ physical vapor deposition (PVD)³² etc. have been proposed. During the last decades, chemical vapor synthesis (CVS) has become more popular for developing high quality, ultra fine, unagglomerated, high purity and air-free nanocrystalline powders. The major limitation of the CVS process is an expensive vacuum apparatus.²⁹ The atmospheric pressure chemical vapor synthesis (APCVS) route is a new, economic and affordable method for synthesizing high purity, well-structured and uniform nanoparticles and is developed by Rahiminezhad et al.³³ In this procedure, some low-cost gasses and precursor will be brought out to a vertical or horizontal quartz tube at different temperatures, and chemical reactions take place at the atmospheric pressure. After synthesis process, nanoparticles will be collected on the surface of a cold trap by passing the exiting gas through a cold trap. As noted above, one of the significant advantages of this procedure is that APCVS operates at atmospheric pressure and thus expensive vacuum apparatus will not be required.

In the past decades, CNT–TiO₂ hybrids have been largely fabricated by the sol–gel method,³⁴ electro-spinning,^13,35 electrophoretic deposition,³⁶ and chemical vapor deposition (CVD).³⁷ Besides the sol–gel method, TiO₂ nanoparticles could be attached onto shortened CNTs by electrostatic attraction. To our knowledge, only one paper was reported about attaching CNTs–TiO₂ by electrostatic attraction in water.³⁸ This physical synthesis has some advantages in comparison with those syntheses of chemical methods. In sol–gel method,³⁴ various reagents, too much time aging and high temperature needed. For electro-spinning method,^13,35 specific instrument, high voltage, suitable viscosity solution, high temperature calcinations, many reagents and too much time processes needed. Electrophoretic deposition³⁶ needs electric field and too much time for synthesis. At last, chemical vapor deposition (CVD),³⁷ needs high temperature calcinations and it takes too much time. Our proposed method for fabrication of MMWCNT–TiO₂ composite offers a simple, fast, efficient, high purity, low time and cost, no consumption of toxic solvents, no specific instrument, ability of synthesis in each laboratory and it was done at room temperature. For investigation of photocatalytic activity of this nanocomposite, malachite green was chosen as a sample model.

Malachite Green (MG), a basic dye has been widely applied for dyeing of leather, silk and wool and in distilleries.³⁸ Its application extents in the aquaculture, commercial fish hatchery and animal farming as an antifungal therapeutic agent, while for human, it is used as antiseptic and fungicidal. Nonetheless, its oral consumption is carcinogenic. The available toxicological information, reveals that in the tissues of fish and mice MG easily reduces to persistable leuco-Malachite Green,^39,40 which behaves as a tumor promoter. Thus, the detection of MG in fishes, animal, milk and other foodstuff designed for human consumption are of great alarm for the human health. Studies also confirm that the products formed after the degradations of MG are also not safe and have been carcinogenic potential.⁴¹ Therefore, it becomes necessary to transfer such a toxic dye from wastewater before it discharged into the aquatic environment.

Our present investigation has been aimed at three different aspects: first, synthesis of TiO₂ nanoparticles by APCVS method, second, attaching of TiO₂ to MMWCNT composite by electrostatic attraction in ethanol and thirdly evaluating its performance towards the photocatalytic degradation of MG. A comparative photocatalytic account of MG degradation was made by MMWCNT–TiO₂ and TiO₂. The preparation, morphology and structure, light absorption, adsorptivity to MG, and photocatalytic activity of the MMWCNT–TiO₂ was investigated in detail.

Experimental

Chemicals

MWCNTs of 95% purity with outer diameter: 10–20 nm and inner diameter: 5–10 nm, length 10–30 μm, numbers of walls 3–15 were purchased from Plasma Chem. GmbH (Germany). MMWCNT was prepared as reported by our previously worked.⁴² Other reagents, including malachite green oxalate (MG oxalate), ammonium ferric sulfate, ammonium ferrous sulfate, titanium tetrachloride, hydrochloric acid, ammonia, were purchased from Merck. All reagents were of analytical grade and used without further purification. All aqueous solutions were prepared using the ultra-pure Milli-Q purification system.

Synthesis of TiO₂ nanoparticle

Synthesis of TiO₂ nanoparticles was carried by the APCVS method as reported by Rahiminezhad et al.³³ Fig. 1 shows a schematic illustration of the experimental apparatus. A hot-wall, atmospheric pressure, a horizontal quartz reactor with the inner diameter of 80 mm and a length of 800 mm was used. It was run at temperature 800 °C and atmospherically pressure. High pure argon (purity > 99.999%) was employed as a carrier gas and split into three sections. Oxygen (purity > 99.999%) was used as an oxidizing gas. Titanium tetrachloride (TiCl₄) (purity > 99.99%) was used as a precursor.

Fig. 1 A diagram of APCVS configuration. (1) Oxygen (2) argon (3) needle valve (4) flow meter (5) water (6) TiCl₄ (7) pressure indicator (8) furnace (9) quartz tube (10) cooling fan.

Before all runs, the internal surface of the quartz reactor was cleaned by laboratory alcohol (ethanol, 96 vol%). The quartz tube was placed inside the furnace and the furnace was brought to temperature 800 °C. Then, argon gas with a flow rate of 0.5 L min⁻¹ and a pressure of 1 atm was fluxed directly into furnace quartz tubes for 10 min to create pure argon and clean the reactor internal surfaces from any impurities. Afterwards that, oxygen gas with a flow rate of 0.5 L min⁻¹ and 1 atm pressure was introduced into a chamber quartz tube. Water vapor was introduced into the reactor by bubbling argon through a water bubbler without vacuum. The flow rate of water vapor was maintained at 0.5 L min⁻¹. The liquid precursor (TiCl₄) was evaporated in a vertical bubbler at oil bath at 800 °C. The precursor was introduced into a hot-wall tubular quartz reactor by bubbling argon through a precursor container without vacuum that was heated by an external resistance furnace. The effective heated region was 20 cm in the middle of quartz reactor. In the reactor, the precursor was oxidized to give TiO₂ monomers. The generated monomers by passing through the reactor underwent coalescence, coagulation, agglomeration and sintering, ending up as titania nanoparticles. The produced nanoparticles were collected in the cold trap which consisted of a chamber, inlet and outlet lines and was kept in an ice-water bathroom.

Synthesis of MMWCNT–TiO₂

MMWCNT (1.0 g) was dispersed in 20 mL ethanol in an ultrasonic bath for 1 hour. TiO₂ nanoparticle (0.2 g) also was dispersed in 10 mL ethanol separately (the ratio of MMWCNT–TiO₂ has been optimized before the final synthesis process). After about 1 h, the MMWCNT and TiO₂ suspensions were mixed ultrasonically for 30 min. Then, the mixture was stirred using a glassware stirrer (1200 rpm) for eight hours. The MMWCNT–TiO₂ composite was isolated from the mixture with the help of a permanent magnet. Separated MMWCNT were washed three times with deionized water followed by ethanol. Finally, MMWCNT was dried at 110 °C for one day. Once the MMWCNT and TiO₂ nanoparticle suspensions are mixed, TiO₂ nanoparticles instantaneously attach onto CNTs due to electrostatic attractive force between them. The suspension also shows a color change from black (for CNT suspension) and white (for TiO₂ suspension) to be dark gray. For attaching TiO₂ nanoparticles onto CNT surfaces the point of zero charges (pzc) should be considered. The pH_pzc of as-produced CNTs is in the range of pH = 4–6. Numerous research works demonstrate that the pH_pzc of acid-modified CNTs and oxidant-modified CNTs are lower than those of as-produced CNTs.⁴³ When pH is <4, the pzc of the original CNT suspension is positive. When pH is >3, the pzc is negative. The TiO₂ suspension has pH_pzc at 6.5. Under this pH, the TiO₂ nanoparticles are positively charged. Above this pH, the TiO₂ nanoparticles are negatively charged. Very desirably, the two suspensions have opposite pzc at pH < 6.5. If CNT and TiO₂ suspensions with opposite pzc are mixed under a suitable pH, TiO₂ nanoparticles and CNTs should attract each other. Attaching TiO₂ to MMWCNT was done between pH = 3–7. As a result, pH 5 is the desired condition for TiO₂ attachment onto CNTs.

Photocatalytic test

The photocatalytic activity of TiO₂–MMWCNT composite was tested in the MG degradation in aqueous solutions under UV light radiation. The experiments were carried out in an open wide glass photochemical reactor charged with a mechanical stirrer. A 250 W high-pressure Hg lamp was placed approximately 20 cm from the reactor, and the wavelength of the emission light was in the range of 200–400 nm (Fig. 2). In a typical experiment; 50 mL of aqueous MG solution with a concentration (C^′₀) of 50 mg L⁻¹ was mixed with the TiO₂–MMWCNT composite (0.2 g L⁻¹), and continually stirred in the dark for 50 min in order to establish the adsorption–desorption equilibrium. Prior to turning on the light, the first sample was taken out at the end of the dark adsorption period to confirm the MG concentration in the solution, which was considered as the initial concentration (C₀) after dark adsorption. The concentration of the solution was determined, and was taken as the initial concentration (C₀) of the MG solution. MMWCNT–TiO₂ was easily separated from the aqueous solutions with the help of a strong permanent magnet with 1 Tesla magnetic fields include the dimensions of the magnet employed (10 × 3 × 2 mm³); so, no filtration or centrifugation was necessary. Subsequently, the cleaning solutions were analyzed regularly by UV-vis spectroscopy (Perkin-Elmer Lambda 20) at scheduled irradiation times (λ_max: 617 nm). The photocatalytic degradation efficiency (η%) could be calculated as the following equation.where C₀ and C are the concentration of MG at t = 0 and t, respectively.

Fig. 2 A schematic illustration of MMWCNT–TiO₂ synthesis and photodegradation of MG.

Characterization

X-ray powder diffraction (XRD) measurements were performed (STADI-MP from a STOE company) with monochromatized Cu K_α radiation (λ = 0.15418 nm). The 2θ scanning angle range was 0–80θ with a step of 0.02°/0.2 s. The Fourier transform infrared spectrophotometer (FT-IR) (4000–400 cm⁻¹) were recorded using KBr pellets (Bruker Vector 22 FT-IR spectrophotometer) with 2 cm⁻¹ resolution. UV-vis diffuse reflectance spectra (DRS) were recorded (UV-3100 spectrophotometer made by Shimadzu Corporation) with spherical diffuse reflectance accessory, using BaSO₄ as a reference. The MMWCNTs were characterized by scanning electronic microscope (SEM) (Philips XL30, Eindhoven), with gold coating. Energy dispersive X-ray spectrometry (EDS) was performed by Oxford ED-2000. Transmission electron microscopic (TEM) image was recorded in a (ZEOL ZEM 2010 TEM) instrument with an accelerating voltage of 200 kV. Specific surface areas were measured by using a NovaWin2 analytical system made by Quantachrome Corporation and were calculated by the Brunauer–Emmett–Teller (BET) and Barrett–Joyner–Halenda (BJH) methods.

Results and discussion

Characterization results

XRD analysis. XRD was used to determine the phase composition of the TiO₂ nanoparticles. Fig. 3 shows the XRD patterns of TiO₂ nanoparticle. The pattern of the TiO₂ sample exhibits several characteristic peaks at 2θ = 25.2°, 37.7°, 48° and 54°, in agreement with anatase phase which was corresponding well with the JCPDS, no. 01-084-1286 data file. Average crystallite size of TiO₂ was estimated according to Scherrer's equation, d = λ/β cos θ, where d is the average crystallite size (nm), λ is the wavelength of the Cu Kα applied (λ = 0.154 nm), β is the peak width at half-maximum and k is the constant usually applied as 0.9. As the precursor temperature of 800 °C, the average nanoparticle size was 50 nm.

Fig. 3 XRD patterns of TiO₂ nanoparticle by APCVS method.

The XRD of the bonding TiO₂–MMWCNT composite is shown in Fig. 4 in order to characterize the crystalline structure of the samples. As can be seen, the peaks appeared at 2θ values of 25.32°, 48.07° and 53.95° could be indexed to the anatase TiO₂, which was corresponding well with the JCPDS, no. 01-084-1286 data file. It is worth noticing that the characterized peaks of CNT and anatase are both present in the bonding TiO₂–MWCNT composite simultaneously, and no other diffraction peaks can be observed. However, the main diffraction peaks of CNT (2θ = 25.4 in comparison with the XRD pattern of CNT reported in our previous work⁴²) are not observed clearly in this composite, this is because the reflection of CNT is overlapped by the reflection of anatase. The XRD results suggest the formation of TiO₂–MWCNT composite. In addition, the dominant peaks appeared at 2θ values of 30.17°, 35.45°, 43.25°, 56.78° and 62.73°, all of these diffraction peaks rest with magnetite of Fe₃O₄ nanoparticles (JCPDS, no. 00-003-0863).

Fig. 4 XRD patterns of MMWCNT–TiO₂ nanocomposite.

Microscopic analysis. To measure the TiO₂ nanoparticle's shape, size and distribution, SEM was used. The morphology, coagulation and agglomeration of the prepared titania nanoparticles were observed by SEM. Fig. 5a shows the SEM images of TiO₂ nanoparticles produced at 800 °C. As it can be seen, measurements are consistent with the results of XRD analysis and their distribution was relatively narrow, which indicates the uniformity and homogeneity of the product. The shape of the particles is generally spherical and quite similar to each other, and the size of the particles varies in the range of 49–70 nm. Fig. 5b shows the TEM image of TiO₂ nanoparticles. This image shows that titania nanoparticles are perfectly round.

Fig. 5 (a) SEM (b) TEM images of TiO₂ nanoparticle.

The morphology of MMWCNT–TiO₂ was investigated by SEM and TEM. As shown in Fig. 6a and b, the SEM and TEM images revealed the presence of two types of particles (TiO₂ and Fe₃O₄) with similar sizes, transparencies, and distributions across the CNT support. In the MMWCNT–TiO₂ photocatalyst, Fe₃O₄ NPs were magnetically aggregated in certain domains, while TiO₂ NPs with were spread homogeneously on the CNT surface. The Fe₃O₄ crystallites were ultrafine, connecting tightly to one another to form spheres. The EDS (Fig. 6c) of the illuminating electron beams on the obtained MMWCNT–TiO₂ composites revealed the existence of Fe, C, Ti and O elements, further confirming the successful modification of MMWCNTs with TiO₂. The quantitative analysis gives weight percentages of Fe (21.71%), C (44.58%), Ti (14.18%) and O (19.53%). Due to the results of the EDS, the optimized concentration of MMWCNT and TiO₂ obtained 8.33 g L⁻¹ and 1.66 g L⁻¹ respectively.

Fig. 6 (a) SEM image (b) TEM image and (c) EDS spectrum of MMWCNT–TiO₂.

Spectroscopic analysis. FT-IR spectrum of TiO₂–MMWCNT was shown in Fig. 7. A peak near 583 cm⁻¹ which belonged to the characterized absorption peaks of Ti–O and Fe–O. The transmittance band at 717 cm⁻¹ is considered to be the signature marker of TiO₂ and is symbolized to be the stretching vibration of Ti–O–Ti. The sharp peak at 1637 cm⁻¹ was a resultant of a bending vibration of a coordinated hydroxyl group as well as Ti. The minor band at 1042 cm⁻¹ in TiO₂–MMWCNT can be ascribed to the stretching vibration of C=O. The vibration bands at 2851 and 2930 cm⁻¹ were assigned to the stretching vibration of –CH. Moreover; the absorption peak at 1391 cm⁻¹ was ascribed to the O–H bond vibration of the surface-adsorbed species of particles (Ti–OH). Additionally, the broadband near 3421 cm⁻¹ is assigned not only to the presence of hydroxyl groups of TiO₂ but also to a strong interaction through hydrogen bonding between the hydroxyl groups on the surface of titanium dioxide.

Fig. 7 FT-IR spectrum of MMWCNT–TiO₂.

DRS technique is a useful technique to characterize the optical absorption properties of nanoparticles. Fig. 8 shows the action spectra of TiO₂, MMWCNT, and MMWCNT–TiO₂. For TiO₂, an absorption edge rising steeply toward the UV below 387 nm can be attributed to band-gap excitation of anatase (∼3.2 eV). TiO₂, which consists of the anatase phase only, exhibits no visible-light absorption and a threshold wavelength around 400 nm. According to the Planck's law and some further calculation, the band gap can be estimated by using E_g = hc/λ_g = 1239.85/λ_g (eV) where h is Planck's constant (4.13566733 × 10⁻¹⁵ eV s); c is the speed of light (2.99792458 × 10¹⁷ nm s⁻¹), E_g was the energy gap and λ_g was absorption threshold. The action regions of MMWCNT and MMWCNT–TiO₂ extended significantly into the visible-light region, suggesting that MMWCNT–TiO₂ can be activated by visible-light. From the calculation, the band gap value for TiO₂, MMWCNT and MMWCNT–TiO₂ was obtained 3.2 eV, 2.3 eV and 1.9 eV, respectively. These observations illustrate that the presence of MMWCNT–TiO₂ not only enhances the visible-light photocatalytic performance in organic dye degradation over TiO₂ or MMWCNT, but also exhibits excellent photostability in the visible-light range.

Fig. 8 DRS spectrum of TiO₂, MMWCNT and MMWCNT–TiO₂.

Photodegradation of MG

Prior to the photocatalytic activity tests, measurements of BET surface area, an important factor with respect to photocatalytic behavior, were obtained for TiO₂, MMWCNT and TiO₂–MMWCNT.⁴⁴ The calculated BET surface areas of MMWCNT and TiO₂–MMWCNT were 180.0 and 173.1 m² g⁻¹, respectively, which were notably higher than that of TiO₂ (65.66 m² g⁻¹). The pore volume analysis showed that the total volume of MMWCNT was 0.708 cm³ g⁻¹, and the total volume of TiO₂–MMWCNT was even decreased (0.625 cm³ g⁻¹) (Table 1). The average pore sizes calculated by the BJH method in TiO₂, MMWCNT and TiO₂–MMWCNT in the order 2.74, 2.73, and 2.74 nm, respectively, indicating that the mesoporous structures of the MMWCNT and TiO₂–MMWCNT composites were retained. As a result, MMWCNT and TiO₂–MMWCNT materials with large specific surface areas and ideal pore size distributions have shown to possess desirable photocatalytic properties for the removal of organic pollutants from wastewater.

Table 1 Physicochemical properties of TiO₂, MMWCNT and MMWCNT–TiO₂

Samples	SBET^a (m^2 g^−1)	Vtot^b (cm^3 g^−1)	Ave. pore diameter^c (nm)
a The calculated BET surface area.b The total pore volume analysis.c Average pore diameter.
TiO2	65.66	0.083	2.74
MMWCNT	180.0	0.708	2.73
MMWCNT–TiO2	173.1	0.625	2.74

---

Fig. 9 compares the photocatalytic degradation of MG in the presence of the neat TiO₂ powder and TiO₂–MMWCNT under irradiation of UV light. It is obvious that TiO₂–MMWCNT present a high photocatalytic activity compared to the neat TiO₂ powder. This behavior may be attributed to the positive effect of MWCNTs: (1) acting as a dispersing agent, the MWCNTs prevent TiO₂ from agglomeration and (2) acting as an adsorbent, the adsorption efficiency of TiO₂–MMWCNTs is better than that of neat TiO₂. It has been confirmed that MWCNTs in TiO₂–MMWCNTs are useful to absorb the MG and transfer the compound to the surface of TiO₂ (3) acting as a photosensitizer, there is a synergetic effect between MWCNTs and TiO₂. The photo-induced electrons in MWCNTs may trigger the formation of radicals in TiO₂ (superoxide radical ion and/or hydroxyl radical), which are responsible for the degradation of the organic compound.³⁴

Fig. 9 Comparison of photocatalytic behavior of TiO₂ and MMWCNT–TiO₂ composite on MG degradation.

The mechanism of photocatalytic can be summarized as follows: TiO₂ nanoparticles on photo-irradiation by UV light can be excited leading to the generation of electrons and holes. These excited electrons can then migrate into the conduction band of MWCNT since its work function is higher than TiO₂ nanoparticles.⁴⁵ This migration is thermodynamically favorable since the conduction band and valence band of TiO₂ are above than that of MWCNT. Further, MWCNT acts as a good electron acceptor for the electrons excited by UV light in the conduction band of TiO₂. Thus, the whole lifetime of the photo-generated electrons and holes is prolonged in the electron transfer process retarding the recombination and inducing higher quantum efficiency. It is hypothesized that the photo-generated electrons in MWCNT might react with the dissolved oxygen molecules, thereby producing oxygen peroxide radicals O₂˙⁻. Holes generated in TiO₂ nanoparticles may react with the OH⁻ which is obtained from water molecules to form hydroxyl radicals OH˙. MG can then be photocatalytically degraded by both oxygen peroxide radicals O₂˙⁻and hydroxyl radicals OH˙.

TiO₂ + hν → TiO₂ (h⁺, e⁻) (2)

TiO₂ (e⁻) + MWCNTs → MWCNTs (e⁻) + TiO₂ (h⁺, e⁻) (3)

TiO₂ (e⁻) + O₂⁻ → TiO₂ + O˙⁻ (4)

MWCNTs (e⁻) + O₂⁻ → MWCNTs + O˙⁻ (5)

TiO₂ (h⁺) + H₂O → TiO₂ + OH˙ + H⁺ (6)

(TiO₂)e + MG → (TiO₂)e⁻ + MG⁺˙ (7)

MG⁺˙ + O₂⁻ → “degraded products” (8)

MG⁺˙ + HO˙ → “degraded products” (9)

The effect of reaction variables such as recycling test, pH of the solution, initial concentration of MG, catalyst concentration, irradiation time, etc. on the degradation efficiency was studied and the results are delineated below.

To investigate whether TiO₂–MMWCNT composites can be recycled and reused for MG degradation, the materials were regenerated by washing with ethanol three times. Noticeably, the catalytic efficiency of the composite photocatalyst was still higher than 90% after be used for five cycles under 240 min UV irradiation. From the experimental results, since the photocatalyst was collected by a magnet, we can infer that the TiO₂ in the composite catalyst did not release or dissolve into solution (Fig. 10a).

Fig. 10 Investigation of recycling test (a), the influence of pH (b) photocatalyst amount (c) MG initial concentration (d) and irradiation time (e) on degradation efficiency of MG using TiO₂ and MMWCNT–TiO₂ photocatalysts (for (a)–(d) irradiation time = 240 min).

The effect of pH on the degradation of MG was investigated by keeping all other experimental conditions constant and varying the initial pH of the MG solution from 3 to 7. Diluted hydrochloride acid solution or sodium hydroxide solution was used to adjust the pH value when necessary. The experimental results reveal that the degree of degradation increases with pH value up to 5, beyond which the photodegradation efficiency starts to decrease, indicating an optimum pH of approximately 5.0 for best performance. This is due to the amount of hydroxyl radical formation on TiO₂ is influenced by the solution pH. The adsorption of MG is quite well at pH near the pzc of photocatalyst (Fig. 10b).

Experiments were performed to study the variations in the rate of degradation at different catalyst concentration ranging from 5 to 15 mg. It is observed that the degradation efficiency increases sharply with the catalyst concentration up to 10 mg. This is due to an increased number of available adsorption and catalytic sites on the surface of the MMWCNT–TiO₂ composite catalyst. A further increase in catalyst concentration, however, may cause light scattering and screening effect and thus reduce the specific activity of the catalyst. The result indicates that the optimum catalyst concentration for degradation of MG is 10 mg (Fig. 10c).

After optimizing the photocatalyst dosage, the effect of initial dye concentration ranging from 10 to 70 mg L⁻¹ on the photodegradation of MG was investigated. It has been observed that the rate of photodegradation increased with increasing in dye up to 20 mg L⁻¹. This may be due to the fact that as the dye concentration was increased, more dye molecules were available for consecutive degradation. The rate of photodegradation was found to decrease with a further increase in dye concentration, i.e. above 20 mg L⁻¹. The reason for this decrease is attributed to the shielding effect of dye at high concentration that retards the penetration of light to the dye molecules deposited over the catalyst surface (Fig. 10d).

The effect of irradiation time on the photocatalytic degradation of MG from its aqueous solution was investigated from 0 to 240 min, at 20 mg L⁻¹ MG concentration, 10 mg catalyst concentration and pH = 5.0. The obtained results are shown in Fig. 10e. The TiO₂–MMWCNT composite can achieve above 90% MG removal for 240 min while neat TiO₂ achieved only 65% MG removal for the same irradiation time. The addition of MMWCNT can enhance photoactivity of TiO₂ remarkably. For the neat TiO₂ or TiO₂–MMWCNT composite catalysts, the photodegradation efficiency increases with time, up to 240 minute. This indicates that photocatalytic degradation of MG with catalysts for 240 min is the optimum irradiation time.

Conclusion

We have shown, for the first time, the successful fabrication of a recoverable and effective TiO₂–MMWCNT photocatalyst by electrostatic attraction from MMWCNT dispersion onto TiO₂ suspension in ethanol. For this aim, firstly, the anatase TiO₂ nanoparticle was synthesized by APCVS method and qualified. Secondly, MMWCNT was prepared and mixed with TiO₂ in ethanol to form MMWCNT–TiO₂ composite and characterized. Finally, the UV lights photocatalytic activity of TiO₂–MMWCNT and TiO₂ as the photocatalysts for degradation of MG were investigated. All the resulting of MMWCNT–TiO₂ exhibited significantly enhanced photocatalytic activities compared with the pure TiO₂ nanoparticle. The photocatalytic stability of MMWCNT–TiO₂ remained almost unchanged after recycling five cycles. In addition, this composite could be easily recovered from water using an external magnet. Because of these attractive, stable and recyclable features, the as-prepared composites may have been promising applications in the photodegradation of organic compounds in aqueous solution under the irradiation of UV light.

Acknowledgements

The financial support from the University of Tehran is gratefully acknowledged.

References

1. S. Gül and O. Ozcan-Yıldırım, Chem. Eng. J., 2009, 155, 684–690.
2. F. Sayilkan, M. Asiltürk, P. Tatar, N. Kiraz, E. Arpaç and H. Sayilkan, J. Hazard. Mater., 2007, 148, 735–744.
3. H. Wang, J. Li, X. Quan, Y. Wu, G. Li and F. Wang, J. Hazard. Mater., 2007, 141, 336–343.
4. A. Fijishima, T. N. Rao, K. Try and D. A. Tryk, J. Photochem. Photobiol., C, 2001, 1, 1–21.
5. R. Asaki, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 2001, 293, 269–271.
6. X. B. Chen and S. S. Mao, Chem. Rev., 2007, 107, 2891–2959.
7. T. H. T. Vu, T. T. T. Nguyen, P. H. T. Nguyen, M. H. Do, H. T. Au and T. B. Nguyen, et al., Mater. Res. Bull., 2012, 47, 308–314.
8. A. de Morais, L. M. D. Loiola, J. E. Benedetti, A. S. Gonçalves, C. A. O. Avellaneda and J. H. Clerici, et al., J. Photochem. Photobiol., A, 2013, 251, 78–84.
9. G. Oza, S. Pandey, A. Gupta, S. Shinde, A. Mewada and P. Jagadale, et al., Mater. Sci. Eng., C, 2013, 33, 4392–4400.
10. X. H. Xia, Z.-J. Jia, Y. Yu, Y. Liang, Z. Wang and L.-L. Ma, Carbon, 2007, 45, 717–721.
11. M. Chen, F. Zhang and W. Oh, Carbon, 2009, 47, 2943.
12. B.-T. Zhang, X. Zheng, H.-F. Li and J.-M. Lin, Anal. Chim. Acta, 2013, 784, 1–17.
13. S. Aryal, C. K. Kim, K.-W. Kim, M. S. Khil and H. Y. Kim, Mater. Sci. Eng., C, 2008, 28, 75–79.
14. A. Jitianu, T. Cacciaguerra, R. Benoit, S. Delpeux, F. Béguin and S. Bonnamy, Carbon, 2004, 42, 1147–1151.
15. Y. Yu, J. C. Yu, J. G. Yu, Y. C. Kwok, Y. K. Che and J. C. Zhao, et al., Appl. Catal., A, 2005, 61, 186–196.
16. H. Wang, X. Quan, H. Yu and S. Chen, Carbon, 2008, 46, 1126–1132.
17. V. Polshettiwar, R. Luque, A. Fihri, H. Zhu, M. Bouhrara and J. M. Basset, Chem. Rev., 2011, 111, 3036–3075.
18. S. Anandan, G. J. Lee, S. H. Hsieh, M. Ashokkumar and J. J. Wu, Ind. Eng. Chem. Res., 2011, 50, 7874–7881.
19. C. F. Chang and C. Y. Man, Ind. Eng. Chem. Res., 2011, 50, 11620–11627.
20. S. Xuan, W. Jiang, X. Gong, Y. Hu and Z. Chen, J. Phys. Chem. C, 2009, 113, 553–558.
21. B. Palanisamy, C. M. Babu, B. Sundaravel, S. Anandan and V. Murugesan, J. Hazard. Mater., 2013, 252, 233–242.
22. F. Mou, L. Xu, H. Ma, J. Guan, D. R. Chen and S. Wang, Nanoscale, 2012, 4, 4650–4657.
23. F. A. Deorsola and D. Vallauri, Powder Technol., 2009, 190, 304–309.
24. C. Wang, Z. X. Deng, G. Zhang, S. Fan and Y. Li, Powder Technol., 2002, 125, 39–44.
25. L. Mao, Q. Li, H. Dang and Z. Zhang, Mater. Res. Bull., 2005, 40, 201–208.
26. S. Mayadevi, S. S. Kulkarni, A. J. Patil, M. H. Shinde, H. S. Potdar and S. B. Deshpande, et al., J. Mater. Sci., 2000, 35, 3943–3949.
27. Y. Kitamura, N. Okinaka, T. Shibayama, O. P. Mahaney, D. Kusano and B. Ohtani, et al., Powder Technol., 2007, 176, 93–98.
28. H. Ishihara, J. P. Bock, R. Sharma, F. Hardcastle, G. K. Kannarpady and M. K. Mazumder, et al., Chem. Phys. Lett., 2010, 489, 81–85.
29. I. Ahmad and S. S. Bhattacharya, J. Phys. D: Appl. Phys., 2008, 41, 155313–155319.
30. A. Simchi, R. Ahmadi, S. M. Seyed Reyhani and A. Mahdavi, Mater. Des., 2007, 28, 850–856.
31. C. S. Kim, K. Nakaso, B. Xia, K. Okuyama and M. Shimada, Aerosol Sci. Technol., 2005, 39, 104–112.
32. C. J. Tavares, S. M. Marques, T. Viseu, V. Teixeira, J. O. Carneiro and E. Alves, et al., J. Appl. Phys., 2009, 106, 113535–113542.
33. M. Rahiminezhad-soltani, K. Saberyan, F. Shahri and A. Simchi, Powder Technol., 2011, 209, 15–24.
34. H. Zhou, C. Zhang, X. Wang, H. Li and Z. Du, Synth. Met., 2011, 161, 2199–2205.
35. J. Cho, S. Schaab, J. A. Roether and A. R. Boccaccini, J. Nanopart. Res., 2007, 10, 99–105.
36. M. Corrias, B. Caussat, A. Ayral, J. Durand, Y. Kihn and P. Kalck, et al., Chem. Eng. Sci., 2003, 58, 4475–4482.
37. H. Dong and K. Lu, Int. J. Appl. Ceram. Technol., 2009, 6, 216–222.
38. J. Zhang, Y. Li, C. Zhang and Y. Jing, J. Hazard. Mater., 2008, 150, 774–782.
39. A. Mittal, J. Hazard. Mater., 2006, 113, 196–202.
40. S. Nethaji, A. Sivasamy, G. Thennarasu and S. Saravanan, J. Hazard. Mater., 2010, 181, 271–280.
41. V. K. Gupta, I. Ali and V. K. Saini, Ind. Eng. Chem. Res., 2004, 43, 1740–1747.
42. G. Daneshvar and F. Shemirani, Talanta, 2013, 115, 744–750.
43. X. Ren, C. Chen, M. Nagatsu and X. Wang, Chem. Eng. J., 2011, 170, 395–410.
44. S. Brunauer, P. H. Emmett and E. Teller, J. Am. Chem. Soc., 1938, 60, 309–319.
45. S. T. Martin, M. R. Hoffmann, W. Choi and D. W. Bahnemannt, Chem. Rev., 1995, 95, 69–96.

---