- Email: [email protected]
Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules
G Model JIEC 3489 No. of Pages 15
Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Journal of Industrial and Engineering Chemistry journal homepage: www.elsevier.com/locate/jiec
Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules Elmira Pajootan* , Mehdi Rahimdokht, Mokhtar Arami Textile Engineering Department, Amirkabir University of Technology, 424 Hafez Ave., Tehran 15875-4413, Iran
A R T I C L E I N F O
A B S T R A C T
Article history: Received 24 August 2016 Received in revised form 20 June 2017 Accepted 21 June 2017 Available online xxx Keywords: TiO2 nanoparticles Electrodeposition Immobilization Solvent evaporation Response surface methodology
We utilize carbon nanotubes fabricated carbon plates with large surface area and high adsorption capacity as the substrate to immobilize TiO2 nanoparticles without any agglomeration using solvent evaporation method. The photocatalytic activity is investigated in a continuous photocatalytic reactor to degrade three acid dyes, which is shown to be stable for 16 successive cycles. Response surface methodology (RSM) is used to optimize and model the binary system for a closer simulation of real industrial wastewater with the electrical energy consumption below 0.5 KWh/gCOD. A fast and complete decolorization was achieved in 100 min by adding H2O2 to the system. © 2017 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
Introduction Nowadays, nanotechnologies are introduced as a preferable solution to current water and wastewater treatment methods to improve the water quality to meet the environmental standards and to conquer the water scarcity resulted from the human activities in both developing and industrialized countries. Titania nanoparticles (TiO2) are one of the most studied semiconductors for the degradation of pollutants existing in water streams, and also in many other applications, such as sunscreen, surface coating, solar cells, environmental remediation, etc. due to their semiconducting properties and photostability [1–5]. When TiO2 is irradiated by UV light, the photoinduced electrons and holes (e/h+) are formed. OH and H2O as electron donors and oxygen as electron acceptor will form the hydroxyl (OH) and
superoxide ion radicals (O 2 ). These generated radicals especially hydroxyl radicals are nonselective strong oxidants which can fully mineralize organic pollutants [6,7]. One of the major drawbacks of the usage of TiO2 nanoparticles (NP) is the problem encountered through their separation and removal from the treated solutions. The UV absorption of these photoactive and nano-sized materials has a potential toxicity to
* Corresponding author. E-mail address: [email protected] (E. Pajootan).
organisms and environment, to which many investigations are dedicated in order to survey the phototoxicity and dramatic effect of TiO2 NP. Therefore, immobilization of these nanoparticles is now the subject of most enquiries pertaining to the area of this science [1–4,8–10]. In addition, the practical application of slurry TiO2 solutions can be very challenging in the continuous wastewater treatment processes. Furthermore, the application of TiO2 NP for photocatalytic processes suffers from a low quantum efficiency because of the rapid recombination of the photoinduced electrons and holes (e/h+), mass transfer and photon transfer limitations [11,12]. Several methods have been developed to enhance the photocatalytic efficiency; in this regard, carbon nanotubes (CNT) with large specific surface area and excellent electrical properties have received tremendous attention for building the composite materials. These nanotubes can also be used as supporting material for TiO2 NP owing to their nanomorphology, high mechanical and chemical stability and high structural integrity [4,11,13–17]. This combination can significantly improve the photocatalytic performance by lowering the recombination of e/h+ couples, and the combination of their electronic, adsorption, mechanical and thermal properties [18–21]. Various techniques have been reported to coat titania NP on carbon-based supports for different purposes such as: hydrolysis of TiCl4 [22], deposition of TiO2 by sol-gel method [23], mixing TiO2 with carbon materials [24], electrodeposition [25], precipitating
http://dx.doi.org/10.1016/j.jiec.2017.06.039 1226-086X/© 2017 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
2
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
[26,27], hydrothermal treatment [28,29], solvent evaporation [30], etc. Nowadays, response surface methodology (RSM) is one of the most popular experimental design and optimization tools being used in a wide range of applications especially wastewater treatment. Reducing the number of experiments while maintaining the accuracy of results, response surface modeling through regression, process optimization and prediction of the variables’ values, exploration the interactions of various combinations of parameters are some of the advantages of RSM technique [31–33]. In this research, we successfully demonstrated the immobilization of TiO2 NP on carbon-based material to decolorize the simulated textile wastewater by photocatalytic process in a continuous reactor. For this purpose, the relatively low-cost and available carbon-based materials and TiO2 NP, with their specific characterizations were taken into consideration. First, CNT were electrodeposited on the surface of the carbon plates via a method reported in our previous study [34]. Then, a solution containing dispersed TiO2 NP were added to the surface of the carbon plates (TiO2-C) and CNT fabricated carbon plates (TiO2-CNT). Finally after the evaporation of the solvent, TiO2 NP were immobilized on the plates. FESEM images and FTIR spectra indicated the characteristics of the prepared plates. The prepared plates were successfully employed for the continuous decontamination of wastewater with appropriate stability for several cycles. The decolorization and further degradation of the contaminated solutions containing C.I. Acid Red 14, C.I. Acid Blue 92 and C.I. Acid Yellow 117 were investigated and the effect of important parameters including pH, initial dye concentration and TiO2 dosage was demonstrated. RSM was applied to design the experiments to evaluate the effect of parameters in the degradation of dyes from binary solutions and to multi-optimize the degradation process considering maximum dye and COD removal (%), as well as minimum electrical energy consumption as the responses to develop a cost effective system. The variation of dye removal efficiency by changing the flow rate and addition of hydrogen peroxide was evaluated as well.
Experimental Materials In all experiments, the reagents were of analytical grade. Titania NP (Degussa P25) was employed as the photocatalyst (average particle size: 30 nm, purity >97% with 80:20 anatase to rutile). Cetyl Trimethyl Ammonium Bromide ((C16H33)N(CH3)3Br, CTAB) and MWCNT (purity >95%, length 10–20 mm and diameter 30–50 nm) were purchased from Merck and Neutrino, respectively. C.I. Acid Red 14 (AR14), C.I. Acid Blue 92 (AB92) and C.I. Acid Yellow 117 (AY117) (Ciba Co.) which are widely used in textile industry, were employed as synthetic dyes (Table 1). H2SO4 (1 M) and NaOH (1 M) (Merck) were used to adjust the pH of dye solutions. Carbon substrates (purity >95%) with the dimension of 50 110 3 mm3 used as the supporter for titania naoparticles were purchased from Seraaj Co. The hydrogen peroxide (H2O2, 30%) was supplied from Loba Chemie. Immobilization of TiO2 NP The carbonaceous support (Carbon and CNT fabricated carbon substrates) were selected as substrates for the immobilization of TiO2 NP due to their excellent adsorption properties and chemical inertness, especially CNT. The fabrication of carbon plates using CNT by a novel, facile and inexpensive electrodeposition technique was fully explained in a separate investigation in our previous study [34]. In summary, the carbon plates were abraded with sand paper and washed with distilled water. Then each electrode was pretreated with 50 mL of NaOH (10% w/v), 50 mL of HNO3 (50%, v/v) and 50 mL of acetone, each step for 5 min, respectively, and then they were rinsed with distilled water for three times. The electrodeposition procedure to form a thin layer of CNT on the surface of carbon plates was performed by applying the DC voltage of 17.5 (V) to a 200-mL solution containing 0.06 g of CNT and 0.04 g of CTAB which had been sonicated for 60 min using Delta D68H
Table 1 Dyes characteristics. C.I. generic name
lmax (nm)
Molecular weight (g/mol)
Formula
C.I. Acid Red 14
515
502.42
C20H12N2O7S22Na
C.I. Acid Blue 92
572
695.58
C26H16N3O10S33Na
C.I. Acid Yellow 117
440
848.83
C39H30N8O8S22Na
Structure
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
3
Fig. 1. Preparation of the photocatalytic plates and a cross section of the photo-reactor.
Ultrasonic prior to the deposition. The positively charged CNT (by the arranged cationic hydrophilic group of CTAB on their surface) would transfer from the bulk to the surface of the negatively charged cathode electrode to form a uniform layer on its surface. The modified electrodes were washed with distilled water and dried in oven for 15 min at 60 C. The structural and electrochemical properties of these carbon plates before and after the fabrication process were also demonstrated and fully reported in our previous study [34]. In order to immobilize TiO2 NP by the solvent evaporation as a fast, simple and cost effective method with no use of chemicals, the carbon plates before and after the fabrication process were placed in oven at 110 C. The specific amount of TiO2 NP (0.01, 0.02, 0.04 and 0.08 g) were dispersed for 30 min in 30 mL distilled water and then dropped on the surface of six carbonaceous hot plates. After the evaporation of water, TiO2 NP formed a uniform layer on the surface of the carbon (TiO2-C) and CNT fabricated carbon substrates (TiO2-CNT). Experimental set-up The dye removal process was carried out in a continuous cylindrical reactor made of Plexiglas having the working volume of
1 L. Six fabricated TiO2-C or TiO2-CNT plates were placed in the cylindrical photo-reactor in the shape of a regular hexagon. A cross section view of the photo-reactor employed for the decolorization is shown in Fig. 1. The UV lamp (Philips, 9 W) in the quartz tube was placed at the center of the reactor to provide near UV radiation; and the aeration output was placed at the center bottom of the reactor. The reactors were placed in an aluminum box to prevent the illumination of UV lamp to the room. The dark adsorption of dyes onto TiO2-C and TiO2-CNT were performed in the same reactors for 30 min excluding the UV irradiation to establish an adsorption–desorption equilibrium and then the UV lamp was turned on to begin the photocatalytic process. The samples were collected over the specific time intervals and the absorbance was measured using UNICO 2100 Spectrophotometer at the maximum wavelength of each dye. Dye removal efficiency was determined according to Eq. (1): DR% = [(A0 A)/A0] 100
(1)
where A0 and A are the absorbance of dye solution before and after the treatment, respectively. For the binary solutions, two equations and two readings of absorbance at two different wavelengths employing the extension of Beer–Lambert law were used to calculate the concentration of each dye in the solution [35,36].
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
4
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Table 2 CCD design and the obtained results. Run order
Time (min)
pH
TiO2 (g/L)
[C0] (mg/L)
AR14 removal (%)
AB92 removal (%)
COD removal (%)
EEC (KWh/gCOD)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
210 180 150 150 120 180 120 120 90 150 120 150 150 150 150 150 150 120 180 150 150 120 180 180 120 150 120 180 150 180 180
4 5 2 4 5 5 3 5 4 4 3 4 4 4 4 4 4 5 3 6 4 3 5 5 5 4 3 3 4 3 3
0.03 0.04 0.03 0.03 0.02 0.02 0.02 0.04 0.03 0.01 0.02 0.03 0.05 0.03 0.03 0.03 0.03 0.04 0.04 0.03 0.03 0.04 0.04 0.02 0.02 0.03 0.04 0.02 0.03 0.02 0.04
30 40 30 30 40 20 40 40 30 30 20 10 30 30 30 30 30 20 20 30 30 40 20 40 20 50 20 40 30 20 40
55.84 27.06 82.28 52.22 19.09 27.52 68.36 22.21 49.69 33.45 68.26 53.12 46.51 52.16 51.90 52.16 52.11 35.96 73.68 12.10 52.06 62.43 37.22 22.86 26.11 40.99 69.09 66.50 52.11 70.47 63.03
66.21 39.24 70.42 66.92 38.58 45.39 45.28 38.67 59.37 56.19 57.86 61.85 58.70 66.71 65.67 67.13 67.10 47.40 71.72 26.45 67.14 55.97 53.25 39.39 43.55 47.64 66.23 53.52 67.10 59.07 60.98
45.31 21.42 67.31 46.18 13.61 28.12 47.10 18.00 43.44 28.69 49.37 51.04 37.95 43.46 43.36 46.28 41.70 26.91 61.26 13.81 40.72 49.64 31.52 17.40 21.15 29.13 58.62 49.50 48.05 54.99 52.99
1.475 2.005 0.709 1.034 2.105 3.056 0.608 1.592 0.659 1.664 1.160 2.806 1.258 1.099 1.101 1.032 1.145 2.129 1.403 3.457 1.173 0.577 2.726 2.469 2.708 0.983 0.977 0.868 0.994 1.563 0.811
The electrical energy consumption (EEC) per unit COD decay (kWh/gCOD) was calculated by using Eq. (2): EEC ¼
EUV t 100 V ðCOD0 CODt Þ
ð2Þ
where EUV is the applied potential (V) for UV lamp; COD0 and CODt are the chemical oxygen demands (g/L) of the solution before and after the time t of the treatment (measured according to the standard method D.5220); and V is the volume of the electrolyte (1 L). The photocatalytic process using TiO2-C and TiO2-CNT are abbreviated as PC and PCNT, respectively.
software version 16.2.4. According to the obtained results from the classical experiments, the variables and their values were determined and reported in Table 2. Initial dye concentration (10, 20, 30, 40 and 50 mg/L), pH (2–6), TiO2 dosage (0.01, 0.02, 0.03, 0.04 and 0.05 g/L) and time (90, 120, 150, 180 and 210 min) were considered as the process variables. AR14 removal (%), AB92 removal (%), COD removal (%) and electrical energy consumption (EEC) were the responses of the degradation procedure. Results and discussions Catalyst characterization
Characterization of catalyst The Fourier transform infrared (FT-IR) spectroscopy of carbon and CNT-carbon plates before and after the deposition of TiO2 NP, and also after the dark adsorption of AR14 were analyzed with a Thermo Nicolet Avatar 360 FT-IR Spectrometer within the range of 500–4000 cm1. The surface morphology of the mentioned substrates was investigated using a field emission scanning electron microscope ((FESEM) JSM-6700F, JEOL, Japan). RSM As a closer approach to simulate the industrial wastewater containing various types of dye molecules, the removal efficiency of the established system was evaluated in binary solutions containing the mixture of two dyes (AB92 and AR14). Also, for more precise and adequate study of the process and to examine the cost effectiveness of the photocatalytic decolorization, the CentralComposite Design (CCD) was used to study the dye degradation in solutions containing two dyes using PCNT by means of Minitab
The FESEM images of carbon plates before and after the deposition of CNT and immobilization of TiO2 NP are shown in Fig. 2. According to these images, carbon plates are successfully fabricated with nanotubes after the electrodeposition. In addition, it can be seen that the TiO2 NP are immobilized on both substrates after the evaporation of water [37]. The corresponding high magnified FESEM images (c0 and d0 ) reveal that titania NP have the diameter of about 33 nm which implies that the photocatalyst NP did not aggregate during the immobilization process. This indicates the minimum impact on the reduction of specific surface area of TiO2 NP after being immobilized. Fig. 3 represents the FTIR spectra of carbon, CNT fabricated carbon, TiO2-C and TiO2-CNT. The broad peak appearing in all samples at 3430 cm1 is related to the OH stretching bond due to the water absorption of carbon materials and TiO2 NP. The weak peak observed in carbon and CNT at 1725 cm1 can be attributed to the C¼O bond as the impurity. The corresponding peaks appearing at 802, 1625 and 2028 cm1 for carbon plates can be assigned to C H (alkene bending), C¼C and CRC, respectively. The peaks at 2923 and 2853 cm1 can be
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
5
Fig. 2. FESEM images of (a, a0 ): carbon substrate, (b, b0 ): CNT fabricated carbon, (c, c0 ): TiO2-C, and (d, d0 ): TiO2-CNT at different magnitude.
related to the symmetric alkane stretching of C H bond. The broad peak in the range of 500–800 cm1 related to the combination vibration of Ti O Ti and Ti O C bonds is
appeared in TiO2-C and TiO2-CNT [38]. Appearance of a peak at 1631 cm1 in TiO2 NP and TiO2 containing substrates is assigned to the deformation vibration of HO H bond from the water
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
6
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Fig. 3. FTIR spectra of carbon, CNT fabricated carbon, TiO2-C, and TiO2-CNT.
adsorption on TiO2 NP. The peaks at 1461 and 1267 cm1 in C-CNT sample that correspond to CH2 bending and C O have shifted to the lower wavenumbers (1444 and 1258 cm1) after the immobilization of TiO2 NP in TiO2-CNT, which may be because of physical interactions between CNT and TiO2 NP after the solvent evaporation [34,39]. Photocatalytic activity tests In the present study, initial dye concentration, pH of the solution and TiO2 loading were changed to investigate the photocatalytic activity of the developed catalysts for the singlecomponent containing solutions. TiO2 loading The effect of TiO2 loading (0.01, 0.02, 0.04 and 0.08 g) on the carbon and CNT fabricated carbon was examined on PC and PCNT processes for the removal of AR14, AB92 and AY117 and the results are shown in Fig. 4. The results indicate that by increasing the TiO2 loading from 0.01 to 0.04 g, the removal efficiencies of AR14, AB92 and AY117 increase from 65%, 60% and 86% to 86%, 68% and 91%, respectively, but further increasing of TiO2 loading will decrease the dye removal (%). This can be explained by the aggregation of TiO2 NP on the surface of carbon materials which results in the lower surface area of nanoparticles and consequently, in the lower performance of the procedure. This result is in great agreement with other studies [40]. The FESEM images of the aggregation of photocatalyst nanoparticles at high dosage are shown in Fig. 5((a, a0 ) and (b,b0 )). However, TiO2 NP show lower aggregation on CNT rather than carbon, which could be due to the higher surface area provided by the nanotubes for the immobilization of TiO2 NP. In case of using the suspension of carbon based-TiO2 composite materials, it has been reported that higher dosages of composite will limit the UV light transmission due to the increase of turbidity of solution; but by using immobilized composite, we overcome this disadvantage. Although, it should be considered that the immobilization process itself decreases the specific surface area of TiO2 to some extent [40]. One of the noticeable advantages of this photocatalytic system is the significantly low dosages of TiO2 NP (0.04 g/L) for
decolorization, which is more than 12 times lower than the optimum dosages in case of the usage of suspension TiO2 NP in most studies (0.5 g/L). Moreover, the continuous treatment of wastewater was possible at extremely low photocatalyst concentration along with the high removal efficiencies [40]. It should be mentioned that lower initial dye concentrations were investigated for AB92 rather than AR14 and AY117 concerning the darker hue and color of AB92, which diminishes the penetration of UV irradiation through the solution to reach the immobilized TiO2. Fig. 5((c,c0 ) and (d,d0 )) also represents the FESEM images of TiO2-C and TiO2-CNT after dark adsorption of AR14. It shows that dye adsorption of nanotubes is higher than carbon plates which attracts dye molecules to concentrate around the photocatalyst NP. This makes them highly accessible to the oxidative radicals formed on the surface of catalyst for faster degradation. On the other hand, the FTIR spectra of the dark adsorption of AR14 on TiO2-C and TiO2CNT are drawn in Fig. 6. The peaks appearing at 1189 and 1047 cm1 are related to the existence of sulfonate (S¼O) and C O groups, and the strong peak at 760 cm1 is corresponding to the S O group in the structure of AR14, which are evident in the spectra of TiO2-C and TiO2-CNT after the adsorption. The intensity of the mentioned peaks are higher at TiO2-CNT rather than TiO2-C, because of the higher adsorption capacity of CNT over carbon plate. Moreover, increasing the light absorption and higher capability of CNT in adsorption of dye molecules to concentrate them around the photocatalyst NP could be responsible for the higher decolorization rate of PCNT at initial stages of the process rather than PC at the same condition [41,42]. pH Fig. 7 displays the effect of pH variation on decolorization efficiencies of AR14, AB92 and AY117 after 210 min of process. It is clear in all cases that the photodecolorization is at its maximum value at pH 3. This phenomenon is related to the higher adsorption tendency between the negatively charged dye molecules and positively charged TiO2 NP and carbonaceous materials as reported in other studies [43]. As pH of the solution increases from 3 to 11, the removal efficiencies of AR14, AB92 and AY117 diminishes nearly about 50%, 27% and 32%, respectively after 210 min of UV irradiation. The less efficient decolorization can be explained by
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
7
Fig. 4. Effect of TiO2 loading on the removal efficiency ([AR14]: 50 mg/L, [AY117]: 50 mg/L, [AB92]: 20 mg/L, pH: 3, reactor volume: 1 L, and flow rate: 0.33 L/h).
the fact that the photocatalyst NP and carbon based substrates become more negative at higher pH values and the repulsion forces between the adsorbent and dye molecules lead to less concentration of anionic dyes around TiO2 which consequently decreases the photocatalytic performance of the system [44]. Initial dye concentration The catalyst activity was also studied by changing the initial dye concentration from 50 to 200 mg/L for AR14 and AY117, and from 50 to 200 mg/L for AB92 because of the low penetration of UV irradiation to reach the photocatalyst NP. All results in Fig. 8 clearly imply that by increasing the concentration of dyes from
50 to 200 mg/L, the process efficiency decreases from 91% and 68% to 48% and 23% for AY117 and AR14, respectively. In case of AB92, the removal efficiency decreased from 68% to 17% by increasing the dye concentration from 20 to 50 mg/L. Also, in order to verify our explanations regarding the darker hue of AB92, the decoloration of another acidic dye (C.I. Acid Black 26) at 75 mg/ L, was investigated and the result was similar to the obtained data earlier. Generally, when the concentration of dye increases, the photon penetration through dye molecules to reach the catalyst surface decreases, which leads to the formation of lower amount of hydroxyl radicals that degrades the dye molecules structure.
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
8
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Fig. 5. FESEM images of (a, a0 ): TiO2-C (TiO2: 0.08 g), (b, b0 ): TiO2-CNT (TiO2: 0.08 g), (c, c0 ): TiO2-C after the dark adsorption of AR14 (TiO2: 0.04 g), and (d, d0 ): TiO2-CNT after the dark adsorption of AR14 (TiO2: 0.04 g) at different magnitude.
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
9
Fig. 6. FTIR spectra of AR14 before and after the dark adsorption on carbon and CNT fabricated carbon plates.
Another reason can be the extreme concentration and aggregation of the adsorbed dye molecules on the surface of TiO2 and carbon substrates, which will reduce the adsorption of UV light and further formation of oxidative radicals. But, this performance drop is less noticeable in PCNT rather than PC, which is probably related to the higher dye adsorption capacity of nanotubes that was confirmed by the FESEM images in Fig. 5(d,d0 ) [32,40]. It is noteworthy that the photocatalytic processes are usually designed for the pretreatment or post treatment of effluents either to degrade the pollutants to less complex compounds or to degrade the trace contaminants at very low concentrations, respectively. The promising results in this study achieved at fairly high concentrations of dye solutions, can offer such a system which benefits from an advanced nanotechnology to textile industry. Stability of TiO2-C and TiO2-CNT To evaluate the photochemical stability of the immobilized catalyst (TiO2-C and TiO2-CNT) in assessing the practical application of the photocatalyst plates in wastewater treatment, the recycle experiments for the photodegradation of AR14 were conducted and the results are shown in Fig. 9. The mean removal (%) of AR14 after 10 and 14 cycles of PC and PCNT processes are 60% and 68%, which decrease to 49% and 62% after 13 and 16 photocatalytic cycles. This means that the titania NP benefit from appropriate photochemical stability on the surface of the carbonaceous substrates. Fig. 9 also indicates that the photocatalyst NP are more stable on the CNT fabricated carbon plates and have lower decreasing rate of efficiency in comparison with PC, which could be the result of higher physical interaction between CNT and TiO2 NP. The FESEM images in Fig. 10 certify the obtained results by showing the existence of titania NP on the surface of the substrates after 13 and 16 successive cycles. The overall results can suggest the cost-effective utilization of the prepared photocatalyst plates in the practical applications due to their high activity and stability. It should be noticed that in this study, the repetitive utilization of photocatalyst for 16 cycles in a continuous process was applicable by a simple and facile procedure, while in other studies, the catalysts are usually shown to be reusable for approximate six cycles [37,43,45].
Fig. 7. Effect of pH on the removal efficiency ([AR14]: 50 mg/L, [AY117]: 50 mg/L, [AB92]: 20 mg/L, TiO2: 0.04 g, Time: 210 min, reactor volume: 1 L, and flow rate: 0.33 L/h).
RSM results The acquired results and data indicated that the more adsorption of the contaminants onto the CNT rather that carbon followed by the generation of oxidative radicals on the surface of TiO2 have improved the photocatalytic activity; and TiO2-CNT revealed higher stability in the repetitive catalytic cycles. Thus, RSM was only applied to PCNT process in binary dye solutions as an approach to the real textile wastewater that generally consists of more than one component. A four-factor, five-level CCD was conducted for the degradation of two acid dyes in a binary dye solution. Table 2 shows the selected parameters, their levels and the corresponding responses for all the conditions based on the response surface model. Analysis of variance (ANOVA) and P-value of each parameter are represented in Table 3.
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
10
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Fig. 8. Effect of initial dye concentration on the removal efficiency (pH: 3, TiO2: 0.04 g, Time: 210 min, reactor volume: 1 L, and flow rate: 0.33 L/h).
The second-order regression equation that relates the response functions to the selected variables is written in Eq. (3): Y i ¼ b0 þ
4 4 3 X 4 X X X b i xi þ bii xi 2 þ bij xi xj i¼1
i¼1
ð3Þ
i¼1 j¼iþ1
where Yi is the response variable, b0 the constant coefficient, bi the regression coefficients for linear effects, bii the quadratic coefficients, bij the interaction coefficients and xi,xj are the coded values of input factors [46,47]. The obtained regression models for the responses are also represented in Table 3. The variables which P-values of 0.05 and/or lower are assigned to, are significant to the response function. The results explore that the quadratic model is statistically significant to the responses and it can be adopted to describe the behavior of each response function. The values of R2 and adjusted R2 are high and the difference between the adjusted R2 and predicted R2 are low in most cases, which means that the model can be applicable for future outcomes. Furthermore, the lack-of-fit terms are not significant determined by their high P-values.
Fig. 9. Stability of TiO2-C and TiO2-CNT after 13 and 16 cycles, respectively.
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
11
Fig. 10. FESEM images of (a, a0 ): TiO2-C, (b, b0 ): TiO2-C after 13 photocatalytic cycles, (c, c0 ): TiO2-CNT and (d, d0 ): TiO2-CNT after 16 photocatalytic cycles (TiO2: 0.04 g) at different magnitude.
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
12
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Table 3 Quadratic model and the related ANOVA. Source
AR14 removal (%) P-value
AB92 removal (%)
COD removal (%)
EEC (KWh/gCOD)
Model Time pH TiO2 [C0] Time*time pH*pH TiO2*TiO2 [C0]*[C0] Time*pH Time*TiO2 Time*[C0] pH*TiO2 pH*[C0] TiO2*[C0] Lack-of-fit R2 (%) R2 adj. (%) R2 pred. (%) Quadratic models
0 0.016 0 0 0 0.492 0.016 0 0.014 0.523 0.522 0.814 0.002 0.121 0.01 0.0.095 99.21% 98.53% 95.48%
0 0.041 0 0.01 0 0.025 0 0.001 0 0.498 0.762 0.988 0.098 0.734 0.364 0.372 94.08% 88.90% 66.07%
0 0.014 0 0.001 0 0.564 0.025 0 0.016 0.666 0.668 0.536 0.719 0.277 0.361 0.491 98.12% 96.47% 92.15%
0 0 0 0 0 0.857 0 0 0 0.393 0.527 0.289 0.006 0.654 0.688 0.073 98.88% 97.91% 94.18%
According to the three dimensional surface plots in Fig. 11, low adsorption tendency between anionic dye molecules and positively charged TiO2 NP, carbon and CNT caused by increasing the pH values has resulted in the lower dye and COD removal (%), which has consequently increased the electrical energy consumption. Also, the existence of a maximum peak for TiO2 dosage clearly indicates that after a specific dosage of catalyst, the aggregation of nanoparticles reducing their specific surface area diminishes the dye and COD removal (%). It can be seen from the results that the EEC per unit COD decay are low (below 3.5 (kWh/ gCOD)). The overall results and changes of removal efficiencies obtained by RSM are in great agreement with the classical experiments. Optimization In order to find the optimal conditions of removal process to maximize the removal of AR14, AB92 and COD removal (%) while consuming the minimum electrical energy, the desirability function optimization has been employed for multi-response optimization. The predicted values of parameters to reach the desired values of responses are summarized in Table 4. The experimental data (the average of three experiments) illustrate that there is an adequate agreement between the experimental data and predicted values, which can be another reason for a proper fit of the quadratic model on the obtained results.
Flow rate and hydrogen peroxide Fig. 12 shows the effect of increasing the solution flow rate fed to the reactor on photocatalysis efficiency at the optimum condition obtained in Table 4. The results demonstrate that by increasing the flow rate from 0.33 to 0.99 L/h, the removal efficiency decreases about 30%, which is due to the reduction of the retention time of dye molecules in the reactor that lowers the reactions between dye molecules and the oxidative agents formed in the solution for dye degradation [48,49]. The addition of hydrogen peroxide as an important parameter in the photocatalytic processes was also investigated in Fig. 12. Various amounts of H2O2 (12.5%, 25%, 37.5%, and 50% of the stoichiometric dosage theoretically required for total mineralization of dyes) were added continuously to the reactor. The stoichiometric dosages of hydrogen peroxide for total mineralization of AR14 and AB92 were calculated according to the following reactions: C20H12N2O7S22Na + 51H2O2 ! 20CO2 + 54H2O + 2H2SO4 + Na2O + 2HNO3
(4)
C26H16N3O10S33Na + 68H2O2 ! 26CO2 + 143/2H2O + 3H2SO4 + 3/ 2Na2O + 3HNO3 (5) The results reveal that the addition of H2O2 to the system has significantly promoted the photocatalytic activity to reach a
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
13
Fig. 11. Three dimensional surface plots from RSM results.
complete decolorization after 100 min. It has been reported that H2O2 can inhibit the recombination of electrons and holes (e/h+), while further provides additional hydroxyl radicals as described by the following mechanisms: H2O2 + hy ! 2HO
destructive hydroxyl radicals formed through Eqs. (10)–(12) [50,51]:
HO2 þ HO ! H2 O þ O2
ð10Þ
2HO ! H2O2
(11)
(6)
H2O2 ! HO2 + 2H+
(7)
HO þ H2 O2 ! HO2 þ H2 O
HO2 þ hy ! HO þ
1=2O 2
O 2 þ H2 O2 ! O2 þ HO þ HO
ð8Þ
ð9Þ
On the other hand, further addition of H2O2 was found to decrease the reaction rate owing to the formation of less
ð12Þ
Conclusion The present study proposed a novel, simple, inexpensive, efficient and stable immobilization of titania NP on the surface of bare and CNT fabricated carbon plates to be employed as photocatalytic plates in a continuous photo-reactor for the
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
14
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Table 4 Predicted and experimental optimized values by response optimizer. Desired (%) AR14 removal (%) AB92 removal (%) COD removal (%) EEC (KWh/gCOD) Optimum condition
Predicted (%)
65–70 (maximize) 71.38 55–60 (maximize) 60.05 50–55 (maximize) 55.67 0.350–0.400 (minimize) 0.338 Time: 108.18 min, pH: 2.85, TiO2: 0.03 and [C0]: 25.95 mg/L
Experimental (%) 69.02 62.45 54.97 0.484
Fig. 12. Effect of flow rate and H2O2 concentration on the removal efficiency at RSM optimum condition.
degradation of C.I. Acid Red 14, C.I. Acid Blue 92, and C.I. Acid Yellow 117. In this regard several important subjects were taken into account: 1. The popular and commercially available TiO2 NP were selected as the photocatalyst. 2. The toxicity and environmental hazards of the discharge of nanomaterials into water streams was considered and therefore, the immobilization of photocatalyst were performed. 3. The available, low-cost, chemically inert and stable carbon and CNT materials with extraordinary adsorption properties were chosen as substrates for the immobilization purpose. 4. A novel, green and facile electrodeposition method was employed for the fabrication of plates with CNT. 5. The solvent evaporation technique was performed as a cost effective, simple with no chemicals involved in the process to immobilize TiO2 NP. 6. Applying photocatalysis in nanoscale in a continuous wastewater treatment process.
TiO2 NP exhibited strong adherence on both carbon and CNT support, and the carbonaceous substrates showed high adsorption affinity towards dye molecules especially CNT. The characterization of the photocatalytic plates were inquired using FESEM images and FTIR analysis, determining the successful and stable immobilization of nanoparticles. The FESEM images, FTIR spectra and stability tests showed that the nanotubes provide high surface area for TiO2 NP to be immobilized without agglomeration along with adequate adhesion to CNT which increased the quantity of dyes to come in contact with TiO2 through adsorption. The catalyst activity was examined by studying the influence of effective parameters such as pH, initial dye concentration, TiO2 dosage and time through classical (single dye solution) and statistical RSM (binary dye solution to get closer to simulate the real textile effluent). The result of the CCD experiments led to the precise fitness of quadratic models that related the variables to the selected responses (dye and COD removal (%) and electrical energy consumption). The multi-response optimization was also conducted and the predicted values by response optimizer expressed a great agreement with
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
the experimental data, which resulted in high removal efficiencies at very low dosages of TiO2 NP and low electrical energy consumption. The fast and complete decolorization were accomplished by the continuous H2O2 feeding system embedded in the reactor.
[23] [24] [25] [26] [27] [28] [29]
References
[30]
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]
X. Qu, P.J.J. Alvarez, Q. Li, Water Res. 47 (2013) 3931. S. Li, L.K. Wallis, H. Ma, S.A. Diamond, Sci. Total Environ. 466–467 (2014) 800. I. El Saliby, L. Erdei, J.-H. Kim, H.K. Shon, Water Res. 47 (2013) 4115. Y. Luo, J. Liu, X. Xia, X. Li, T. Fang, S. Li, Q. Ren, J. Li, Z. Jia, Mater. Lett. 61 (2007) 2467. Z. Wu, H. Wang, Y. Liu, Z. Gu, J. Hazard. Mater. 151 (2008) 17. S.-H. Lee, S. Pumprueg, B. Moudgil, W. Sigmund, Colloids Surf. B 40 (2005) 93. H. Li, S. Gao, M. Cao, R. Cao, J. Colloid Interface Sci. 394 (2013) 434. B. Wang, F.C. de Godoi, Z. Sun, Q. Zeng, S. Zheng, R.L. Frost, J. Colloid Interface Sci. 438 (2015) 204. M.A. Nawi, S. Sabar, J. Sheilatina, Colloid Interface Sci. 372 (2012) 80. A. Carabin, P. Drogui, D. Robert, J. Taiwan Inst. Chem. Eng. 54 (2015) 109. H. Yu, X. Quan, S. Chen, H. Zhao, J. Phys. Chem. C 111 (2007) 12987. H.U. Farouk, A.A.A. Raman, W.M.A.W. Daud, J. Ind. Eng. Chem. 33 (2016) 11. Y.-J. Kim, Y. Horie, Y. Matsuzawa, S. Ozaki, M. Endo, M.S. Dresselhaus, Carbon 42 (2004) 2423. C.-Y. Kuo, J. Hazard. Mater. 163 (2009) 239. L. Guardia, S. Villar-Rodil, J.I. Paredes, R. Rozada, A. Martínez-Alonso, J.M.D. Tascón, Carbon 50 (2012) 1014. S.C. Smith, D.F. Rodrigues, Carbon 91 (2015) 122. M.A. Andrade, A.S. Mestre, R.J. Carmona, A.P. Carvalho, C.O. Ania, Appl. Catal. A 507 (2015) 91. J. Cho, S. Schaab, J. Roether, A. Boccaccini, J. Nanopart. Res. 10 (2008) 99. W. Wang, P. Serp, P. Kalck, J.L. Faria, Appl. Catal. B 56 (2005) 305. P. Serp, M. Corrias, P. Kalck, Appl. Catal. A 253 (2003) 337. K. Woan, G. Pyrgiotakis, W. Sigmund, Adv. Mater. 21 (2009) 2233. A. Matsumoto, K. Tsutsumi, K. Kaneko, Langmuir 8 (1992) 2515.
[31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51]
15
A. Omri, M. Benzina, F. Bennour, J. Ind. Eng. Chem. 21 (2015) 356. J. Matos, J. Laine, J.-M. Herrmann, Appl. Catal. B 18 (1998) 281. I. Zhitomirsky, L. Gal-Or, A. Kohn, H.W. Hennicke, J. Mater. Sci. 30 (1995) 5307. B. Tryba, A.W. Morawski, M. Inagaki, Appl. Catal. B 41 (2003) 427. K.E. Tettey, M.Q. Yee, D. Lee, ACS Appl. Mater. Interfaces 2 (2010) 2646. H. Meng, W. Hou, X. Xu, J. Xu, X. Zhang, Particuology 14 (2014) 38. H. Wu, J. Fan, E. Liu, X. Hu, Y. Ma, X. Fan, Y. Li, C. Tang, J. Alloys Compd. 623 (2015) 298. S. Minnikanti, N. Peixoto, in: Jose Mauricio Marulanda (Ed.), Carbon Nanotubes Applications on Electron Devices, InTech, 2011. L. Eskandarian, E. Pajootan, M. Arami, Ind. Eng. Chem. Res. 53 (2014) 14841. E. Pajootan, M. Arami, M. Rahimdokht, Ind. Eng. Chem. Res. 53 (2014) 16261. R.D.C. Soltani, A. Rezaee, A.R. Khataee, M. Safari, J. Ind. Eng. Chem. 20 (2014) 1861. E. Pajootan, M. Arami, Electrochim. Acta 112 (2013) 505. D. Hornero-Méndez, M.I. Mínguez-Mosquera, J. Agric. Food. Chem. 49 (2001) 3584. E. Pajootan, M. Arami, N.M. Mahmoodi, J. Taiwan Inst. Chem. Eng. 43 (2012) 282. B. Appavu, K. Kannan, S. Thiripuranthagan, J. Ind. Eng. Chem. 36 (2016) 184. A.H. Noroozi, S. Safa, J. Adv. Mater. Proc. 2 (2014) 31. S.-J. Bao, C.M. Li, J.-F. Zang, X.-Q. Cui, Y. Qiao, J. Guo, Adv. Funct. Mater. 18 (2008) 591. H. Zangeneh, A.A.L. Zinatizadeh, M. Habibi, M. Akia, M. Hasnain Isa, J. Ind. Eng. Chem. 26 (2015) 1. M.-Q. Yang, N. Zhang, Y.-J. Xu, ACS Appl. Mater. Interfaces 5 (2013) 1156. D. Feng, S. Xu, G. Liu, Chemosphere 125 (2015) 102. J.R. Kim, E. Kan, J. Ind. Eng. Chem. 21 (2015) 644. I.K. Konstantinou, T.A. Albanis, Appl. Catal. B 49 (2004) 1. H. Guo, Z. Zheng, J. Chen, W. Weng, M. Huang, J. Ind. Eng. Chem. 36 (2016) 306. E. Errais, J. Duplay, F. Darragi, I. M'Rabet, A. Aubert, F. Huber, G. Morvan, Desalination 275 (2011) 74. M. Fathinia, A.R. Khataee, J. Ind. Eng. Chem. 19 (2013) 1525. Y. Li, J. Chen, J. Liu, M. Ma, W. Chen, L. Li, J. Environ. Sci. 22 (2010) 1290. E. Pajootan, M. Arami, M. Rahimdokht, Sep. Purif. Technol. 130 (2014) 34. M.N. Chong, B. Jin, C.W.K. Chow, C. Saint, Water Res. 44 (2010) 2997. S. Rodriguez, A. Santos, A. Romero, Desalination 280 (2011) 108.
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Journal of Industrial and Engineering Chemistry journal homepage: www.elsevier.com/locate/jiec
Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules Elmira Pajootan* , Mehdi Rahimdokht, Mokhtar Arami Textile Engineering Department, Amirkabir University of Technology, 424 Hafez Ave., Tehran 15875-4413, Iran
A R T I C L E I N F O
A B S T R A C T
Article history: Received 24 August 2016 Received in revised form 20 June 2017 Accepted 21 June 2017 Available online xxx Keywords: TiO2 nanoparticles Electrodeposition Immobilization Solvent evaporation Response surface methodology
We utilize carbon nanotubes fabricated carbon plates with large surface area and high adsorption capacity as the substrate to immobilize TiO2 nanoparticles without any agglomeration using solvent evaporation method. The photocatalytic activity is investigated in a continuous photocatalytic reactor to degrade three acid dyes, which is shown to be stable for 16 successive cycles. Response surface methodology (RSM) is used to optimize and model the binary system for a closer simulation of real industrial wastewater with the electrical energy consumption below 0.5 KWh/gCOD. A fast and complete decolorization was achieved in 100 min by adding H2O2 to the system. © 2017 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
Introduction Nowadays, nanotechnologies are introduced as a preferable solution to current water and wastewater treatment methods to improve the water quality to meet the environmental standards and to conquer the water scarcity resulted from the human activities in both developing and industrialized countries. Titania nanoparticles (TiO2) are one of the most studied semiconductors for the degradation of pollutants existing in water streams, and also in many other applications, such as sunscreen, surface coating, solar cells, environmental remediation, etc. due to their semiconducting properties and photostability [1–5]. When TiO2 is irradiated by UV light, the photoinduced electrons and holes (e/h+) are formed. OH and H2O as electron donors and oxygen as electron acceptor will form the hydroxyl (OH) and
superoxide ion radicals (O 2 ). These generated radicals especially hydroxyl radicals are nonselective strong oxidants which can fully mineralize organic pollutants [6,7]. One of the major drawbacks of the usage of TiO2 nanoparticles (NP) is the problem encountered through their separation and removal from the treated solutions. The UV absorption of these photoactive and nano-sized materials has a potential toxicity to
* Corresponding author. E-mail address: [email protected] (E. Pajootan).
organisms and environment, to which many investigations are dedicated in order to survey the phototoxicity and dramatic effect of TiO2 NP. Therefore, immobilization of these nanoparticles is now the subject of most enquiries pertaining to the area of this science [1–4,8–10]. In addition, the practical application of slurry TiO2 solutions can be very challenging in the continuous wastewater treatment processes. Furthermore, the application of TiO2 NP for photocatalytic processes suffers from a low quantum efficiency because of the rapid recombination of the photoinduced electrons and holes (e/h+), mass transfer and photon transfer limitations [11,12]. Several methods have been developed to enhance the photocatalytic efficiency; in this regard, carbon nanotubes (CNT) with large specific surface area and excellent electrical properties have received tremendous attention for building the composite materials. These nanotubes can also be used as supporting material for TiO2 NP owing to their nanomorphology, high mechanical and chemical stability and high structural integrity [4,11,13–17]. This combination can significantly improve the photocatalytic performance by lowering the recombination of e/h+ couples, and the combination of their electronic, adsorption, mechanical and thermal properties [18–21]. Various techniques have been reported to coat titania NP on carbon-based supports for different purposes such as: hydrolysis of TiCl4 [22], deposition of TiO2 by sol-gel method [23], mixing TiO2 with carbon materials [24], electrodeposition [25], precipitating
http://dx.doi.org/10.1016/j.jiec.2017.06.039 1226-086X/© 2017 The Korean Society of Industrial and Engineering Chemistry. Published by Elsevier B.V. All rights reserved.
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
2
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
[26,27], hydrothermal treatment [28,29], solvent evaporation [30], etc. Nowadays, response surface methodology (RSM) is one of the most popular experimental design and optimization tools being used in a wide range of applications especially wastewater treatment. Reducing the number of experiments while maintaining the accuracy of results, response surface modeling through regression, process optimization and prediction of the variables’ values, exploration the interactions of various combinations of parameters are some of the advantages of RSM technique [31–33]. In this research, we successfully demonstrated the immobilization of TiO2 NP on carbon-based material to decolorize the simulated textile wastewater by photocatalytic process in a continuous reactor. For this purpose, the relatively low-cost and available carbon-based materials and TiO2 NP, with their specific characterizations were taken into consideration. First, CNT were electrodeposited on the surface of the carbon plates via a method reported in our previous study [34]. Then, a solution containing dispersed TiO2 NP were added to the surface of the carbon plates (TiO2-C) and CNT fabricated carbon plates (TiO2-CNT). Finally after the evaporation of the solvent, TiO2 NP were immobilized on the plates. FESEM images and FTIR spectra indicated the characteristics of the prepared plates. The prepared plates were successfully employed for the continuous decontamination of wastewater with appropriate stability for several cycles. The decolorization and further degradation of the contaminated solutions containing C.I. Acid Red 14, C.I. Acid Blue 92 and C.I. Acid Yellow 117 were investigated and the effect of important parameters including pH, initial dye concentration and TiO2 dosage was demonstrated. RSM was applied to design the experiments to evaluate the effect of parameters in the degradation of dyes from binary solutions and to multi-optimize the degradation process considering maximum dye and COD removal (%), as well as minimum electrical energy consumption as the responses to develop a cost effective system. The variation of dye removal efficiency by changing the flow rate and addition of hydrogen peroxide was evaluated as well.
Experimental Materials In all experiments, the reagents were of analytical grade. Titania NP (Degussa P25) was employed as the photocatalyst (average particle size: 30 nm, purity >97% with 80:20 anatase to rutile). Cetyl Trimethyl Ammonium Bromide ((C16H33)N(CH3)3Br, CTAB) and MWCNT (purity >95%, length 10–20 mm and diameter 30–50 nm) were purchased from Merck and Neutrino, respectively. C.I. Acid Red 14 (AR14), C.I. Acid Blue 92 (AB92) and C.I. Acid Yellow 117 (AY117) (Ciba Co.) which are widely used in textile industry, were employed as synthetic dyes (Table 1). H2SO4 (1 M) and NaOH (1 M) (Merck) were used to adjust the pH of dye solutions. Carbon substrates (purity >95%) with the dimension of 50 110 3 mm3 used as the supporter for titania naoparticles were purchased from Seraaj Co. The hydrogen peroxide (H2O2, 30%) was supplied from Loba Chemie. Immobilization of TiO2 NP The carbonaceous support (Carbon and CNT fabricated carbon substrates) were selected as substrates for the immobilization of TiO2 NP due to their excellent adsorption properties and chemical inertness, especially CNT. The fabrication of carbon plates using CNT by a novel, facile and inexpensive electrodeposition technique was fully explained in a separate investigation in our previous study [34]. In summary, the carbon plates were abraded with sand paper and washed with distilled water. Then each electrode was pretreated with 50 mL of NaOH (10% w/v), 50 mL of HNO3 (50%, v/v) and 50 mL of acetone, each step for 5 min, respectively, and then they were rinsed with distilled water for three times. The electrodeposition procedure to form a thin layer of CNT on the surface of carbon plates was performed by applying the DC voltage of 17.5 (V) to a 200-mL solution containing 0.06 g of CNT and 0.04 g of CTAB which had been sonicated for 60 min using Delta D68H
Table 1 Dyes characteristics. C.I. generic name
lmax (nm)
Molecular weight (g/mol)
Formula
C.I. Acid Red 14
515
502.42
C20H12N2O7S22Na
C.I. Acid Blue 92
572
695.58
C26H16N3O10S33Na
C.I. Acid Yellow 117
440
848.83
C39H30N8O8S22Na
Structure
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
3
Fig. 1. Preparation of the photocatalytic plates and a cross section of the photo-reactor.
Ultrasonic prior to the deposition. The positively charged CNT (by the arranged cationic hydrophilic group of CTAB on their surface) would transfer from the bulk to the surface of the negatively charged cathode electrode to form a uniform layer on its surface. The modified electrodes were washed with distilled water and dried in oven for 15 min at 60 C. The structural and electrochemical properties of these carbon plates before and after the fabrication process were also demonstrated and fully reported in our previous study [34]. In order to immobilize TiO2 NP by the solvent evaporation as a fast, simple and cost effective method with no use of chemicals, the carbon plates before and after the fabrication process were placed in oven at 110 C. The specific amount of TiO2 NP (0.01, 0.02, 0.04 and 0.08 g) were dispersed for 30 min in 30 mL distilled water and then dropped on the surface of six carbonaceous hot plates. After the evaporation of water, TiO2 NP formed a uniform layer on the surface of the carbon (TiO2-C) and CNT fabricated carbon substrates (TiO2-CNT). Experimental set-up The dye removal process was carried out in a continuous cylindrical reactor made of Plexiglas having the working volume of
1 L. Six fabricated TiO2-C or TiO2-CNT plates were placed in the cylindrical photo-reactor in the shape of a regular hexagon. A cross section view of the photo-reactor employed for the decolorization is shown in Fig. 1. The UV lamp (Philips, 9 W) in the quartz tube was placed at the center of the reactor to provide near UV radiation; and the aeration output was placed at the center bottom of the reactor. The reactors were placed in an aluminum box to prevent the illumination of UV lamp to the room. The dark adsorption of dyes onto TiO2-C and TiO2-CNT were performed in the same reactors for 30 min excluding the UV irradiation to establish an adsorption–desorption equilibrium and then the UV lamp was turned on to begin the photocatalytic process. The samples were collected over the specific time intervals and the absorbance was measured using UNICO 2100 Spectrophotometer at the maximum wavelength of each dye. Dye removal efficiency was determined according to Eq. (1): DR% = [(A0 A)/A0] 100
(1)
where A0 and A are the absorbance of dye solution before and after the treatment, respectively. For the binary solutions, two equations and two readings of absorbance at two different wavelengths employing the extension of Beer–Lambert law were used to calculate the concentration of each dye in the solution [35,36].
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
4
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Table 2 CCD design and the obtained results. Run order
Time (min)
pH
TiO2 (g/L)
[C0] (mg/L)
AR14 removal (%)
AB92 removal (%)
COD removal (%)
EEC (KWh/gCOD)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
210 180 150 150 120 180 120 120 90 150 120 150 150 150 150 150 150 120 180 150 150 120 180 180 120 150 120 180 150 180 180
4 5 2 4 5 5 3 5 4 4 3 4 4 4 4 4 4 5 3 6 4 3 5 5 5 4 3 3 4 3 3
0.03 0.04 0.03 0.03 0.02 0.02 0.02 0.04 0.03 0.01 0.02 0.03 0.05 0.03 0.03 0.03 0.03 0.04 0.04 0.03 0.03 0.04 0.04 0.02 0.02 0.03 0.04 0.02 0.03 0.02 0.04
30 40 30 30 40 20 40 40 30 30 20 10 30 30 30 30 30 20 20 30 30 40 20 40 20 50 20 40 30 20 40
55.84 27.06 82.28 52.22 19.09 27.52 68.36 22.21 49.69 33.45 68.26 53.12 46.51 52.16 51.90 52.16 52.11 35.96 73.68 12.10 52.06 62.43 37.22 22.86 26.11 40.99 69.09 66.50 52.11 70.47 63.03
66.21 39.24 70.42 66.92 38.58 45.39 45.28 38.67 59.37 56.19 57.86 61.85 58.70 66.71 65.67 67.13 67.10 47.40 71.72 26.45 67.14 55.97 53.25 39.39 43.55 47.64 66.23 53.52 67.10 59.07 60.98
45.31 21.42 67.31 46.18 13.61 28.12 47.10 18.00 43.44 28.69 49.37 51.04 37.95 43.46 43.36 46.28 41.70 26.91 61.26 13.81 40.72 49.64 31.52 17.40 21.15 29.13 58.62 49.50 48.05 54.99 52.99
1.475 2.005 0.709 1.034 2.105 3.056 0.608 1.592 0.659 1.664 1.160 2.806 1.258 1.099 1.101 1.032 1.145 2.129 1.403 3.457 1.173 0.577 2.726 2.469 2.708 0.983 0.977 0.868 0.994 1.563 0.811
The electrical energy consumption (EEC) per unit COD decay (kWh/gCOD) was calculated by using Eq. (2): EEC ¼
EUV t 100 V ðCOD0 CODt Þ
ð2Þ
where EUV is the applied potential (V) for UV lamp; COD0 and CODt are the chemical oxygen demands (g/L) of the solution before and after the time t of the treatment (measured according to the standard method D.5220); and V is the volume of the electrolyte (1 L). The photocatalytic process using TiO2-C and TiO2-CNT are abbreviated as PC and PCNT, respectively.
software version 16.2.4. According to the obtained results from the classical experiments, the variables and their values were determined and reported in Table 2. Initial dye concentration (10, 20, 30, 40 and 50 mg/L), pH (2–6), TiO2 dosage (0.01, 0.02, 0.03, 0.04 and 0.05 g/L) and time (90, 120, 150, 180 and 210 min) were considered as the process variables. AR14 removal (%), AB92 removal (%), COD removal (%) and electrical energy consumption (EEC) were the responses of the degradation procedure. Results and discussions Catalyst characterization
Characterization of catalyst The Fourier transform infrared (FT-IR) spectroscopy of carbon and CNT-carbon plates before and after the deposition of TiO2 NP, and also after the dark adsorption of AR14 were analyzed with a Thermo Nicolet Avatar 360 FT-IR Spectrometer within the range of 500–4000 cm1. The surface morphology of the mentioned substrates was investigated using a field emission scanning electron microscope ((FESEM) JSM-6700F, JEOL, Japan). RSM As a closer approach to simulate the industrial wastewater containing various types of dye molecules, the removal efficiency of the established system was evaluated in binary solutions containing the mixture of two dyes (AB92 and AR14). Also, for more precise and adequate study of the process and to examine the cost effectiveness of the photocatalytic decolorization, the CentralComposite Design (CCD) was used to study the dye degradation in solutions containing two dyes using PCNT by means of Minitab
The FESEM images of carbon plates before and after the deposition of CNT and immobilization of TiO2 NP are shown in Fig. 2. According to these images, carbon plates are successfully fabricated with nanotubes after the electrodeposition. In addition, it can be seen that the TiO2 NP are immobilized on both substrates after the evaporation of water [37]. The corresponding high magnified FESEM images (c0 and d0 ) reveal that titania NP have the diameter of about 33 nm which implies that the photocatalyst NP did not aggregate during the immobilization process. This indicates the minimum impact on the reduction of specific surface area of TiO2 NP after being immobilized. Fig. 3 represents the FTIR spectra of carbon, CNT fabricated carbon, TiO2-C and TiO2-CNT. The broad peak appearing in all samples at 3430 cm1 is related to the OH stretching bond due to the water absorption of carbon materials and TiO2 NP. The weak peak observed in carbon and CNT at 1725 cm1 can be attributed to the C¼O bond as the impurity. The corresponding peaks appearing at 802, 1625 and 2028 cm1 for carbon plates can be assigned to C H (alkene bending), C¼C and CRC, respectively. The peaks at 2923 and 2853 cm1 can be
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
5
Fig. 2. FESEM images of (a, a0 ): carbon substrate, (b, b0 ): CNT fabricated carbon, (c, c0 ): TiO2-C, and (d, d0 ): TiO2-CNT at different magnitude.
related to the symmetric alkane stretching of C H bond. The broad peak in the range of 500–800 cm1 related to the combination vibration of Ti O Ti and Ti O C bonds is
appeared in TiO2-C and TiO2-CNT [38]. Appearance of a peak at 1631 cm1 in TiO2 NP and TiO2 containing substrates is assigned to the deformation vibration of HO H bond from the water
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
6
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Fig. 3. FTIR spectra of carbon, CNT fabricated carbon, TiO2-C, and TiO2-CNT.
adsorption on TiO2 NP. The peaks at 1461 and 1267 cm1 in C-CNT sample that correspond to CH2 bending and C O have shifted to the lower wavenumbers (1444 and 1258 cm1) after the immobilization of TiO2 NP in TiO2-CNT, which may be because of physical interactions between CNT and TiO2 NP after the solvent evaporation [34,39]. Photocatalytic activity tests In the present study, initial dye concentration, pH of the solution and TiO2 loading were changed to investigate the photocatalytic activity of the developed catalysts for the singlecomponent containing solutions. TiO2 loading The effect of TiO2 loading (0.01, 0.02, 0.04 and 0.08 g) on the carbon and CNT fabricated carbon was examined on PC and PCNT processes for the removal of AR14, AB92 and AY117 and the results are shown in Fig. 4. The results indicate that by increasing the TiO2 loading from 0.01 to 0.04 g, the removal efficiencies of AR14, AB92 and AY117 increase from 65%, 60% and 86% to 86%, 68% and 91%, respectively, but further increasing of TiO2 loading will decrease the dye removal (%). This can be explained by the aggregation of TiO2 NP on the surface of carbon materials which results in the lower surface area of nanoparticles and consequently, in the lower performance of the procedure. This result is in great agreement with other studies [40]. The FESEM images of the aggregation of photocatalyst nanoparticles at high dosage are shown in Fig. 5((a, a0 ) and (b,b0 )). However, TiO2 NP show lower aggregation on CNT rather than carbon, which could be due to the higher surface area provided by the nanotubes for the immobilization of TiO2 NP. In case of using the suspension of carbon based-TiO2 composite materials, it has been reported that higher dosages of composite will limit the UV light transmission due to the increase of turbidity of solution; but by using immobilized composite, we overcome this disadvantage. Although, it should be considered that the immobilization process itself decreases the specific surface area of TiO2 to some extent [40]. One of the noticeable advantages of this photocatalytic system is the significantly low dosages of TiO2 NP (0.04 g/L) for
decolorization, which is more than 12 times lower than the optimum dosages in case of the usage of suspension TiO2 NP in most studies (0.5 g/L). Moreover, the continuous treatment of wastewater was possible at extremely low photocatalyst concentration along with the high removal efficiencies [40]. It should be mentioned that lower initial dye concentrations were investigated for AB92 rather than AR14 and AY117 concerning the darker hue and color of AB92, which diminishes the penetration of UV irradiation through the solution to reach the immobilized TiO2. Fig. 5((c,c0 ) and (d,d0 )) also represents the FESEM images of TiO2-C and TiO2-CNT after dark adsorption of AR14. It shows that dye adsorption of nanotubes is higher than carbon plates which attracts dye molecules to concentrate around the photocatalyst NP. This makes them highly accessible to the oxidative radicals formed on the surface of catalyst for faster degradation. On the other hand, the FTIR spectra of the dark adsorption of AR14 on TiO2-C and TiO2CNT are drawn in Fig. 6. The peaks appearing at 1189 and 1047 cm1 are related to the existence of sulfonate (S¼O) and C O groups, and the strong peak at 760 cm1 is corresponding to the S O group in the structure of AR14, which are evident in the spectra of TiO2-C and TiO2-CNT after the adsorption. The intensity of the mentioned peaks are higher at TiO2-CNT rather than TiO2-C, because of the higher adsorption capacity of CNT over carbon plate. Moreover, increasing the light absorption and higher capability of CNT in adsorption of dye molecules to concentrate them around the photocatalyst NP could be responsible for the higher decolorization rate of PCNT at initial stages of the process rather than PC at the same condition [41,42]. pH Fig. 7 displays the effect of pH variation on decolorization efficiencies of AR14, AB92 and AY117 after 210 min of process. It is clear in all cases that the photodecolorization is at its maximum value at pH 3. This phenomenon is related to the higher adsorption tendency between the negatively charged dye molecules and positively charged TiO2 NP and carbonaceous materials as reported in other studies [43]. As pH of the solution increases from 3 to 11, the removal efficiencies of AR14, AB92 and AY117 diminishes nearly about 50%, 27% and 32%, respectively after 210 min of UV irradiation. The less efficient decolorization can be explained by
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
7
Fig. 4. Effect of TiO2 loading on the removal efficiency ([AR14]: 50 mg/L, [AY117]: 50 mg/L, [AB92]: 20 mg/L, pH: 3, reactor volume: 1 L, and flow rate: 0.33 L/h).
the fact that the photocatalyst NP and carbon based substrates become more negative at higher pH values and the repulsion forces between the adsorbent and dye molecules lead to less concentration of anionic dyes around TiO2 which consequently decreases the photocatalytic performance of the system [44]. Initial dye concentration The catalyst activity was also studied by changing the initial dye concentration from 50 to 200 mg/L for AR14 and AY117, and from 50 to 200 mg/L for AB92 because of the low penetration of UV irradiation to reach the photocatalyst NP. All results in Fig. 8 clearly imply that by increasing the concentration of dyes from
50 to 200 mg/L, the process efficiency decreases from 91% and 68% to 48% and 23% for AY117 and AR14, respectively. In case of AB92, the removal efficiency decreased from 68% to 17% by increasing the dye concentration from 20 to 50 mg/L. Also, in order to verify our explanations regarding the darker hue of AB92, the decoloration of another acidic dye (C.I. Acid Black 26) at 75 mg/ L, was investigated and the result was similar to the obtained data earlier. Generally, when the concentration of dye increases, the photon penetration through dye molecules to reach the catalyst surface decreases, which leads to the formation of lower amount of hydroxyl radicals that degrades the dye molecules structure.
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
8
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Fig. 5. FESEM images of (a, a0 ): TiO2-C (TiO2: 0.08 g), (b, b0 ): TiO2-CNT (TiO2: 0.08 g), (c, c0 ): TiO2-C after the dark adsorption of AR14 (TiO2: 0.04 g), and (d, d0 ): TiO2-CNT after the dark adsorption of AR14 (TiO2: 0.04 g) at different magnitude.
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
9
Fig. 6. FTIR spectra of AR14 before and after the dark adsorption on carbon and CNT fabricated carbon plates.
Another reason can be the extreme concentration and aggregation of the adsorbed dye molecules on the surface of TiO2 and carbon substrates, which will reduce the adsorption of UV light and further formation of oxidative radicals. But, this performance drop is less noticeable in PCNT rather than PC, which is probably related to the higher dye adsorption capacity of nanotubes that was confirmed by the FESEM images in Fig. 5(d,d0 ) [32,40]. It is noteworthy that the photocatalytic processes are usually designed for the pretreatment or post treatment of effluents either to degrade the pollutants to less complex compounds or to degrade the trace contaminants at very low concentrations, respectively. The promising results in this study achieved at fairly high concentrations of dye solutions, can offer such a system which benefits from an advanced nanotechnology to textile industry. Stability of TiO2-C and TiO2-CNT To evaluate the photochemical stability of the immobilized catalyst (TiO2-C and TiO2-CNT) in assessing the practical application of the photocatalyst plates in wastewater treatment, the recycle experiments for the photodegradation of AR14 were conducted and the results are shown in Fig. 9. The mean removal (%) of AR14 after 10 and 14 cycles of PC and PCNT processes are 60% and 68%, which decrease to 49% and 62% after 13 and 16 photocatalytic cycles. This means that the titania NP benefit from appropriate photochemical stability on the surface of the carbonaceous substrates. Fig. 9 also indicates that the photocatalyst NP are more stable on the CNT fabricated carbon plates and have lower decreasing rate of efficiency in comparison with PC, which could be the result of higher physical interaction between CNT and TiO2 NP. The FESEM images in Fig. 10 certify the obtained results by showing the existence of titania NP on the surface of the substrates after 13 and 16 successive cycles. The overall results can suggest the cost-effective utilization of the prepared photocatalyst plates in the practical applications due to their high activity and stability. It should be noticed that in this study, the repetitive utilization of photocatalyst for 16 cycles in a continuous process was applicable by a simple and facile procedure, while in other studies, the catalysts are usually shown to be reusable for approximate six cycles [37,43,45].
Fig. 7. Effect of pH on the removal efficiency ([AR14]: 50 mg/L, [AY117]: 50 mg/L, [AB92]: 20 mg/L, TiO2: 0.04 g, Time: 210 min, reactor volume: 1 L, and flow rate: 0.33 L/h).
RSM results The acquired results and data indicated that the more adsorption of the contaminants onto the CNT rather that carbon followed by the generation of oxidative radicals on the surface of TiO2 have improved the photocatalytic activity; and TiO2-CNT revealed higher stability in the repetitive catalytic cycles. Thus, RSM was only applied to PCNT process in binary dye solutions as an approach to the real textile wastewater that generally consists of more than one component. A four-factor, five-level CCD was conducted for the degradation of two acid dyes in a binary dye solution. Table 2 shows the selected parameters, their levels and the corresponding responses for all the conditions based on the response surface model. Analysis of variance (ANOVA) and P-value of each parameter are represented in Table 3.
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
10
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Fig. 8. Effect of initial dye concentration on the removal efficiency (pH: 3, TiO2: 0.04 g, Time: 210 min, reactor volume: 1 L, and flow rate: 0.33 L/h).
The second-order regression equation that relates the response functions to the selected variables is written in Eq. (3): Y i ¼ b0 þ
4 4 3 X 4 X X X b i xi þ bii xi 2 þ bij xi xj i¼1
i¼1
ð3Þ
i¼1 j¼iþ1
where Yi is the response variable, b0 the constant coefficient, bi the regression coefficients for linear effects, bii the quadratic coefficients, bij the interaction coefficients and xi,xj are the coded values of input factors [46,47]. The obtained regression models for the responses are also represented in Table 3. The variables which P-values of 0.05 and/or lower are assigned to, are significant to the response function. The results explore that the quadratic model is statistically significant to the responses and it can be adopted to describe the behavior of each response function. The values of R2 and adjusted R2 are high and the difference between the adjusted R2 and predicted R2 are low in most cases, which means that the model can be applicable for future outcomes. Furthermore, the lack-of-fit terms are not significant determined by their high P-values.
Fig. 9. Stability of TiO2-C and TiO2-CNT after 13 and 16 cycles, respectively.
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
11
Fig. 10. FESEM images of (a, a0 ): TiO2-C, (b, b0 ): TiO2-C after 13 photocatalytic cycles, (c, c0 ): TiO2-CNT and (d, d0 ): TiO2-CNT after 16 photocatalytic cycles (TiO2: 0.04 g) at different magnitude.
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
12
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Table 3 Quadratic model and the related ANOVA. Source
AR14 removal (%) P-value
AB92 removal (%)
COD removal (%)
EEC (KWh/gCOD)
Model Time pH TiO2 [C0] Time*time pH*pH TiO2*TiO2 [C0]*[C0] Time*pH Time*TiO2 Time*[C0] pH*TiO2 pH*[C0] TiO2*[C0] Lack-of-fit R2 (%) R2 adj. (%) R2 pred. (%) Quadratic models
0 0.016 0 0 0 0.492 0.016 0 0.014 0.523 0.522 0.814 0.002 0.121 0.01 0.0.095 99.21% 98.53% 95.48%
0 0.041 0 0.01 0 0.025 0 0.001 0 0.498 0.762 0.988 0.098 0.734 0.364 0.372 94.08% 88.90% 66.07%
0 0.014 0 0.001 0 0.564 0.025 0 0.016 0.666 0.668 0.536 0.719 0.277 0.361 0.491 98.12% 96.47% 92.15%
0 0 0 0 0 0.857 0 0 0 0.393 0.527 0.289 0.006 0.654 0.688 0.073 98.88% 97.91% 94.18%
According to the three dimensional surface plots in Fig. 11, low adsorption tendency between anionic dye molecules and positively charged TiO2 NP, carbon and CNT caused by increasing the pH values has resulted in the lower dye and COD removal (%), which has consequently increased the electrical energy consumption. Also, the existence of a maximum peak for TiO2 dosage clearly indicates that after a specific dosage of catalyst, the aggregation of nanoparticles reducing their specific surface area diminishes the dye and COD removal (%). It can be seen from the results that the EEC per unit COD decay are low (below 3.5 (kWh/ gCOD)). The overall results and changes of removal efficiencies obtained by RSM are in great agreement with the classical experiments. Optimization In order to find the optimal conditions of removal process to maximize the removal of AR14, AB92 and COD removal (%) while consuming the minimum electrical energy, the desirability function optimization has been employed for multi-response optimization. The predicted values of parameters to reach the desired values of responses are summarized in Table 4. The experimental data (the average of three experiments) illustrate that there is an adequate agreement between the experimental data and predicted values, which can be another reason for a proper fit of the quadratic model on the obtained results.
Flow rate and hydrogen peroxide Fig. 12 shows the effect of increasing the solution flow rate fed to the reactor on photocatalysis efficiency at the optimum condition obtained in Table 4. The results demonstrate that by increasing the flow rate from 0.33 to 0.99 L/h, the removal efficiency decreases about 30%, which is due to the reduction of the retention time of dye molecules in the reactor that lowers the reactions between dye molecules and the oxidative agents formed in the solution for dye degradation [48,49]. The addition of hydrogen peroxide as an important parameter in the photocatalytic processes was also investigated in Fig. 12. Various amounts of H2O2 (12.5%, 25%, 37.5%, and 50% of the stoichiometric dosage theoretically required for total mineralization of dyes) were added continuously to the reactor. The stoichiometric dosages of hydrogen peroxide for total mineralization of AR14 and AB92 were calculated according to the following reactions: C20H12N2O7S22Na + 51H2O2 ! 20CO2 + 54H2O + 2H2SO4 + Na2O + 2HNO3
(4)
C26H16N3O10S33Na + 68H2O2 ! 26CO2 + 143/2H2O + 3H2SO4 + 3/ 2Na2O + 3HNO3 (5) The results reveal that the addition of H2O2 to the system has significantly promoted the photocatalytic activity to reach a
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
13
Fig. 11. Three dimensional surface plots from RSM results.
complete decolorization after 100 min. It has been reported that H2O2 can inhibit the recombination of electrons and holes (e/h+), while further provides additional hydroxyl radicals as described by the following mechanisms: H2O2 + hy ! 2HO
destructive hydroxyl radicals formed through Eqs. (10)–(12) [50,51]:
HO2 þ HO ! H2 O þ O2
ð10Þ
2HO ! H2O2
(11)
(6)
H2O2 ! HO2 + 2H+
(7)
HO þ H2 O2 ! HO2 þ H2 O
HO2 þ hy ! HO þ
1=2O 2
O 2 þ H2 O2 ! O2 þ HO þ HO
ð8Þ
ð9Þ
On the other hand, further addition of H2O2 was found to decrease the reaction rate owing to the formation of less
ð12Þ
Conclusion The present study proposed a novel, simple, inexpensive, efficient and stable immobilization of titania NP on the surface of bare and CNT fabricated carbon plates to be employed as photocatalytic plates in a continuous photo-reactor for the
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
14
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
Table 4 Predicted and experimental optimized values by response optimizer. Desired (%) AR14 removal (%) AB92 removal (%) COD removal (%) EEC (KWh/gCOD) Optimum condition
Predicted (%)
65–70 (maximize) 71.38 55–60 (maximize) 60.05 50–55 (maximize) 55.67 0.350–0.400 (minimize) 0.338 Time: 108.18 min, pH: 2.85, TiO2: 0.03 and [C0]: 25.95 mg/L
Experimental (%) 69.02 62.45 54.97 0.484
Fig. 12. Effect of flow rate and H2O2 concentration on the removal efficiency at RSM optimum condition.
degradation of C.I. Acid Red 14, C.I. Acid Blue 92, and C.I. Acid Yellow 117. In this regard several important subjects were taken into account: 1. The popular and commercially available TiO2 NP were selected as the photocatalyst. 2. The toxicity and environmental hazards of the discharge of nanomaterials into water streams was considered and therefore, the immobilization of photocatalyst were performed. 3. The available, low-cost, chemically inert and stable carbon and CNT materials with extraordinary adsorption properties were chosen as substrates for the immobilization purpose. 4. A novel, green and facile electrodeposition method was employed for the fabrication of plates with CNT. 5. The solvent evaporation technique was performed as a cost effective, simple with no chemicals involved in the process to immobilize TiO2 NP. 6. Applying photocatalysis in nanoscale in a continuous wastewater treatment process.
TiO2 NP exhibited strong adherence on both carbon and CNT support, and the carbonaceous substrates showed high adsorption affinity towards dye molecules especially CNT. The characterization of the photocatalytic plates were inquired using FESEM images and FTIR analysis, determining the successful and stable immobilization of nanoparticles. The FESEM images, FTIR spectra and stability tests showed that the nanotubes provide high surface area for TiO2 NP to be immobilized without agglomeration along with adequate adhesion to CNT which increased the quantity of dyes to come in contact with TiO2 through adsorption. The catalyst activity was examined by studying the influence of effective parameters such as pH, initial dye concentration, TiO2 dosage and time through classical (single dye solution) and statistical RSM (binary dye solution to get closer to simulate the real textile effluent). The result of the CCD experiments led to the precise fitness of quadratic models that related the variables to the selected responses (dye and COD removal (%) and electrical energy consumption). The multi-response optimization was also conducted and the predicted values by response optimizer expressed a great agreement with
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039
G Model JIEC 3489 No. of Pages 15
E. Pajootan et al. / Journal of Industrial and Engineering Chemistry xxx (2017) xxx–xxx
the experimental data, which resulted in high removal efficiencies at very low dosages of TiO2 NP and low electrical energy consumption. The fast and complete decolorization were accomplished by the continuous H2O2 feeding system embedded in the reactor.
[23] [24] [25] [26] [27] [28] [29]
References
[30]
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22]
X. Qu, P.J.J. Alvarez, Q. Li, Water Res. 47 (2013) 3931. S. Li, L.K. Wallis, H. Ma, S.A. Diamond, Sci. Total Environ. 466–467 (2014) 800. I. El Saliby, L. Erdei, J.-H. Kim, H.K. Shon, Water Res. 47 (2013) 4115. Y. Luo, J. Liu, X. Xia, X. Li, T. Fang, S. Li, Q. Ren, J. Li, Z. Jia, Mater. Lett. 61 (2007) 2467. Z. Wu, H. Wang, Y. Liu, Z. Gu, J. Hazard. Mater. 151 (2008) 17. S.-H. Lee, S. Pumprueg, B. Moudgil, W. Sigmund, Colloids Surf. B 40 (2005) 93. H. Li, S. Gao, M. Cao, R. Cao, J. Colloid Interface Sci. 394 (2013) 434. B. Wang, F.C. de Godoi, Z. Sun, Q. Zeng, S. Zheng, R.L. Frost, J. Colloid Interface Sci. 438 (2015) 204. M.A. Nawi, S. Sabar, J. Sheilatina, Colloid Interface Sci. 372 (2012) 80. A. Carabin, P. Drogui, D. Robert, J. Taiwan Inst. Chem. Eng. 54 (2015) 109. H. Yu, X. Quan, S. Chen, H. Zhao, J. Phys. Chem. C 111 (2007) 12987. H.U. Farouk, A.A.A. Raman, W.M.A.W. Daud, J. Ind. Eng. Chem. 33 (2016) 11. Y.-J. Kim, Y. Horie, Y. Matsuzawa, S. Ozaki, M. Endo, M.S. Dresselhaus, Carbon 42 (2004) 2423. C.-Y. Kuo, J. Hazard. Mater. 163 (2009) 239. L. Guardia, S. Villar-Rodil, J.I. Paredes, R. Rozada, A. Martínez-Alonso, J.M.D. Tascón, Carbon 50 (2012) 1014. S.C. Smith, D.F. Rodrigues, Carbon 91 (2015) 122. M.A. Andrade, A.S. Mestre, R.J. Carmona, A.P. Carvalho, C.O. Ania, Appl. Catal. A 507 (2015) 91. J. Cho, S. Schaab, J. Roether, A. Boccaccini, J. Nanopart. Res. 10 (2008) 99. W. Wang, P. Serp, P. Kalck, J.L. Faria, Appl. Catal. B 56 (2005) 305. P. Serp, M. Corrias, P. Kalck, Appl. Catal. A 253 (2003) 337. K. Woan, G. Pyrgiotakis, W. Sigmund, Adv. Mater. 21 (2009) 2233. A. Matsumoto, K. Tsutsumi, K. Kaneko, Langmuir 8 (1992) 2515.
[31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51]
15
A. Omri, M. Benzina, F. Bennour, J. Ind. Eng. Chem. 21 (2015) 356. J. Matos, J. Laine, J.-M. Herrmann, Appl. Catal. B 18 (1998) 281. I. Zhitomirsky, L. Gal-Or, A. Kohn, H.W. Hennicke, J. Mater. Sci. 30 (1995) 5307. B. Tryba, A.W. Morawski, M. Inagaki, Appl. Catal. B 41 (2003) 427. K.E. Tettey, M.Q. Yee, D. Lee, ACS Appl. Mater. Interfaces 2 (2010) 2646. H. Meng, W. Hou, X. Xu, J. Xu, X. Zhang, Particuology 14 (2014) 38. H. Wu, J. Fan, E. Liu, X. Hu, Y. Ma, X. Fan, Y. Li, C. Tang, J. Alloys Compd. 623 (2015) 298. S. Minnikanti, N. Peixoto, in: Jose Mauricio Marulanda (Ed.), Carbon Nanotubes Applications on Electron Devices, InTech, 2011. L. Eskandarian, E. Pajootan, M. Arami, Ind. Eng. Chem. Res. 53 (2014) 14841. E. Pajootan, M. Arami, M. Rahimdokht, Ind. Eng. Chem. Res. 53 (2014) 16261. R.D.C. Soltani, A. Rezaee, A.R. Khataee, M. Safari, J. Ind. Eng. Chem. 20 (2014) 1861. E. Pajootan, M. Arami, Electrochim. Acta 112 (2013) 505. D. Hornero-Méndez, M.I. Mínguez-Mosquera, J. Agric. Food. Chem. 49 (2001) 3584. E. Pajootan, M. Arami, N.M. Mahmoodi, J. Taiwan Inst. Chem. Eng. 43 (2012) 282. B. Appavu, K. Kannan, S. Thiripuranthagan, J. Ind. Eng. Chem. 36 (2016) 184. A.H. Noroozi, S. Safa, J. Adv. Mater. Proc. 2 (2014) 31. S.-J. Bao, C.M. Li, J.-F. Zang, X.-Q. Cui, Y. Qiao, J. Guo, Adv. Funct. Mater. 18 (2008) 591. H. Zangeneh, A.A.L. Zinatizadeh, M. Habibi, M. Akia, M. Hasnain Isa, J. Ind. Eng. Chem. 26 (2015) 1. M.-Q. Yang, N. Zhang, Y.-J. Xu, ACS Appl. Mater. Interfaces 5 (2013) 1156. D. Feng, S. Xu, G. Liu, Chemosphere 125 (2015) 102. J.R. Kim, E. Kan, J. Ind. Eng. Chem. 21 (2015) 644. I.K. Konstantinou, T.A. Albanis, Appl. Catal. B 49 (2004) 1. H. Guo, Z. Zheng, J. Chen, W. Weng, M. Huang, J. Ind. Eng. Chem. 36 (2016) 306. E. Errais, J. Duplay, F. Darragi, I. M'Rabet, A. Aubert, F. Huber, G. Morvan, Desalination 275 (2011) 74. M. Fathinia, A.R. Khataee, J. Ind. Eng. Chem. 19 (2013) 1525. Y. Li, J. Chen, J. Liu, M. Ma, W. Chen, L. Li, J. Environ. Sci. 22 (2010) 1290. E. Pajootan, M. Arami, M. Rahimdokht, Sep. Purif. Technol. 130 (2014) 34. M.N. Chong, B. Jin, C.W.K. Chow, C. Saint, Water Res. 44 (2010) 2997. S. Rodriguez, A. Santos, A. Romero, Desalination 280 (2011) 108.
Please cite this article in press as: E. Pajootan, et al., Carbon and CNT fabricated carbon substrates for TiO2 nanoparticles immobilization with industrial perspective of continuous photocatalytic elimination of dye molecules, J. Ind. Eng. Chem. (2017), http://dx.doi.org/10.1016/j. jiec.2017.06.039