Polyaniline composite for the visible-light driven hydrogen generation

Accepted Manuscript Enhanced photocatalytic performance of Ni-ZnO/Polyaniline composite for the visiblelight driven hydrogen generation Mohamed Faouzi Nsib, Samira Saafi, Ali Rayes, Noomen Moussa, Ammar Houas PII:

S1743-9671(14)20457-0

DOI:

10.1016/j.joei.2015.05.001

Reference:

JOEI 154

To appear in:

Journal of the Energy Institute

Received Date: 20 December 2014 Revised Date:

1 May 2015

Accepted Date: 6 May 2015

Please cite this article as: M.F. Nsib, S. Saafi, A. Rayes, N. Moussa, A. Houas, Enhanced photocatalytic performance of Ni-ZnO/Polyaniline composite for the visible-light driven hydrogen generation, Journal of the Energy Institute (2015), doi: 10.1016/j.joei.2015.05.001. 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 Enhanced photocatalytic performance of Ni-ZnO/Polyaniline composite for the visiblelight driven hydrogen generation Mohamed Faouzi Nsib a,b, Samira Saafi a, Ali Rayes a,b, Noomen Moussaab, Ammar Houas a,c URCMEP (UR11ES85), Faculty of Sciences, University of Gabès, 6029 Gabès, Tunisia

b

National School of Engineers (ENIG), University of Gabès, 6029 Gabès, Tunisia

RI PT

a

c

Al Imam Mohammad Ibn Saud Islamic University (IMSIU), College of Sciences, Department of Chemistry, Riyadh 11623, Saudi Arabia

Fax: (+216) 75 392 190

SC

* Corresponding author. Tel.: (+216) 97770192

M AN U

E-mail address: [email protected] (M. F. Nsib)

Abstract

NixZn1-xO/Polyaniline hybrid photocatalysts are synthesized and used for the experiments of

TE D

hydrogen production from water-splitting under visible irradiation. XRD, UV-Vis absorption and SEM-EDX are used to characterize the prepared materials. It is shown that the Ni2+ amount doped into ZnO controls its morphology and enhances its photoactivity for H2

EP

generation. Polyaniline (PANI) is shown to sensitize ZnO and to extend its light absorption toward the visible region. The hybrid photocatalyst with 10 mol. % Ni2+ and 10 wt. % PANI

AC C

shows the maximum photocatalytic H2 production for one hour of visible irradiation: ~ 558 µmole while only ~ 178 µmole in the presence of pure ZnO. Additives such as sacrificial electron donors are found to play a key role in the improvement of H2 evolution. Thus, the hydrogen photoproduction efficiency increases in the following order: thiosulfate > sulfide > propanol. All the obtained results show that Ni-ZnO/PANI hybrid photocatalyst is a good candidate to promote the H2 production using visible/solar lights. Key words: hydrogen production; photocatalysis; Ni doped ZnO; polyaniline; additives.

1

ACCEPTED MANUSCRIPT 1. Introduction Hydrogen is considered the future fuel that may solve the global energy and environmental problems. It is clean and energy-efficient [1, 2]. Produced without carbon dioxide or any

RI PT

other greenhouse gas affecting the climate, hydrogen can be the basis of a truly sustainable energy [3, 4]. Currently, there are several hydrogen-producing technologies, such as those based on fossil fuels, natural gas reforming and bio-derived liquids reforming, besides to coal and biomass gasification, thermochemical and nuclear production, and water electrolysis [5-

SC

9]. These technologies are either expensive or environmentally harmful.

M AN U

Heterogeneous photocatalysis is a cheap and environmentally-safe procedure that has been well-studied to treat the wastewater/air [10-13]. In recent years, hydrogen production using the photocatalytic water splitting has attracted many researchers. When developing photocatalysts, it is important to find a material with a lower energy band gap (Eg) that enables the efficient use of solar energy in order to produce hydrogen [14-17].

TE D

Zinc oxide (ZnO) has been considered a promising semiconductor to support the future hydrogen economy since it has distinctive optoelectronic, catalytic, and photochemical

EP

properties [18]. The quantum efficiency of ZnO is also noticeably larger than titanium dioxide (TiO2) [19, 20]. The energy conversion efficiency from solar to hydrogen by

AC C

photocatalytic water splitting, however, is still low, due to these main reasons: (i) Recombination of photogenerated electron/hole pairs, (ii) Fast backward recombination of hydrogen and oxygen into water, and (iii) Inability to use the visible part of the sunlight. To solve the above problems and make possible the solar photocatalytic production of hydrogen, we need, first, to use chemical additives (electron donors, noble metals, carbonate salts,..) in order to avoid the electron/hole and O2/H2 recombinations [21] and second, to change the photocatalyst in order to make it active under visible light [22, 23]. Particularly, it is reported that modifying ZnO with transition metal ions such as Ni2+ can restrain the 2

ACCEPTED MANUSCRIPT electron/hole recombination [24]. The Ni2+ ions can increase the surface defects and shift the light absorption towards the visible region [25]. On the other hand, the hybridization of ZnO with a functional polymer improves its morphology and allows better coverage of the solar emission spectrum, especially when the and

inorganic

networks

are

interpenetrated

[26-28].

The

ZnO/polymer

RI PT

organic

nanocomposites reveal some novel properties such as improved electrical conductivity and thermal stability. Besides, the synergistic effects of physical and chemical interactions

SC

between the organic and inorganic components can allow the use of ZnO/polymer hybrid as a photocatalyst for efficient H2 production under visible and solar lights [29-31]. Polyaniline

M AN U

(PANI) is among the most convenient conductive polymers [32]. It has an extended πconjugated electron system. PANI is an efficient electron donor and a good hole transporter upon visible-light excitation [33]. It has environmental, electrical, optical and electrochemical properties and can be chemically and electrochemically-synthesized [34-37]. Besides, it has a

TE D

high mobility of charge carriers and strong absorption in the visible spectrum due to its low band gap (2.8 eV) [33]. Consequently, PANI is a well-synthesizing candidate for the organicinorganic hybrid photocatalyst to produce H2 from water splitting under visible/solar lights

EP

[38]. Particularly, strong interactions exist between ZnO and PANI allowing the synthesis of

AC C

a stable ZnO/PANI hybrid photocatalyst [39, 40]. The main purposes of this research is to prepare an organic-inorganic hybrid photocatalyst based on Ni doped ZnO and PANI; and to investigate the effect of some synthetic parameters (Ni and PANI amounts) and additives (electron donors) on its structure and hydrogen production activity under visible irradiation.

2. Experimental

3

ACCEPTED MANUSCRIPT

2.1. Synthesis of Ni-ZnO materials The Ni-doped ZnO is prepared by a modified homogeneous precipitation using zinc acetate,

RI PT

oxalic acid and nickel chloride as precursors [41]. 4.8 g of zinc acetate and 2.5 g of oxalic acid are dissolved in 50 ml of distilled water under stirring for 2 h at room temperature. An amount of nickel chloride, which corresponds to Ni to ZnO molar ratio equal to 1, 5 and 10 %, respectively, is then added to the solution. The obtained precipitate is filtered, washed

SC

with methanol, dried in an oven at 80 °C for 24 h, and then heated at 400 °C for 3 h to get a

M AN U

Ni doped zinc oxide (Ni-ZnO).

Pure ZnO is prepared using the same method, but without adding the nickel chloride

TE D

precursor.

2.2. Synthesis of polyaniline (PANI)

The method used in our preparation of PANI is the oxidative polymerization of aniline

EP

monomer by ammonium persulfate (APS) in aqueous solution of H2SO4 at a pH of 1 to 3 [38]. In a typical synthesis, 50 mL of H2SO4 solution containing 1.4 g of APS was added

AC C

dropwise to 50 mL of H2SO4 solution containing 1.82 mL of aniline. The polymerization was allowed to proceed by stirring the mixture for 2 h at ambient temperature. The obtained dark green precipitate corresponding to PANI was then filtered and washed with a large amount of distilled water followed by methanol. The resulting PANI was finally dried for 24 h at 80 °C.

2.3. Synthesis of Ni-ZnO/PANI composites

4

ACCEPTED MANUSCRIPT The Ni-ZnO/PANI composites were prepared by the direct impregnation method. In a typical procedure, 1 g of Ni-ZnO nanoparticles was dispersed in 30 mL of aqueous PANI solution during one hour-stirring. The concentration of the PANI aqueous solution was such as the wt.% of PANI in the final Ni-ZnO/PANI hybrid was 3, 6 and 10 %, respectively. The

RI PT

obtained Ni-ZnO/PANI composite powders were filtered, washed three times with ethanol and water respectively, and dried for 24 h at 80 °C. The Ni-ZnO/PANI composites were labeled NixZn1-xO/PANIy where x (x = 1, 5 and 10 %) and y (y = 3, 6 and 10) are the Ni to

M AN U

SC

ZnO molar ratio and the wt.% of PANI in the Ni-ZnO/PANI composite, respectively.

2.4. Characterization of Ni-ZnO nanoparticles and Ni-ZnO/PANI hybrids The Ni-ZnO nanoparticles and Ni-ZnO/PANI hybrids were characterized by XRD, SEM and UV-Vis absorption.

TE D

The X-ray diffraction (XRD) patterns obtained on a SIEMENS HT/BT X-ray diffractometer using Cu Kα radiation at a scan rate of 0.02◦ s−1 were used to decide the identity of ZnO and

EP

the crystallite size of the powders. The average crystallite size was determined according to Scherrer’s equation: d = k.λ / β.cosƟ , where k is the shape factor of particles equal to 0.89, β

AC C

is the peak width at half maximum (in radians), λ is the X-ray wavelength (0.15418 nm), and Ɵ is the Bragg angle.

The SEM-EDX analysis was carried out with the help of a Neoscope JCM-5000 (JEOL Company, Japan) scanning electron microscope (SEM) equipped with an energy dispersive X-ray system. UV-Vis absorbance spectra of the samples were recorded using an UV-Vis 3101 PC (UV Probe Shimadzu) spectrophotometer. 5

ACCEPTED MANUSCRIPT 2.5. Photocatalytic hydrogen production experiments The typical liquid-phase photocatalytic reactor [42] used in our laboratory for H2 production is shown in the Fig.1. This photocatalytic reactor of 100 mL was irradiated with an outer 250 W halogen visible lamp. In each experiment, 1 g.L-1 of a photocatalyst and 0.2 M of Na2S2O3

RI PT

as a sacrificial electron donor were used. Before illumination, the aqueous slurry was deaerated with N2 gas and kept in the dark for 30 min. During irradiation, the evolved H2 gas was collected and volumetrically-measured in a water displacement graduated burette. It is

SC

worth pointing out that no chemical tests were done to ensure the presence of hydrogen, but every time, it was checked that the produced gas was a combustible one. All experiments

M AN U

were carried out under the same conditions and blank test was performed to determine the H2 photolysis volume. The dark control performed in the absence of light showed no noticeable H2 production. In order to confirm the reaction efficiency in the presence of the prepared photocatalysts, the apparent quantum yield (Q.Y.(H2)) was measured using the following

Q.Y.(H2) = 2 x

TE D

equation:
      
      

x 100

EP

The average photon flux of the incident light was determined by the method of the liquid-

AC C

phase chemical actinometry (Reinecke’s chemical actinometry).

3. Results and discussion

3.1. Materials characterization

3.1.1. X-ray diffraction The Fig. 2 shows the XRD patterns of the prepared pure and Ni-doped ZnO. They show peaks at positions 2 Ɵ = 31.63°, 34.22°, 36.03°, 47.3°, 56.43°, 62.79°, 66.31°, 67.84° and 6

ACCEPTED MANUSCRIPT 69.05°, which correspond to the reflection planes (100), (002), (101), (102), (110) (103), (200), (112) and (201), respectively. They are in well-agreement with the standard JCPDS file no. 36-1451 (a = b = 3.249 and c = 5.206 Ǻ) for the hexagonal wurtzite structure of ZnO. Moreover, no change in the wurtzite structure of ZnO by the Ni substitution is observed.

RI PT

Particularly, the lattice constants of Ni0.1Zn0.9O (a = b = 3.241 and c = 5.198 Ǻ) are slightly smaller than those of pure ZnO, showing that Ni2+ occupies the Zn2+ site into the crystl lattice [43]. Also, it should be noticed that wurtzite peaks of ZnO are slightly shifted (∆Ɵ = 4°) by

SC

Ni doping, which is a proof that Ni2+ is incorporated into ZnO lattice. No additional peaks corresponding to the secondary phases of nickel oxides are obtained when the nickel content

M AN U

is less than 10 % mol, but an additional small peak assigned to the (200) plane of NiO is shown at 2 Ɵ = 43.20° in the Ni0.1Zn0.9O pattern. This new peak clearly shows that a segregation phase occurred in the Ni0.1Zn0.9O sample. A possible reason of the absence of the NiO signal when Ni2+ concentration is less than 10 mol% may be the combination of its low

TE D

content and small particle size, due to the low sensitivity of the XRD analysis. Furthermore, the absence of diffraction peaks of NiO may possibly illustrate that NiO is highly dispersed in the bulk phase of the Ni-ZnO photocatalysts.

EP

The formation of NiO in a segregated phase make possible a p-n heterojunction between NiO

AC C

and ZnO oxides when the Ni content in the Ni-doped ZnO sample is high enough (10 mol %). This is expected to enhance the electron/hole separation and promote the photocatalytic activity. The possibility of NiO-ZnO heterojunction will be more discussed on the basis of UV-Vis absorption spectra. For the most part, the diffraction peaks are sharp, narrow and symmetrical with a low and stable baseline, suggesting that the samples are well-crystallized [44]. The average values of the crystalline size of the samples are calculated using the DebyeScherrer equation and the half width of the (101) crystal plane. Interestingly, the diffraction’

7

ACCEPTED MANUSCRIPT peaks of undoped ZnO are larger than those of the Ni-ZnO samples. The estimated average size of the samples from the XRD patterns is of the order of 22-26 nm. The increase of the ZnO crystalline size with the Ni doping is unexpected since the ionic radius of Ni2+ (0.69 Å) is smaller than that of Zn2+ (0.74 Å). This result can be explained taking into consideration

RI PT

the alteration of the ZnO morphology from nanoparticles to nanorods and the role of Ni2+ ions which may act as a catalyst to promote the growth of ZnO nanocrystals.

SC

3.1.2. SEM-EDX analysis

The influence of Ni2+ molar ratios on the photocatalysts’ surface morphologies was studied

M AN U

by SEM (Fig.3). The SEM images of pure ZnO (a) and 1 mol.% Ni-doped ZnO (b) show a spherical morphology of particles which are low-aggregated and evenly-distributed on the whole surface. However, the increase of the Ni content to 5 mol. % changes drastically the morphology of the Ni-ZnO particles (c). Thus, the Ni0.05Zn0.95O particles are more aggregated

TE D

and have nanorods-like morphology. The nanorods are quite uniform in size. Similar morphology is observed in the Ni0.1Zn0.9O sample (not shown here). These observations provide evidences that Ni content controls the growth of Ni doped ZnO crystals. Ni2+ ions

EP

may act as a catalyst to promote the one dimensional growth of ZnO nanocrystals. In this

AC C

case, Ni2+ may act as a nucleation site and help the growth of arrays of ZnO when the Ni2+ content reaches 5 mol. %. The nanorods-like morphology is maintained when Ni0.05Zn0.95O is hybridized with PANI (d). The nanorod morphology is expected to enhance the photosensitivity as well as the photoresponse [45]. These observations are more supported by the EDX analysis (Fig.4) which clearly shows the presence of Ni and C in the Ni0.1Zn0.9O/PANI10 sample compared to the ZnO one.

3.1.3. UV-Vis absorption spectra

8

ACCEPTED MANUSCRIPT UV–Vis absorption spectroscopy, which is generally used to detect the presence of framework and non-framework incorporated transition metal species in the structures and to distinguish the coordination states of the metal ions [46, 47], was used here to provide some insights into the interactions of the photocatalyst materials with photon energies.

RI PT

Fig.5a illustrates the UV-Vis absorption full spectra of PANI, bare ZnO, Ni0.01Zn0.99O, Ni0.01Zn0.99O/PANI3 and Ni0.1Zn0.9O/PANI10 photocatalysts. It is obvious that the bare ZnO absorbs light with wavelengths only below 390 nm. The spectrum of ZnO consists of a single

SC

absorption usually ascribed to charge-transfer from the valence band (formed by 2p orbitals of the oxide anions) to the conduction band (formed by 3d orbitals of the Zn2+ cations).

M AN U

It is also clear in Fig.5a that the absorbance of Ni-doped ZnO nanoparticles in the visible region is higher than that of pure ZnO nanoparticles. Particularly, it can be seen that the band edge at 390 nm of the undoped ZnO is shifted towards higher wavelength (405 nm) for Ni0.01Zn0.99O. The red shift of the Ni-ZnO absorption peaks compared to the undoped ZnO

TE D

can be ascribed to the increase of particle size of Ni-ZnO samples, as it is shown in XRD analysis. Moreover, the absorption bands of Ni0.01Zn0.99O appeared at 430 and 680 nm correspond to the d-d transition bands, which are characteristics of Ni2+ with tetrahedral

EP

symmetry [33]. This result shows, again, the incorporation of Ni ions into the ZnO lattice and

AC C

their substitution in the tetrahedral sites of the wurtzite structure when the Ni content is weak. It is assumed that the enhanced optical activity of lower Ni-doped ZnO, such as Ni0.01Zn0.99O is owed to the increase in the surface imperfections due to Ni doping in ZnO nanoparticles. This is expected to be driven by the interaction of the Ni 3d and O 2p states in both the valence and conduction bands [48]. It means that defect sites are generated in the vicinity of valence band due to doping of ZnO with Ni ions. As a result, the electron-hole pair is generated within the effective band gap following the photoexcitation of particles. Therefore,

9

ACCEPTED MANUSCRIPT the electron transition occurs from the defect valence state to the defect conduction one. This transition requires much lower energy than the band gap of ZnO. On the contrary, the photoactivity of Ni-ZnO sample with high Ni-content (10 mol.%) is thought to be enhanced owing to the p-n heterojunction which can be established between

RI PT

NiO and ZnO oxides. This thought is supported by XRD and UV-Vis absorption results. In particular, the UV-Vis absorption spectrum of Ni0.1Zn0.9O/PANI10 shows peak with maximum at 500 nm, attributed to the charge transfer from Ni2+ to Zn2+. These transitions are

SC

enabled by the overlap of the conduction band of 3d level of Zn2+ with 3d level of Ni2+ in the junction region of NiO/ZnO [49]. The electronic conduction band’ interaction of the Zn 3d

M AN U

orbitals of zinc oxide and the Ni 3d orbitals of NiO can lead to the decrease in the energy gap between the Zn d and O p orbitals of ZnO. It leads to a visible light absorption. All these observations reveal that NiO and ZnO are well-contacted.

The UV-Vis absorption spectrum of PANI reveals maximums at 600, 370 and 300 nm, which

TE D

originate from the charge-transfer excitation from the highest occupied energy level to the lowest unoccupied one and the  −  ∗ transition [49], respectively. Besides, the peak from PANI at 600 nm is observed to be shifted to 570 nm, indicating the interaction of PANI with

EP

Ni-ZnO nanoparticles. After being sensitized by PANI, the Ni0.01Zn0.99O/PANI3 composite

AC C

photocatalyst cannot absorb only the UV light, but can also absorb visible light more than pure ZnO and Ni0.01Zn0.99O. We observed a noticeable shift of the optical absorption edge of the Ni-ZnO/PANI system towards the visible region. Surely, this shift towards the longer wavelengths originates from the band gap narrowing of Ni-ZnO by PANI hybridization. The band gap as calculated for pure ZnO, Ni0.01Zn0.99O, Ni0.01Zn0.99O/PANI3 and Ni0.1Zn0.90O/PANI10 is 3.12, 3.07, 2.98 and 2.90 eV, respectively. The band gap energy of the photocatalysts samples was determined using the Tauc equation [50]: αhν = A (hν – Eg)n

10

ACCEPTED MANUSCRIPT Where α is the absorption coefficient, A is a constant and n = ½ for direct band gap semiconductor. The extrapolation of the plot of (αhν)2 vs. hν (Fig.4b, inset in the figure 4) gives the value of the band gap [51, 52]. The above results indicate that PANI is able to sensitize ZnO efficiently. The resulted

RI PT

composite photocatalyst can be excited by absorbing both UV and visible lights in order to produce more electron-hole pairs and give a maximum visible-light harvesting. This, in turn, can result in higher photocatalytic activities, and therefore it can be a promising

SC

photocatalytic material for the efficient use of light, mainly sunlight in the photoproduction of

M AN U

hydrogen.

3.2. Photocatalytic hydrogen production in the presence of NixZn1-xO/PANIy 3.2.1. Effect of Ni2+ content

TE D

The effect of Ni2+ contents on the photocatalytic activity of the Ni-ZnO/PANI material under visible light is examined while keeping constant the amount of PANI in the hybrid

EP

composition. Fig.6 presents the time course of the photocatalytic hydrogen generation by NixZn1-xO/PANI3 photocatalyst with different Ni2+ content. It is clearly shown that the

AC C

increase of Ni content in the Ni-ZnO/PANI photocatalyst significantly increase its activity for the hydrogen photoproduction under visible illumination. For one hour of visible irradiation, the hydrogen production of NixZn1-xO/PANI3 photocatalysts reached the points of 8.5 mL (~ 379 µmole), 9.1 mL (~ 406 µmole) and 9.8 mL (~ 437 µmole) for x = 0.01, 0.05 and 0.1, respectively. These productions are much higher than that of pure ZnO on the same reaction condition, which is only 4 mL (~ 178 µmole). This tendency can be explained by two main factors: The ability to absorb visible light and to separate the photogenerated charges.

11

ACCEPTED MANUSCRIPT As shown above by UV-Vis absorption results, Ni doped ZnO absorbs more visible light than the undoped one, and so, act as a better photocatalyst under visible light irradiation. Besides enhancing the ability to absorb visible light, Ni doping is expected to enhance the separation of electron and hole and avoid their fast recombination, but the extent of this enhancement

RI PT

increase with the Ni content. At low doping of ZnO with Ni ions, defect sites are suggested to be generated in the vicinity of valence band which may facilitate the electron transition between bands. This mechanism seems to be less effective for the electron-hole separation

SC

than that considered when the Ni content is higher (10 mol%). As shown above, NiO-ZnO heterostructure is formed at high Ni content. Consequently, the NiO can act as a co-catalyst

M AN U

to improve the photocatalytic activity of the material. In addition, a p-n heterojunction is likely to occur at the p-NiO/n-ZnO interface. The p-n junction is expected to induce an internal electric field that can extend the probability of electron-hole separation and, then

TE D

improve the photoactivity of the composite material [53, 54].

3.2.2. Effect of PANI amount

EP

The effect of the PANI amount on the hydrogen production is investigated using

AC C

Ni0.1Zn0.9O/PANIy photocatalysts. Fig.7 describes the obtained results when y values are respectively 3, 6 and 10 wt.%. It is found that the hybridization of Ni-ZnO with PANI improves its photocatalytic activity under visible light irradiation. As a result, the volume of hydrogen photoproduced increases in the presence of Ni0.1Zn0.9O/PANIy photocatalysts compared to the non-hybridized Ni0.1Zn0.9O. This enhancement is more noticeable when the amount of PANI in the Ni0.1Zn0.9O/PANIy composite increases. Thus, for one hour of visible irradiation, the volume of photoproduced hydrogen reaches 9.8 mL (~ 437 µmole) with Ni0.1Zn0.9O/PANI3, 10.9 mL (~ 486 µmole) with Ni0.1Zn0.9O/PANI6, 12.5 mL (~ 558 µmole)

12

ACCEPTED MANUSCRIPT with Ni0.1Zn0.9O/PANI10 and only 6.6 mL (~ 294 µmole) with Ni0.1Zn0.9O. In addition, it is found that the apparent quantum yield (Q.Y.H2) of hydrogen evolution increases with the amount of PANI in the Ni0.1Zn0.9O/PANIy photocatalysts (Table 1). The amounts of hydrogen production measured in this work in the presence of Ni-ZnO/PANI hybrid photocatalysts are

RI PT

significantly higher than those found when using some other photocatalysts [55-57]. Two reasons may explain the increase of the photocatalytic activity of Ni-ZnO/PANI hybrids compared to that of pure ZnO or Ni-doped ZnO: First of all, Chemical interactions are

SC

expected to occur between ZnO nanoparticles and PANI conducting polymer. Consequently, electrons transfer from PANI to ZnO, leading to an increase of electrons density in ZnO. As a

M AN U

result, the amount of photoexcited electrons is increased and the photocatalytic activity of the composites for hydrogen production is improved. Second, the disordered structure of PANI may be improved and has some order after addition of ZnO nanoparticles, which may make easier the transfer of electrons from PANI to ZnO. Subsequently, electrons are easily

TE D

photoexcited in ZnO and the photocatalytic activity of Ni-ZnO/PANI composites for hydrogen evolution is improved.

EP

In other words, it is suggested that the hybridization of ZnO with the conducting polymer leads to the increase of the rate of electron transfer to the ZnO surface. This behavior plays a

AC C

considerable role of increasing hydrogen production. In fact, as the surface of ZnO is saturated by the accumulated photoexcited electrons, the Fermi level shifts nearer to the conduction band, and the energy moves into the more negative level. Besides, the accumulated electrons will be transferred into the protons which are adsorbed at the catalyst surface. Hence, the protons are reduced into hydrogen.

3.2.3. Effect of the nature of the sacrificial electron donor

13

ACCEPTED MANUSCRIPT Fig.8 shows the photocatalytic hydrogen evolution activities over a Ni0.1Zn0.9O/PANI10 photocatalyst from various sacrificial electron donor solutions. Here, sulfide (S2-), thiosulfate (S2O32-) and propanol (C3H7OH) are used as electron donor, respectively. The reactions were conducted with electron donor concentrations of 0.2 mol.L-1.It is shown that the hydrogen

RI PT

evolution over Ni0.1Zn0.9O/PANI10 photocatalyst is quite stable in the presence of the tested sacrificial electron donors. This result shows that these electron donors react irreversibly with the photoinduced holes. Moreover, it is observed that the hydrogen photoproduction

SC

efficiency depends strongly on the nature of the sacrificial electron donor and increases in this order: thiosulfate > sulfide > propanol. For one visible irradiation-hour, the volumes of

M AN U

photoproduced hydrogen are: 12.5 mL ~ 558 µmole (thiosulfate), 9.3 mL ~ 415 µmole (sulfide), and 7.7 mL ~ 343 µmole (propanol).

The photocatalytic hydrogen generation in the presence of propanol may take place stepwise involving intermediates such as aldehydes and acids, which may compete between them and

TE D

propanol with the valence band holes (h+) [58]. This may partially explain the decrease of the amount of hydrogen photoproduced in the presence of propanol compared to S2- and S2O32-.

EP

Under our experimental conditions (pH = 7-8), sulfide ions (S2-) are less abundant in the electron donor solution than bisulfide ions HS- (pka2 (HS-/S2-) = 13). HS- can trap easily the

AC C

valence band holes during a single stepwise reaction [59]: HS- + h+

HS.

(1)

The small amount of S2- present in the solution may diffuse slower than HS- to the photocatalyst surface and react with holes as follows: S2- + h+

S.-

(2)

14

ACCEPTED MANUSCRIPT Thiosulfate anions (S2O32-) are the most efficient electron donors experimented here; they potentially increase the reaction rate of the photocatalytic hydrogen generation. The photogenerated semiconductor holes (h+) are scavenged by the added S2O32- anions. S2O32anions are oxidized by direct electron transfer to the photogenerated holes during a single

2 S2O32- + 2 h+

S4O62-

RI PT

stepwise reaction: (3)

This is supposed to lead to decrease the electron/hole recombination. Therefore, the

2H+ + 2 e-cb

H2

(4)

M AN U

delivered to the pre-adsorbed H+ to produce H2:

SC

photoinjected electrons in the semiconductor conduction band are increased and rapidly

Because the standard redox potentials of HS-/HS. and S2O32-/S4O62- couples are 1.08 and 0.08 V, respectively, thiosulfate oxidation can take place more efficiently than that of sulfide/bisulfide. Thus, photocatalytic hydrogen evolution is more effective in the presence of

TE D

thiosulfate anions as sacrificial electron donors.

The above results suggest that the rate of H2 production is mainly governed by the reduction potential of the sacrificial electron donor and by the kinetic of the electron transfer process.

EP

The adsorption behavior of the electron donors is may be another important factor that can

AC C

explain the difference in hydrogen evolution between the three donors. However, their exact adsorption nature is remained uncertain, so an investigation of the surface reactions is ongoing.

3.2.4. Effect of photocatalyst loading One of the operational parameters which can affect the photocatalytic hydrogen generation is the catalyst loading. To illustrate this effect, experiments were done with Ni0.1Zn0.9O/PANI3 amount of 0.5, 1 and 2 g.L-1, respectively. S2O32- 0.2 M was used as the electron donor. It is 15

ACCEPTED MANUSCRIPT shown that the amount of H2 photogenerated increases with increasing the catalyst loading. For one irradiation-hour, the H2 volume reaches about 8, 10 and 15 mL in the presence of 0.5, 1 and 2 g.L-1 of catalyst concentration, respectively. The above results show that the suspended Ni0.1Zn0.9O/PANI3 did not reduce the penetration of the light in the solution and no

RI PT

severe scattering-light was occurred for larger catalyst loading (2 g.L-1). It means that the

catalyst.

M AN U

3.2.5. Photocatalytic mechanism of H2 generation

SC

number of available photons inside the photoreactor increased with the amount of the

The increase in the photocatalytic activity of Ni-ZnO/PANI compared to bare ZnO is supposed to be essentially attributed to the efficient absorption of visible light and the electron-hole separation. Both Ni2+ ions and PANI have significant contribution and great

TE D

influence on the photocatalytic activity of Ni-ZnO/PANI composite for the hydrogen production. It can be hypothesized that at low Ni content, Ni2+ doping into the ZnO lattice introduces an additional electron energy level into the valence band, which may reduce the

EP

energy band gap [60, 61]. When Ni-ZnO/PANI composite is illuminated under visible light,

AC C

both Ni-ZnO and PANI absorb the photons at their interface. Since the lowest unoccupied molecular orbital (LUMO) level of PANI is more negative than that of the CB of ZnO, the photogenerated electrons can be transferred from PANI to the CB of ZnO where protons can be reduced to hydrogen. At higher Ni content, NiO-ZnO heterostructure is shown to be formed. Consequently, NiO is likely to act as a cocatalyst or/and to establish a p-n junction with ZnO. As a result, electrons are transferred from both NiO and PANI to the CB of ZnO whereas holes are transferred to the VB of NiO. This proposed mechanism may explain the

16

ACCEPTED MANUSCRIPT better results obtained with the Ni-ZnO/PANI photocatalyst of high Ni content (10 mol %) compared to the lower one.

Conclusion

RI PT

4.

Ni-ZnO/PANI hybrid photocatalysts were prepared by a simple impregnation method. As the organic partner, PANI can extend the photoresponse of ZnO to the visible region and narrow its band gap energy. The Ni2+ ions were found to substitute Zn2+ into the ZnO lattice and its

SC

content was shown to control the ZnO morphology. NiO-ZnO heterostructure was shown to

M AN U

be formed at high Ni content (10 mol %). Throughout this research, it is found that both Ni2+ ions and PANI have a great influence on the photocatalytic activity of Ni-ZnO/PANI composite. The Ni0.1Zn0.9/PANI10 displayed the highest photoactivity for hydrogen production under visible-light irradiation. The nature of the added sacrificial electron donor was found to increase the photoproduction efficiency of hydrogen in this order: thiosulfate >

TE D

sulfide > propanol. Moreover, a photocatalytic mechanism of H2 generation is proposed depending on the Ni content. Subsequently, the Ni-ZnO/PANI hybrid photocatalyst is a good

AC C

EP

candidate to promote the H2 production using visible/solar lights.

References

[1] C-H Liao, C-W Huang, JCS Wu. Catalysts, 2012, 2: 490.

[2] CE Thomas, BD James, Jr FD Lomax, Jr IF Kuhn. Int J Hydrogen Energy, 2000, 25: 551. [3] V Preethi, S Kanmani. Mat Sci Semicon Proc, 2013, 16 (3):561. [4] X Chen, S Shen, L Guo, S S Mao. Chem Rev, 2010, 110: 6503. [5] A Damien. Techniques de l’Ingénieur, 1992, J6366. 17

ACCEPTED MANUSCRIPT [6] S Turn, C Kinoshita, Z Zhang, D Ishimura, J Zhou. Int J Hydrogen Energy, 1998, 23: 641.

[7] Y Chen, Y Wang, H Xu, G Xiong. App Catal B: Environ, 2008, 80: 283.

RI PT

[8] GF Naterer. Hydrogen Production from Nuclear Energy. London: Springer-Verlag, 2013. [9] C Gaudillere, L Naverrete, J M Serra. Int J Hydrogen Energy, 2014, 39: 3047.

[10] F Han, V S R Kambala, M Srinivasan, D Rajarathnam, R Naidu. Appl Catal A-General,

SC

2009, 359: 25.

M AN U

[11] MN Chong, B Jin, CWK Chow, C Saint. Water Res, 2010, 44: 2997.

[12] AA Tellez, R Masson, D Robert, N Keller, V Keller. J Photochem Photobiol A-Chem, 2012, 250: 58-65.

TE D

[13] MF Nsib, A Mayoufi, N Moussa, N Tarhouni, A Massouri, A Houas, Y Chevalier. J Photochem Photobiol A-Chem, 2013, 251: 10.

EP

[14] A Kudo. Pure Appl Chem, 2007, 79 (11): 1917.

AC C

[15] A Kudo. Int J Hydrogen Energy, 2007, 32 (14): 2673.

[16] D Jing, L Guo, L Zhao, X Zhang, H Liu, M Li, S Shen, G Liu, X Hu, X Zhang, K Zhang, L Ma, P Guo. Int J Hydrogen Energy, 2010, 35 (13): 7087.

[17] K Villa, X Domènech, S Malato, M I Maldonado, J Peral. Int J Hydrogen Energy, 2013, 38 (29): 12718.

[18] AL Stroyuk, VV Shvalagin, S Ya Kuchmii. J Photochem Photobiol A-Chem, 2005, 173(2): 185. 18

ACCEPTED MANUSCRIPT [19] FD Mai, CS Lu, CW Wu, CH Huang, JY Chen, CC Chen. Sep Purif Technol, 2008, 62: 423. [20] Y Li, W Xie, X Hu, G Shen, X Zhou, Y Xiang, X Zhao, P Fang. Langmuir, 2010, 26: 591.

RI PT

[21] M Ni, M KH Leung, DYC Leung, K Sumathy. Renew Sust Energ Rev, 2007, 11(3): 401.

[22] H Park, Y Park, W Kim, W Choi. J Photochem Photobiol C-Rev, 2013, 15: 1.

SC

[23] R Qiu, D Zhang, Y Mo, L Song, E Brewer, X Huang, Y Xiong. J Haz Mater, 2008, 156: 80.

M AN U

[24] Z Yin, N Chen, F Yang, S Song, C Chai, J Zhong. Solid State Commun, 2005, 135: 430. [25] S Kant, A Kumar. Adv Mat Lett, 2012, 3(4): 350.

[26] G Kickelbick. Hybrid Materials. Synthesis, Characterization, and Applications, Weinheim: Wiley-VCH Verlag GmbH & Co. KGaA, 2007.

TE D

[27] J Huang, Z Yin, Q Zheng, Energ Environ Sci, 2011, 4(10): 3861. [28] S Kango, S Kalia, A Celli, J Njuguna, Y Habibi, R Kumar. Prog Polymer Sci, 2013, 38 :1232.

EP

[29] Y Yu, J Ren, M Meng. Int J Hydrogen Energy, 2013, 38 (28): 12266.

AC C

[30] Y Wang, X Guo, L Dong, G Jin, Y Wang, X-Y Guo. Int J Hydrogen Energy, 2013, 38 (29): 12733.

[31] X Zhang, Z Jin, Y Li, S Li, G Lu. J Coll Interface Sci, 2009, 333 (1): 285. [32] A Pron, P Rannou. Prog Polymer Sci, 2002, 27 (1): 135-190. [33] F Wang, S Min, Y Han, L Feng. Superlattice Microst, 2010, 48 (2): 170. [34] S B Kondawar, S A Acharya1, S R Dhakate. Adv Mat Lett, 2011, 2(5): 362. [35] A Katoch, M Burkhart, T Hwang, S S Kim. Chem Eng J, 2012, 192 (1): 262. [36] A Pron, F Genoud, C Menardo, M Nechtschein. Synthetic Met, 1988, 24 (3): 193.

19

ACCEPTED MANUSCRIPT [37] H Okamoto, M Okamoto, T Kotaka. Polymer, 1998, 39 (18): 4359. [38] K He, M Li, L Guo, Int J Hydrogen Energy, 2012, 37 (1): 755. [39] S B Kondawar, S A Acharya, S R Dhakate, Adv Mat Lett, 2011, 2(5): 362-367.

RI PT

[40] S M Giripunje, J Ghushe, Int J App Phys Math, 2012, 2(4): 234-236. [41] S Khatoon, T Ahmad, J Mater Sci Eng B, 2012, 2 (6): 325-333. [42] A Koca, M Şahin. Int J Hydrogen Energy, 2002, 27 (4): 363.

SC

[43] G Murugadoss, J Mater Sci Technol, 2012, 28 (7): 587.

[44] RJ Wen, X Fan, Z Yang, Z Tan, B Yang. Electrochim Acta, 2012, 83: 376.

M AN U

[45] O Yalçın. Nanorods, Croatia: Published by InTech, Janeza Trdine 9, 51000 Rijeka, 2012.

[46] I Fechete, O Ersen, F Garin, L Lazar, A Rach. Catal Sci Technol, 2013, 3: 444. [47] S Haddoum, I Fechete, B Donnio, F Garin, D Lutic, CE Chitour. Catal Comm, 2013, 27:

TE D

141.

[48] B Pall, M Sharon. Mater Chem Phys, 2002, 76: 82.

EP

[49] S Singh, N Rama, MSR Rao. Appl Phys Lett, 2006, 88 (22): 222111. [50] S Suwanboon, P Amornpitoksuk, A Haidoux, JC Tedenac. J Allys Compd, 2008, 462:

AC C

335.

[51] H Yang, S Nie. Mater Chem Phys, 2009, 114: 279. [52] IK El Zawawi, RA Abd Alla. Thin solid films, 1999, 339 (1-2): 314. [53] N Helaili, Y Bessekhouad, A Bouguelia, M Trari. Solar energy, 2010, 84 (7): 1187. [54] C Luo, D Li, W Wu, Y Zhang, C Pan. RSC Adv, 2014, 4: 3090.

[55] T Lyubina, D V Markovskaya, E A Kozlova, VN Parmon. Int J Hydrogen Energy, 2013, 38: 14172. 20

ACCEPTED MANUSCRIPT [56] X Zhang, D Jing, L Guo. Int J Hydrogen Energy, 2010, 35: 7051. [57] AA Nada, MH Barakat, HA Hamed, NR Mohamed, TN Veziroglu. Int J Hydrogen Energy, 2005, 30: 687.

RI PT

[58] B Zielinska, EB Palen, RJ Kalenczuk. Int J Hydrogen Energy, 2008, 33: 1797. [59] Y Li, F He, S Peng, G Lu, S Li. Int J Hydrogen Energy, 2011, 36 (17): 10565.

SC

[60] J Wang, Z Wang, B Huang, Y Ma, Y Liu, X Qin, X Zhang, Y Dai. ACS Appl Mater Interfaces, 2012, 4 (8): 4024.

M AN U

[61] SC Das, RJ Green, J Podder, TZ Regier, GS Chang, A Moewes. J Phys Chem C, 2013,

AC C

EP

TE D

117 (24):12745.

21

ACCEPTED MANUSCRIPT Caption of table Table.1. The quantum Yield (Q.Y.H2) of H2 generation in the presence of undoped ZnO and

RI PT

Ni0.1Zn0.9O/PANIy photocatalysts.

Caption of figures

Fig.1. Schematic diagram showing liquid-phase photocatalytic H2 production from water-

SC

splitting.

M AN U

Fig.2. XRD profiles of prepared pure ZnO, Ni0.01Zn0.99O and Ni0.1Zn0.9O.

Fig.3. SEM images of bare ZnO nanoparticles (a), Ni0.01Zn0.99O (b), Ni0.05Zn0.95O (c) and Ni0.05Zn0.95O/PANI3 (d) photocatalysts.

Fig.4. Energy-dispersive X-ray (EDX) of (a) ZnO and (b) Ni0.1Zn0.9O/PANI10 samples. UV–vis

spectra

of

PANI,

ZnO,

TE D

Fig.5.

Ni0.01Zn0.99O,

Ni0.01Zn0.99O/PANI3

and

Ni0.1Zn0.9O/PANI10 nanocomposite (a) and the corresponding plot of (αhν)2 vs. hν (b).

EP

Fig.6. Effect of Ni2+ content on the volume of hydrogen evolution (Vsolution = 100 mL, catalyst = 0.1 g NixZn1-xO/PANI3; x = 0, 0.01, 0.05 and 0.1, electron donor: Na2S2O3 0.2 M).

AC C

Fig.7. Effect of PANI wt.% on the volume of hydrogen evolution (Vsolution = 100 mL, catalyst = 0.1 g Ni0.1Zn0.9O/PANIy, y = 3, 6 and 10, electron donor: Na2S2O3 0.2 M). Fig.8. Effect of the nature of the added electron donor on the volume of hydrogen evolution (Vsolution = 100 mL, catalyst = 0.1 g Ni0.1Zn0.9O/PANI10).

22

ACCEPTED MANUSCRIPT

Q.Y.H2

ZnO

4.69

Ni0.1Zn0.9O/PANI3

11.5

Ni0.1Zn0.9O/PANI6

12.8

Ni0.1Zn0.9O/PANI10

14.7

AC C

EP

TE D

M AN U

SC

Photocatalyst

RI PT

Table 1

23

ACCEPTED MANUSCRIPT

SC

RI PT

Figures

AC C

Fig.2

EP

TE D

M AN U

Fig.1

24

RI PT

ACCEPTED MANUSCRIPT

AC C

Fig.4

EP

TE D

M AN U

SC

Fig.3.

25

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

Fig.6.

TE D

M AN U

SC

Fig. 5.

Fig.7.

26

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

Fig.8.

27

ACCEPTED MANUSCRIPT

Highlights

EP

TE D

M AN U

SC

RI PT

NixZn1-xO/PANIy hybrids are synthesized and used in H2 photocatalytic generation. Ni2+ amount controls the morphology of ZnO and enhances its photoactivity. Both Ni2+ and PANI extend the light absorption of ZnO toward the visible region. Both Ni2+ and PANI enhance the electron-hole separation.

AC C