Abstract

The multiwall carbon nanotubes (MWCNTs)/titanium dioxide (P25) composite in different ratios was prepared using simple evaporation and drying process. The composite was characterized by Raman spectroscopy, X-ray diffraction (XRD), UV-visible diffuse reflectance spectroscopy, and scanning electron microscopy (SEM). The photocatalytic activity of this composite was investigated using degradation of the Bismarck brown R dye (BBR). An optimal MWCNTs/TiO2 ratio of 0.5% (w/w) was found to achieve the maximum rate of BBR degradation. It was observed that the composite exhibits enhanced photocatalytic activity compared with TiO2. The enhancement in photocatalytic activity performance of the MWCNTs/P25 composite is explained in terms of recombination of photogenerated electron-hole pairs. In addition, MWCNTs act as a dispersing support to control the morphology of TiO2 particles in the MWCNTs/TiO2 composite.

1. Introduction

Titanium dioxide (TiO2) is photocatalyst material using light to destroy environmental contaminants from wastewater due to its ability to convert them into innocuous end products such as CO2 and H2O [1]. TiO2 is one of the most prominent oxide semiconductor materials due to its high photocatalytic activity, chemical stability, nontoxicity, physical properties, and low cost [2]. However, TiO2 has some disadvantage such as low surface area and easy recombination of the electron-hole pair [3]. The photocatalytic activity of TiO2 can be improved by reducing the electron-hole recombination rate, preventing the particles agglomeration, and increasing the adsorption capacity. The photocatalytic efficiency can be enhanced by the modification of TiO2 surface [4]. Unique properties of carbon nanotubes (CNTs) make them promising material for the fabrication of CNTs/P25 composite for a broad range of applications in photocatalysis [5].

In recent years, titanium dioxide nanoparticles supported on carbon nanotubes have been extensively studied and found to be effective photocatalyst for the removal of hazardous organic chemicals from wastewater. CNTs/TiO2 nanocomposite can be prepared by different methods which are classified into two basic classes. The first class involves the prior synthesis of TiO2 nanoparticles thereafter connected to surface functionalized MWCNTs by either covalent or noncovalent interactions [6]. The second class is the one step method which involves direct deposition of TiO2 nanoparticles onto MWCNTs surface and in situ formation of nanoparticles through redox reactions or electrochemical deposition on CNTs [7, 8]. The important issues that must be considered to obtain composite with enhanced properties are efficient chemical functionalization of CNTs, homogeneous dispersions in solvents, and good interconnectivity with TiO2. The effective utilization of CNTs in composite applications strongly depends on the ability to disperse them homogeneously. Chemical modifications have become an important issue due to the poor solubility of the CNTs in almost any solvent. Chemical modification of CNTs ensures good dispersion of nanotubes in a medium and enhances interconnectivity with titanium dioxide [9, 10]. After chemical modification or functionalization, the nanotube surface contains polar groups like hydroxyl or carboxyl groups which are able to interact with the oxygen of the titanium dioxide through hydrogen bonding. In CNTs/TiO2 composite the photocatalytic enhancement is generally attributed to electron capture by the CNTs and in the surface recombination rate [11]. The bond of carbon-oxygen-titanium can extend the light absorption to longer wavelengths and therefore potentially lead to the improvement of the photocatalytic activity [12].

In the present study, MWCNTs/P25 composite was prepared by simple evaporation and drying process. The resultant composite was characterized by Raman spectroscopy, X-ray diffraction (XRD), UV-visible reflectance spectroscopy, and scanning electron microscopy (SEM). The effect of MWCNTs on the optical properties of TiO2 was observed by measuring the degree of photocatalytic degradation of aqueous Bismarck brown R (BBR) dye solution.

2. Experiment

2.1. Chemicals

Multiwall carbon nanotubes (MWCNTs) and BBR (C21H24N82HCl) were purchased from Sigma-Aldrich in Germany. For the oxidation of the surfaces of MWCNTs, sulphuric and nitric acid were chosen as the oxidizing agents. These oxidizing agents were purchased from Sigma-Aldrich also. Titanium dioxide was obtained from Degussa AG.

2.2. Functionalization of Multiwall Carbon Nanotubes

100 mg of MWCNTs was suspended in 50 mL of a 3 : 1 mixture of concentrated H2SO4/HNO3. The MWCNTs suspension was sonicated in a water bath for 7 hrs at room temperature to open the agglomeration of the nanotubes and to anchor the acid solution uniformly on the carbon’s surface. Figure 1 shows the oxidation of the CNTs. The resultant suspension was then diluted with 1000 mL of distilled water. The oxidized MWCNTs were filtered and washed several times with distilled water by centrifuge. A dialysis process was used to obtain a neutral pH, and then the oxidized MWCNTs were dried at 100°C for 10 hrs.

2.3. Synthesis of the Composite

Different ratios from the MWCNTs/P25 composite were prepared using a simple evaporation and drying process. That is, commercial titanium dioxide (P25) and O-MWCNTs were dispersed into 100 and 20 mL of distilled water and sonicated for 30 and 15 min, respectively. The O-MWCNTs solution was added to the TiO2 suspension along with stirring. The suspension containing MWCNTs and TiO2 particles was heated to 80°C to evaporate the water. After the water had evaporated, the composite was dried overnight in an oven at 100°C. Figure 2 shows the experimental procedure for preparation of the MWCNTs/P25 composite by using simple evaporation and drying process.

2.4. Characterization Methods

X-ray diffraction (XRD) spectra were collected on a STADI P diffractometer (STOE, Germany) with a positional sensitive detector using monochromatic Cu K radiation.

XRD data were employed to calculate the mean crystallite sizes () by Scherrer’s formula in the following equation [13]: where is the average crystal size, is the constant crystal lattice,  nm is the X-ray wavelength of Cu K, is the full width of the peak measured at half maximum intensity, and is Bragg’s angle of the peak.

Band gap energies of P25 and MWCNTs/P25 composites were determined via the measurement of reflectance data using a Cary 100 Scan UV-visible spectrophotometer system. This is equipped with a Labsphere integrating sphere diffusing reflectance accessory for diffuse reflectance spectra by employing BaSO4 as a reference material. The measured reflectance data were transformed to the Kubelka-Munk function by the following equations [14]: The band gap energy for all the samples was measured from the plot of versus energy of light () in eV. This depended on the intersection of the tangent via the point of inflection in the absorption band and the photon energy axis.

Scanning electron microscopy (SEM) measurements were carried out on a JEOL JSM-6700F instrument, using a secondary electron detector (SE) at an accelerating voltage of 2.0 kV.

2.5. Photocatalytic Activity

The photocatalytic activity of MWCNTs/P25 composite was determined using BBR decomposition in aqueous solution under an UV lamp (0.7 mW/cm2). The suspensions solutions were prepared by adding 50 mg of nanocomposite catalysts to 100 mL of 5 × 10−5 M BBR aqueous solution. Prior to irradiation, the suspensions were stirred in darkness for 30 min using a magnetic stirrer to ensure adsorption equilibrium. During adsorption and irradiation, the suspensions were sampled at regular intervals. 3 mL of the reaction mixture was collected and centrifuged for 15 min. The clear solution was carefully removed using a syringe and centrifuged again at the same speed and the same period time. The second centrifuge was found necessary to remove fine particles of the catalyst. After the second centrifuge, the absorbance at the maximum wavelength of 459 nm of the BBR was measured with a UV-visible spectrophotometer.

The photocatalytic reactions on the surfaces of MWCNTs/P25 composites can be expressed by the Langmuir-Hinshelwood model. The rate of the photocatalytic reaction can be expressed as where and are the reactant concentration at time and , respectively, and and are the apparent reaction rate constant and time, respectively. A plot of ln() versus will yield a slope of .

3. Results and Discussion

3.1. Characterization of the MWCNTs/P25 Composite

Raman spectra for MWCNTs and O-MWCNTs are shown in Figure 3. Three distinct bands found in the MWCNTs Raman spectrum, which originate from different aspects of the nanotube, are the radial breathing mode (RBM), the disordered mode (D mode), and the tangential mode (G band) [15]. For MWCNTs, the radial breathing mode at 105 cm−1 depends on the tube’s diameter. The shape and intensity of the D mode at 1340 cm−1 correspond to the sp3 hybridized carbon atoms [16], which is correlated with the extent of the nanotubes sidewall defects and the chemical sidewall functionalization. The higher frequency tangential G mode at 1574 cm−1 and the first overtone of the D band, which is called the G’ band at 2694 cm−1, are sensitive to the charge exchanged between the nanotubes and the guest moiety. The G band is thus an intrinsic feature of carbon nanotubes, which is closely related to vibrations in all sp2 carbon materials. The second order G’ band does not require an elastic defect-related scattering process and is observable for defect-free sp2 carbons. Chemical oxidation produces defects on the sidewalls of the nanotubes and attaches some functional groups onto the defective areas of the nanotubes. In the Raman spectrum of the O-MWCNTs, these bands were shifted to 106.5, 1350.5, 1584, and 2662.5 cm−1 for the RBM, D, G, and for the first overtone of the D bands, respectively. Also, the results indicate that the intensity of the G mode slightly increased, while the D mode intensity decreased after oxidative treatment. The ratio of the D to the G band intensity () is usually used for a measurement of the disordered sites on carbon nanotubes walls and it is the indicator of the level of the covalent functionalization of the MWCNTs [17]. On comparing the modified and nonmodified tubes a decrease in the ratio of intensities from 1.2748 to 1.1437 was observed because of the elimination of amorphous carbon during the acid treatment on the modified tubes as is shown in Table 1. The full width at half the maximum (FWHM) of the D and G bands was analyzed for pristine and functionalized samples of MWCNTs (Table 1). For the D peak, the FWHM of the functionalized MWCNTs ( cm−1) was much larger than that of the untreated MWCNTs ( cm−1), and for G peak, the FWHM of the functionalized MWCNTs ( cm−1) was smaller than that of the untreated MWCNTs ( cm−1). The results are in good agreement with [18]. Table 1 shows a comparison of the Raman spectra features of the pristine and functionalized MWCNTs samples.

Raman spectroscopy was used to investigate the interaction of titanium dioxide P25 with the MWCNTs. The spectra of the P25, MWCNTs, and MWCNTs (0.5%)/P25 composite are illustrated in Figure 4. The characteristic bands at 144, 197, 397.5, 515.5, and 638.5 cm−1 correspond to the anatase phase of the P25 [19]. The main four bands (197, 397.5, 515.5, and 638.5 cm−1) in the Raman spectrum are representative of the anatase P25 being broadened and shifted to 195.5, 397, 514, and 637 cm−1 in MWCNTs/P25 composite, as compared to the pure P25. The band broadening and decreasing in intensity can be attributed to the effects of smaller particle sizes [20, 21]. The composite sample shows the D and G bands of MWCNTs. These bands shift to 1356 and 1598.5 cm−1 in the MWCNTs/P25 composite, clearly indicating that the strong interaction between the P25 and the MWCNTs may enhance a charge transfer from the P25 to the MWCNTs in order to separate and stabilize the charge and thereby hinder charge recombination [2224].

XRD characterization method is used to measure the extent of graphitization, as well as providing information about the degree of nanotube alignment. Figure 5 shows comparison between the X-ray diffraction of the MWCNTs and the O-MWCNTs. The pattern of the O-MWCNTs shows a high intense peak at ° and a low intense peak at °, corresponding to the (002) and (100) reflections, respectively. Compared to the MWCNTs, ° and °, these peaks show a downward shift, which is attributed to an increase in the sp2 C=C layers spacing [25] and suggests that crystallinity is not lost due to oxidative acid treatment. Furthermore, the XRD patterns were used to characterize the crystalline structure of the MWCNT/P25 composites. Figure 6 shows the XRD patterns of the P25 compared with the MWCNT/P25 composite. The patterns demonstrate the highly crystalline nature of the composite. The same peaks in P25 and composite at 25.3°, 37.8°, 48.0°, 54.9°, and 62.5° were the diffractions of (101), (004), (200), (211), and (204) planes off anatase, respectively, and peaks at 27.4°, 36.1°, 41.2°, and 54.3° belonging to the diffraction peaks of (110), (101), (111), and (211) planes of rutile, thus indicating that the P25 and the composite contain a mixed structure of anatase and rutile. It was also observed that the MWCNTs/P25 composite had a weaker intensity compared with the P25. When comparing the XRD patterns of the MWCNTs and the MWCNTs/P25 composites, the characteristic peaks for the MWCNTs at the positions of 25.2° and 44.0° might disappear or become thinner in the XRD pattern of the composites. The reason for this is that these peaks in the composites overlapped with the main peak of the anatase phase of the P25 at 25.3°.

The crystalline extent of the MWCNTs is lower than the crystalline extent of the P25, leading to the shielding of the peaks of the MWCNTs by those of TiO2. Also, the small content in the composite, as well as the absence of MWCNTs aggregated in the pores, was supported by the disappearance of the characteristic peaks from the MWCNTs in the XRD patterns [26]. The average crystallite sizes of the P25 and MWCNTs/P25 composite were calculated by Debye-Scherer formula. According to the calculation, the average crystallite sizes of the P25 and composite were 22.7 and 21.4 nm, respectively [27].

The results of the UV-visible diffuse reflectance spectra of the P25 and the composite, as well as UV-visible Kubelka-Munk transformed diffuse reflectance spectra, indicate that the band gap of the P25 is wider than that of the MWCNTs/P25 composite. Figure 7 shows the optical absorption spectra of the P25 and the MWCNTs (0.5%)/P25 composite. The MWCNTs/P25 composite can exhibit better absorbency than the P25 within a wavelength region of 200–800 nm. Compared with the P25, the MWCNTs/P25 composite particles can cause an obvious red shift of the UV-visible spectra. The P25 particles are clearly absorbed at wavelengths above 380 nm by adding MWCNTs to obtain MWCNTs/P25 composites. The effective band gap of the P25 3.0 eV is reduced to 2.3 eV for MWCNTs (0.5%)/P25. Therefore the MWCNTs/P25 composite can be excited to produce more electron-hole pairs under solar irradiation, which may result in higher photocatalytic activity. For all the composites, their enhanced absorbance extends broadly over wavelengths of >400 nm, which is in agreement with the black color of the sample [2830].

The morphology of the MWCNTs/P25 composite was studied using an SEM, as shown in Figure 8. The SEM images indicate that the MWCNTs are homogenously distributed throughout the P25 matrix with an apparent agglomeration of the P25 particles. An explanation was provided that the thorough dispersion of the small MWCNTs particles into the P25 aggregates could provide the information on the presence of more reactive sites, due to a considerable portion of the P25 being enclosed in the three-dimensional matrix. So it was considered that the MWCNTs/P25 composite could show much more activity, including an excellent photocatalytic activity. The average sizes of the catalyst particles were 25.1 nm, which agrees with the XRD measurements [3133].

3.2. Photocatalytic Activity of the MWCNTs/P25 Composite

The photocatalytic removal of the BBR aqueous solution was investigated using a source of UV light irradiation. Figure 9 shows the changes in relative concentration of the BBR in aqueous solution on UV irradiation times for the MWCNTs/P25 composite in different ratios. The results indicate that the MWCNTs enhanced the adsorption properties of TiO2 (P25) due to enhancement in surface area [34, 35]. After 60 min of UV irradiation, the MWCNTs (0.5%)/P25 in Figure 10 had the highest degradation of the BBR solution, from which 69% was almost removed. The BBR degradation of the P25, MWCNTs (0.25%)/P25, MWCNTs (1%)/P25, and MWCNTs (5%)/P25 also achieved 38, 42, 66, and 63%, respectively. It clearly indicates that all ratios of MWCNTs favor the separation of photogenerated electron-hole pairs and decrease of their recombination rate and thus display significantly enhanced photocatalytic activity for degrading BBR dye in aqueous solution under UV irradiation due to the chemically bonded between P25 and MWCNTs, as compared with commercial P25 photocatalyst [36].

The photocatalytic degradation of the BBR in aqueous suspension of the composite catalysts containing MWCNTs and P25 follows apparent first order kinetics. The kinetic plots are shown in Figures 11 and 12.

4. Conclusions

The MWCNTs/P25 composite photocatalysts with different ratios were prepared using simple evaporation and drying process. Composite characterization indicates a relatively homogeneous MWCNTs dispersion in the P25 matrix with modified UV-visible absorption properties, suggesting a strong interphase structure effect between the MWCNTs and the P25 in the composite. According to the results it could be suggested that all of the samples had a good photocatalytic activity for the BBR degradation. The rate constant for the photodegradation of BBR dye by the P25 increased by 2.5 times in the presence of the 0.5% MWCNTs. The MWCNTs (0.5%)/P25 composite had higher activity of about 2 times than that of the P25 alone. The enhanced photocatalytic efficiency of composite suggests that the MWCNTs act as an adsorbent, dispersing agent and electron reservoir to trap electrons from the P25 particles due to UV irradiation and hence hindering the electron-hole pair recombination.

Conflict of Interests

The authors declare that there is no conflict of interests regarding the publication of this paper.

Acknowledgment

The authors wish to thank Professor Dr. Detlef W. Bahnemann group in the Institute of Technical Chemistry, Leibniz Universität Hannover, Germany, for their assistance.