lable at ScienceDirect

Energy 73 (2014) 926e932

Contents lists avai

Energy

journal homepage: www.elsevier .com/locate/energy

Development and performance of bench-scale reactor for thephotocatalytic generation of hydrogen

Priya Ruban*, Kanmani SellappaCentre for Environmental Studies, Anna University, Chennai 600 025, India

a r t i c l e i n f o

Article history:Received 13 August 2013Received in revised form7 June 2014Accepted 28 June 2014Available online 23 July 2014

Keywords:Hydrogen generationHydrogen sulfidePhotocatalysisPollution control

* Corresponding author. Tel.: þ91 9841634405.E-mail addresses: Chempriya_2001@yahoo.co.in (P

com (K. Sellappa).

http://dx.doi.org/10.1016/j.energy.2014.06.1070360-5442/© 2014 Elsevier Ltd. All rights reserved.

a b s t r a c t

In this study, a new novel bench-scale (5 L) tubular photocatalytic reactor was developed and itsfeasibility studies were conducted for optimizing the operating variables, namely concentration of sulfideion, concentration of sulfite ion, pH, catalyst concentration, lamp power, volume of wastewater andrecycle flow rates at batch recycle mode for the generation of hydrogen from aqueous sodium sulfideusing CdSeZnS/TiO2 coreeshell NPs (nanoparticles). The maximum H2 generation was found at 0.05 Mconcentration of sulfide ion, 0.2 M concentration of sulfite ion, pH 11.3, 0.5 g/L catalyst concentration andrecycle flow rate of 18 L/h. Reusability studies were conducted for analyzing stability of photocatalyst.The results showed that the generation of hydrogen depends on light intensity, photoreactor used, na-ture of photocatalysts and the operating conditions.

© 2014 Elsevier Ltd. All rights reserved.

1. Introduction

Hydrogen sulfide (H2S) is produced in large quantities as abyproduct in coal and petroleum industries, natural gas and oilwells, and geothermal plants. Sulfide wastewater is generated fromSTPs (sewage treatment plants), tanning of hides and pulp andpaper industries. In most cases, this toxic H2S gas and sulfidewastewater has to be converted into an environmentally lessharmful form to comply with environmental regulations [1].Various treatment processes are available for the removal of H2Sgas and sulfide wastewater. As these processes have some draw-backs, new emerging technologies viz., biological [2], thermo-chemical, electrochemical and photochemical are applied for thehydrogen (H2) production through waste minimization. Amongthese approaches, photocatalysis has received attention as apossible method for photochemical conversion and storage of solarenergy. Solar energy is the primary source for clean and renewableenergy alternatives. Thus, simultaneous hydrogen (H2) generationand H2S decomposition in an alkaline solution is a highly desirableprocess that could satisfy both energy and environment re-quirements [3e5].

Hydrogen (H2) is regarded as the fuel of the future and is one ofthe most suitable energy carriers in the context of sustainable

. Ruban), skanmani@hotmail.

development. Recently, biomass-based energy technologiesemployed for CO2 capture and also for the production of H2 [6]. H2is produced from fossil fuels as well as from biomass by estab-lished technologies [7] but the conventional sources should bereplaced by alternative sources such as solar energy. H2 generationfrom cheap raw materials like H2S is a challenging method thatcould provide a solution for future energy needs. So far, manyphotocatalytic research studies have been focused on exploringnew active photocatalysts on H2 generation. The reported oxidephotocatalysts, such as NaTaO3 [8] and Sr2Nb2O7 [9e10] exhibithigh photocatalytic activities for H2 generation from water, how-ever, these oxide photocatalysts are only active under UV lightwhich constitutes only 5% of the solar spectrum [11]. Therefore, itis indispensable to develop photocatalyst responding to visibleportion of solar spectrum. Sulfide semiconductors, such as CdS andZnS with narrow band gaps and valence bands at relativelynegative potentials compared to oxide semiconductors can begood candidates for visible light driven photocatalysts. Titania(TiO2) has attracted extensive attention for many applicationssuch as photocatalysis, gas sensing, solar water splitting, dye-sensitized solar cells and solid state dye-sensitized solar cells. 1D(One dimensional) TiO2 materials including nanotubes, andnanotube arrays are currently of great interest owing to the pos-sibility for suppression of charge recombination [12e13]. Thecoupling of two or more semiconductors and nanomaterials canbe effectively used for the generation of H2 and also it eliminatesthe use of noble metals [14]. Employing artificial lamps for this

P. Ruban, K. Sellappa / Energy 73 (2014) 926e932 927

technology is not advantageous. Harvesting of renewable andabundantly available renewable solar energy would be beneficial.Using of solar UVevisible light responsive photocatalysts wouldbe a promising method to achieve the commercial application ofthis technology. For the transfer of this technology in society, thedevelopment of a suitable photocatalytic reactor is needed.Particularly in a subtropical country like India where sunlight isabundant almost throughout the year, research on solar photo-catalytic generation of H2 from H2S in an alkaline solution will bequite promising.

Several parameters govern or influence the kinetics of thephotocatalytic H2 generation process. Kinetics of the process isneeded for the scale-up of photocatalytic reactors. Very few studieswere focused on the development of photoreactor for the genera-tion of H2 [15e20]. Linkous et al. [15e16] conducted pioneeringwork on development of photocatalytic reactors for the generationof H2. Earlier batch recycle photocatalytic reactor with N2 bubblingwas developed in our laboratory for the decomposition of H2S inalkaline solution for the generation of H2 [17]. Jing et al. (2010)developed solar photocatalytic hydrogen reactor for the productionof H2. The maximum production rate amounted to 1.88 L/h [18].Katherine Villa et al. (2013) CPC (Compound Parabolic Concen-trator) solar reactor has been used to generate hydrogen fromaqueous solutions of formic acid, glycerol, or a real wastewater, andusing Pt/(TiO2eN) and Pt/(CdSeZnS) as photocatalysts. Theapproximate energy efficiencies calculated with the highesthydrogen production rates were 2.5% and 1.6% for Pt/(TiO2eN) andPt/(CdSeZnS), respectively [19]. Recently, Jing et al. (2013) devel-oped fluidized fluidized-bed photocatalytic reactor with H2Sbubbling. The reaction rate constant was 1.12 E�1 [20].

The influence of several experimental variables such as effect ofconcentration of sulfide ion [21e27], concentration of sulfite ion[24,25,28e30], pH [27,31], catalyst concentration [32e35], lamppower [32,34e35], volume of wastewater [36e37] and recycle flowrates [17,38] on the rate of photocatalytic H2 generation fromaqueous sodium sulfide was reported for various catalytic systems.Developing photoreactors and conducting kinetic studies for opti-mization of operating variables are an important issues for scale up.Compared to other configurations, a cylindrical shape is bettersuited for transmitting the light radiation. The need of the hour is togenerate H2 by the utilization of solar energy using efficientnanophotocatalysts and appropriate photoreactors.

In this present study, CdSeZnS/TiO2 coreeshell NPs (nano-particles) was synthesized for the photocatalytic H2 generationfrom aqueous sodium sulfide solution and the performance ofphotocatalytic reactor with respect to various process parameterssuch as concentration of sulfide ion, concentration of sulfite ion, pH,catalyst concentration, lamp power, volume of wastewater andrecycle flow rates were studied.

2. Material and methods

All chemicals used were reagent grade or high quality used assupplied. The solutions of sodium sulfide and sodium sulfite werepurged with high purity N2 gas for 30 min prior to illumination.All the studies were conducted with distilled and de-ionizedwater.

2.1. Materials

Cadmium acetate (Cd(CH3COO)2), zinc acetate (Zn(CH3COO)2),sodium sulfide (Na2S.9H2O), sodium sulfite (Na2SO3), titanium di-oxide (Degussa P25), thiourea, sodium hydroxide and hydrochloricacid from MERCK were procured. Other chemicals were used ofanalytical reagent grade.

2.2. Development of BTR (bench-scale tubular photocatalytic reactor)

The schematic diagram of the experimental setup for bench scaletubular photocatalytic reactor operated in batch recycle mode isshown in Fig. 1. The reactor shape was chosen to optimize the expo-sure of the catalyst to the light. By building the reactor in a cylindricalshape and surrounding it with the lamps, most of the light energycould be used. The plexiglass made tubular reactor was developedwith a capacity of 5 L (diameter: 10 cm and height: 63.7 cm). Thereactor consists of two inlets, i.e., one for purging of N2 gas andanother for solution feeding and two outlets, i.e., for collection of H2gas and solution. In thebatch recyclemodeofoperation, anadditionalport was used for recirculating the Na2S/Na2SO3 solution. The irra-diation was carried out by using three sun lamps (Philips) of power100 W each (wavelength ranges from 400 to 700 nm and light in-tensity ranges from510 to 646W/m2) andoneUV lampof power 8W(365 nm). The emission intensity of UV lamp was 183 mW/m2.

2.3. Synthesis of CdSeZnS/TiO2 coreeshell NPs

The CdSeZnS/TiO2 coreeshell NPs was synthesized by twosteps. It included hydrothermal synthesis of TiO2 NTs [39] and co-precipitation of CdSeZnS nanoparticles on TiO2 NTs. In a 2 Lbeaker, each 500 mL of 0.1 M cadmium acetate and zinc acetatewere stirred. 7.6 g thiourea, 3.5 g of TiO2 NTs and 0.2 M Na2S wereadded to it. The crystallized products were separated by ultracen-trifugation, washed thoroughly with de-ionized water, isopropanoland dried at 110 �C in an oven. The weight ratio of the synthesizedcatalyst was 1:1:1 (CdS:ZnS:TiO2 NTs). Typical particle sizes andband gap energy of the synthesized coreeshell NPs as estimatedfrom XRD studies and diffuse reflectance UVevisible absorptionspectroscopy, respectively were about 3.2 nm and 2.88 eV.

2.4. Conducting feasibility studies for the optimization of operatingvariables

The nanophotocatalysts were suspended in the solution con-taining sodium sulfide and sodium sulfite (Na2S and Na2SO3). Thesuspensions were deairated with N2 gas for 30 min to prevent up-take of photogenerated electrons by dissolved oxygen. The air spaceabove the solution in the reactor was flushedwith N2 for 1 h in eachexperiment. The flow rate of the N2 gas was controlled by the ro-tameters. The part of the liquidwas recycled by the peristaltic pump.The temperature of the photoreactor (25 �C) was maintained byusing exhaust fans. The effluent gas was collected in the collectiontank by downward displacement of water. The gas samples wereanalyzed for hydrogen with a Gas chromatograph (Chromatographand Instruments Company) having a PorapakQ column and thermalconductivity detector. N2 was used as a carrier gas. The influence ofoperating variables viz., concentration of sulfide ion, concentrationof sulfite ion, pH, catalyst concentration, lamp power, volume ofwastewater and recycle flow rate were studied. At optimized con-ditions, solar studies were conducted for the generation of H2. Inoutdoor solar studies, temperature of the photoreactor was main-tained byexternalwater circulation. The percentage conversion (XA)was calculated by using the following equation (1) [40],

XA ¼ Moles of H2 generatedMoles of Na2S fed

� 100 (1)

The apparent energy conversion efficiency of the system (ɳc) canbe calculated by following equation (2) [12],

hc ¼GH2

RH2

WS A� 100 (2)

Fig. 1. Schematic diagram of the tubular photocatalytic reactor operated under batch recycle mode.

P. Ruban, K. Sellappa / Energy 73 (2014) 926e932928

where, GH2is the Gibbs free energy of formed H2, 237.20 kJ/mol in

normal condition; RH2is the generation rate of H2 (mol/s); WS is

the intensity of incident radiation (W/m2), A radiation area (m2).

3. Results and discussion

Bench-scale feasibility studies were carried out in order to studythe influence of operating variables viz., Direct photolysis, darkadsorption, concentration of sulfide ion, concentration of sulfiteion, pH, catalyst concentration, lamp power, volume of wastewaterand recycle flow rates for the photocatalytic generation of H2. Theworking volume of bench-scale reactor was 1 L.

During direct photolysis 36 mL of H2 was generated. From theobservations, it was evident that there was no more H2 generationafter 1 h irradiation time. Na2S absorption started at 290 nm. Thisreaction appeared to direct water reduction by excited sulfide ions(S2�) [35]. The effect of dark adsorption (without light) on H2generation was studied. From the study, it was observed that thereis no H2 generation during the reaction time. Similar results wereobserved by Ueno et al. (1985) [32] and Silva et al. (2008) [41].

Fig. 2. Effect of concentration of sulfide ion.

3.1. Effect of concentration of sulfide ion

The effect of concentration of sulfide ion on the generation of H2was studied by varying the concentration of sulfide ion in the rangeof 0e0.125 M. Concentration of sulfite ion, pH, catalyst concentra-tion, lamp power and recycle flow rates were kept constant as0.175 M, 11.3, 0.5 g/L, 3 sun lamps and 1 UV lamp and 24 L/hrespectively. The influence of the concentration of sulfide ion on thephotocatalytic generation of H2 is shown in Fig. 2. Without sulfideion, 66 mL of H2 was generated. The maximum conversion of 19.8%(250 mL of H2 in 90 min) was obtained at 0.05 M.

Similar rate of H2 generation was observed by Zheng et al.(2009) [42]. The direct irradiation of alkaline sulfide in the presenceof (CdSeZnS)/TiO2 coreeshell NPs, generated H2. The generation ofH2 was decreased with the increase in irradiation time and ulti-mately reached zero after 90 min. It was due to the formation ofyellow polysulfides which absorbed part of the visible-wavelengthphotons by which it could contribute to CdS photocatalysis[24e25]. The formations of polysulfide caused catalyst deactivationas well as consumption of the donor. The poisoning of active

catalyst surfaces was due to the adsorption of sulfur containingcompounds.

It is important, from application points of view, to study thedependence of the photocatalytic H2 generation on the concen-tration of sulfide ion. With an increase in the concentration ofsulfide ion, the photocatalytic H2 generation was increased andreached a maximum at the concentration of sulfide ion 0.05 M.Further increase in concentration of sulfide ion led to a decrease inthe photocatalytic generation of H2. Under the studied conditions,the optimum concentration of sulfide ionwas about 0.05M. Beyondthe optimum point, the number of adsorbed sulfide ions wasincreased. But the active sites available are constant for a fixedcatalyst concentration. It was due to the blockage of the adsorptionof hydronium cations at the surface active sites. Sulfide and sulfiteions behaved as a quenching agent of ions and radicals [29] and H2generation became thermodynamically difficult in a strong alkalinemedium. In the present study, the reaction solution was basicbecause of the hydrolysis of SO3

2� and S2� ions [26]. Therefore, thehighest H2 evolution rate was not attained in the presence of eitherdiluted or concentrated sacrificial reagents. Higher solution con-centration was decreased the generation of H2, due to thedecreased in contact between the solution and active sites on thephotocatalyst surface [37,43].

The mechanism of the process was explained by Sabate et al.(1989) [25] and it is explained in equations (3)e(6). The reaction on

Fig. 4. Effect of pH.

P. Ruban, K. Sellappa / Energy 73 (2014) 926e932 929

a molecule comprised the following events viz., absorption of aphoton (3), reactions on the catalyst surface equations (4) and (5),and desorption of intermediates and/or products equation (6). Theoverall reaction is presented in equation (7).

Absorption of photon

Catalyst þ hn/Catalyst þ e� þ hþ (3)

Reactions on the catalyst surface.

Cathodic: 2H2O þ 2e�/H2 þ 2OH� (4)

Anodic: 2HS� þ OH� þ 2hþ/HS2� þ H2O (5)

Reactions in the liquid phase

HS2� þ SO32�/S2O3

2� þ HS� (6)

Overall reaction

HS� þ SO32� þ H2O/H2 þ S2O3

2� þ OH� (7)

3.2. Effect of concentration of sulfite ion

The effect of concentration of sulfite ion on the generation of H2was studied by varying concentration of sulfite ion in the range of0e0.25 M. Concentration of sulfide ion, pH, catalyst concentration,lamp power and recycle flow rates were kept constant as 0.05 M,11.3, 0.5 g/L, 3 sun lamps and 1 UV lamp and 24 L/h respectively. Theresults are illustrated in Fig. 3. Without sulfite ion, 176 mL of H2 wasgenerated. The generation of H2 was increased with the increase inconcentration of sulfite ion from 0.05 to 0.2 M. With an increase ofconcentration of sulfite ion from 0.2 M, the generation of H2 wasdecreased. It was due to the competitive adsorption of SO3

2� andS2O3

2� on the active sites of the catalyst [24]. The competitiveadsorption processes thus led to inhibition of the rate by thesespecies. It was evident that the H2 generation stopped at 90 min.

The high concentration of SO32� ions prevented the poisoning of

catalyst by elemental sulfur. The rate of H2 generation wasdecreased over time with a lower sulfite ion concentration [22]. Inthe absence of sulfite ion, HS� was oxidized to disulfide ions(equation (8)).

2HS� þ 2hþ/S22� þ 2Hþ (8)

Disulfide ions acted as an optical filter and it reduced the gen-eration of H2 with the increase in disulfide ion concentration. In the

Fig. 3. Effect of concentration of sulfite ion.

presence of sulfite ions, disulfide anions combined with sulfite ionsto form thiosulphate (equation (9)).

S22� þ SO32�/S2O3

2� þ S2� (9)

The formed thiosulphate ions in solution can regenerate thesulfide ions which can establish more H2 generation rate over time[28,30].

3.3. Effect of pH

In order to study the effect of pH on H2 generation, experimentswere conducted by varying the pH in the range of 9.5e12.5. Con-centration of sulfide ion, concentration of sulfite ion, catalyst con-centration, lamp power and recycle flow rates were kept constantas 0.05 M, 0.2 M, 0.5 g/L, 3 sun lamps and 1 UV lamp and 24 L/hrespectively. The results were illustrated in Fig. 4. At each pH, H2generation was increased with an increase in irradiation time. TheH2 generation was fast in the initial stages of the reaction. A non-linear variation in the H2 generation was clearly evident.

FrompH9.5 to 11.3, an increase inH2 generationwas observed forevery interval of time. It is due to the increase in the dissociation ofHS� to S2�, the pKa2 ofH2S is nearer to 12.AtpH11.3, theremight be asignificant amount of S2�. Hence higher generation of H2 wasobserved at 11.3, since the concentration of S2� at pH12.5was higherthan all other pHs. As the hole neutralization is largely enhanced bythe negative charge of the anion, S2� could bemore active than HS�.Hence the mediumwith S2� can give higher H2 generation than theHS�. At pH12.5, the generation ofH2was decreased. Itwas due to theincreased OH� ion concentration which apparently acted as an in-termediate of the photocatalytic reduction process. When the hy-droxide ion concentration became too high, apparently, manyphotogenerated hydrogen ions interacted with hydroxide ions pro-ducingwater. Thus thehydrogen generation rate declined at toohighNaOHconcentration [27]. Baoet al. (2008) [26] reportedadecrease inH2 generation in higher pHwas due to hydrolysis of SO3

2� and S2�. Inorder to study the influence of OH� on the absorption spectrum ofcoreeshell nanoparticles, DR UVevisible spectra was taken and it isillustrated in Fig. 5. The spectrum appeared similar to that of parentcatalyst illustrating options of catalyst degradation. Hence, thedecrease in H2 generationwas not due to catalyst but to the reagentspresent in themedium.Hence in the present study, the same processmight be caused for the reduced level of H2 generation.

3.4. Effect of catalyst concentration

The effect of the concentration of the synthesized CdSeZnS/TiO2coreeshell NPs on the photocatalytic H2 generation was

Fig. 5. DR UVevisible spectra of CdSeZnS/TiO2 coreeshell NPs (pH: 11.3 and 12.5). Fig. 7. Effect of lamp power.

P. Ruban, K. Sellappa / Energy 73 (2014) 926e932930

investigated by varying the catalyst concentration in the range of0e1 g/L. Studies were conducted with an aqueous 0.05 M Na2S and0.2 M Na2SO3 solutions (pH 11.3) without pH adjustment. Fig. 6illustrated the photocatalytic H2 generation as a function of pho-tocatalyst concentration. Without catalyst, 58 mL of H2 wasgenerated. The generation of H2 increased as catalyst concentrationwas increased from 0.1 to 0.5 g/L. It was due to the increase in thenumber of active sites in the added photocatalyst.

As the catalyst concentration was raised from 0.6 g/L, the gen-eration of H2 reduced. The availability of active sites increased withthe increase in catalyst concentration in the suspension, but thepenetration of light decreased in the reactor by the screening andshadowing effect of the photocatalyst in the aqueous solution.Consequently, it reduced the light absorption of inner particles ofthe solution and it reduced the generation of H2 [32e35]. An op-timum photocatalyst concentration also greatly depends on thephotoreactor geometry, the working conditions of the photo-reactor, degree of mixing, lamp power, and lamp geometry. Theoptimum amount of the photocatalyst was added to the reactor toimprove total absorption of light photons for the efficient photo-catalytic H2 generation reaction.

3.5. Effect of lamp power

The light intensity has significant influence on the photo-catalytic generation of H2, which controls the generation of e�/hþ

pairs. The results of the effect of the lamp power on the generationof H2 are illustrated in Fig. 7. An increase in lamp power by using 1sun lamp to 3 sun lamps and 1 UV lamp, the generation of H2 wasincreased from 78 to 315 mL. It was due to the fact that with anincrease in power of lamp, large number of e�/hþ pairs might begenerated. Thus, more e� might be available for reducing the Hþ

Fig. 6. Effect of catalyst concentration.

ions and hence, the higher in the rate of photocatalytic generationof H2. The reaction rate constant usually follows a power-lawdependence on light intensities. It has been found for manymodel compounds that the reaction rate constant is proportional tothe square root of the light intensity at high intensity, while at asufficiently low level of illumination, the reaction rate constantfollows first-order dependence [32,34e35].

3.6. Effect of volume of wastewater

In order to study the effect of volume of wastewater on H2generation, experiments were carried out by varying the volume ofwastewater as 2 L, 1 L, 750 mL, 500 mL and 250 mL. Concentrationof sulfide ion, concentration of sulfite ion, pH, catalyst concentra-tion, lamp power and recycle flow rates were kept constant as0.05 M, 0.2 M, 11.3, 0.5 g/L, 3 sun lamp and 1 UV lamp and 24 L/hrespectively. The results are illustrated in Fig. 8. The generation ofH2 was increased with decrease in volume of wastewater. It wasdue to decrease of light penetration into the solution.

The volume of the solution in photoreactor also had a significanteffect on removal efficiency. As increase in volume of wastewater,increases the number of available substrate molecules under thesame conditions. Moreover, mass diffusion as the controlling stepdepends on the depth of the liquid volume. Increase in volume ofsulfide/sulfite solution decreased the mass diffusion and conse-quently it reduced the process efficiency [36,37].

3.7. Effect of recycle flow rates

The liquid flow rate has significant influence on the residencetime of liquid and the mass transfer in the reactor, which in turnaffects the photocatalytic generation of H2. When the flow rate is

Fig. 8. Effect of volume of wastewater.

P. Ruban, K. Sellappa / Energy 73 (2014) 926e932 931

increased, two antagonistic effects are brought into play. It de-creases the residence time in the photocatalytic reactor and in-creases the mass transfer rate. An augmentation of the flow rateresulted in a higher mass transfer and a smaller concentrationgradient between the bulk and the catalyst surface. The results ofthe effect of the recycle flow rates on the generation of H2 areillustrated in Fig. 9.Without recycle,158mL of H2 was generated. Asthe increase in the recycle flow rate from 5 to 18 L h�1, the H2generation was increased from 238 to 408 mL. As the increase inflow rate from 18 to 33 L h�1, the H2 generationwas decreased from408 to 158 mL. From this observation, it was evident that therecycling of solution is necessary during H2 generation. At lowerflow rates (less than 18 L h�1), settling of catalyst particles on thebottom of the reactor surface occurred, thus reducing the genera-tion of H2. Above the recycle flow rate of 18 L h�1, decrease in H2generationwas ascribed to the less residence time of the catalyst inthe reactor. Similar workswere reported by Ray,1998 [38] and Priyaand Kanmani [17]. The optimal liquid flow rate in this system was18 L h�1.

At optimized conditions, in the normal sunny day (at 543W/m2)384 mL of H2 was generated. It corresponds to 2.19% of energeticefficiency. Similar studies were reported by Borgarello et al., 1986[44], Grzyll et al. (1989) [24] and Huang et al., 2011 [37].

The generation of H2 stopped at 90 min, due to the catalystinactivity. This might be due to the fact that the formation of sulfurand its conversion to polysulfide was not rapidly retarded by theadded sulfite. Therefore, in order to ensure continuous generationof H2, it was necessary to renew the catalyst system.

The nanophotocatalysts were continuously reused for 8 runs.For each consequent trial, the efficiency of the catalyst wasdecreased. The nanophotocatalyst was reactivated by washingwith distilled water. The stability of the catalysts is an importantfactor for industrial implications. The photocatalytic experimentswere repeated eight times with the same catalyst. After eachexperiment, the catalyst was filtered, washed, and recycled. Foreach consequent trial, the generation of H2 was decreased grad-ually. In the last run, 43% decrease in H2 generation was observed.The decrease of photocatalyst activity might be due to the loss ofthe catalyst during washing and filtration process combined with asmall quantity of photocorrosion in the catalysts [45]. This sug-gests that a fresh catalyst loading might be required to compen-sate the loss after a few runs. This result indicated that theCdSeZnS/TiO2 coreeshell NPs catalysts were fairly photostableand suitable for practical application. Cost of the photocatalyst was89.2 US$/kg [46,47]. Compare to the cost of treatment with com-mercial value of H2 and thiosulphate, cost input was practicallynegligible. Similar observations were reported by Bhattacharyaet al. (1996) [48].

Fig. 9. Effect of recycle flow rates.

4. Conclusions

The use of CdSeZnS/TiO2 coreeshell NPs for photocatalytic H2generation from sodium sulfide solutionwas studied under variousreaction conditions. The experimental results showed that sodiumsulfite was found to be the most efficient sacrificial reagent foreffective suppression of photocorrosion. Strong basic pH value 11.3was favorable for the reaction. The experimental results showedthat the concentration of sulfide ion, concentration of sulfite ion,pH, catalyst concentration, lamp power, volume of sulfide/sulfiteand recycle flow rates have significant influence on H2 generation.

At optimized conditions, energetic efficiency of the system is2.19%. The estimated cost of the photocatalyst was 89.2 US$/kg.Further reactor studies are recommended, to treat real wastewaterand scale-up studies are conducted to improve the economicfeasibility of the system, to reduce the operational costs. Furtherdesign and lab studies are also recommended to focus on increasingthe efficiency of the reactor, to make the photocatalytic systems abetter alternative way of hydrogen production.

Acknowledgment

The authors gratefully acknowledge the financial supports fromthe Council of Scientific and Industrial Research, New Delhi, India.(Award No.: 9/468 (0389)/2008-EMR-I)

References

[1] Jang JS, Hwang DW, Lee JS. CdS-AgGaS2 photocatalytic diodes for hydrogenproduction from aqueous Na2S/Na2SO3 electrolyte solution under visible light(l�420nm). Catal Today 2007;120:174e81.

[2] Kvesitadze G, Sadunishvili T, Dudauri T, Zakariashvili N, Partskhaladze G,Ugrekhelidze V, et al. Two-stage anaerobic process for bio-hydrogen and bio-methane combined production from biodegradable solid wastes. Energy2012;7(1):94e102.

[3] Supriya VT, Subrahmanyam M. Photocatalytic generation of hydrogen fromhydrogen sulfide: an energy bargain. Int J Hydrogen Energy 1997;22:959e65.

[4] Penner SS. Steps towards hydrogen economy. Energy 2006;31:33e43.[5] Guo LJ, Zhao L, Jing DW, Lu YJ, Yang HH, Bai EF, et al. Solar hydrogen pro-

duction and its development in China. Energy 2009;34(9):1073e90.[6] Budzianowski WM. Negative net CO2 emissions from oxy-decarbonization of

biogas to H2. Int J Chem React Eng 2010;8:A156.[7] Tseng P, Lee J, Eriley P. A hydrogen economy: opportunities and challenges.

Energy 2005;30:2703e20.[8] Kato H, Asakura K, Kudo A. Highly efficient water splitting into H2 and O2 over

lanthanum-doped NaTaO3 photocatalysts with high crystallinity and surfacenanostructure. J Am Chem Soc 2003;125:3082e9.

[9] Kudo A. Photocatalyst materials for water splitting. Catal Surv Asia 2003;7:31e8.

[10] Kudo A. Recent progress in the development of visible light driven powderedphotocatalysts for water splitting. Int J Hydrogen Energy 2007;32:2673e8.

[11] Ikeda M, Kusumoto Y, Somekawa S, Ngweniform P, Ahmmad B. Effect ofgraphite silica on TiO2 photocatalysis in hydrogen production from water-methanol solution. J Photochem Photobio A: Chem 2006;184:306e12.

[12] Hsu S-H, Hung S-F, Chien S-H. CdS sensitized vertically aligned single crystalTiO2 nanorods on transparent conducting glass with improved solar cell ef-ficiency and stability using ZnS passivation layer. J Power Sources 2013;23:236e43.

[13] Yoong LS, Chong FK, Dutta BK.Development of copper-doped TiO2 photocatalystfor hydrogen production under visible light. Energy 2009;34(10):1652e61.

[14] Yu J, Zhang J, Jaroniec M. Preparation and enhanced visible-light photo-catalytic H2-production activity of CdS quantum dots-sensitized Zn1-xCdxSsolid solution. Green Chem 2010;12:1611e4.

[15] Linkous CA, Muradov NZ, Ramser N. Consideration of reactor design for solarhydrogen production from hydrogen sulfide using semiconductor particu-lates. Int J Hydrogen Energy 1995;20:701e9.

[16] Linkous CV, Mingo TM, Muradov NZ. Aspects of solar hydrogen productionfrom hydrogen sulfide using semiconductor particulates. Int J Hydrogen En-ergy 1994;19:203e8.

[17] Priya R, Kanmani S. Batch slurry photocatalytic reactors for the generation ofhydrogen from sulfide and sulfite streams under solar irradiation. Sol Energy2009;83:1802e5.

[18] Jing D, Guo L, Zhao L, Zhang X, Liu H, Li M, et al. Efficient solar hydrogenproduction by photocatalytic water splitting: from fundamental study to pilotdemonstration. Int J Hydrogen Energy 2010;35:7087e97.

P. Ruban, K. Sellappa / Energy 73 (2014) 926e932932

[19] Villa Katherine, Dom�enech Xavier, Malato Sixto, Maldonado Manuel I,Peral Jos�e. Heterogeneous photocatalytic hydrogen generation in a solar pilotplant. Int J Hydrogen Energy 2013;38(29):12718e24.

[20] Jing D, Li Jing, Liu H, Yao S, Guo L. Photocatalytic hydrogen production fromrefinery gas over a fluidized-bed reactor II: parametric study. Ind Eng ChemRes 2013;52:1992e9.

[21] Bulher N, Meier K, Reber JF. Photochemical hydrogen production with cad-mium sulfide suspensions. J Phys Chem 1984;88:3261e8.

[22] Reber JF, Rusek M. Photochemical production of hydrogen with zinc sulfidesuspensions. J Phys Chem 1986;90:824e34.

[23] Neil Furlong D, Franz G, David H, Wolfgang S, Darrell W. Kinetics of hydrogenproduction from illuminated CdS/Pt/Na2S dispersions. J Phys Chem 1986;90:2388e96.

[24] Grzyll LR, Thomas JJ, Barile RG. Photo electrochemical conversion of hydrogensulfide to hydrogen using artificial light and solar radiation. Int J HydrogenEnergy 1989;14:647e51.

[25] Sabate J, Cervera e March S, Simarro R, Gimenez J. Photocatalytic productionof hydrogen from sulfide and sulfite streams: a kinetic model for reactionsoccurring in illuminated suspensions of CdS. Chem Eng Sci 1990;45:3089e96.

[26] Bao N, Shen L, Takata T, Domen K. Self-Templated synthesis of nanoporousCdS nanostructures for highly efficient photocatalytic hydrogen productionunder visible light. Chem Mater 2008;20:110e7.

[27] Strataki N, Antoniadou M, Dracopoulos V, Lianos P. Visible-light photocatalytichydrogen production from ethanol-water mixtures using a Pt-CdS-TiO2photocatalyst. Catal Today 2010;151:53e7.

[28] Muradov NZ, Rustamov MI, Guseinova AD. Bazhutin YuV. Photocatalyticproduction of hydrogen from H2S solutions over CdS/Pt colloids. React KinetCatal Lett 1987;33:279e83.

[29] Cui W, Feng L, Xu C, Lu S, Qiu F. Hydrogen production by photocatalytic decom-position of methanol gas on Pt/TiO2 nano-film. Catal Commun 2004;5:533e6.

[30] Li Y, Xie Y, Peng S, Lu G, Li S. A composite visible light photocatalyst forhydrogen production. Chemosphere 2006;63:1312e8.

[31] Reber JF, Meier K. Photochemical hydrogen production with platinized sus-pensions of cadmium sulfide and cadmium zinc sulfide modified by silversulfide. J Phy Chem 1984;88:5903e13.

[32] Ueno A, Kakuta N, Park KH, Finlayson MF, Bard AJ, Campion A, et al. Silicasupported ZnS.CdS mixed semiconductor catalysts for photogeneration ofhydrogen. J Phys Chem 1985;89:3828.

[33] Bangun J, Adesina AA. The photodegradation kinetics of aqueous sodiumoxalate solution using TiO2 catalyst. Applied Catalysis A: General 1998;175:221e35.

[34] Parrek VK, Brungs MP, Adesina AA. Continuous process for photodegradationof industrial bayer liquor. Ind Eng Chem Res 2001;40:5122e5.

[35] Linkous CA, Huang C, Fowler Randy. UV photochemical oxidation of aqueoussodium sulfide to produce hydrogen and sulfur. J Photochem Photobiol A:Chem 2004;168:153e60.

[36] Faramarzpour M, Vossoughi M, Borghei M. Photocatalytic degradation offurfural by titania nanoparticles in a floating-bed photoreactor. Chem Eng J2009;146:79e85.

[37] Huang C, Yao W, T-Raissi A, Muradov N. Development of efficient photo-reactors for solar hydrogen production. Sol Energy 2011;85:19e27.

[38] Ray AK. A new photocatalytic reactor for destruction of toxic water pollutantsby advanced oxidation process. Catal Today 1998;44:357e68.

[39] Zhang YJ, Yan W, Wu YP, Wang ZH. Synthesis of TiO2 nanotubes coupled withCdS nanoparticles and production of hydrogen by photocatalytic waterdecomposition. Mater Lett 2008;62:3846e8.

[40] Kalyasundaram K, Borgarello E, Gratzel M. Visible induced water cleavage inCdS dispersions loaded with Pt and RuO2, hole scavenging by RuO2. HelveticaChim Acta 1981;64:362e6.

[41] Silva LA, Ryu SY, Choi J, Choi W, Hoffmann MR. Photocatalytic hydrogenproduction with visible light over Pt-Interlinked hybrid composites of cubic-phase and hexagonal-phase CdS. J Phy Chem Lett 2008;112(9):12069e73.

[42] Zheng XJ, Wei LF, Zhang ZH, Jiang QJ, Wei YJ, Xie B, et al. Research on pho-tocatalytic H2 production from acetic acid solution by Pt/TiO2 nanoparticlesunder UV irradiation. Int J Hydrogen Energy 2009;34:9033e41.

[43] Matsumura M, Furukawa S, Saho Y, Tsubomura H. Cadmium sulfide photo-catalysed hydrogen production from aqueous solutions of sulfite. Effect ofCrystal structure and preparation method of the catalyst. J Phy Chem 1985;89.1327e1329.

[44] Borgarello E, Serpone N. Photodecomposition of H2S in aqueous alkalinemedia catalysed by RuO2-loaded alumina in the presence of cadmium sulfide.Application of the inter-particle electron transfer mechanism. InorganicaChimica Acta 1986;112:197e201.

[45] Bai S, Li H, Guan Y, Jiang YS. Photocatalytic decomposition of H2S to produceH2 over CdS nanoparticles formed in HY-zeolite pore. Appl Surf Sci 2011;257:6406e9.

[46] Priya R, Kanmani S. Design of pilot-scale solar photocatalytic reactor for thegeneration of hydrogen from alkaline sulfide wastewater of sewage (17e20):2817e2823.

[47] Priya R. Solar photocatalytic generation of hydrogen from hydrogen sulfidein an alkaline solution using CdS-ZnS/TiO2 nanocomposites. Ph.D. thesis;2011.

[48] Bhattarcharya RG, Mandal DP, Bera SC, Mukherjee Rohatgi. Photoelectrosyn-thesis of dihydrogen via water-splitting using Sx2� (X¼1,2,3 ….) as an analyte:a first step for a viable solar rechargeable battery. Int J Hydrogen Energy1996;21:343e7.