Synthesis of Ni2P/CdS and Pt/TiO2 nanocomposite for photoreduction of CO2 into methanol

It is a great challenge to convert thermochemically stable CO2 into value-added products such as CH4, CH3OH, CO via utilizing solar energy. It is also a difficult task to develop an efficient catalyst for the reduction of CO2. We have designed and synthesized noble metal-free photocatalytic nanostructure Ni2P/CdS and Pt/TiO2 for conversion of CO2 to methanol in the presence of sacrificial donor triethylamine (TEA) and hydrogen peroxide. The synthesised catalysts physicochemical properties were studied by using several spectroscopic techniques like; XRD, UV-DRS, XPS, TEM, SEM and PL. Quantification of methanol by GC–MS showed encouraging results of 1424.8 and 2843 μmol g−1 of catalyst for Pt/TiO2 and 5 wt% Ni2P/CdS composites, respectively. Thus, Ni2P/CdS is a promising catalyst with higher productivity and significant selectivity than in-vogue catalysts.

From the past few decades, substantial changes have been witnessed in the atmosphere due to the burning of fossil fuels 1 leading to an increase in the demand for renewable energy 2 . Also, the natural resources of fossil fuels are decreasing day by day, thus increasing the demand for an alternate source of energy 3,4 . At present, much of our energy demands are met by fossil fuel, but this resource is not a renewable source and is associated with contemporary environmental issues. Thus, renewable sources of energy shall play a critical role in fulfilling the requirement of energy demand 5 . More efforts have been focused on substituting renewable biomass sources for chemicals and fuels, which possess high energy density and environmentally friendly properties 6 . Mixing alcohol with petroleum products increases the fuel's combustion efficiency due to oxygen and reduces the emission of pollutants into the atmosphere 7,8 . The regular trend for methanol synthesis uses syngas conversion with high pressure and high thermal energy with heterogeneous catalytic reaction over alumina supported Cu/ZnO catalyst 9 . In the petrochemical industry, the production of methanol from CO 2 reduction an important achievement. In this research area, efforts are continuing to increase the selectivity for methanol by reverse-water-gas-shift (RWGS) reaction in thermal catalytic CO 2 hydrogenation via Cu/ZnO-based catalyst. The great challenge in this reaction is to maintain the operating temperature and pressure. Due to the competing RWGS reaction, the exothermic nature of methanol synthesis, equilibrium-limited conversion, and the very high activation energy barrier of the CO 2 hydrogenation to methanol reaction, the process must balance operation temperature megapascal pressures to achieve high selectivity and production rate 10,11 . Fujishima first reported the photocatalysis study in 1972; in recent years, it's become a common word in chemistry and also various technology products 12 . The theoretical target is to get chemical energy by using light energy with semiconductor materials. The electron present on the conduction band gets activated and migrates to the valance band of the semiconductor with the holes in the conduction band. Thus, the charge carriers facilitate the reaction wherein the electrons react with dissolved CO 2 present in the reactant stream 13 . The reduction of CO 2 is mainly depending on the highly reducing electron, which has a high reduction ability to convert CO 2 into useful methanol 14 . As per Gibb's free energy values, the CO 2 and methanol having − 394.4 and − 166.4 kJ/mol energy indicating that the reduction of CO 2 is an exothermic reaction 15 . Many processes are used for the conversion of CO 2 into methanol such as (1) chemical conversion method, (2) electrochemical conversion method, (3) photochemical reduction method and (4) photoelectrochemical reduction method (Eq. 1).
(1)  16 studied the hydrogenation of CO 2 to MeOH and CH 4 which disclosed that thermodynamically CH 4 formation is more favorable than MeOH generation. Fiordaliso and Diez-ramiez 17,18 used Pd 2 Ga and Cu/ ZnO catalyst, reported 100% selectivity for MeOH at ambient pressure but these gave a low yield of conversion. TiO 2 is widely studied as a photocatalyst with a wide bandgap semiconductor (3.2 eV) which can only be excited by UV light. CdS has a narrow bandgap of 2.4 eV and has received considerable interest in the field of water splitting and environmental remediation 19 . It can be a suitable candidate for visible light absorption and fits the thermodynamics requirements for H 2 O/H 2 20-22 . However, the photocatalytic activity over pure CdS is relatively low due to the instability and recombination of the electron-hole pairs. To overcome the above-mentioned problems, investigators are making a series of approaches that can be applied to improve the photocatalytic activity of CdS and TiO 2 23-25 . Combining Ni 2 P with CdS has proved to be an efficient way to enhance photocatalytic performance. With this background, we have synthesized two nano-systems Pt/TiO 2 and Ni 2 P/CdS to convert CO 2 into methanol in the presence of sacrificial donor triethylamine. The loading Ni 2 P on CdS is also studied by increasing concentration of Ni 2 P from 1 to 5% weight to find the optimum concentration of Ni 2 P. H 2 O 2 can act as a reducing as well as oxidizing agent depending on pH value. It works as a potent reducing agent in the presence of a basic medium and produces O 2 and 2H + ions. The present manuscript discusses the significant conversion of CO 2 to methanol with H 2 O 2 , the selectivity of methanol production (no by-products were observed) and the excellent yield of methanol (2843 μmol/g cat using 5 wt%-Ni 2 P/CdS catalyst). The finding of this has been compared with the conventional catalyst reported in the literature 26 .
Synthesis of Ni 2 P/CdS. The solvothermal method was used to prepare cadmium sulphide nanorods.
10 mmol and 30 mmol of cadmium nitrate tetrahydrate and thiourea were dissolved in ethylenediamine (30 ml) respectively. These were transferred into 80 ml of Teflon-lined autoclave (stainless steel), which was then kept for 48 h at 160 °C in a hot air oven. After 48 h, an autoclave is set aside to cool at ambient temperature. The yellow colored product is rinsed with deionized (DI-water) water and ethyl alcohol several times. The yellow-colored cadmium sulphide was dried in a vacuum oven at around 60 °C for 6 h 27,28 . Preweighed quantity of nickel phosphide and synthesized cadmium sulphide were dissolved in ethylenediamine. The same autoclave is used to heat the mixture at 140 °C for 12 h. The yellow-green colored product obtained was collected from the cooled autoclave, which was rinsed with ethanol and DI-water several times. The final Ni 2 P/CdS catalyst was dried into a vacuum oven at around 60 °C for 5 h.

Synthesis of Pt/TiO 2 nanoparticles.
The photo-deposition technique was used to deposit platinum over titanium dioxide using hydro chloroplatinic acid. The experiment was conducted in a 500 ml irradiation type (inner) reactor. A 350 w medium pressure mercury vapor lamp was used as an irradiation light source and cooled to room temperature with cold circulating water. This photochemical reaction was performed in 400 ml of aqueous reactant solution with 150 mg of titanium dioxide and was continuously stirred magnetically for about 30-45 min to accomplish complete dispersion. Before starting the photo-deposition reaction, nitrogen was purged within a glass reactor to withdraw the air for 25 to 30 min (Scheme 1) 29 .
Characterization. Morphology of the photocatalysts was determined by scanning electron microscope (SEM) (Tescan, HiPace 10) and HRTEM using JEM-2100F JEOL Japan. The XRD patterns were recorded by using a benchtop X-ray Diffractometer (Model: Rigaku Miniflex II at 30 kV) having a scintillation counter detector. X-ray photoelectron spectroscopy (XPS) analysis has been done by using KRATOS AXIS 165 with Mg Kα irradiation. An Agilent Cary 5000 UV/VIS/NIR spectrophotometer was used to determine the UV/VIS absorption at ambient conditions. Photoluminescence was recorded at Hitachi F-7000 spectrofluorometer. Liquid samples were taken at a different time interval with an airtight syringe and separated by offline GC detected by a flame ionized detector (FID) using helium as carrier gas (Perkin Elmer Clarus 680).

Results and discussion
XRD studies. The XRD patterns of the TiO 2 , Ni 2 P, CdS, Pt/TiO 2 , CdS, 1 wt% Ni 2 P/CdS, 3 wt% Ni 2 P/CdS, 5 wt% Ni 2 P/CdS and 6 wt% Ni 2 P/CdS are shown in Fig. 1 SEM and HRTEM studies. SEM and HRTEM analysis were carried out to investigate the morphology and microstructure of prepared photocatalysts. In Fig. 2a-d shows crystalline CdS nanorods with an average diameter of 100-5 nm while nickle phosphide deposited over CdS via solvothermal process showed similar geometry of nanorods and their diameter has not apparent change. The distance of lattice fringe of 0.33 and 0.22 nm for CdS and Ni 2 P, respectively, clearly shows that they are in close contact. The images clearly showed that the length of Ni 2 P/CdS is a little smaller than pure CdS while the structure and the results confirmed that morphology had no effect of the deposition of Ni 2 P on CdS. In Fig. 2e-h Pt/TiO 2 crystal clearly shows platinum deposited on the TiO 2 crystals which are seen as black spots at high resolution in TEM images 32 . In Fig. 3 SEM images of CdS (a-d) and Ni 2 P/CdS (e-h) from 200 to 1 μm indicates that there is no change after deposition of nickel on CdS nanorods, especially in d and h clearly showing that after and before deposition of Ni 2 P the nanorods structures are clearly observed in both images. SEM results corroborate the finding of TEM that deposition of co-catalysts like Ni 2 P and Pt nanostructure on CdS and TiO 2 has minimal impact on the structural morphology of the materials.

UV-Vis diffuse reflectance spectroscopic (DRS) studies.
The diffuse reflectance UV-Vis absorption spectra of synthesized composites were assembled for an area of 200-700 nm. The strong absorption peak ~ 264 to 296 nm indicates the interband transition between Ni 2 P-CdS and Pt-TiO 2 (Fig. 4A). The absorption at a wavelength shorter than 390 and 580 nm in Ni 2 P/CdS and Pt/TiO 2 can be ascribed as intrinsic bandgap absorption. These catalysts show the broad absorption in a visible range due to Ni 2 P and Pt deposition on CdS and TiO 2

33
. The bandgap E g , of both the samples estimated using the Tauc's relation given below in Eq. (2) is as follows; where A is a constant, Eg the semiconductor bandgap and n is a number equal to 1/2 for the direct gap. The (αhυ) 2 versus energy graphs were plotted. In Fig. 4, a straight line was fitted for the straight segment, this straight line was extrapolated to the X-axis to get the band gap values. The catalysts Ni 2 P, TiO 2 and Ni 2 P/CdS have shown a bandgap of 1.05 eV, 3.3 eV and ~ 2.33 eV, respectively (Fig. 4B). The bandgap of CdS is shifted to left after loading of Ni 2 P and maximum shift has been observed in 5 wt% Ni 2 P/CdS indicating deposition of Ni 2 P on CdS and improved photocatalytic efficiency for CO 2 reduction.
A series of experiments were conducted to understand the catalyst better. UV-DRS and photoluminescence (PL), was performed to investigate the photocatalytic mechanism for Ni 2 P/CdS composite. In the Fig. 4A, it is seen that CdS shows an absorption edge of approximately 490 nm 34 . The UV and visible light absorption intensity increased compared to pure CdS and 1-6 wt% Ni 2 P/CdS samples, which was attributed to the reduction of reflectivity 35 . The photoabsorption property has been increased by increasing the amount of Ni 2 P because of its dark colour. A slight redshift is seen in the samples of Ni 2 P/CdS and thus, the Ni 2 P is loaded onto the surface www.nature.com/scientificreports/ rather than the crystal lattice. In the present work, the optimal value was attained in 5 wt% Ni 2 P/CdS sample. In addition, as the diameter grows, it takes longer for the electrons to pass nanoparticles into the surface of Ni 2 P and less electrons to produce methanol. The maximum yield of methanol was obtained (2843 μmol/g cat) by using 5 wt% Ni 2 P/CdS catalyst under a visible light source of 300 W Xe lamp with 420 nm cutoff filter and 300 W tungsten light for 1 h. The Hg lamp with 300 W was the light source for Pt/TiO 2 catalyst. As expected, increased Ni 2 P loading resulted in increased photoactivity. Ni 2 P served as an active centre for the production of methanol by electron trapping to decrease the recombination rate of electron-hole pairs during photocatalyst excitation. The adsorption of reactant CO 2 can be supported by surface hydroxyl (OH) of CdS to improve photoreaction 36 . The amount of hydroxyl on the surface of CdS increased with increasing Ni 2 P loading, and the total CO 2 photoreduction increased significantly. However, excess of Ni 2 P on the CdS surface (6 wt%-Ni 2 P/CdS), resulted in less catalyst light exposure. As a result, electron and hole pair photoexcitation was decreased because less photo-energy was absorbed. Therefore, an optimal Ni 2 P loading on the catalyst is a 5 wt% Ni 2 P and the highest methanol yields were found at this percentage under experimental conditions. The photoluminescence (PL) is obtained from the radiative recombination of the photo-generated electron-hole pair. The (PL) emission experiment was carried out at an excitation wavelength of 350 nm. A series of studies includes loading of Ni 2 P on CdS (1 wt% Ni 2 P/CdS, 3 wt% Ni 2 P/CdS, 5 wt% Ni 2 P/CdS and 6 wt% Ni 2 P/ CdS) as shown in Fig. 4C were restricted to the recombination of photoexcited electron-hole pairs. It appears that all samples exhibit a high emission at 535 nm because free electrons are radially recombined with free holes  . UV-Vis diffuse reflectance spectrum (A), Tauc plots (B) and photoluminescence spectrum of (C) of synthesized catalysts CdS, 1 wt% Ni 2 P/CdS, 3 wt% Ni 2 P/CdS, 3 wt% Ni 2 P/CdS, 5 wt% Ni 2 P/CdS and 6 wt% Ni 2 P/CdS. XPS spectra. The valence state of the elements and the chemical composition of Ni 2 P/CdS photocatalyst were studied through XPS measurements. Figure 5 represents the binding energy (BE) curve for the core level spectra of Cd 3d, S 2p, Ni 2p, and P 2p constituent elements in the photocatalyst. Figure 5a displays the core level spectrum of Cd 3d area where 403.04 eV and 408.80 eV peaks correspond to Cd 3d 5/2 and Cd 3d 3/2 , respectively. These values are close to earlier reported values 37 . In Fig. 5b, S2p has two peaks positioned at 160.54 eV and 159.38 eV are corresponding to S 2p 1/2 and S 2p 3/2 orbitals of divalent sulfide ions (S 2− ), respectively, which are in line with the formation of CdS 38 . The characteristic binding energies of peaks located at 854.35 eV, 871.72 eV, 130.82 eV and 131.15 eV correspond to Ni 2p 3/2 , Ni 2p 5/2 , P 2p 3/2 and P 2p 1/2 respectively, shown in Fig. 5c,d 39,40 . These results clearly demonstrate the existence and strong interaction between Ni 2 P and CdS, which is in good agreement with the TEM analysis.
Photocatalytic reduction of CO 2 to methanol. A known amount of CO 2 is purged into DMF after removal of moisture from the solvent with nitrogen purging at least for 30 min. 20 mg of the prepared photocatalyst was well dispersed in the mixture of DMF with TEA as a sacrificial electron donor. Carbon dioxide is added to the reaction mixture, followed by hydrogen peroxide and irradiated with light for the different time intervals. Liquid samples were filtered with a microsyringe filter and injected in offline GC to detect products (Scheme 2). The photocatalytic reduction of CO 2 to methanol was performed for both prepared catalysts. The catalysis test results showed that Ni 2 P/CdS is highly active than Pt/TiO 2 under the influence of light (Fig. 6). An increment in the concentration of H 2 O 2 shows a positive effect on methanol generation until the concentration researches 1.5 ml. There is no detection of methanol over oxidation products like formaldehyde and formic acid etc., in the GC and GC-MS spectrum indicting high selectivity towards methanol. There is less variation in methanol formation between 1 and 1.5 ml of H 2 O 2 , indicating that the reaction is completed at this point. However, the additional increment in H 2 O 2 concentration of more than 1.5 ml shows the reduction in the formation of methanol. This may be attributed to a decrease in the availability of CO 2 in the reaction mixture.
Therefore the ideal concentration of H 2 O 2 for the reduction of CO 2 is 1.5 ml/20 mg of catalyst. The effect of light over a different time is also done using the same catalyst. It is concluded that as the time duration increases the concentration of methanol increases. After an hour, no significant impact on the generation of methanol     (Fig. 7A). Nanocomposites typically have specific features that support the catalyst and are reusable. To illustrate this characteristic, we have studied the performance of selected catalyst 5 wt%-Ni 2 P/CdS to reduce CO 2 to CH 3 OH. The obtained results were strongly supporting to the production of methanol after 3-cycle at 1 h time. After completion of 1 h reaction, the catalyst was removed from the reaction mix and dried and the next reaction was conducted. Based on the findings, the formation of CH 3 OH after each run does not effect a significant loss in efficiency of catalyst (Fig. 7B).
Quantum efficiency for CO 2 reduction. Photochemical efficiency describes the percent of absorbed photons that reduce CO 2 to products. CO 2 adsorption on photocatalysts affects the efficacy of photochemical efficiency. The quantum outcome of the reaction is widely called photochemical efficiency. The photo-reduction of CO 2 to methanol requires two electrons, the photochemical efficiency of the reaction is obtained by the equation given below 41 .
Here we calculate the quantum efficiency at λ = 420 ± 15 nm. The catalyst mixture was irradiated by a 300 W Xe lamp for 1 h. The average incident irradiation was determined to be 2.75 W/cm 2 and the area was 17.59 cm 2 . The amount of methanol produced in 1 h was 2843 μmol. All the calculations are given below.
The number of incident photons (N) in 1 h over 17.59 cm 2 area: Photochemical efficiency depends on the intensity and wavelength of radiation. According to experiment results, a maximum quantum yield of 27.91% was obtained by 5 wt% Ni 2 P/CdS (Table 1).
Mechanism. Based on the experimental results a plausible schematic mechanism basic the robust methanol production over 5 wt%-Ni 2 P/CdS composite was proposed (Fig. 8). It is well known that the CB edge of CdS is more negative than that of Ni 2 P. Thus on irradiation with visible light the photogenerated electrons can efficiently transfer from the conduction band (CB) of CdS to Ni 2 P. The electrons accumulating on the Ni 2 P particles can reduce CO 2 into CO 2 − while the holes on CdS can oxidize H 2 O 2 to produce O 2 and H + . Then CO 2 − reacts with H + to produce CH 3 OH in presence of sacrificial electron donor TEA (Eq. 2). The effective separation of the photo-generated electrons and holes in CdS further improves photocatalytic activity.

Conclusion
The reduction of CO 2 via visible light to methanol is achieved successfully. Here we are reporting methanol generation over noble metal-free hybrid semiconductor photocatalyst (Ni 2 P/CdS) under visible-light-driven CO 2 reduction. The photocatalytic activity of the catalyst was enhanced by the homogenous dispersion of the co-catalyst. The synthesized Pt/TiO 2 photocatalyst was adopted as the reference for comparison of the photocatalytic activity of the Ni 2 P/CdS nanocomposites under similar experimental conditions with an increase in Ni 2 P loading, the photocatalysts showed increased crystallinity. The CdS loaded with Ni 2 P showed greater efficiency than CdS for the formation of methanol from CO 2 . The 5 wt% loading of N i2 P on CdS was found to be optimal among the different compositions and afforded the highest product yield (2843 μmol/g cat). The reaction was found to be completely photocatalytic, as in the absence of visible light, no conversion was observed. The synthesized photocatalyst was heterogeneous and showed clear 3 recycle runs with no change in catalytic efficiency and also no significant leaching and change in morphologywas observed. Also, the reduction rate of CO 2 is much higher than the previously reported catalysts. These results may provide a new avenue for various photocatalytic applications.   Figure 8. Probable mechanism of charge transfer and CO 2 reduction by 5 wt%-Ni 2 P/CdS composite and chemical equations. Comparision of Ni 2 P/CdS and Pt/TiO 2 for photoreduction of CO 2 to methanol is provided in Table 2. The synthesised 5% Ni 2 P/CdS photocatalyst in this work appears to outperforming the other reported photocatalysts.