Role of palladium crystallite size on CO oxidation over ...

Molecular Catalysis 455 (2018) 1–5

Contents lists available at ScienceDirect

Molecular Catalysis journal homepage: www.elsevier.com/locate/mcat

Role of palladium crystallite size on CO oxidation over CeZrO4-δ supported Pd catalysts ⁎

T

⁎

Anand S. Burangea,b, , Kasala Prabhakar Reddyb, Chinnakonda S. Gopinathb, , Rakesh Shuklac, ⁎ Avesh K. Tyagic, a

Department of Chemistry, Wilson College, Chowpatty, Mumbai, 400 007, India Catalysis Division, National Chemical Laboratory, Dr. Homi Bhabha Road, Pune, 411 008, India c Chemistry Division, Bhabha Atomic Research Centre, Mumbai, 400 085, India b

A R T I C LE I N FO

A B S T R A C T

Keywords: Role of defects Heterogeneous catalysis Photodeposition Ceria-zirconia Anionic vacancies

Photodeposited palladium on nanocrystalline CeZrO4-δ support was evaluated for CO oxidation, as a function of particle size and pre-treatment. All the catalysts were well characterized by XRD, SEM, EDX, TEM, CO chemisorption and Raman spectroscopy. It was observed that change in photodeposition time alters the average crystallite size of Pd, and the same was confirmed by CO chemisorption. 7 nm average crystallite size of Pd exhibited excellent catalytic activity for CO oxidation. In order to understand the effect of support and metalsupport interaction, 1 wt% Pd was deposited deliberately onto reduced and oxidized CeZrO4-δ support. The reduced support with anionic vacancies and more Ce3+ species exhibited an improvement in CO oxidation at onset temperature. Detailed analysis shows that Pd-deposition occurs preferentially at oxygen-vacancy sites and subsequent metal-support interaction influences the catalysis.

1. Introduction Carbon monoxide is a poisonous gas and has toxic effects on human health at concentration above 35 ppm [1]. CO oxidation has received much attention in the last few decades because of its importance in automotive exhaust treatment, fuel cells, air purification as well as in gas sensors [2–4]. Variety of materials like metal oxides [5–7], mixed metal oxides [8,9] and metal supported materials [10], etc. have been reported as catalysts for the CO oxidation in the literature. Among various oxides, ceria (CeO2) was one of the major components of automotive catalysts since its crystal structure does not change during storage and release process of oxygen i.e. oxygen storage capacity (OSC) and thus added to three way catalysts (TWC) as an OSC material [11]. OSC is one of the key requirements for TWC to remove NOx, CO and HCs. Ceria exists as redox material Ce3+/Ce4+ and oxygen (anionic) vacancies, where cerium bears +4 and +3 oxidation states under oxygen-rich and oxygen-lean conditions, respectively. To improve OSC of pure ceria, effect of addition of other oxides to it, was well studied. It was observed that addition of zirconia (forming solid solution) to ceria not only improves OSC but also its thermal stability [12]. To develop automotive catalysts, ceria zirconia (CZ) solid solutions were further developed to second and third generation CZ materials [12]. There are

⁎

various ceria-zirconia based materials including solid solutions, mixed oxides, physical mixtures, exists; among them, mixed metal oxides are the most popular class of materials. It was observed that when ZrO2 is incorporated in the lattice of the ceria, it enhances OSC and thermal stability [12]. Ceria based noble metal catalysts for CO oxidation reported in the literature mainly includes Pt/CeO2 [13,14], Au-CeO2 composite [15,16] while other than ceria comprise Pt/SnO2, Pd/SnO2, Au supported on γ-Fe2O3, MnOx and Co3O4 [17–20]. Materials for TWC, particularly for the total oxidation of volatile organic compounds (VOCs) and CO, supported palladium (Pd) based catalyst is considered as an active ingredient. Though there are noble metals other than palladium, like Pt, Rh, they have a serious disadvantage that they become passive (specifically Rh and Pt) in oxygen rich atmosphere due to formation of oxides when subjected to elevated temperatures [21–24]. Pd/CeO2 catalysts, with 0.25 −2.0% Pd loading exhibited the significant low temperature activity for CO oxidation [25] while with 1 wt % Pd loading catalysts like Pd/CeO2-TiO2 [26] and Pd/Al2O3, Pd/CeO2Al2O3 [27], etc. are also reported in the literature. These ceria based catalysts showed significant CO conversion, however, the role of Pd crystallite size and support on CO oxidation is not thoroughly evaluated. Recently, CeZrO4-δ, a new material was explored for oxidative dehydrogenation of ethylbenzene to styrene [28] and Pd/CeZrO4-δ for

Corresponding authors. E-mail addresses: [email protected] (A.S. Burange), [email protected] (C.S. Gopinath), [email protected] (A.K. Tyagi).

https://doi.org/10.1016/j.mcat.2018.05.021 Received 16 April 2018; Received in revised form 19 May 2018; Accepted 21 May 2018 2468-8231/ © 2018 Elsevier B.V. All rights reserved.


A.S. Burange et al.

the CO oxidation reaction.

Suzuki cross coupling reaction in water [29]. CeZrO4-δ has robust fluorite structure and exhibits high redox properties [28]. Exploitation of new materials in catalysis is always an arena of great interest for researchers worldwide. In continuation of our efforts in exploring new materials in catalysis [29–31], herein an attempt was made to explore Pd/CeZrO4-δ for carbon monoxide oxidation reaction. Present manuscript, mainly emphasizes on effect of crystallite size of metal (Pd) dispersion and role of defects present on the support on catalytic activity. In the present work, Pd catalyst was synthesized wherein; Pd was deposited onto CeZrO4-δ support by photo-deposition technique with varying irradiation time. Change in irradiation time affects the crystallite size and thereafter the catalytic activity for the said reaction. In the present work, we tried to find out the optimum Pd crystallite size for CO oxidation reaction with 1% Pd loading over CeZrO4-δ support. The effect of oxidation state of cerium from the support on the catalytic activity was also investigated.

2.5. General procedure for CO oxidation The evaluation of catalytic activity of various prepared catalyst samples was performed using fixed bed reactor with 14 mm diameter at atmospheric pressure. The activity was evaluated for CO oxidation at various temperatures from room temperature to 300° C. CO, O2 and N2 gases were used and their flow rate was controlled by using mass flow controller (MFC). In this set up, vertical reactor was kept in tubular furnace having uniform temperature zone of 4 cm equipped with radix 4000 temperature controller. The catalyst bed temperature was measured by using K-type thermocouple. The reaction was carried out with 50 mL/min flow rate (CO: O2: N2 = 1:5:19). The conversion was measured by gas chromatography (GC) equipped with online gas sampling valve, molecular sieve column and thermal conductivity detector (TCD) in a steady state with a ramping rate of 2⁰/min, kept at different temperatures used for analysis for 10 min for the equilibration. CO oxidation to CO2 was calculated by formula COin − COout/COin × 100.

2. Experimental section 2.1. Materials and methods

3. Results and discussion All the chemicals were procured from Sigma Aldrich and used for the catalyst preparations without further purification. For XPS analysis, X-ray photoelectron spectrum (XPS) with custom built APPES unit equipped with R3000HP (VG Scienta) analyzer was used. Transmission Electron Microscopy (TEM) imaging was done using TECNAI 20 s-twin electron microscope. Raman spectra of the samples were carried out using Horiba JY LabRAM HR 800 Raman spectrometer (microscope in reflectance mode). CO chemisorption analysis; analysis of average crystallite size of palladium samples were carried out by using autosorb iQ instrument.

3.1. Characterization of catalysts The X-ray diffraction pattern was recorded for the as prepared, reduced as well as oxidized CeZrO4-δ support. All the peaks could be indexed for the fluorite structure and no impurity phases were observed [28,29]. After reduction (refer section 2.4) of the said support, the decrease in the line broadening was observed which attributed to the increase in the average crystallite size and release of micro-strain in the support to some extent. XRD patterns of PCZ-20, PCZ-40 and PCZ-60 catalysts also showed prominent peaks of above said fluorite support with undetectable Pd phases (Fig. 1b). The presence of Pd was confirmed by EDX and XPS analysis well discussed below. On reduction of CeZrO4-δ anionic vacancies/defects are generated due to partial reduction of Ce4+ to Ce3+. Reduced and unreduced CeZrO4-δ showed clearly distinct Raman spectra from each other (SI Fig. 6). In case of reduced CeZrO4-δ large broadening was observed along with shifts near ∼ 460 and 610 cm−1 due to generated anionic vacancies. It was observed that upon reduction, the colour of CeZrO4-δ became bluish (cyan) from yellow due to decrease in the band gap as well as partial filling of Ce 4f orbital [29]. PCZ-40 and PCZ-40 reduced catalyst were further analyzed by scanning electron microscopy (SEM) and transmission electron microscopy (TEM). SEM image of PCZ-40 showed layered morphology and can be clearly seen in SI Fig. 1(a). EDX (energy dispersive X-ray analysis) and TEM image is shown for reduced PCZ-40 material in SI Fig. 1c and d, respectively. From the TEM analysis, all the particles are of the order of 20 nm (SI Fig. 1b). TEM analysis (SI Fig. 4) shows the presence of micro and mesoporous morphology of the material which is an essential factor for diffusion and mass transfer of reactants as well as products. On cursory look at various TEM images (see SI), nano size particles with random shapes with spherical to ellipsoidal geometry were observed. Random polygonal morphology was observed in case of PCZ-40 reduced (SI Fig. 1b) whereas in case of PCZ-40, hexagonal morphology was also evidenced from TEM (SI Fig. 1d). Elemental composition obtained from EDX was in accordance with the expected nominal input concentration. BET surface area of this material was observed to be 48 m2/g (refer supplementary information). After the deposition of palladium onto the support by changing time of irradiation, average palladium crystallite size, % metal dispersion and active metal surface area were calculated by CO chemisorptions method. The data of CO chemisorptions is well tabulated in Table 1 below. It was observed that for all the catalysts the average crystallite size was in the range of 6.6–8.5 nm where 17% maximum metal dispersion was achieved in case of a catalyst PCZ-40 and PCZ-60 (Table 1,

2.2. Preparation of support CeZrO4-δ CeZrO4-δ was prepared by gel-combustion method using cerium nitrate and zirconyl nitrate as metal precursors and glycine as a fuel. In a typical catalyst preparation method, desired amount of both the precursors were added in distilled water and to this glycine (60% fueldeficient ratio) was added and further evaporated to dryness on hot plate to yield a transparent gel. The obtained transparent gel was heated further to give a fluffy powder which was further calcined at 600° C for 6 h (4⁰/min ramping rate) to give CeZrO4-δ [28]. 2.3. Preparation of support Pd/CeZrO4-δ Palladium (1 wt% loading) was deposited onto CeZrO4-δ support by photo-deposition technique [32]. In a typical catalyst preparation method, palladium nitrate was first dissolved in 120 mL of methanol to this desired amount of CeZrO4-δ and 30 mL water was added. The formed mixture was further purged with Argon with continuous stirring for 90 min for degassing. Finally the mixture was kept for photo-deposition by irradiating it with 400W UV light for different times. Catalysts prepared by 20, 40 and 60 min irradiation time are designated as PCZ-20, PCZ-40 and PCZ-60 respectively (where P stands for palladium and CZ for CeZrO4-δ). 2.4. Reduction and oxidation of the support (CeZrO4-δ) To investigate the effect of reduced and oxidized support on catalytic activity, the said support was reduced in a tubular furnace under H2 atmosphere at 550° C for 1 h with flow rate of 50 mL/min and ramping rate of 5°/min. For oxidation, material CeZrO4-δ was kept in tubular furnace using air flow at 500° C with 50 mL/min, 4°/min, flow rate and ramping rate, respectively for 1 h. On as prepared, reduced and oxidized support, 1 wt % Pd was photo-deposited and their activity was further evaluated for 2


A.S. Burange et al.

respectively. The enhancement in the percentage of Ce3+ is attributed to the formation of anion vacancies after reduction. The anion vacancies formation can also be seen by examining O 1 s XPS spectra. The O 1 s spectra of both (reduced and unreduced) samples show three peaks due to the different electronic environment around the oxygen. The binding energy (BE) at 529.5 eV is due to the 4-fold oxygen bonded to Ce4+, where as the feature appears at 531.3 eV is attributed to 3 fold oxygen bonded to Ce3+ and/or near to O-vacancy. High BE component at 532.8 eV is attributed to the hydroxyl groups present on the surface, which is to be expected due to interaction with atmospheric moisture. The relative increase in the intensity of peak at 531.3 compared with 529.5 eV in the reduced sample is due to formation of Ce3+ ions due to reduction. As expected, Zr 3p and Pd 3d core levels overlap and the resulting xps spectrum is shown in Fig. 2(II). To distinguish the peaks we deconvoluted into 6 peaks. Two peaks are due to spin orbit splitting of Zr 3p into Zr 3p3/2 and 3p1/2 appearing at binding energy 332.4 and 346.1 eV respectively. Remaining 4 peaks belong to the Pd 3d region showing doublet structure. Among the pairs, peaks at 335 and 340.7 eV represents metallic Pd 3d 5/2 and 3d 3/2 core levels respectively, and in good agreement with the literature; [34] whereas peaks at 336.7 and 342.4 eV are due to the presence of palladium in Pd2+. 3.2. Evaluation of catalytic activity for CO oxidation reaction The various prepared catalysts (CeZrO4-δ, PCZ-20, PCZ-40, PCZ-40 reduced, PCZ-40 oxidized and PCZ-60) for evaluation of their catalytic activity were first pelletized and then sewed to the micron size. Typical CO oxidation was carried out using fixed bed reactor. 100 mg of catalyst was used for the evaluation of catalytic activity. Pure CeZrO4-δ did not show any activity up to 240° C while the onset of oxidation at 250° C with 2% CO conversion was recorded. Maximum CO conversion achieved with Pure CeZrO4-δ at 300° C was 19.5% and it is in agreement with literature report [35]. In the next part of our study, catalysts prepared by photo-deposition techniques with different photodeposition times were screened for the said reaction. For the catalyst PCZ-20, PCZ-40 and PCZ-60, 10% CO oxidation (onset) activity was recorded at 140, 120 and 125° C respectively (Fig. 3). 100% CO oxidation for PCZ-20 catalyst was recorded at 160° C compared to 140° C for PCZ-40 and PCZ-60 (refer Fig. 3).

Fig. 1. XRD patterns of various CeZrO4-δ supports designated as Ce-Zr-O (as prepared, oxidized and reduced) and of various catalysts PCZ-20, PCZ-40 and PCZ-60. Table 1 CO Chemisorption Analysis.a Entry

1 2 3 4 a

Catalyst

PCZ-20 PCZ-40 PCZ-40 reduced PCZ-60

% Metal dispersion

Average Pd crystallite size (nm)

Active metal surface area per gram of sample (m2/g)

13 17 16

8.5 6.7 7.0

0.60 0.74 0.70

17

6.6

0.75

3.3. Role of palladium crystallite size and support Catalyst preparation found profound role on the catalytic activity which is further attributed to% dispersion of metal (Pd) and its average crystallite size. Catalysts PCZ-40 and PCZ-60 both showed nearly same catalytic activity where similar onset activity and 100% CO oxidation were recorded at 120 and 140° C, respectively. Comparable activity of PCZ-40 and PCZ-60 is due to similar Pd-dispersion on the support and average Pd crystallite size. Both of them have around 17% metal dispersion and nearly 7 nm average crystallite size (Refer Table 1). Catalyst PCZ-20 synthesized with 20 min of irradiation showed an average crystallite size around 8.5 nm with least metal dispersion (13%) compared to other catalysts. It was clearly seen that, as average crystallite size increases above 8 nm, the catalytic activity decreases. Among PCZ40 and PCZ-60, for PCZ-40 maximum conversion was recorded at onset temperature and hence used for the further studies (Table 1). To check the effect of role of support; we tried to find out role of strong metal support interaction (SMSI) on catalytic activity. For the same, first as prepared CeZrO4-δ was deliberately reduced or oxidized on which Pd was deposited with 40 min irradiation time designated as PCZ–40 reduced and PCZ-40 oxidized respectively. PCZ-40, PCZ-40 reduced and PCZ-40 oxidized all of them showed almost same catalytic activity with 100% conversion at 140° C (Fig. 4). But conversions recorded at 120° C for PCZ-40, PCZ-40 reduced and PCZ-40 oxidized were 11, 17 and 5.5% respectively.

Pd content for all catalysts: 1 wt%.

entries 2 and 4). X-ray photoelectron spectrum (XPS) was collected on an custom built APPES unit equipped with R3000HP (VG Scienta) analyzer at a pass energy of 100 eV using non monochromatized Al K alpha X-ray as the excitation source and C 1 s (285 eV) as the reference line [33]. Fig. 2(A) shows Ce 3d and O 1 s core level spectra for as-prepared and reduced support. Complex nature of Ce 3d features, splits the 3d peaks into 3d5/2 and 3d 3/2 designated as V and U respectively, in the figure. Ce 3d core level results obtained are in good correspondence with partially reduced ceria (CeO2-x). Ce3+/Ce4+ ratio in unreduced and reduced CeZrO4-δ samples was calculated, according to literature report [33]. The enhancement in the concentration of Ce3+ relative to the Ce+4 species was observed. Before and after reduction of CeZrO4-δ, concentration of Ce3+ was observed to be 14.1% and 21.2% 3


A.S. Burange et al.

Fig. 2. (I) The XPS core level spectra of Ce 3d (a) CeZrO4-δ (b) reduced CeZrO4-δ catalyst and O 1 s in (c) CeZrO4-δ (d) reduced CeZrO4-δ. The Ce 3d core level spectrum and fitting results, V denotes Ce 3d5/2, U denotes Ce 3d3/2, where the fitting peaks of V0,V′, U0 and U′ belong to Ce3+ ions and fitting peaks of V, V′′, V′′′, U, U′′ and U′′′ belong to Ce4+ ions; (II) The XPS core level spectra of reduced sample for Pd 3d5/2 and 3d3/2 and Zr 3p.

nanometer level can alter their activity due to change in reactivity of Ce3+/Ce4+ along with oxygen release/uptake characteristics [36]. Other than ceria support, recently synthesis of sintering resistant Pt NPs by the application of wide mouthed compartment have also been explored [37]. Incorporation of ZrO2 to ceria lattice is found to be a key factor for the stability of the material at elevated temperature. In this context various MOx-ceria based materials [38,39] and new materials [40,41] have been explored in the literature and therefore we focused our studies on novel CeZrO4-δ material. It was observed that, the photodeposition onto support preferentially takes place at anionic vacancy sites [42]. Therefore to check the effect of SMSI, we compared both PCZ-40 and PCZ-40 reduced samples for the said conversion. In present work, we observed decrease in % metal dispersion with increase in irradiation time during photodeposition. Where after 40 min of irradiation no further change in % dispersion or average crystallite site was observed. Since shape of cerium based support can alter catalytic activity, we observed ellipsoidal and hexagonal shaped particles by TEM which could also be another factor for the observed catalytic activity. Low temperature, CO oxidation is well reported on supported nanoparticles [43,44]. In case of 1% Pd/CeO2 for the CO oxidation, 100% conversion was achieved at 175° C. The average crystallite size of the Pd was 10 nm where it was observed that Pd ions were incorporated into ceria to form a solid solution [45]. In our case, with same metal loading we could achieve 100% conversion at 140° C (lesser than ref. no. 45) by reducing the average crystallite size from 10 nm to 7 nm. Thus, little change in the average crystallite size can also affect the catalytic activity where by enhancing the amount of Pd onto support; low temperature catalytic activity can be achieved. To the best of our knowledge, this is the first report where we clarified the role of the average crystallite size on carbon monoxide oxidation reaction.

Fig. 3. CO oxidation activity for PCZ-20, PCZ-40 and PCZ-60 catalysts.

4. Conclusions In continuation of our efforts in catalysis [46–48], herein we report CO oxidation over novel Pd/CeZrO4-δ catalyst. In present work, we restricted our studies to 1 wt%, to find out the role of support, metal dispersion, metal area and effect of pre-treatment. With increase in further Pd content, low temperature CO oxidation activity can be achieved. With only 1 wt% Pd loading, 100% conversion was achieved at 140° C with sustainability of the catalyst for 12 h. Nearly 7 nm average crystallite size of Pd (by 40 min of photo-deposition) found best for CO oxidation whereas further increase in crystallite size to 8.5 nm decreases the activity. In support pre-treatment studies, upon reduction of support, formation of anionic vacancies or more Ce3+ species (confirmed by XPS) increases the SMSI which also affected the conversion at onset temperature. In nutshell, 7 nm crystallite size and more surface

Fig. 4. CO oxidation activity for PCZ-40, PCZ-40 reduced and PCZ-40 oxidized (air treated).

To check the sustainability and robustness of the catalyst, the catalyst activity of PCZ-40 was evaluated. In the typical studies, the activity was recorded after every hour for 12 h and no change in the activity was recorded. PCZ-40 reduced catalyst exhibited comparatively more conversion at 120° C; this activity is attributed to formation of anionic vacancies and Ce3+ species (Fig. 2). It clearly indicates that Ce3+ species plays vital role in the catalytic activity. It was observed that change in the shape of support like ceria, at 4


A.S. Burange et al.

Ce3+ species found to be a suitable combination for CO oxidation reaction. Thus, by simple photo-deposition technique not only high metal dispersion was achieved but also we could fine tune the average crystallite size of Pd, simply by changing the irradiation time. Thus by employing anionic vacancies along with tuned crystallite size, we can better activity even with low metal loading.

ChemCatChem 6 (2014) 3116–3124. [17] G.B. Hoflund, S.D. Gardner, D.R. Schryer, B.T. Upchurch, E.J. Kielin, Appl. Catal. B 6 (1995) 117–126. [18] T. Engel, G. Ertl, Adv. Catal. 28 (1979) 1–78. [19] M. Haruta, S. Tsubota, T. Kobayashi, H. Kageyama, M.J. Genet, B. Delmon, J. Catal. 144 (1993) 175–192. [20] D.R. Schryer, B.T. Upchurch, J.D. Van Norman, K.G. Brown, J. Schryer, J. Catal. 122 (1990) 193–197. [21] C. S.Gopinath, F. Zaera, J. Catal. 186 (1999) 387–404. [22] C.S. Gopinath, F. Zaera, J. Catal. 200 (2001) 270–287. [23] J.F. Weaver, J.J. Chen, A.L. Gerrard, Surf. Sci. 592 (2005) 83–103. [24] R.B. Shumbera, H.H. Kan, J.F. Weaver, Surf. Sci. 601 (2007) 4809–4816. [25] M.F. Luo, Z.H. Hou, X.X. Yuan, X.M. Zheng, Catal. Lett. 50 (1998) 205–209. [26] H. Zhu, Z. Qin, W. Shan, W. Shen, J. Wang, J. Catal 233 (2005) 41–50. [27] R.S. Monteiro, L.C. Dieguez, M. Schmal, Catal. Today 65 (2001) 77–89. [28] K. Periyasamy, V.T. Aswathy, V. Ashok kumar, M. Manikandan, R. Shukla, A.K. Tyagi, T. Raja, RSC Adv. 5 (2015) 3619–3626. [29] A.S. Burange, R. Shukla, A.K. Tyagi, C.S. Gopinath, ChemistrySelect 1 (2016) 2673–2681. [30] S.S. Kahandal, A.S. Burange, S.R. Kale, P. Prinsen, R. Luque, R.V. Jayaram, Catal. Commun. 97 (2017) 138–145. [31] K. Roy, C.P. Vinod, C.S. Gopinath, J. Phys. Chem. C 117 (2013) 4717–4726. [32] Ambrose A. Melvin, Kavya Illath, T. Das, T. Raja, S. Bhattacharyya, C.S. Gopinath, Nanoscale 7 (2015) 13477–13488. [33] R. Jain, A. Dubey, M.K. Ghosalya, C.S. Gopinath, Catal. Sci. Technol. 6 (2016) 1746–1756. [34] C. Murli, M.S. Shashidhar, C.S. Gopinath, Tetrahedron 63 (2007) 4149–4155. [35] A. Dubey, S.K. Kolekar, E.S. Gnanakumar, K. Roy, C.P. Vinod, C.S. Gopinath, Catal. Struct. React. 2 (2016) 1–12. [36] A. Trovarelli, J. Llorca, ACS Catal. 7 (2017) 4716–4735. [37] J. Liu, Q. Ji, T. Imani, K. Ariga, H. Abe, Sci. Rep. 7 (41773) (2017) 1–7. [38] M. Jeong, N. Nunotani, N. Imanaka, Bull. Chem. Soc. Jpn. 91 (2018) 158–164. [39] K. Mori, Y. Miyauchi, Y. Kuwahara, H. Yamashita, Bull. Chem. Soc. Jpn. 90 (2017) 556–564. [40] C.S. Gopinath, F. Zaera, J. Catal. 186 (1999) 387–404. [41] T. Mathew, S. Kumarsrinivasan, E.S. Gnanakumar, Y. Yamada, T. Kobayashi, C.S. Gopinath, J. Mater. Chem. 22 (2012) 13484–13493. [42] P.A. Bharad, K. Sivaranjani, C.S. Gopinath, Nanoscale 7 (2015) 11206–11215. [43] M. Haruta, S. Tsubota, T. Kobayashi, H. Kageyama, M.J. Genet, B. Delmon, J. Catal. 144 (1993) 175–192. [44] J.M. Campelo, D. Luna, R. Luque, J.M. Marinas, A.A. Romero, ChemSusChem 2 (2009) 18–45. [45] P. Bera, K.C. Patil, V. Jayaram, G.N. Subbanna, M.S. Hegde, J. Catal. 196 (2000) 293–301. [46] A.J. Sahani, R.V. Jayaram, A.S. Burange, Mol. Catal. 450 (2018) 14–18. [47] A.S. Burange, M.B. Gawande, F.L.Y. Lam, R.V. Jayaram, R. Luque, Green Chem. 17 (2015) 146–156. [48] C.S. Gopinath, K. Roy, S. Nagarajan, ChemCatChem 7 (2015) 588–594.

Acknowledgement Author ASB is thankful to CSIR-Nehru postdoctoral fellowship. CSG is thankful for the partial financial support from CSIR, New Delhi through CSC-0404 project. All authors are thankful to Dr. Raja and Dr. Vinod, National Chemical Laboratory for instrumental facilities. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.mcat.2018.05.021. References [1] T. Gerhardt, M.G. Öthert, G. Malorny, H. Wilke, Int. Arch. Arbeitsmed. 28 (1971) 127–140. [2] J. Qi, K. Zhao, G.D. Li, Y. Gao, H.J. Zhao, R.B. Yu, Z.Y. Tang, Nanoscale 6 (2014) 4072–4077. [3] A. Xie, J. Guo, W. Liu, Y. Yang, RSC Adv. 4 (2014) 11357–11359. [4] S.W. Yang, L. Gao, J. Am. Chem. Soc. 128 (2006) 9330–9331. [5] G.N. Li, L. Li, B.H. Wu, J.X. Li, Y. Yuang, J.L. Shi, Nanoscale 7 (2015) 17855–17860. [6] B.P. Mandal, V. Grover, M.R. Pai, A.K. Tyagi, J. Mater. Res. 24 (2009) 2845–2854. [7] V. Iabolkov, R. Barbosa, G. Pollefeyt, I.V. Driessche, S. Chenakin, N. Kruse, ACS Catal. 5 (2015) 5714–5718. [8] M. Kang, M.W. Song, C.H. Lee, Appl. Catal. A 251 (2003) 143–156. [9] J.Y. Luo, M. Meng, X. Li, X.G. Li, Y.Q. Zha, T.D. Hu, Y.N. Xie, J. Zhang, J. Catal. 254 (2008) 310–324. [10] R. Sѐbastien, D. Daniel, ChemCatChem 3 (2011) 24–65. [11] H.C. Yao, Y.F. Yu Yao, J. Catal. 86 (1984) 254–265. [12] M. Sugiura, Catal. Surv. Asia 7 (2003) 77–87. [13] T. Jin, T.O. Kuhara, G.J. Mains, J.M. White, J. Phys. Chem. 91 (1987) 3310–3315. [14] B. Engler, E. Koberstein, P. Schubert, Appl. Catal. 48 (1989) 71–92. [15] W. Liu, M. Flytzanistephanopoulos, J. Catal. 153 (1995) 304–316. [16] E.S. Gnanakumar, J. Madhusudhan Naik, M. Manikandan, T. Raja, C.S. Gopinath,

5