Photocatalytic CO2 Reduction Using Ni2P Nanosheets

Zhiming Pan, Minghui Liu, Pingping Niu, Fangsong Guo, Xianzhi Fu, Xinchen Wang

State Key Laboratory of Photocatalysis on Energy and Environment, College of Chemistry, Fuzhou University, Fuzhou 350116,P.R.China.

Abstract:Artificial photosynthesis is an ideal method for solar-to-chemical energy conversion, wherein solar energy is stored in the form of chemical bonds of solar fuels.In particular, the photocatalytic reduction of CO2 has attracted considerable attention due to its dual benefits of fossil fuel production and CO2 pollution reduction.However, CO2 is a comparatively stable molecule and its photoreduction is thermodynamically and kinetically challenging.Thus, the photocatalytic efficiency of CO2 reduction is far below the level of industrial applications.Therefore,development of low-cost cocatalysts is crucial for significantly decreasing the activation energy of CO2 to achieving efficient photocatalytic CO2 reduction.Herein,we have reported the use of a Ni2P material that can serve as a robust cocatalyst by cooperating with a photosensitizer for the photoconversion of CO2.An effective strategy for engineering Ni2P in an ultrathin layered structure has been proposed to improve the CO2 adsorption capability and decrease the CO2 activation energy, resulting in efficient CO2 reduction.A series of physicochemical characterizations including X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS),transmission electron microscopy (TEM), high-resolution transmission electron microscopy (HRTEM), and atomic force microscopy (AFM) were used to demonstrate the successful preparation of ultrathin Ni2P nanosheets.The XRD and XPS results confirm the successful synthesis of Ni2P from Ni(OH)2 by a low temperature phosphidation process.According to the TEM images, the prepared Ni2P nanosheets exhibit a 2D and near-transparent sheet-like structure, suggesting their ultrathin thickness.The AFM images further demonstrated this result and also showed that the height of the Ni2P nanosheets is ca 1.5 nm.The photoluminescence (PL) spectroscopy results revealed that the Ni2P material could efficiently promote the separation of the photogenerated electrons and holes in [Ru(bpy)3]Cl2·6H2O.More importantly, the Ni2P nanosheets could more efficiently promote the charge transfer and charge separation rate of [Ru(bpy)3]Cl2·6H2O compared with the Ni2P particles.In addition, the electrochemical experiments revealed that the Ni2P nanosheets, with their high active surface area and charge conductivity, can provide more active centers for CO2 conversion and accelerate the interfacial reaction dynamics.These results strongly suggest that the Ni2P nanosheets are a promising material for photocatalytic CO2 reduction, and can achieve a CO generation rate of 64.8 μmol·h-1, which is 4.4 times higher than that of the Ni2P particles.In addition, the XRD and XPS measurements of the used Ni2P nanosheets after the six cycles of the photocatalytic CO2 reduction reaction demonstrated their high stability.Overall, this study offers a new function for the 2D transition-metal phosphide catalysts in photocatalytic CO2 reduction.

Key Words:Photocatalysis;Nanosheet;Ni2P;CO2 reduction;CO

1 Introduction

Two of the world’s problems, the energy dilemma and global warming, stem from the abundant use of chemical fuels accompanied by the release of the greenhouse gas, CO21-10.The conversion CO2into available carbon forms has drawn particular attention, for which the photoreduction of CO2into chemical fuels has been generally considered an attractive solution to simultaneously solve both environmental problems and energy issues11-15.Over the past decade, tremendous effort has been dedicated to developing an efficient catalyst for achieving efficient CO2reduction16-24.Although progress has certainly been made, the CO2photocatalytic efficiency is still far from satisfactory, primarily because of the huge CO2 activation energy and the slow kinetics of the complicated multiple electron transfer and reaction processes involved25-28.Hence, it is necessary to design a new artificial material based on proper architectures, high selectivities and high activities by using inexpensive materials for solar-driven CO2reduction.

Transition-metal phosphides (TMPs) have attracted a significant attention, due to their low price, unique electronic structure, and prominent catalytic features5,28,29.In past few years, notable efforts have been devoted to the study of TMPs as potential noble-metal free electrocatalyst for hydrogen evolution reaction (HER)30,31and oxygen evolution reaction (OER)32-35.In additional, the TMPs as HER or OER cocatalysts were widely investigated and applied in the field of photocatalytic water splitting5,36,37.However, the TMPs were rarely reported to reduce CO2 in past few years.Until recently, it was demonstrated that TMPs has a noteworthy performance for CO2 electroreduction.For example, MoP nanoparticles supported on Indium-doped porous carbon was applied to electrochemical reduction of CO238.Meanwhile, the nickel phosphides also were reported as electrocatalyst for the reduction of CO2at overpotential as low as 10 mV39.More importantly, Fuet al.40reported that a highly efficient photocatalytic system constructed from CoP/carbon nanotubes or graphene for visible light driven CO2reduction.Thus, it is highly desired that TMP materials can be further explored as an efficient catalyst for the photocatalytic reduction of CO2using sustainable light energy to run the reaction.Because the catalytic efficiency is affected by the morphology of the catalyst41,42, it is necessary to design proper architectures for the TMPs to maximize their cocatalytic functions.

Nanosheets with unique physical and chemical properties have shown great advantages in activating and driving the CO2conversion43-46.For example, nanosheets could shorten the propagation path of charge carriers, offer gigantic surface areas and expose more active sites on the surface47,48.In addition,nanosheets can maximize the exposure of the catalytically active facet for CO2reduction by crystal facet engineering49-51.Nanosheets can introduce more defects on the surface to activate and drive the CO2conversion52.As a result, a variety of nanosheets, including metals53, layered double hydroxides54,metal sulphides1, and transition-metal oxides44,49, have been studied for catalytic reduction of CO2.For example, ultrathin 2D antimony nanosheets53showed excellent catalytic performance for the electrocatalytic reduction of CO2into formate with low overpotential.This excellent CO2reduction property, which is no active in bulk antimony, is possibly because of the ultrathin 2D layer structure with more active sites for CO2reduction.Therefore, the synthesis of nanosheets is considered to be an efficient strategy to improve photocatalytic CO2reduction performance.

Here, we developed Ni2P nanosheets as an efficient cocatalyst to split CO2in collaboration with a ruthenium-based photosensitizer (Scheme 1).The Ni2P nanosheets exhibited a high activity of 64.8 μmol·h-1.Compared to that of Ni2P particles,the photocatalytic activity of the Ni2P nanosheets improved by a factor of 4.6.The electrochemical experiments demonstrated that the Ni2P nanosheets possess more activity sites for CO2adsorption and activation, realized an improved efficiency for the photocatalytic reduction of CO2.

2 Experimental and computational section

2.1 Synthesis of TMPs nanosheets

The TMP nanosheets was synthesized using a water solution templating salts method reported before35.100 mg of MOx(Co3O45,Fe2O37and Ni(OH)2 (Alfa Aesar., Ni ＞ 61%)), 500 mg of NaH2PO2(Sinopharm Chemical Reagent Co., ＞ 99%) and 1200 mg LiCl (Sinopharm Chemial Reagent Co., ＞ 95%) were ground in a mortar to form a fine powder.Then, the mixtures were heated to 300 °C for 2 h under N2atmosphere.The product was washed with deionized water and dried in a vacuum oven at 60 °C for 24 h.Finally, the TMPs nanosheets were collected by centrifugation at 2000 r·min-1to remove the residual big TMPs particles.

2.2 Characterization

The crystal phases of the samples were detected by X-ray diffraction (XRD) on a Bruker D8 Advance diffractometer(Bruker, Germany) with CuKα1 radiation (λ= 1.5406 ?, 1 ? = 0.1 nm).X-Ray photoelectron spectroscopy (XPS) was collected on a Thermo Scientific ESCALAB250 instrument (VG Scintic Ltd.,England) with a monochromatized AlKαline source (200 W).Binding energies (BE) were analyzed using the C 1speak at 284.6 eV.The morphology of the samples was analyzed by transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM) on a FEI TECNAI G2F20 instrument (FEI CO.USA).AFM (Agilent Technologies,USA) were used to examine to the thickness of the 2D nanosheets.

2.3 Electrochemical measurements

All the electrochemical measurements were performed in a three-electrode system at an electrochemical station in an airtight home-made “H” type electrochemical cell with 60 mL volume,which was separated by a cation exchange membrane (Nafion 117).30 mL Ar-saturated or CO2-saturated 0.1 mol·L-1KHCO3(Alfa Aesar., ＞ 99%) electrolyte was elected as electrolyte.The counter and the reference electrodes were a Pt sheet and the Ag/AgCl reference electrode, respectively.The catalysts modified-Ti foil served as the working electrode.Then, 20 μL of the dispersion (5 mg·mL-1Ni2P DMF solution) was deposited onto 0.25 cm2pieces of Ti foil.The work electrode was then airdried at room temperature.The Ti foil is coated with epoxy except for the catalysts films.

For electrochemical CO2reduction experiments, linear sweep voltammetry at a scan rate of 20 mV·s-1was performed in CO2-saturated 0.1 mol·L-1KHCO3solution (30 mL).For comparison,linear sweep voltammetry with a scan rate of 20 mV·s-1was also carried out in Ar-saturated 0.1 mol·L-1KHCO3 solution.

The electrochemically active surface areas of the catalysts were tested from double layer charging curves using cyclic voltammograms.The electrochemically active surface area(ECSA) =RfS, whereSwas generally equal to the geometric area of Ti foil (In this work,S= 0.25 cm-2).The roughness factor (Rf)was determined by the relationRf= Cdl/20 μF·cm-2based on the double-layer capacitance (Cdl) of a smooth metal electrode(assuming that the average double-layer capacitance of a smooth metal surface is 20 μF·cm-2).TheCdlwas determined by plotting the Δj(ja-jc) at -0.36 VvsAg/AgCl against the scan rate, where the Δjcould be acquired by cyclic voltammetry measurement under the potential windows from -0.35 to -0.45 VvsAg/AgCl(1 mol·L-1KHCO3 solution).Electrochemical impedance spectroscopy (EIS) was analyzed at -0.8vsAg/AgCl from 100 mHz to 200 kHz in 1 mol·L-1KHCO3solution.

2.4 Photocatalytic CO2 reduction

The photocatalytic CO2reduction reactions were performed in an 80 ml reactor at 30 °C controlled by cooling water and 1 atm(1 atm = 100 kPa) CO2 partial pressure.The reaction system contained Ni2P (1 mg), [Ru(bpy)3]Cl2·6H2O (bpy = 2’2-bipyridine, 10 μmol), solvent [5 mL, acetonitrile: H2O = 3 : 2(vol/vol)] and triethanolamine (TEOA) (1 mL), stirred with a magnetic stirrer, and irradiated under a 300 W Xe lamp with a 420 nm cutoff filter.After reactions, the generated gases were detected and quantified by an Agilent 7820A gas chromatograph.

3 Results and discussion

Scheme 1 Schematic process for the photosplitting of CO2 into CO using Ni2P nanosheets as a catalyst.

Considering that it is difficult to carry out the XRD and XPS characterization on a single ultrathin structure, the layer-by-layer assembly strategy35,55was performed for the XRD and XPS characterization of Ni2P nanosheets.To evaluate the chemical composition and phase of the Ni(OH)2and Ni2P samples,powder X-ray diffraction (XRD) patterns were obtained.As shown in Fig.1a, the diffraction peaks for the precursor could be assigned to the Ni(OH)2phase.After phosphidation, the diffraction peaks of the Ni2P phase were observed56.Obviously,these results imply the successful synthesis of Ni2P from Ni(OH)2by a low temperature phosphidation process.In addition, the Ni2P nanosheets show a additional XRD peak at 26.7°, which are correspond to the (101) of quartz holder.To further demonstrate that the Ni2P material was successfully synthesized, XPS spectra were obtained.As shown in Fig.S1(Supporting Information (SI)), the XPS survey spectra of the Ni2P nanosheets clearly show the appearance of elemental Ni and P.To gain further insight into the chemical states of the Ni2P nanosheets, high resolution spectra were analyzed.As shown in Fig.1b, the Ni 2pXPS spectrum shows four characteristic peaks at 852.9, 856.6, 870.3 and 874.8 eV57.In addition, the two additional peaks at 861.4 and 879.5 eV correspond to the satellite peaks.The P 2pXPS spectrum (Fig.1c) displays the two distinct peaks at 129.4 and 130.3 eV related to the metal phosphide signal.In addition, we find that one broad peak at 133.4 eV could be ascribed to phosphorus species.This is possible because that the Ni2P surface was oxidized because of air contact.

Transmission electron microscope (TEM) images were obtained to check the morphology of Ni2P nanosheets.In Fig.1d,the TEM image reveals that the product possesses a 2D sheetlike structure, and the near-transparency of the Ni2P nanosheets suggests their ultrathin thickness.The high resolution TEM(HRTEM) image (Fig.1e) shows a clear crystal lattice with interplanar spacings of 0.22 and 0.29 nm, corresponding to the(111) and (110) planes of Ni2P, respectively58.This result further confirms that Ni2P nanosheets have been successfully synthesized.In addition, the colloidal suspension of the asobtained sample (Fig.1e inset) showed the Tyndall phenomenon,suggesting the formation of homogeneous nanosheets in water.The thickness of the Ni2P nanosheets was evaluated by atomic force microscopy (AFM).As shown in Fig.1f and Fig.S2 (SI),the height of the Ni2P nanosheets is ca.1.5 nm.

The photocatalytic reduction of CO2was performed in CH3CN/H2O solvent by visible light irradiation (λ＞ 420 nm)under mild reaction condition (30 °C and 1 atm CO2) with triethanolamine (TEOA) and [Ru(bpy)3]Cl2 as the electron donor and photosensitizer, respectively.As expected, the Ni2P nanosheets (Fig.2a) exhibited excellent catalytic properties for CO2reduction with a CO-evolving rate of 64.8 μmol·h-1(32.4 μmol/0.5 h).Compared to commercial Ni2P particles (Figs.S3 and S4 (SI)) with a CO-evolving rate of 7.4 μmol/0.5 h, the activity of the Ni2P nanosheets had improved by a factor of 4.4.This result was possibly attributed to the ultrathin 2D layer structure of Ni2P, which had a huge amount of coordinatively unsaturated surface atoms that could act as the catalytically active centers to split CO2into CO.Meanwhile, Ni(NO3)2·6H2O was selected as a catalyst for CO2 reduction, and the results show that its catalytic performance is far below that of the Ni2P nanosheets.These results suggest that the Ni2P nanosheets can act as efficient catalysts for CO2reduction.

Fig.1 (a) XRD patterns for Ni2P nanosheets and Ni(OH)2 nanoparticles.XPS spectra of Ni2P nanosheets in the (b) Ni 2p and(c) P 2p regions.(d) TEM image of Ni2P nanosheets.(e) HRTEM image of Ni2P nanosheet and photograph of the corresponding colloidal dispersion displaying the Tyndall effect (inset).(f) AFM image of Ni2P nanosheets.

Fig.2 (a) Production of H2 and CO using different samples and results of the GC-MS analysis of the generation of CO using13 CO2 as the gas source (inset).(b) Generation of H2 and CO from the photocatalytic CO2 reduction system under various reaction conditions.(c) Evolution of H2 and CO from the photocatalytic CO2 reduction system in various solvents (THF, tetrahydrofuran; DMF, N,N-dimethylformamide;DCM, dichloromethane).(d) Wavelength dependence of the formation of H2 and CO, and the light adsorption spectrum to the[Ru(bpy)3]Cl2 photosensitizer.The wavelength of the incident light is regulated by applying relative long-pass cut-off filters.

To further study the key factors in the CO2reduction of Ni2P nanosheets, control experiments (Fig.2b) were performed.No CO or H2 gas can be observed when tested in the dark or without[Ru(bpy)3]Cl2, suggesting that this reaction is photocatalytic.Meanwhile, if the CO2reduction reaction was carried out without TEOA, no CO or H2 gas evolution could be observed.Besides, no obvious CO2-to-CO conversion reactivity was observed without catalyst, demonstrating the vital role of the Ni2P catalyst in promoting the reduction of CO2.No CO evolution was observed when the reaction was carried out by substituting CO2with Ar under similar conditions, suggesting that the CO gas stems from photosplitting of the CO2 reactant.To further demonstrate that the CO gas came from the splitting of CO2, a13CO2isotope experiment (Fig.2a inset) was performed.Only13CO can be observed when13CO2is used as the gas source, demonstrating that the CO gas comes from the splitting of CO2.

To study the effect of solvent on the catalytic performance of the system, we performed the CO2 photoreduction reaction at various solvents.As shown in Fig.2c, the reaction solvents have a significant effect on the catalytic performance.When the MeCN, DMF and THF were selected as the reaction solvents,the catalytic system shown moderate catalytic activities.However, when the catalytic reactions were performed in the presence of DCM and BTF, the system produced only a small or none amount of H2and CO production.This result mainly attributed to the different chemical affinities between the CO2molecules and the solvent used, because the MeCN, DMF and THF contain nitrogen or oxygen atoms that are benefit for solubilizing CO2through Lewis acid-base interactions59.Meanwhile, this result implied that selecting proper reaction solvents play a significant role in CO2 conversion.To further evaluated the catalytic activities of the Ni2P nanosheets, the photocatalytic CO2reduction experiments were carried out under different wavelengths.As shown in Fig.2d, the rates of H2and CO evolution both decrease with increasing wavelength, and the trends of H2and CO evolution rates match well with the DRS absorption spectrum of the Ru complex.It suggested that the CO2 reduction reaction is driven photocatalytically by the harvested light photons of the Ru photosensitizer.

To assess their stability, Ni2P nanosheets were repeatedly used to in the CO2reduction reaction for six cycles.The used Ni2P nanosheets were washed with CH3CN and collected by cenrifugation at 11000 r·min-1, subsequently redispersion in a fresh dye solution for 0.5 reaction.As shown in Fig.3a, no evident decline in the catalytic performance was observed during the catalytic activity test, confirming the high stability of the ultrathin Ni2P nanosheets.In addition, XRD (Fig.4a) and XPS(Fig.4b and c) measurements of the used Ni2P nanosheets after the six photocatalytic cycles also demonstrate the stability of the catalyst.The time curves of CO and H2 evolution are shown in Fig.3b.The formation of CO and H2increased noticeably over the first 1 h of the CO2reduction reaction, but the subsequent CO2-to-CO conversion and H2formation rates decreased rapidly.The decrease in the generation rates of CO and H2after the long-term reaction is primarily ascribed to the photodegradation of[Ru(bpy)3]Cl260,61.The accumulated product of a four hour reaction is 48.6 μmol and therefore afforded a catalytic TON of 7.2 with respect to the Ni2P nanosheets, revealing the catalytic nature of the reaction.

Fig.3 (a) The stability test of Ni2P nanosheets in six of photocatalytic operations.(b) Time-yield plots of H2 and CO produced from the Ni2P nanosheets and ruthenium-based photosensitizer CO2 photoreduction system under visible light.

Fig.4 XRD patterns (a) and XPS (b and c) spectra of the Ni2P nanosheets: fresh sample and used sample after the photocatalytic CO2 reduction reaction.

Some other TMP nanosheets (Figs.S5 and S6 (SI)) were also studied as catalysts in the system to gain a better understanding of Ni2P nanosheets in promoting CO2reduction catalysis.When FeP nanosheets were used instead of Ni2P nanosheets (Fig.5),both the CO2-to-CO conversion and H2formation rates reduced obviously.Interesting, when Co2P nanosheets were adopted, the system catalyzes CO and H2production, but with poor selectivity.These results strongly implied that the Ni2P nanosheets have more advantaged than FeP nanosheets and Co2P nanosheets in activating and driving the CO2 conversion.

To further demonstrate that the ultrathin layer structure of Ni2P could greatly improve the CO2reduction properties, linear sweep voltammetry (LSV) was performed in Ar- and CO2-saturated 0.1 mol·L-1KHCO3solutions using Ni2P particles and Ni2P nanosheet-decorated Ti foil as the working electrodes (Fig.6a).As expected, both the Ni2P particles and Ni2P nanosheets showed higher current densities under the CO2atmosphere than they did under the Ar atmosphere.In addition, the Ni2P nanosheets exhibited a higher current density and noticeably decreased onset potential compared to those of the Ni2P particles.These results implied that the Ni2P nanosheets can more efficiently decrease activation energy barrier of CO2reduction and drive CO2conversion compared to the Ni2P particles.

To investigate the intrinsic reason for the enhanced CO2activity of the Ni2P nanosheets, the electrochemically active surface area (ECSA) of the Ni2P nanosheets and Ni2P particles was measured by employing the double layer capacitance.As shown in Fig.6b and Fig.S7 (SI), the Ni2P nanosheets showed a largerCdlthan that of the Ni2P particles, demonstrating that the ultrathin layer structure endows the Ni2P nanosheets with more catalytic sites to activate CO2.To study the interface electron transfer of the Ni2P nanosheets during CO2 reduction, an electrochemical impedance experiment was carried out.Fig.6c shows that the Ni2P nanosheets have a lower interfacial transport resistance than the Ni2P particles.This could be attributed to the 2D transport paths, which endow the Ni2P nanosheets with much faster electron transport properties.In additional, the charge transfer efficiency between the [Ru(bpy)3]2+and Ni2P material was analyzed by room temperature photoluminescence (PL).The strong PL of [Ru(bpy)3]2+was quenched in the presence of Ni2P particles (Fig.6d), revealing that the Ni2P particles can facilitate charge transfer and suppress the recombination of charge carriers.Interestingly, Ni2P nanosheets can obvious improve the charge transfer and charge separation efficiency of [Ru(bpy)3]2+compared to that of the Ni2P particles, confirming that Ni2P nanosheets can efficiently promote charge transfer and separation of photogenerated charges.Overall, the enhanced photocatalytic CO2reduction performances of the ultrathin Ni2P nanosheets could be ascribed to their increased number of catalytic active centers, and their facilitation of charge transfer and separation of photogenerated charges.

Fig.5 (a) Formation of H2 and CO over TMPs.AFM images of the FeP nanosheets (b) and Co2P nanosheets (c).

Fig.6 (a) LSV curves of Ni2P particles and Ni2P nanosheets in Ar-saturated and CO2-saturated 0.1 mol·L-1 KHCO3 solutions.Scan rate of 20 mV·s-1.(b) Charging current density differences plotted against scan rates.(c) Nyquist plots of the Ni2P particles and Ni2P nanosheets recorded at -0.8 V vs an Ag/AgCl electode.(d) Room-temperature PL of the photocatalytic CO2 reduction systems with and without Ni2P(Ni2P particles or Ni2P nanosheets) as a catalyst under 500 nm light irradiation.

4 Conclusions

In summary, we have proposed an effective strategy to boost the CO2adsorption capability and CO2reduction performance of TMPs by synthesizing the ultrathin nanosheets.For example,ultrathin Ni2P nanosheets with an average thickness of 1.5 nm were synthesized by utilizing a salt surface as the growth templates.ECSA experiments demonstrated that the ultrathin layered structure endows the Ni2P nanosheets with more active sites relative to the Ni2P particles.In addition, the PL spectra and electrochemical impedance experiments revealed that the ultrathin thickness promoted charge transfer and suppressed the recombination of charge carriers.Importantly, electrochemical experiments demonstrated that Ni2P nanosheets can capture and activate CO2molecules more efficiently than Ni2P particles.As a result, the Ni2P nanosheets achieve a CO generation rate of 64.8 μmol·h-1, which is 4.4 times higher than that of Ni2P particles.In brief, our study paves the way for designing the nanostructured TMP catalysts for photocatalytic CO2reduction.

Supporting Information:available free of chargeviathe internet at http://www.whxb.pku.edu.cn.