- Email: [email protected]
TiO2 nanosheet-graphene oxide based photocatalytic hierarchical membrane for water purification
Accepted Manuscript TiO2 nanosheet-graphene oxide based photocatalytic hierarchical membrane for water purification
Abhinav K. Nair, P.E. JagadeeshBabu PII: DOI: Reference:
S0257-8972(17)30022-1 doi: 10.1016/j.surfcoat.2017.01.022 SCT 21998
To appear in:
Surface & Coatings Technology
Received date: Revised date: Accepted date:
31 August 2016 28 November 2016 8 January 2017
Please cite this article as: Abhinav K. Nair, P.E. JagadeeshBabu , TiO2 nanosheet-graphene oxide based photocatalytic hierarchical membrane for water purification. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Sct(2017), doi: 10.1016/j.surfcoat.2017.01.022
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT TiO2 nanosheet-graphene oxide based photocatalytic hierarchical membrane for water purification Abhinav .K. Nair, P. E. JagadeeshBabu* Department of Chemical Engineering, National Institute of Technology Karnataka,
PT
Surathkal, Mangalore 575 025, India Abstract
RI
There is a rising concern over the toxicity of nanomaterials which emphasizes the need for
SC
reforms in application of nanomaterials in water treatment. A hierarchical membrane with a
NU
thin layer of photocatalyst coated on top of the membrane surface has emerged as a better alternate for immobilization of photocatalyst. Studies have revealed that nanoparticles are not
MA
pliablefor synthesis of hierarchical membranes due to their smaller size and low stability after deposition. TiO2 nanosheets serve as better alternate due their thin structure which enables
D
stable layer formation. Integrating the nanosheets with modifiers like graphene oxide can
PT E
further enhance the photocatalytic activity. The sheet structure of graphene oxide enhances stable film formation and also acts as support for interconnecting TiO2 nanosheets. In the
CE
present work, TiO2 nanosheets are modified with graphene oxide and used to develop a hierarchical membrane by depositing a catalyst coating on a support membrane. The
AC
hierarchical membrane performance was studied using congo red dye as model pollutant and the effect of catalyst loading on the permeate flux and dye removal were analyzed. KEYWORDS Titanium dioxide nanosheets; Graphene oxide; photocatalysis; hierarchical membrane; congo red dye. Correspondence to: P. E. JagadeeshBabu (E-mail: [email protected], [email protected] )
1
ACCEPTED MANUSCRIPT 1. Introduction Membrane separation processes have emerged as an efficient technology for water purification in the last few decades. Isothermal operation, low installation foot print and lack of chemical agents give it an edge over other conventional water purification techniques. Membrane separation cannot degrade the contaminants; it can only remove it from a media.
PT
The accumulation of contaminant on the membrane surface due to fouling adversely affects
RI
the membrane performance [1]. Alternatively, photocatalysis is capable of completely
SC
eliminating organic contaminant by degrading it in to smaller molecules [2]. TiO2 nanomaterials have been extensively studied for photocatalytic degradation of various
NU
organic contaminants in water. Nanocatalyst possess high specific surface but due to the small size catalyst recovery becomes a challenge. Researchers in nanotoxicology have raised
MA
serious doubts over the non-toxicity of TiO2 nanomaterials [3]. Hence the emphasis is on developing better techniques to immobilize the photocatalyst. Immobilization of
D
photocatalyst on ceramic membranes for coupling photocatalysis with filtration has been well
PT E
established [4]. In case of polymer membranes the incorporation of catalyst in polymer matrices drastically reduces the exposed catalyst area. Most of the common polymer
CE
materials are also vulnerable to UV irradiation. A novel approach to solve these setbacks was reported by Bai et al., they developed a hierarchical membrane with a coating of
AC
photocatalyst on top of a polymer base membrane. Hierarchical membranes have the advantages of good catalyst exposure, catalyst immobilization and can operate continuously. Various TiO2 structures like nanoparticles, nanowires, nanofibers, nanothorns and microspheres have been reported in the form of hierarchical membranes [5]. TiO2 nanosheets (TNS) are 2D nanostructures with nanoscale thickness. The thin sheet structure enables stable layer formation on deposition. These nanostructures are difficult to synthesize with effective size control. Hydrothermal methods have been recently
2
ACCEPTED MANUSCRIPT used to produce good anatase nanosheets [6]. Little work has been reported in literature on their modification to enhance catalytic activity. To the best of our knowledge there have been no reports on the application of TiO2 nanosheet in hierarchical membranes. Graphene oxide (GO) is a widely studied modifier to enhance the photocatalytic activity of TiO2 [7]. When employed in hierarchical membranes GO has also shown to impart better mechanical stability
PT
to photocatalyst layer [8]. In this study, we report the synthesis of TNS-GO hierarchical
RI
membrane for simultaneous water filtration and photodegrdation.
SC
2. Experimental methods 2.1 Materials
NU
Graphene oxide sheets and tetrabutyltitanate (TBT, 98%) were purchased from SigmaAldrich Co, Bangalore, India. Hydrofluoric acid (HF, 40%) and Congo red dye were
MA
procured from Nice Chemicals, Cochin, India. 2.2 Synthesis of TNS
D
TBT (10 mL) was mixed thoroughly with HF (3 mL) and transferred in to a Teflon lined
PT E
stainless steel autoclave. Solvothermal treatment was carried out in a hot air oven at 200 °C for 24 h. The product obtained was then washed repeatedly with distilled water and ethanol.
CE
The residue obtained was dried at 60 °C in a hot air oven [6]. 2.3 Synthesis of TNS-GO
AC
GO (20 mg) was dispersed in 100 mL distilled water under constant stirring. The dispersion was sonicated for 30 min to enable exfoliation of the GO sheets. A dispersion of TNS (100 mg) was prepared in 100 ml water with constant stirring. Both the dispersions were mixed and stirred vigorously for 2 h. The obtained product was dried at 60°C in a hot air oven [8]. 2.4 Synthesis of photocatalyst coated hierarchical membrane Hierarchical membranes were synthesized using cellulose acetate base membrane (Sartorius Stedim Biotech, diameter - 47 mm and pore size - 0.2 µm). The base membrane was
3
ACCEPTED MANUSCRIPT positioned properly in a dead end filtration cell. Varying amount of TNS-GO photocatalyst; 50 mg (M-50), 100 mg (M-100), 200 mg (M-200), 300 mg (M-300) and 400 mg (M-400) respectively were dispersed in 100 mL of distilled water and loaded in to the filtration cell. During high pressure filtration the photocatalyst got evenly coated on the surface of the base membrane.
PT
2.5 Characterization
RI
Synthesized samples were observed under field emission scanning electron microscope
SC
(FESEM, Zeiss Sigma) and scanning electron microscope (SEM, Jeol JSM-6380LA). All samples were sputtered with gold before analysis. Further Transmission electron microscope
NU
(TEM, JEOL JEM-2100) was used to analyze the TNS samples. X-ray Diffractometer (Rigaku Miniflux 6000) equipped with monochromatized high intensity Cu Kα radiation (λ =
MA
1.54178 Å) was used to analyze the crystalline nature of the samples. UV-Visible spectrometer (Hitachi, U-2900) was used to estimate congo red dye concentration at a
D
wavelength of 498 nm.
PT E
2.4 Performance study of the membranes The permeation properties of the membranes were studied using a lab scale dead end
CE
filtration cell. The cell consists of a Quartz glass window on the top to enable UV irradiation. The active area of the membrane cell was 14.6 cm2. The membrane cell was connected to a
AC
feed tank which was pressurised using an oxygen cylinder. The flux studies were carried out at a transmembrane pressure of 0.2 MPa. Permeate flux (J) was calculated using the following equation: J = Q/(Δt A) (1) Where J is the flux expressed in Lm-2h-1 and Q (L) is the amount of water collected for Δt (h) time duration using a membrane of area A (m2).
4
ACCEPTED MANUSCRIPT Dye removal studies were carried out using the same dead end filtration cell. 50 mgL-1 of Congo red dye solution was prepared in distilled water and 100 mL dye solution was used for each dye removal study. Dye removal studies where done in the presence and absence of UV radiation to evaluate the effects of adsorption and photocatalytic degradation. Permeate obtained after dye removal study was subjected to UV-Vis spectroscopy to estimate residual
PT
dye concentration.
RI
3. Results and Discussion
PT E
D
MA
NU
SC
3.1Characterization of TNS
Fig. 1. a) FESEM image of TNS b) TEM image of TNS
CE
FESEM image of TNS is shown in Fig. 1 (a) and from the figure individual square flakes of TiO2 nanosheets can be observed. TEM image (Fig. 1 b) of TNS reveals the thin structure of
AC
the TiO2 nanosheets. Based on the semi transparency seen in the TEM image, it is inferred that the thickness of TNS is in the range of a few nanometers. X-ray diffraction patterns of TNS are shown in Fig. 2. The major XRD peaks were observed at 25.8°, 37.6°, 48.6° and 55.4° which correspond to anatse phase of TiO2 (JCPDS No. 21-1272) [9]. In general, anatase phase of TiO2 is preferable for superior photocatalytic activity. Hence TNS can serve as a good photocatalyst for the synthesis of hierarchical membrane.
5
SC
RI
PT
ACCEPTED MANUSCRIPT
NU
Fig. 2. XRD pattern of TNS
CE
PT E
D
MA
3.2 Characterization of TNS-GO hierarchical membrane
AC
Fig.3. a) FESEM image of TNS-GO b) SEM image of M-100 membrane cross section The hierarchical membrane was synthesized by depositing TNS-GO on the polymer membrane. Fig. 3(a) shows the FESEM image of TNS-GO and the presence of TNS on the GO sheet can be observed from the figure. In the cross sectional image of the hierarchical membrane Fig.3 (b), the porous cellulose acetate support membrane can be seen completely coated with TNS-GO. TNS-GO has formed a uniform stable coating over the membrane surface. The strength of the coating was verified by repeated washing of the coated 6
ACCEPTED MANUSCRIPT membrane with water and it was observed that TNS-GO was intact on membrane surface. Energy band gap estimation of the TNS-GO was carried out by analyzing the UV-Vis spectral absorption peaks as reported in literature [10]. The Tauc plots indicate lowering of
MA
NU
SC
RI
PT
band gap from 3.2 eV to 2.98 eV on addition of GO to TNS (Fig. 4).
Fig. 4. Tauc plots for band gap estimation of photocatalysts
D
3.3 Performance study of TNS-GO membranes
PT E
PWF and dye removal traits of various membranes are given in Fig .5. Theoretically, addition of TNS-GO on the membrane surface blocks the surface pores and results in permeate flux
CE
reduction [11]. A gradual reduction in PWF has been observed as TNS-GO loading was gradually increased but the dye removal gradually increased. In a hierarchal membrane, lower
AC
flux enhances the interaction between the dye and the catalyst and improves the dye removal. But very low flux will unnecessarily prolong the filtration process. Hence in a hierarchal membrane, there is always a tradeoff between the permeate flux and dye removal. As the TNS-GO loading increases there is an appreciable increase in dye removal, complete dye removal was obtained in case of M-400 with the lowest flux of 29.3 Lm-2h-1.
7
SC
RI
PT
ACCEPTED MANUSCRIPT
NU
Fig.5. Flux and dye removal (DR) of various membranes
MA
The TNS-GO can remove the dye in two ways, either by adsorption or by photocatalytic degradation [12]. The presence of reactive oxygen atoms, carboxyl and hydroxyl groups
D
enable GO to interact effectively with Congo red resulting in good adsorption [13]. In
PT E
aqueous media, the surface hydroxyl groups of TiO2 facilitate the adsorption of Congo red dye [14]. Up on UV irradiation, the contact between TNS and GO enables electron transfer from conduction band of TiO2 to GO. Electron transfer improves the photocatalytic activity
CE
by lowering the charge recombination rate. The electrons and holes generated during UV
AC
irradiation react with oxygen and water molecules in the media to form, form hydroxyl radicals (•OH) which enable photodegradation of the dye [8]. When the catalyst loading is increased the thickness of the catalyst coating becomes substantial. UV radiation can’t penetrate in to the entire thickness of the catalyst coating. Hence, the top portion of the catalyst coating which receives radiation will involve in photodegradation. The remaining underneath portion will adsorb the dye. Dye adsorption is undesirable since it adversely affects catalyst reusability. The thickness of the membrane
8
ACCEPTED MANUSCRIPT should be such that it photodegrades the maximum amount of dye and still offers minimal
MA
NU
SC
RI
PT
flux drop [13].
Fig.6. Reuse cycles (RC) of the membrane for Dye removal The dye removal study done in absence of UV irradiation evaluates the affect of adsorption.
PT E
D
It is observed that amount of dye absorbed is directly proportional to the catalyst loading. Beyond 100 mg catalyst loading adsorption played a much significant role. In case of M-400 which had given 100% dye removal under UV irradiation, is capable of removing 94.5% of
CE
the dye just by adsorption. It is evident that beyond M-100, increased loading didn’t improve
AC
photodegrdation. The membrane surface area exposed to UV radiation is constant and does not further increase with catalyst loading. By comparing the difference between dye removal in presence and absence of UV light, it is evident that M-100 membrane could achieve maximum photodegradation of dye. Also when compared to higher catalyst loaded membranes, M-100 exhibited superior permeate flux. The M-100 membrane is thus more efficient and reusable. The membrane dye removal reuse cycles were carried out and the results are shown below in Fig. 6. It is observed that except M-50 membrane all other membranes showed consistent dye removal traits. The adsorption phenomenon weakens with 9
ACCEPTED MANUSCRIPT continuous reuse of the photocatalyst. At higher catalyst loading the lower flux provides sufficient residence time for efficient dye degradation via photocatalysis. 4. Conclusions In summary, TNS-GO coated hierarchical membranes were successfully synthesised for simultaneous water filtration and photocatalytic degradation. At higher catalyst loading the
PT
permeate flux dropped heavily due to the additional cake resistance of the catalyst coating.
RI
Dye adsorptions played a major role at higher catalyst loading where as photodegradation
SC
was significant at lower catalyst concentration. Although M-400 could completely remove the dye, it exhibited lowest flux and greater dye adsorption. M-100 membrane showed best
NU
result with comparatively higher flux and better photodegradation. Photocatalyst reuse cycles revealed that higher TNS-GO loaded membranes could maintain dye removal capacity since
MA
the flux decline provided sufficient time for photocatlytic degradation. References
D
[1] A.K. Nair, P.M. Shalin, P.E. JagadeeshBabu, Performance enhancement of polysulfone
PT E
ultrafiltration membrane using TiO2 nanofibers, Desalin. Water Treat. 3994 (2015) 1–9. [2] L.M. Pastrana-Martínez, S. Morales-Torres, J.L. Figueiredo, J.L. Faria, A.M.T. Silva,
CE
Graphene oxide based ultrafiltration membranes for photocatalytic degradation of organic
AC
pollutants in salty water, Water Res. 77 (2015) 179–190. [3] H. Shi, R. Magaye, V. Castranova, J. Zhao, Titanium dioxide nanoparticles: a review of current toxicological data., Part. Fibre Toxicol. 10 (2013) 15. [4] Q. Zhang, H. Wang, X. Fan, F. Lv, S. Chen, X. Quan, Fabrication of TiO 2 nanofiber membranes by a simple dip-coating technique for water treatment, Surf. Coatings Technol. 298 (2016) 45–52. [5] H. Bai, Z. Liu, D.D. Sun, Hierarchically multifunctional TiO2 nano-thorn membrane for water purification, Chem. Commun. 46 (2010) 6542–6544. 10
ACCEPTED MANUSCRIPT [6] K. Chen, Z. Jiang, J. Qin, Y. Jiang, R. Li, H. Tang, X. Yang, Synthesis and improved photocatalytic activity of ultrathin TiO2 nanosheets with nearly 100 % exposed (001) facets, Ceram. Int. 40 (2014) 16817–16823. [7] Y. Gao, M. Hu, B. Mi, Membrane surface modification with TiO2-graphene oxide for enhanced photocatalytic performance, 455 (2014) 349–356.
PT
[8] P. Gao, Z. Liu, M. Tai, D. Delai, W. Ng, Applied Catalysis B: Environmental
RI
Multifunctional graphene oxide–TiO2 microsphere hierarchical membrane for clean water
SC
production, Applied Catal. B, Environ. 138-139 (2013) 17–25.
[9] Y. Masuda, K. Kato, Synthesis and phase transformation of TiO2 nano-crystals in aqueous
NU
solutions, J. Ceram. Society Japan. 117 (2009) 373–376.
[10] P. Uddandarao, R.M. B, ZnS semiconductor quantum dots production by an endophytic
MA
fungus Aspergillus flavus, Mater. Sci. Eng. B. 207 (2016) 26–32. [11] A. Dipareza, C. Lin, C. Wu, Removal of natural organic matter by ultrafiltration with
D
TiO2-coated membrane under UV irradiation, Journal of Colloid and Interface Science 323
PT E
(2008) 112–119.
[12] H. Bai, X. Zan, J. Juay, D. Delai, Hierarchical heteroarchitectures functionalized
CE
membrane for high efficient water purification, J. Membr. Sci. 475 (2015) 245–251. [13] S. Debnath, A. Maity, K. Pillay, Impact of process parameters on removal of Congo red
AC
by graphene oxide from aqueous solution, J. Environ. Chem. Eng. 2 (2014) 260–272. [14] E. Kordouli, K. Bourikas, A. Lycourghiotis, C. Kordulis, The mechanism of azo-dyes adsorption on the titanium dioxide surface and their photocatalytic degradation over samples with various anatase/rutile ratios, Catal. Today. 252 (2015) 128–135.
11
ACCEPTED MANUSCRIPT Highlights TNS-GO coated hierarchical membranes were synthesised.
At higher catalyst loading dye removal improved but permeate flux declined.
Dye adsorption played a major role at higher catalyst loading.
Photodegradation was significant at lower catalyst loading.
An optimum photodegradation with good flux was obtained in case of M-100 membrane
AC
CE
PT E
D
MA
NU
SC
RI
PT
12
Abhinav K. Nair, P.E. JagadeeshBabu PII: DOI: Reference:
S0257-8972(17)30022-1 doi: 10.1016/j.surfcoat.2017.01.022 SCT 21998
To appear in:
Surface & Coatings Technology
Received date: Revised date: Accepted date:
31 August 2016 28 November 2016 8 January 2017
Please cite this article as: Abhinav K. Nair, P.E. JagadeeshBabu , TiO2 nanosheet-graphene oxide based photocatalytic hierarchical membrane for water purification. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Sct(2017), doi: 10.1016/j.surfcoat.2017.01.022
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT TiO2 nanosheet-graphene oxide based photocatalytic hierarchical membrane for water purification Abhinav .K. Nair, P. E. JagadeeshBabu* Department of Chemical Engineering, National Institute of Technology Karnataka,
PT
Surathkal, Mangalore 575 025, India Abstract
RI
There is a rising concern over the toxicity of nanomaterials which emphasizes the need for
SC
reforms in application of nanomaterials in water treatment. A hierarchical membrane with a
NU
thin layer of photocatalyst coated on top of the membrane surface has emerged as a better alternate for immobilization of photocatalyst. Studies have revealed that nanoparticles are not
MA
pliablefor synthesis of hierarchical membranes due to their smaller size and low stability after deposition. TiO2 nanosheets serve as better alternate due their thin structure which enables
D
stable layer formation. Integrating the nanosheets with modifiers like graphene oxide can
PT E
further enhance the photocatalytic activity. The sheet structure of graphene oxide enhances stable film formation and also acts as support for interconnecting TiO2 nanosheets. In the
CE
present work, TiO2 nanosheets are modified with graphene oxide and used to develop a hierarchical membrane by depositing a catalyst coating on a support membrane. The
AC
hierarchical membrane performance was studied using congo red dye as model pollutant and the effect of catalyst loading on the permeate flux and dye removal were analyzed. KEYWORDS Titanium dioxide nanosheets; Graphene oxide; photocatalysis; hierarchical membrane; congo red dye. Correspondence to: P. E. JagadeeshBabu (E-mail: [email protected], [email protected] )
1
ACCEPTED MANUSCRIPT 1. Introduction Membrane separation processes have emerged as an efficient technology for water purification in the last few decades. Isothermal operation, low installation foot print and lack of chemical agents give it an edge over other conventional water purification techniques. Membrane separation cannot degrade the contaminants; it can only remove it from a media.
PT
The accumulation of contaminant on the membrane surface due to fouling adversely affects
RI
the membrane performance [1]. Alternatively, photocatalysis is capable of completely
SC
eliminating organic contaminant by degrading it in to smaller molecules [2]. TiO2 nanomaterials have been extensively studied for photocatalytic degradation of various
NU
organic contaminants in water. Nanocatalyst possess high specific surface but due to the small size catalyst recovery becomes a challenge. Researchers in nanotoxicology have raised
MA
serious doubts over the non-toxicity of TiO2 nanomaterials [3]. Hence the emphasis is on developing better techniques to immobilize the photocatalyst. Immobilization of
D
photocatalyst on ceramic membranes for coupling photocatalysis with filtration has been well
PT E
established [4]. In case of polymer membranes the incorporation of catalyst in polymer matrices drastically reduces the exposed catalyst area. Most of the common polymer
CE
materials are also vulnerable to UV irradiation. A novel approach to solve these setbacks was reported by Bai et al., they developed a hierarchical membrane with a coating of
AC
photocatalyst on top of a polymer base membrane. Hierarchical membranes have the advantages of good catalyst exposure, catalyst immobilization and can operate continuously. Various TiO2 structures like nanoparticles, nanowires, nanofibers, nanothorns and microspheres have been reported in the form of hierarchical membranes [5]. TiO2 nanosheets (TNS) are 2D nanostructures with nanoscale thickness. The thin sheet structure enables stable layer formation on deposition. These nanostructures are difficult to synthesize with effective size control. Hydrothermal methods have been recently
2
ACCEPTED MANUSCRIPT used to produce good anatase nanosheets [6]. Little work has been reported in literature on their modification to enhance catalytic activity. To the best of our knowledge there have been no reports on the application of TiO2 nanosheet in hierarchical membranes. Graphene oxide (GO) is a widely studied modifier to enhance the photocatalytic activity of TiO2 [7]. When employed in hierarchical membranes GO has also shown to impart better mechanical stability
PT
to photocatalyst layer [8]. In this study, we report the synthesis of TNS-GO hierarchical
RI
membrane for simultaneous water filtration and photodegrdation.
SC
2. Experimental methods 2.1 Materials
NU
Graphene oxide sheets and tetrabutyltitanate (TBT, 98%) were purchased from SigmaAldrich Co, Bangalore, India. Hydrofluoric acid (HF, 40%) and Congo red dye were
MA
procured from Nice Chemicals, Cochin, India. 2.2 Synthesis of TNS
D
TBT (10 mL) was mixed thoroughly with HF (3 mL) and transferred in to a Teflon lined
PT E
stainless steel autoclave. Solvothermal treatment was carried out in a hot air oven at 200 °C for 24 h. The product obtained was then washed repeatedly with distilled water and ethanol.
CE
The residue obtained was dried at 60 °C in a hot air oven [6]. 2.3 Synthesis of TNS-GO
AC
GO (20 mg) was dispersed in 100 mL distilled water under constant stirring. The dispersion was sonicated for 30 min to enable exfoliation of the GO sheets. A dispersion of TNS (100 mg) was prepared in 100 ml water with constant stirring. Both the dispersions were mixed and stirred vigorously for 2 h. The obtained product was dried at 60°C in a hot air oven [8]. 2.4 Synthesis of photocatalyst coated hierarchical membrane Hierarchical membranes were synthesized using cellulose acetate base membrane (Sartorius Stedim Biotech, diameter - 47 mm and pore size - 0.2 µm). The base membrane was
3
ACCEPTED MANUSCRIPT positioned properly in a dead end filtration cell. Varying amount of TNS-GO photocatalyst; 50 mg (M-50), 100 mg (M-100), 200 mg (M-200), 300 mg (M-300) and 400 mg (M-400) respectively were dispersed in 100 mL of distilled water and loaded in to the filtration cell. During high pressure filtration the photocatalyst got evenly coated on the surface of the base membrane.
PT
2.5 Characterization
RI
Synthesized samples were observed under field emission scanning electron microscope
SC
(FESEM, Zeiss Sigma) and scanning electron microscope (SEM, Jeol JSM-6380LA). All samples were sputtered with gold before analysis. Further Transmission electron microscope
NU
(TEM, JEOL JEM-2100) was used to analyze the TNS samples. X-ray Diffractometer (Rigaku Miniflux 6000) equipped with monochromatized high intensity Cu Kα radiation (λ =
MA
1.54178 Å) was used to analyze the crystalline nature of the samples. UV-Visible spectrometer (Hitachi, U-2900) was used to estimate congo red dye concentration at a
D
wavelength of 498 nm.
PT E
2.4 Performance study of the membranes The permeation properties of the membranes were studied using a lab scale dead end
CE
filtration cell. The cell consists of a Quartz glass window on the top to enable UV irradiation. The active area of the membrane cell was 14.6 cm2. The membrane cell was connected to a
AC
feed tank which was pressurised using an oxygen cylinder. The flux studies were carried out at a transmembrane pressure of 0.2 MPa. Permeate flux (J) was calculated using the following equation: J = Q/(Δt A) (1) Where J is the flux expressed in Lm-2h-1 and Q (L) is the amount of water collected for Δt (h) time duration using a membrane of area A (m2).
4
ACCEPTED MANUSCRIPT Dye removal studies were carried out using the same dead end filtration cell. 50 mgL-1 of Congo red dye solution was prepared in distilled water and 100 mL dye solution was used for each dye removal study. Dye removal studies where done in the presence and absence of UV radiation to evaluate the effects of adsorption and photocatalytic degradation. Permeate obtained after dye removal study was subjected to UV-Vis spectroscopy to estimate residual
PT
dye concentration.
RI
3. Results and Discussion
PT E
D
MA
NU
SC
3.1Characterization of TNS
Fig. 1. a) FESEM image of TNS b) TEM image of TNS
CE
FESEM image of TNS is shown in Fig. 1 (a) and from the figure individual square flakes of TiO2 nanosheets can be observed. TEM image (Fig. 1 b) of TNS reveals the thin structure of
AC
the TiO2 nanosheets. Based on the semi transparency seen in the TEM image, it is inferred that the thickness of TNS is in the range of a few nanometers. X-ray diffraction patterns of TNS are shown in Fig. 2. The major XRD peaks were observed at 25.8°, 37.6°, 48.6° and 55.4° which correspond to anatse phase of TiO2 (JCPDS No. 21-1272) [9]. In general, anatase phase of TiO2 is preferable for superior photocatalytic activity. Hence TNS can serve as a good photocatalyst for the synthesis of hierarchical membrane.
5
SC
RI
PT
ACCEPTED MANUSCRIPT
NU
Fig. 2. XRD pattern of TNS
CE
PT E
D
MA
3.2 Characterization of TNS-GO hierarchical membrane
AC
Fig.3. a) FESEM image of TNS-GO b) SEM image of M-100 membrane cross section The hierarchical membrane was synthesized by depositing TNS-GO on the polymer membrane. Fig. 3(a) shows the FESEM image of TNS-GO and the presence of TNS on the GO sheet can be observed from the figure. In the cross sectional image of the hierarchical membrane Fig.3 (b), the porous cellulose acetate support membrane can be seen completely coated with TNS-GO. TNS-GO has formed a uniform stable coating over the membrane surface. The strength of the coating was verified by repeated washing of the coated 6
ACCEPTED MANUSCRIPT membrane with water and it was observed that TNS-GO was intact on membrane surface. Energy band gap estimation of the TNS-GO was carried out by analyzing the UV-Vis spectral absorption peaks as reported in literature [10]. The Tauc plots indicate lowering of
MA
NU
SC
RI
PT
band gap from 3.2 eV to 2.98 eV on addition of GO to TNS (Fig. 4).
Fig. 4. Tauc plots for band gap estimation of photocatalysts
D
3.3 Performance study of TNS-GO membranes
PT E
PWF and dye removal traits of various membranes are given in Fig .5. Theoretically, addition of TNS-GO on the membrane surface blocks the surface pores and results in permeate flux
CE
reduction [11]. A gradual reduction in PWF has been observed as TNS-GO loading was gradually increased but the dye removal gradually increased. In a hierarchal membrane, lower
AC
flux enhances the interaction between the dye and the catalyst and improves the dye removal. But very low flux will unnecessarily prolong the filtration process. Hence in a hierarchal membrane, there is always a tradeoff between the permeate flux and dye removal. As the TNS-GO loading increases there is an appreciable increase in dye removal, complete dye removal was obtained in case of M-400 with the lowest flux of 29.3 Lm-2h-1.
7
SC
RI
PT
ACCEPTED MANUSCRIPT
NU
Fig.5. Flux and dye removal (DR) of various membranes
MA
The TNS-GO can remove the dye in two ways, either by adsorption or by photocatalytic degradation [12]. The presence of reactive oxygen atoms, carboxyl and hydroxyl groups
D
enable GO to interact effectively with Congo red resulting in good adsorption [13]. In
PT E
aqueous media, the surface hydroxyl groups of TiO2 facilitate the adsorption of Congo red dye [14]. Up on UV irradiation, the contact between TNS and GO enables electron transfer from conduction band of TiO2 to GO. Electron transfer improves the photocatalytic activity
CE
by lowering the charge recombination rate. The electrons and holes generated during UV
AC
irradiation react with oxygen and water molecules in the media to form, form hydroxyl radicals (•OH) which enable photodegradation of the dye [8]. When the catalyst loading is increased the thickness of the catalyst coating becomes substantial. UV radiation can’t penetrate in to the entire thickness of the catalyst coating. Hence, the top portion of the catalyst coating which receives radiation will involve in photodegradation. The remaining underneath portion will adsorb the dye. Dye adsorption is undesirable since it adversely affects catalyst reusability. The thickness of the membrane
8
ACCEPTED MANUSCRIPT should be such that it photodegrades the maximum amount of dye and still offers minimal
MA
NU
SC
RI
PT
flux drop [13].
Fig.6. Reuse cycles (RC) of the membrane for Dye removal The dye removal study done in absence of UV irradiation evaluates the affect of adsorption.
PT E
D
It is observed that amount of dye absorbed is directly proportional to the catalyst loading. Beyond 100 mg catalyst loading adsorption played a much significant role. In case of M-400 which had given 100% dye removal under UV irradiation, is capable of removing 94.5% of
CE
the dye just by adsorption. It is evident that beyond M-100, increased loading didn’t improve
AC
photodegrdation. The membrane surface area exposed to UV radiation is constant and does not further increase with catalyst loading. By comparing the difference between dye removal in presence and absence of UV light, it is evident that M-100 membrane could achieve maximum photodegradation of dye. Also when compared to higher catalyst loaded membranes, M-100 exhibited superior permeate flux. The M-100 membrane is thus more efficient and reusable. The membrane dye removal reuse cycles were carried out and the results are shown below in Fig. 6. It is observed that except M-50 membrane all other membranes showed consistent dye removal traits. The adsorption phenomenon weakens with 9
ACCEPTED MANUSCRIPT continuous reuse of the photocatalyst. At higher catalyst loading the lower flux provides sufficient residence time for efficient dye degradation via photocatalysis. 4. Conclusions In summary, TNS-GO coated hierarchical membranes were successfully synthesised for simultaneous water filtration and photocatalytic degradation. At higher catalyst loading the
PT
permeate flux dropped heavily due to the additional cake resistance of the catalyst coating.
RI
Dye adsorptions played a major role at higher catalyst loading where as photodegradation
SC
was significant at lower catalyst concentration. Although M-400 could completely remove the dye, it exhibited lowest flux and greater dye adsorption. M-100 membrane showed best
NU
result with comparatively higher flux and better photodegradation. Photocatalyst reuse cycles revealed that higher TNS-GO loaded membranes could maintain dye removal capacity since
MA
the flux decline provided sufficient time for photocatlytic degradation. References
D
[1] A.K. Nair, P.M. Shalin, P.E. JagadeeshBabu, Performance enhancement of polysulfone
PT E
ultrafiltration membrane using TiO2 nanofibers, Desalin. Water Treat. 3994 (2015) 1–9. [2] L.M. Pastrana-Martínez, S. Morales-Torres, J.L. Figueiredo, J.L. Faria, A.M.T. Silva,
CE
Graphene oxide based ultrafiltration membranes for photocatalytic degradation of organic
AC
pollutants in salty water, Water Res. 77 (2015) 179–190. [3] H. Shi, R. Magaye, V. Castranova, J. Zhao, Titanium dioxide nanoparticles: a review of current toxicological data., Part. Fibre Toxicol. 10 (2013) 15. [4] Q. Zhang, H. Wang, X. Fan, F. Lv, S. Chen, X. Quan, Fabrication of TiO 2 nanofiber membranes by a simple dip-coating technique for water treatment, Surf. Coatings Technol. 298 (2016) 45–52. [5] H. Bai, Z. Liu, D.D. Sun, Hierarchically multifunctional TiO2 nano-thorn membrane for water purification, Chem. Commun. 46 (2010) 6542–6544. 10
ACCEPTED MANUSCRIPT [6] K. Chen, Z. Jiang, J. Qin, Y. Jiang, R. Li, H. Tang, X. Yang, Synthesis and improved photocatalytic activity of ultrathin TiO2 nanosheets with nearly 100 % exposed (001) facets, Ceram. Int. 40 (2014) 16817–16823. [7] Y. Gao, M. Hu, B. Mi, Membrane surface modification with TiO2-graphene oxide for enhanced photocatalytic performance, 455 (2014) 349–356.
PT
[8] P. Gao, Z. Liu, M. Tai, D. Delai, W. Ng, Applied Catalysis B: Environmental
RI
Multifunctional graphene oxide–TiO2 microsphere hierarchical membrane for clean water
SC
production, Applied Catal. B, Environ. 138-139 (2013) 17–25.
[9] Y. Masuda, K. Kato, Synthesis and phase transformation of TiO2 nano-crystals in aqueous
NU
solutions, J. Ceram. Society Japan. 117 (2009) 373–376.
[10] P. Uddandarao, R.M. B, ZnS semiconductor quantum dots production by an endophytic
MA
fungus Aspergillus flavus, Mater. Sci. Eng. B. 207 (2016) 26–32. [11] A. Dipareza, C. Lin, C. Wu, Removal of natural organic matter by ultrafiltration with
D
TiO2-coated membrane under UV irradiation, Journal of Colloid and Interface Science 323
PT E
(2008) 112–119.
[12] H. Bai, X. Zan, J. Juay, D. Delai, Hierarchical heteroarchitectures functionalized
CE
membrane for high efficient water purification, J. Membr. Sci. 475 (2015) 245–251. [13] S. Debnath, A. Maity, K. Pillay, Impact of process parameters on removal of Congo red
AC
by graphene oxide from aqueous solution, J. Environ. Chem. Eng. 2 (2014) 260–272. [14] E. Kordouli, K. Bourikas, A. Lycourghiotis, C. Kordulis, The mechanism of azo-dyes adsorption on the titanium dioxide surface and their photocatalytic degradation over samples with various anatase/rutile ratios, Catal. Today. 252 (2015) 128–135.
11
ACCEPTED MANUSCRIPT Highlights TNS-GO coated hierarchical membranes were synthesised.
At higher catalyst loading dye removal improved but permeate flux declined.
Dye adsorption played a major role at higher catalyst loading.
Photodegradation was significant at lower catalyst loading.
An optimum photodegradation with good flux was obtained in case of M-100 membrane
AC
CE
PT E
D
MA
NU
SC
RI
PT
12