==== Front
Nanomicro Lett
Nanomicro Lett
Nano-Micro Letters
2311-6706
2150-5551
Springer Nature Singapore Singapore

37386207
1132
10.1007/s40820-023-01132-3
Article
Efficient CO2 Reduction to Formate on CsPbI3 Nanocrystals Wrapped with Reduced Graphene Oxide
Hoang Minh Tam 12
Han Chen 3
Ma Zhipeng 3
Mao Xin 12
Yang Yang 12
Madani Sepideh Sadat 12
Shaw Paul 4
Yang Yongchao 3
Peng Lingyi 3
Toe Cui Ying 35
Pan Jian 3
Amal Rose 3
Du Aijun 12
Tesfamichael Tuquabo 12
Han Zhaojun zhaojun.han@unsw.edu.au

3
Wang Hongxia hx.wang@qut.edu.au

12
1 https://ror.org/03pnv4752 grid.1024.7 0000 0000 8915 0953 School of Chemistry and Physics, Faculty of Science, Queensland University of Technology, Brisbane, QLD 4001 Australia
2 https://ror.org/03pnv4752 grid.1024.7 0000 0000 8915 0953 Centre for Materials Science, Queensland University of Technology, Brisbane, QLD 4001 Australia
3 https://ror.org/03r8z3t63 grid.1005.4 0000 0004 4902 0432 School of Chemical Engineering, The University of New South Wales, Kensington, NSW 2052 Australia
4 https://ror.org/00rqy9422 grid.1003.2 0000 0000 9320 7537 Centre for Organic Photonics & Electronics (COPE), School of Chemistry and Molecular Biosciences, The University of Queensland, Brisbane, QLD 4072 Australia
5 https://ror.org/00eae9z71 grid.266842.c 0000 0000 8831 109X School of Engineering, The University of Newcastle, Callaghan, NSW 2038 Australia
29 6 2023
29 6 2023
12 2023
15 16124 3 2023
22 5 2023
© The Author(s) 2023
https://creativecommons.org/licenses/by/4.0/ Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
Highlights

A rational design of metal halide perovskites for achieving efficient CO2 reduction reaction was demonstrated.

The stability of CsPbI3 perovskite nanocrystal (NCs) in aqueous electrolyte was improved by compositing with reduced graphene oxide (rGO).

The CsPbI3/rGO catalyst exhibited > 92% Faradaic efficiency toward formate production with high current density which was associated with the synergistic effects between the CsPbI3 NCs and rGO.

Abstract

Transformation of greenhouse gas (CO2) into valuable chemicals and fuels is a promising route to address the global issues of climate change and the energy crisis. Metal halide perovskite catalysts have shown their potential in promoting CO2 reduction reaction (CO2RR), however, their low phase stability has limited their application perspective. Herein, we present a reduced graphene oxide (rGO) wrapped CsPbI3 perovskite nanocrystal (NC) CO2RR catalyst (CsPbI3/rGO), demonstrating enhanced stability in the aqueous electrolyte. The CsPbI3/rGO catalyst exhibited > 92% Faradaic efficiency toward formate production at a CO2RR current density of ~ 12.7 mA cm−2. Comprehensive characterizations revealed the superior performance of the CsPbI3/rGO catalyst originated from the synergistic effects between the CsPbI3 NCs and rGO, i.e., rGO stabilized the α-CsPbI3 phase and tuned the charge distribution, thus lowered the energy barrier for the protonation process and the formation of *HCOO intermediate, which resulted in high CO2RR selectivity toward formate. This work shows a promising strategy to rationally design robust metal halide perovskites for achieving efficient CO2RR toward valuable fuels.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40820-023-01132-3.

Keywords

Perovskite nanocrystal
Electrocatalyst
Inorganic perovskite
CO2 reduction
Formate production
Shanghai Jiao Tong UniversityOpen access funding provided by Shanghai Jiao Tong University.

issue-copyright-statement© Shanghai Jiao Tong University 2023
==== Body
pmcIntroduction

Metal halide perovskites (MHPs) are recognized as one of the most effective materials for solar-to-electrical energy conversion [1–3]. Recently, the interesting electrical properties of MHPs have inspired the exploration of their applications in catalytic reactions including CO2 reduction, water splitting, organic synthesis, and dye degradation [4–12]. The versatility of chemical tailoring and ion combination of the MHPs makes them suitable for many different catalytic reactions [13–15]. However, the application of MHPs for electrocatalysis remains challenging due to their low stability, particularly under electrocatalytic conditions [16, 17].

The catalytic property of MHPs was firstly demonstrated in 2016 by Park et al. They used methylammonium lead triiodide (MAPbI3) MHP for solar-driven electrochemical hydrohalic acid splitting for H2 production, which achieved a Faradaic efficiency (FE) of almost 100% [18], where the MAPbI3 was stabilized by utilizing the precipitation–solubility equilibrium between the perovskite phase and the soluble ionic species in aqueous electrolyte. This opens an appealing chapter of metal halide perovskites for catalytic applications [19]. Especially, in the field of electrochemical reduction of CO2, the inorganic MHP CsPbBr3 exhibited significant electrocatalytic activity with FE of 32% for CH4 and 40% for CO [20]. In addition, mixed halides CsPb(Br0.5Cl0.5)3 MHP catalyst boosted solar-driven electrochemical production of CO from CO2 with a yield of 875 μmol g−1 and selectivity up to 99% [21], while the MAPbI3 catalyst incorporated in iron-porphyrin derived metal–organic frameworks delivered a high total yield of 1559 μmol g−1 for photocatalytic CO2 reduction to CO (34%) and CH4 (66%) [22]. These studies have shown the application potential of MHPs in electrochemical CO2RR toward valuable fuels. Although there are potential environmental and health risks associated with the use of lead-based MHP catalysts, research of this type of material for CO2RR is still important fundamentally. Pb-based catalysts such as Pb nanoparticles or PtnPb nanocrystals have shown excellent effectivity and selectivity in CO2RR [23–25], however the study of the potential of Pb-based metal halide perovskites for CO2 reduction is very limited. Moreover, from economical point of view, the much lower material cost and the ease of scalability of Pb-based MHP materials by solution method also make them very attractive compared to the precious metal-based catalysts for CO2RR, such as platinum which have been widely studied for CO2RR application.

Nevertheless, significant research challenges for the development of MHPs CO2RR catalysts remain [26]. One daunting issue hindering their application is the instability of MHPs in the majority of common CO2RR electrolytes. MHPs are highly unstable in polar solvents, especially under catalytic reaction conditions where radical species are involved. Hence, low-polarity solvents, such as ethyl acetate, acetonitrile, or toluene were previously used as electrolyte solvent to preserve the phase and optoelectrical properties of MHPs catalysts [4, 11, 27]. However, the catalytic performance of MHPs in these electrolytes is unfavourable due to the low solubility of CO2 in these low polarity solvents [28]. Therefore, catalyst design strategies based on hybrid MHPs with different materials, i.e., graphene oxide [29], Pt nanoparticles [30], metal–organic frameworks [31, 32], have been used to enhance the stability of the MHPs. For example, the hybrid CsPbBr3 nanocrystals (NCs) with graphitic carbon nitride showed enhanced stability in both acetonitrile/water and ethyl acetate/water solvent systems, while exhibited enhanced photocatalytic CO production rate of 149 µmol h−1 g−1 [27]. However, the lifetime of MHP-based catalyst is still far from the requirement for a substantial electrolysis system [33]. It is thus critical to resolve the instability of MHPs in the catalytic reaction conditions. In addition, most of the efforts have been devoted into the study of CsPbBr3 catalysts with the primary reduction products of CH4 and CO; whereas electrochemical CO2 reduction toward other valuable chemicals has not been demonstrated.

Herein, we demonstrate an inorganic CsPbI3 perovskite NCs wrapped with reduced graphene oxide (rGO) for CO2RR in aqueous electrolyte, which exhibited increased stability compared with the bare CsPbI3 perovskite NCs. In addition, the CsPbI3/rGO catalyst exhibited outstanding activity (jCO2RR of 12.7 mA cm−2) and selectivity (FEformate of 92%), which is attributed to the accumulated electron density near the active CsPbI3 perovskite structure induced by the rGO wrapping, as evidenced by experimental characterization and density functional theory (DFT) modelling. Together with the unravelled CO2RR mechanism of CsPbI3/rGO catalyst, this work unleashes the potential of inorganic MHPs as robust and efficient CO2RR catalysts for sustainable fuels production.

Experimental Section

Materials

Chemicals including Cs2CO3 (99.95%, Sigma Aldrich), Oleic Acid (≥ 99%, Sigma Aldrich), Octadecene (Sigma Aldrich), PbI2 (99.99%, Sigma Aldrich), KHCO3, Oleylamine (Sigma Aldrich), reduced graphene oxide (Sigma Aldrich), Methyl acetate, Hexane was used as received without further purification.

Synthesis of Perovskite NCs

The synthesis procedures were carried out following a published method in literature with some modifications [34]. Cs-Oleate was prepared by adding 202.8 mg of Cs2CO3 and 10 mL Octadecene (ODE) into a 100 mL three-neck flask. The flask was then degassed and dried under vacuum at 120 °C for 30 min. Subsequently, 0.63 mL of Oleic Acid (OA) was also added into the solution and the system was degassed for another 30 min. After that, the mixture was heated to 160 °C under N2 gas and stirred until all the Cs2CO3 was dissolved to make a clear solution. The Cs-oleate solution is kept at 100 °C under N2 gas for further use.

In another three-neck flask, 0.3 mg of PbI2 and 20 mL ODE were loaded. The flask also was degassed and dried under vacuum for 1 h at 120 °C. A mixture of 1.5 mL OA and 1.5 mL Oleylamine (OLA) was added into the flask during this time. After PbI2 was completely dissolved, the temperature was raised to 170 °C and was kept stirring in N2 gas for 20 min. At this state, 1.5 mL of as-prepared Cs-Oleate (preheated at 100 °C) was quickly injected into the PbI2 solution. About 5 s later, the reaction was quenched in an ice bath. After the flask was cooled down to room temperature, 20 mL of methyl acetate was added to precipitate the NCs. The NCs were then collected by centrifugation at 10,000 rpm for 10 min. After that, the supernatant was discarded, and the collected NCs were dispersed in 10 mL of Hexane. The solution then goes through another centrifuging step at 8000 rpm for 5 min. The precipitate is collected and kept in vacuum for 6 h to remove the residual solvents. The final NCs were dispersed in Hexane for further measurement and applications.

Catalyst Preparation

The electrodes were prepared by mixing carbon black and catalysts with a mass ratio of 1 and painted on carbon fiber paper.

To specific, the electrode for the H-cell test was prepared through a drop-casting method. Firstly, 5 mg CsPbI3/rGO catalyst was dispersed in 1 mL hexane, and the catalyst ink was then mixed with 5 mg carbon black and ultrasonicated for 10 min. Finally, the obtained catalyst ink was drop-casted onto a carbon fiber paper (Toray Carbon Paper 090) and used as the electrode for the H-cell CO2 reduction test after drying in a vacuum desiccator for 5 h. For preparing the electrode for flow-cell test, 5 mg CsPbI3/rGO catalyst dispersed in 1 mL hexane, 5 mg Polytetrafluoroethylene (PTFE) powder (diameter of 1 μm) and 5 mg carbon black, then the obtained catalyst ink was sprayed onto a carbon fiber paper (AvCarb GDS2230) coated with PTFE and used as the electrode for flow-cell CO2 reduction test after drying in a vacuum desiccator for 5 h.

Electrochemical CO2 Reduction Experiments

The electrochemical CO2 reduction experiments in H-cell were conducted in a three-electrode H-Cell reactor filled with CO2-saturated 0.1 M KHCO3 solution (Sigma Aldrich, 99.7%). The synthesized CsPbI3/rGO electrode was used as the working electrode, while Pt foil and Ag/AgCl electrodes were used as counter and reference electrodes, respectively. The working electrode and counter electrode were separated by a Nafion 117 membrane (Dupont) and connected to Autolab potentiostat (PGSTAT204). Each sample was evaluated for 1.5 h before the gas phase products were collected for analysis using gas chromatography (GC-2010, Shimadzu). The liquid phase products were analyzed by nuclear magnetic resonance (Avance III 600 MHz NMR, Bruker) using a pre-saturation method for water suppression.

The electrochemical CO2 reduction experiments in flow cell were conducted in a two-electrode flow-cell reactor with continuously pumped-in 1.0 M KHCO3 solution (Sigma Aldrich, 99.7%) at flow rate of 3 mL min−1 in the cathode (CsPbI3/rGO) side with 1.0 M KOH (Sigma Aldrich, 85.0%) in the anode (Ni foam) side. The CO2 gas continuously flowed through the back side of the cathode at flow rate of 50 mL min−1 via a custom-designed gas gate. The working electrode and counter electrode were separated by a Sustainion® anion exchange membranes and connected to Autolab potentiostat (PGSTAT204) equipped with a BOOSTER10A module. The gas phase and liquid phase products were collected for analysis using gas chromatography (GC-2010, Shimadzu) and nuclear magnetic resonance (Avance III 600 MHz NMR, Bruker) using the same method as the above-mentioned for the H-cell CO2 reduction test. The electrochemical impedance spectroscopy (EIS) was tested in CO2-saturated 0.1 M KHCO3 aqueous solution under 0 V vs RHE with frequency range of 0.01–105 Hz.

Characterization

Characterization was conducted to reveal the structure, optical and electrical properties of the synthesized materials. UV–visible absorbance spectrum of the samples was measured by a UV–visible spectrometer (Carry 60), and the photoluminescence spectrum was recorded with a Cary Eclipse Fluorescence Spectrophotometer. The TEM images were captured by a JEOL 2100 microscope operated at 200 kV. STEM-EDS was obtained using an EDS detector coupled with JEOL 2100 TEM. A Kratos AXIS Supra photoelectron spectrometer (He I radiation, hν = 21.22 eV) was used to measure the XPS spectra and the UPS energy state of the materials. The crystal structure of the NCs were measured by X-ray diffraction equipment (Rigaku Smartlab) using a monochromatic CuKα (λ = 0.154 nm) as a radiation source. Fourier transform infrared (FTIR) spectrum was recorded by a Bruker model Alpha-P FTIR with ATR accessory. Time-resolved photoluminescence (TRPL) was measured by an Edinburgh fluorescence spectrometer at room temperature. The photo-excited source was provided by a laser with exciting wavelength of 474 nm and a pulse of 82.4 ps. In-situ and ex-situ Raman spectroscopy was performed with a Renishaw inVia Qontor spectrometer with exciting wavelength of 514 nm.

DFT Calculation

All the calculations for the optimization of CsPbI3 (001) plane and graphene-decorated CsPbI3 (001) were performed by using the density functional theory (DFT) method as implemented in the Vienna Ab initio Simulation Package (VASP). The Perdew–Burke–Ernzerhof (PBE) function of the generalized gradient approximation (GGA) was used for the calculation of electron exchange–correlation with the projector augmented wave (PAW) method [1–3]. Spin-polarization was also included through the calculations, and the cut-off energy of 500 eV for plain-wave basis sets was adopted. The convergence threshold was set to 10–5 eV, and 5 × 10–3 eV Å−1 for energy and force, respectively. The weak interaction was treated by DFT + D3 method using an empirical correction in Grimme’s scheme [4]. The reaction Gibbs free energy changes (ΔG) for each elementary steps were calculated using the computational hydrogen electrode model shown in the following equation:1 ΔG=ΔE+ΔZPE-TΔS+ΔGU+ΔGpH

where ΔE is obtained directly from DFT calculations, ΔZPE is the change of zero-point energies (ZPE), T is the temperature of 298.15 K, and ΔS is the change in entropy of products and reactants. ΔGU=-eU, which is the contribution of the electrode potential to ΔG, and ΔGpH=-kBTln10×pH.

Result and Discussion

Growth of CsPbI3 on rGO

CsPbI3 NCs were synthesized using a hot-injection synthetic method developed by Protesescu et al. [34]. To synthesize the CsPbI3/rGO composites, rGO was introduced into the PbI2 precursor solution before the injection of Cs-oleate precursor to initiate the growth of perovskite NCs. Previous studies have shown that perovskite materials have a strong binding affinity to rGO, leading to perovskite crystals readily growing and attaching onto the graphene surface [5, 29, 35]. Transmission electron microscopy (TEM) was used to reveal the structure of the NCs. Figure 1a shows the morphology of reference CsPbI3 NCs which are well dispersed cubic particles with an average size of ∼ 13 nm. The observed lattice spacing of ∼ 0.62 nm corresponds to (100) plane of cubic phase CsPbI3 perovskite as revealed in the high-resolution TEM image (inset of Fig. 1a). For the CsPbI3/rGO composite (Fig. 1b), the NCs are distributed on the rGO substrate with a slightly larger size and varying degrees of aggregations, implying that the presence of rGO can induce non uniform nucleation and growth of CsPbI3 NCs. Nevertheless, the CsPbI3 NCs grown on rGO still exhibit crystallinity of cubic phase perovskite. The existence and distribution of CsPbI3 perovskite on the rGO were also confirmed in the energy dispersive X-ray (EDX) mapping, which shows that Cs, Pb and I are well distributed on the rGO substrate (Fig. S1). In addition, the crystallinity of perovskite NCs was determined by X-ray diffraction (XRD). As seen in Fig. 1c, both CsPbI3 NCs and CsPbI3/rGO composite showed similar XRD patterns with characteristic peaks of cubic phase CsPbI3 perovskite (PDF No. 01-080-4039). The CsPbI3 NCs exhibited a preferential orientation at the (100) and (200) planes. In the CsPbI3/rGO composite, besides the peaks attributed to the cubic phase CsPbI3 perovskite, an additional peak at 23.2° was observed, which corresponds to the (002) plane of rGO [36].Fig. 1 TEM images reveal the morphology of a CsPbI3 NCs and b CsPbI3/rGO composite (insets are the high-resolution TEM showing lattice fringe of the CsPbI3 NCs). c XRD pattens of CsPbI3 and CsPbI3/rGO composite together with the standard cubic phase CsPbI3 (PDF 01-080-4039). d Raman spectra of rGO and CsPbI3/rGO

Since the surface properties of the nanocrystals catalyst are critical for the electrocatalytic reaction [13, 37], we characterized the chemical composition and chemical state of the synthesized CsPbI3/rGO composite by X-ray photoelectron spectroscopy (XPS). The XPS survey scan clearly identified the C 1s, N 1s, O 1s, Cs 3d, Pb 4f, and I 3d peaks (Fig. S2). Fitting of the C 1s spectra has shown peaks at binding energies of 285.0, 286.2, 286.8 and 289.1 eV, which can be assigned to C–C, C=C, C–O and C=O bonds, respectively [38]. The C 1s features should originate from the surface ligands (oleic acid, oleylamine) used in the NCs synthesis and rGO. High resolution XPS of Cs, Pb and I are shown in Fig. S3. The fitted Cs 3d (725.4 and 739.3 eV), Pb 4f (139.1 and 143.8 eV) and I 3d (619.8 and 631.4 eV) peaks are consistent with the characteristic binding energy of CsPbI3 NCs [39]. Compared to the XPS spectra of CsPbI3, it shows that the introduction of rGO slightly alters the surface property of the CsPbI3 NCs (Fig. S3) with a small XPS peak shift (up to 0.4 eV) in all Cs 3d, Pb 4f and I 3d peaks, implying a change in the distribution of electrons on the surface of CsPbI3. The origin and consequence of this change will be discussed later. The existence of surface ligands of the NCs was also confirmed by FTIR measurement (Fig. S4). The FTIR spectra show the vibration peaks of the C–C and C–H stretching modes of the CH2 group and the C=O and O–H characteristic vibrations of the carboxylate ligand. There is a small increase in the intensity of C=O and O–H vibrations in the CsPbI3/rGO composite, which is possibly due to the contribution of rGO. All the above results have confirmed that the cubic phase CsPbI3 NCs have successfully grown on the rGO layers.

Raman spectroscopy is a powerful method to characterize carbon materials [40]. The pristine rGO shows two distinguishable peaks at 1350 (D band) and 1586 (G band) cm−1 (Fig. 1d) with the intensity ratio of ID/IG ~ 1.16. The G band arises from the C–C stretching in the sp2-carbon networks and the D band originates from the presence of disorder or defects in the rGO [40, 41]. In the CsPbI3/rGO composite, the two characteristic peaks of D and G bands were also detected, validating the presence of rGO. Interestingly, the band ratio of ID/IG is reduced to 1.08 in the CsPbI3/rGO composite. This implies that the loaded CsPbI3 NCs possibly passivate the defects on the surface of rGO.

Optical Properties and Stability

Perovskite nanocrystals are remarkable in terms of their light emitting properties which is closely related to the crystal structure and electronic properties of the materials [42]. Thus, optical properties of the as-synthesized perovskite NCs were measured to investigate the structural and electronic properties of the materials. As can be seen in Fig. 2a inset, the CsPbI3 NCs solution exhibits strong red emission under UV light (365 nm) excitation. The absorption and photoluminescence characteristics of the as-synthesized CsPbI3 (Fig. 2a) show an absorption onset of ~ 720 nm and an emission peak at 682 nm with full-width-at-half-maximum (FWHM) of 35 nm. The calculated band gap is ~ 1.71 eV (Fig. S5), which is typical for the cubic phase CsPbI3 perovskite [43]. By using a reference dye Rhodamine 6G [44, 45], we determined the photoluminescence quantum yield (PLQY) of the NCs to be ~ 73% (Table S1). In contrast, the CsPbI3/rGO exhibits much lower photoluminescence intensity (Fig. 2a) with a PLQY of 51%. A small red-shift in the absorption onset and PL photoemission peak position (688 nm) are also observed with the composite. The red-shift of 6 nm (~ 15.9 meV) could originate from the aggregation of NCs on the rGO sheets as observed in the TEM measurement. The lower PLQY implies that there is significant PL quenching in the material. This is confirmed by the time-resolved PL (TRPL) measurements (Fig. 2b). It shows that the presence of rGO has shortened the PL lifetime of the perovskite NCs. The pristine CsPbI3 NCs exhibit multi-exponential decays with an average lifetime of 62.1 ns, whereas the CsPbI3/rGO composite shows a 17% reduction of the average PL lifetime to 51.5 ns. Detailed calculation of the TRPL data is shown in Table S2. Apparently, addition of rGO has created an efficient charge transfer channel in the CsPbI3 NCs, which suppresses the radiative recombination of the excited excitons. Measurement of the energy band alignment of CsPbI3 and rGO by ultraviolet photoelectron spectroscopy (UPS) (Fig. S6a) has confirmed the feasibility for charge transfer between the two materials. The CsPbI3 NCs has a valence band energy edge at − 5.6 eV (vs vacuum) while rGO has a work function at − 5.2 eV, allowing an efficient hole transfer from CsPbI3 to rGO at the interface between the two materials (Fig. S6b). Fig. 2 a PL and absorption spectra of CsPbI3 NCs and CsPBI3/rGO composite (insets are the photos of two solutions in hexane under UV excitation); b TR-PL decay spectra of CsPbI3 NCs and CsPBI3/rGO composite; c Plots of the change in PL intensity of CsPbI3 NCs and CsPBI3/rGO solutions during storage in ambient condition; d Photos showing the stability of CsPbI3/rGO thin film immersed in water for different durations. The film is exhibiting red emission under UV excitation. (Color figure online)

Stability is a major problem for CsPbI3 NCs. The pristine NCs inks turned from dark brown to a light-yellow with yellow precipitate after 20 days storage in ambient condition (relative humidity 35% ~ 60%) (Fig. S7). This is resulted from the loss of surface ligands, formation of halide vacancy, and eventually the transformation of CsPbI3 NCs to PbI2 as reported previously [46, 47]. In contrast, the ink of CsPbI3/rGO still maintained dark brown colour even after 5 months of storage under the same condition (Fig. S7). The stability of both NCs was further assessed by the PL measurement as shown in Fig. 2c. A dramatical loss in PL intensity was observed in the pristine CsPbI3 NCs ink during storage, and the PL intensity dropped to zero in 20 days, whereas the CsPbI3/rGO composite ink still maintained more than 90% of its initial PL intensity after 2 months of storage. The XRD measurement shows that the CsPbI3/rGO composite largely retained its cubic phase after 2 months storage, with only a small portion of orthorhombic CsPbI3 detected (Fig. S8). We further tested the stability of the composite in water. Figure 2d shows the film of CsPbI/rGO composite still exhibited strong light emissions after being immersed in water for 3 h. The water contact angle measurement on the thin film of CsPbI3 and CsPbI3/rGO shows that the CsPbI3/rGO surface is more hydrophobic with a larger contact angle of 81° comparing to CsPbI3 film which shows a contact angle of approximate 73° (Fig. S9). Clearly, the rGO sheets provide significant protection to the CsPbI3 against aqueous environment.

Catalytic CO2 Reduction Performance

To explore the CO2RR performance of the CsPbI3/rGO, the CO2RR experiments over CsPbI3 and CsPbI3/rGO were performed in a typical electrochemical H-cell configuration. The CO2RR experiments were conducted at different applied potentials ranging from − 1.05 to − 1.45 V vs reversible hydrogen electrode (RHE). The FE toward CO2 reduction products is shown in Fig. 3a. It shows that the CsPbI3/rGO catalysts has a higher FECO2RR under a more negative potential. The FE reaches to more than 90% at − 1.45 V vs RHE. Moreover, the CsPbI3/rGO catalysts exhibit a higher selectivity towards formate compared with that of the pristine CsPbI3 catalyst. It should be noted that formate is considered a promising hydrogen energy carrier because of its high volumetric hydrogen storage capacity, low toxicity and low flammability, thus efficient production of formate is desirable in practice [48]. A higher content of H2 was observed with the pristine CsPbI3 catalyst, which suggests that the interaction between the rGO and the CsPbI3 NCs suppresses the HER process. The gas phase products (i.e., CO and H2) were quantified by gas chromatography (GC) according to a calibration curve (Fig. S10). The excellent R2 value (R2 > 0.998) of the calibrate curve ensures the accuracy of the experimental results. The total current density and partial current density of formate are shown in Fig. 3b. The CsPbI3/rGO catalyst exhibits a total current density of 12.7 mA cm−2, in which the formate partial current density accounts for 11.7 mA cm−2. The electrochemical impedance spectroscopy (EIS) of the samples (Fig. S11) were measured to investigate the charge transfer resistance (Rct) of the catalysts in the presence of the CO2 saturated 0.1 M KHCO3 aqueous solution. The Nyquist plot of EIS was fitted according to the equivalent circuit shown in Inset of Fig. S11. Detail of the fitting results of the EIS spectrum is shown in Table S3. The much lower Rct of the CsPbI3/rGO composite indicates that the introduction of rGO into CsPbI3 has significant benefit for the electrochemical reaction of CO2RR.Fig. 3 Electrochemical CO2 reduction performance of CsPbI3 and CsPbI3/rGO in CO2-saturated 0.1 M KHCO3. a FE of main products under different applied potentials. b Total CO2RR current density (solid symbols) and partial current density toward formate (dash symbols) under different applied potentials. c I–t curves and FEformate in stability test under − 1.45 VRHE

The CO2RR performance of the CsPbI3/rGO catalyst was further evaluated in a 2-electrode flow-cell system. We achieved ~ 81.0 mA partial current toward CO2RR products over a 3 cm2 electrode. This is more than twofold of that in the H-cell system (Fig. S12). The CO2RR performance of the rGO was tested under the same condition (Fig. S13). However, no CO2RR product was detected with the bare rGO. Instead, H2 was the main product with FE ~ 100%. This implies that CsPbI3 is the main active material for the CO2 reduction in the composite.

We further tested the long-time stability of the catalysts at − 1.45 VRHE for 10.5 h continuously. As shown in Fig. 3c, the CsPbI3 alone showed a significant deterioration on the selectivity to formate, with an FEformate of only 52.5% at the end of test. In contrast, the CsPbI3/rGO maintained a FEformate of 76.4% after 10.5 h continuous stability test. This confirms the improved CO2RR stability of the CsPbI3/rGO catalyst compared with bare CsPbI3. Moreover, the performance of our CsPbI3/rGO catalyst is competitive among other perovskite-based and metal-based catalysts for CO2 reduction reported in the recent literature such as sulfide-derived Pb or SnO2 porous nanowires catalyst (Table S4) [49, 50].

To understand this phenomenon, we have analysed the morphology and composition of the fresh and the spent samples. As shown in Fig. 4a, the CsPbI3 without rGO appear in the form of nanoparticles with well-defined crystalline structure dispersed in a carbon matrix. After the first cycle of EC CO2RR reaction under − 1.45 VRHE, the NCs of CsPbI3 without rGO transformed to Pb aggregations (Fig. 4b), as evidenced by the TEM EDX mapping (Fig. 4c). In case of CsPbI3 with rGO, the CsPbI3/rGO catalyst also exhibit the morphology of well-crystalline particles dispersed in a carbon matrix (Fig. 4d), but this morphology remained intact after the same EC CO2RR condition (Fig. 4e). From the EDS measurement (Fig. 4f), it can be seen that the chemical composition of CsPbI3 NCs in the CsPbI3/rGO was well maintained. The obtained atomic percentages of C, Cs, Pb and I were 98.4%, 0.34%, 0.34% and 0.92%, respectively, corresponding to a Cs/Pb/I ratio of approximately 1/1/2.7, which is close to that of pristine CsPbI3 perovskite phase. After the long-time stability test, despite of the emergence of nanoparticles, the CsPbI3/rGO still preserved the original crystal phase and chemical compounds (Fig. S14), as opposed to the pristine CsPbI3 catalyst which completely changed to Pb particles (Fig. S15). It is noted that the aggregation of CsPbI3 NCs in the CsPbI3/rGO composite led to the formation of nanorods after the long reaction time, which is expected to reduce the surface area of the catalyst and adversely affect the overall electrocatalytic performance.Fig. 4 a TEM images of fresh electrodes CsPbI3; b TEM image of CsPbI3 after one EC CO2RR run under − 1.45 VRHE and c the corresponding EDS elemental mapping image; d TEM image of fresh CsPbI3/rGO electrode; e TEM of CsPbI3/rGO electrode after one EC CO2RR run under − 1.45 VRHE and f the corresponding EDS elemental mapping of composite for Cs (green), Pb (red), I (yellow) elements. (Color figure online)

The structural stability of the samples was further investigated by in-situ Raman spectroscopy, which is a powerful method to monitor phase transformation of materials. The pristine CsPbI3 (α-CsPbI3) exhibited an indiscernible Raman peak (Fig. 5a), which is consistent with the previous study that α-CsPbI3 only shows Raman features at a low frequency region [51]. After being aged in the ambient condition, an obvious vibration mode appeared at 108 cm−1, which matches the characteristic fingerprint of δ-CsPbI3, implying that the perovskite undergoes a phase transformation from α-CsPbI3 phase to δ-CsPbI3 phase [51]. The in-situ Raman spectroscopy was used to track phase transition during the EC reaction of CsPbI3/rGO catalyst. As shown in Fig. 5b, there is no change in the Raman spectra of the CsPbI3/rGO catalyst under − 1.45 VRHE EC CO2RR conditions in the 1200 s testing period, indicating that the phase transformation to δ-CsPbI3 did not occur and the CsPbI3 in CsPbI3/rGO remained in the α phase.Fig. 5 a Raman spectra of pristine CsPbI3 (black) and aged CsPbI3 (red); b In-situ Raman spectra of CsPbI3/rGO under EC CO2RR conditions with applied potential of − 1.45 VRHE; The DFT models of α-CsPbI3 c and δ-CsPbI3 d with rGO supported and the corresponding calculated binding energy. e XPS Pb 4f spectra of fresh CsPbI3 and CsPbI3/rGO electrodes. g Electron transfer between α-CsPbI3 and rGO from DFT calculations, where yellow area means charge accumulation, and cyan area means charge depletion. (Color figure online)

We carried out density functional theory (DFT) calculations of binding energy of rGO with α- and δ-CsPbI3 to understand the improved stability of CsPbI3/rGO. As shown in Fig. 5c, d, the binding energy of rGO with α-CsPbI3 (black phase) is − 1.533 eV, which is significantly lower than that of rGO with δ-CsPbI3 (yellow phase) (− 0.894 eV), suggesting that the CsPbI3 NCs formed on rGO sheets preferably maintains the α-CsPbI3 phase thermodynamically. This also explains the enhanced stability of the α-CsPbI3 in the CsPbI3/rGO composite compared to the pristine CsPbI3 NCs. In addition, comparing the XPS spectra of fresh CsPbI3 and CsPbI3/rGO electrodes (Figs. 5e and S16), we have observed a distinct shift toward lower-binding energy in the XPS spectra of Pb 4f, Cs 3d and I 3d of CsPbI3/rGO. This indicates an increased electron density in the CsPbI3 material in the presence of rGO. The observation agrees well with the DFT calculation results (Fig. 5f), which show electron transfer between α-CsPbI3 and rGO, and the electron tends to accumulate around the α-CsPbI3. The unique charge distribution in CsPbI3/rGO material plays an important role in the EC CO2RR performance of the material as discussed below.

Reaction Pathway

The above results show that the integration of rGO to CsPbI3 significantly improves the performance, selectivity, and stability of α-CsPbI3 phase in EC CO2 reduction. To understand the catalysis mechanism, we calculated the activation energy of adsorbed species formed in the CO2 reduction reaction on CsPbI3 (100) facet with and without rGO, since this is the dominant phase in our material as shown in the XRD (Fig. 1c). Previous theoretical investigations have suggested that Cs cannot be the active site because Cs–C bonding is less favoured than Cs–O bonding [20]. In contrast, the Pb–C bonding is favourable to form. Therefore, Pb is selected as the active site in the DFT calculations. The activation energy of adsorbed species on Pb site was calculated on models of α-CsPbI3 with and without rGO supported as illustrated in Fig. 6a, b. The resulting free-energy diagrams are shown in Fig. 6c, d, which include two possible pathways for CO2 reduction, i.e., (i) reduction of CO2 to the COOH intermediate toward CO and (ii) reduction of CO2 to the *HCOO intermediate toward HCOOH. The formation of *HCOO intermediate is thermodynamically favourable according to the simulations. This is consistent with the experimental result of high selectivity toward HCOO− product rather than CO product. Interestingly, when CsPbI3 is anchored on rGO, the activation energy barrier for *HCOO intermediate formation is reduced from 0.92 to 0.68 eV. It means the CsPbI3/rGO composite can further facilitate the reaction toward the formate production. This is attributed to the electron accumulation formed around Pb atoms in the presence of rGO (Fig. 5f), which strengthens the Pb–C bonding and the adsorption of *H species, followed by the acceleration of the protonation of CO2 for formate formation. It should be noted that the HER is not preferred in both material systems due to the high energy barrier according to our DFT simulation (Fig. S17).Fig. 6 Models of a CsPbI3 and b CsPbI3/rGO composite with HCOOH species in the DFT calculations; c, d Free-energy diagrams showing the CO and HCOOH formation pathways in the CO2 reduction reaction when c CsPbI3 and d CsPbI3/rGO catalyst were used

Additionally, we have observed the formation of Pb metal in the spent CsPbI3 catalyst (Fig. 4b). It suggests that the CsPbI3 particles slowly lost Pb atoms during the CO2 reduction reaction. Therefore, a model of CsPbI3 with Pb vacancy was also built to simulate the CO2 reduction reactions. The calculation shows that with Pb vacancy in the CsPbI3 crystal, the energy barrier for *HCOO intermediate formation increases to 1.31 eV (Fig. S18). This explains the decrease of formate selectivity during the long-term stability test in CsPbI3 catalyst. Our results have demonstrated an effective strategy to prevent the phase degradation or Pb defect formation in CsPbI3, which is critical for maintaining the superior CO2RR performance and formate selectivity.

Conclusion

In summary, we have devised an efficient CsPbI3/rGO composite catalyst for CO2RR in aqueous electrolyte. Compared to the pristine CsPbI3 NCs, the α-CsPbI3 perovskite structural degradation in CsPbI3/rGO composite is thermodynamically mitigated, hindering the formation of less-active Pb defects or even Pb particles. Furthermore, by redistributing electrons near the CsPbI3/rGO interface, the rGO lowers the energy barrier for the formation of *HCOO intermediates and protonation process in the CsPbI3/rGO catalyst, thus promoting the CO2RR pathway toward formate product with a high selectivity (> 90% FEformate) at applied voltage of − 1.45 V vs RHE. The study has revealed the perspective application of metal halide perovskite material in electrochemical CO2RR while paves the way for the development of selective, stable, and active new catalysts.

Supplementary Information

Below is the link to the electronic supplementary material.Supplementary file1 (PDF 1461 KB)

Acknowledgements

H.W acknowledges the financial support by Australian Research Council (ARC). M.T. H. thank Queensland University of Technology (QUT) postgraduate scholarship. The data of SEM, TEM, fluorescence, XRD and Raman reported in this paper were obtained at the Central Analytical Research facility (CARF), QUT and the Mark wainwright Analytic Centre (MWAC), UNSW Sydney. Access to CARF was supported by the generous funding from Science and Engineering faculty, QUT. C. H. and Z. M. acknowledge the facilities and the scientific and technical assistant of Microscopy Australia at the Electron Microscope Unit (EMU) within the MWAC, UNSW Sydney.

Funding

Open access funding provided by Shanghai Jiao Tong University.

Declarations

Conflict of interest

The authors declare no interest conflict. They have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Minh Tam Hoang, Chen Han, Zhipeng Ma, and Xin Mao equally contributed to this work.
==== Refs
References

1. NREL, Best Research-Cell Efficiencies, https://www.nrel.gov/pv/assets/pdfs/best-research-cell-efficiencies.20190802.pdf
2. Yoo JJ Seo G Chua MR Park TG Lu Y Efficient perovskite solar cells via improved carrier management Nature 2021 590 587 593 10.1038/s41586-021-03285-w 33627807
3. Jeong J Kim M Seo J Lu H Ahlawat P Pseudo-halide anion engineering for α-FAPbI3 perovskite solar cells Nature 2021 592 381 385 10.1038/s41586-021-03406-5 33820983
4. Hou J Cao S Wu Y Gao Z Liang F Inorganic colloidal perovskite quantum dots for robust solar CO2 reduction Eur. J. Chem. 2017 23 9481 9485 10.1002/chem.201702237
5. Wu Y Wang P Zhu X Zhang Q Wang Z Composite of CH3NH3PbI3 with reduced graphene oxide as a highly efficient and stable visible-light photocatalyst for hydrogen evolution in aqueous HI solution Adv. Mater. 2018 30 1704342 10.1002/adma.201704342
6. Mollick S Mandal TN Jana A Fajal S Desai AV Ultrastable luminescent hybrid bromide perovskite@ mof nanocomposites for the degradation of organic pollutants in water ACS Appl. Nano Mater. 2019 2 1333 1340 10.1021/acsanm.8b02214
7. Méndez-Galván M Alcántar-Vázquez B Diaz G Ibarra IA Lara-García HA Metal halide perovskites as an emergent catalyst for CO2 photoreduction: a minireview React. Chem. Eng. 2021 6 828 838 10.1039/D1RE00039J
8. Zhu X Lin Y San Martin J Sun Y Zhu D Lead halide perovskites for photocatalytic organic synthesis Nat. Commun. 2019 10 2843 10.1038/s41467-019-10634-x 31253792
9. Huang H Yuan H Janssen KP Solís-Fernández G Wang Y Efficient and selective photocatalytic oxidation of benzylic alcohols with hybrid organic–inorganic perovskite materials ACS Energy Lett. 2018 3 755 759 10.1021/acsenergylett.8b00131
10. Zhu X Lin Y Sun Y Beard MC Yan Y Lead-halide perovskites for photocatalytic α-alkylation of aldehydes J. Am. Chem. Soc. 2019 141 733 738 10.1021/jacs.8b08720 30602111
11. Chen K Deng X Dodekatos G Tüysüz H Photocatalytic polymerization of 3,4-ethylenedioxythiophene over cesium lead iodide perovskite quantum dots J. Am. Chem. Soc. 2017 139 12267 12273 10.1021/jacs.7b06413 28787135
12. Wu Y Wu Q Zhang Q Lou Z Liu K An organometal halide perovskite supported Pt single-atom photocatalyst for H2 evolution Energy Environ. Sci. 2022 15 1271 1281 10.1039/D1EE03679C
13. Hwang J Rao RR Giordano L Katayama Y Yu Y Perovskites in catalysis and electrocatalysis Science 2017 358 751 756 10.1126/science.aam7092 29123062
14. X. Hou, J. Ren, F. Li, C. Ma, X. Zhang et al., Research progress of perovskite materials as catalysts, in IOP Conference Series: Earth and Environmental Science (2019), p. 032020.
15. Kuhl KP Hatsukade T Cave ER Abram DN Kibsgaard J Electrocatalytic conversion of carbon dioxide to methane and methanol on transition metal surfaces J. Am. Chem. Soc. 2014 136 14107 14113 10.1021/ja505791r 25259478
16. Adinolfi V Peng W Walters G Bakr OM Sargent EH The electrical and optical properties of organometal halide perovskites relevant to optoelectronic performance Adv. Mater. 2018 30 1700764 10.1002/adma.201700764
17. Xiang W Liu SF Tress W A review on the stability of inorganic metal halide perovskites: challenges and opportunities for stable solar cells Energy Environ. Sci. 2021 14 2090 2113 10.1039/D1EE00157D
18. Park S Chang WJ Lee CW Park S Ahn H-Y Photocatalytic hydrogen generation from hydriodic acid using methylammonium lead iodide in dynamic equilibrium with aqueous solution Nat. Energy 2017 2 16185 10.1038/nenergy.2016.185
19. Luo J Zhang W Yang H Fan Q Xiong F Halide perovskite composites for photocatalysis: a mini review EcoMat 2021 3 e12079 10.1002/eom2.12079
20. Chen K Qi K Zhou T Yang T Zhang Y Water-dispersible CsPbBr3 perovskite nanocrystals with ultra-stability and its application in electrochemical CO2 reduction Nano-Micro Lett. 2021 13 172 10.1007/s40820-021-00690-8
21. Guo S-H Zhou J Zhao X Sun C-Y You S-Q Enhanced CO2 photoreduction via tuning halides in perovskites J. Catal. 2019 369 201 208 10.1016/j.jcat.2018.11.004
22. Wu LY Mu YF Guo XX Zhang W Zhang ZM Encapsulating perovskite quantum dots in iron-based metal–organic frameworks (MOFs) for efficient photocatalytic CO2 reduction Angew. Chem. Int. Ed. 2019 58 9491 9495 10.1002/anie.201904537
23. Zhao CX Bu YF Gao W Jiang Q CO2 reduction mechanism on the Pb(111) surface: effect of solvent and cations J. Phys. Chem. C 2017 121 19767 19773 10.1021/acs.jpcc.7b04375
24. García J Jiménez C Martínez F Camarillo R Rincón J Electrochemical reduction of CO2 using Pb catalysts synthesized in supercritical medium J. Catal. 2018 367 72 80 10.1016/j.jcat.2018.08.017
25. Wu X Jiang Y Yan Y Li X Luo S Tuning surface structure of Pd3Pb/PtnPb nanocrystals for boosting the methanol oxidation reaction Adv. Sci. 2019 6 1902249 10.1002/advs.201902249
26. Gong E Ali S Hiragond CB Kim HS Powar NS Solar fuels: research and development strategies to accelerate photocatalytic CO2 conversion into hydrocarbon fuels Energy Environ. Sci. 2022 15 880 937 10.1039/D1EE02714J
27. Ou M Tu W Yin S Xing W Wu S Amino-assisted anchoring of CsPbBr3 perovskite quantum dots on porous g-C3N4 for enhanced photocatalytic CO2 reduction Angew. Chem. Int. Ed. 2018 57 13570 13574 10.1002/anie.201808930
28. Miller MB Chen D-L Luebke DR Johnson JK Enick RM Critical assessment of CO2 solubility in volatile solvents at 298.15 K J. Chem. Eng. Data 2011 56 1565 1572 10.1021/je101161d
29. Xu Y-F Yang M-Z Chen B-X Wang X-D Chen H-Y A CsPbBr3 perovskite quantum dot/graphene oxide composite for photocatalytic CO2 reduction J. Am. Chem. Soc. 2017 139 5660 5663 10.1021/jacs.7b00489 28385017
30. Chen Y-X Xu Y-F Wang X-D Chen H-Y Kuang D-B Solvent selection and Pt decoration towards enhanced photocatalytic CO2 reduction over CsPbBr3 perovskite single crystals Sustain. Energy Fuels 2020 4 2249 2255 10.1039/C9SE01218D
31. Kong Z-C Liao J-F Dong Y-J Xu Y-F Chen H-Y Core@ Shell CsPbBr3 @ Zeolitic imidazolate framework nanocomposite for efficient photocatalytic CO2 reduction ACS Energy Lett. 2018 3 2656 2662 10.1021/acsenergylett.8b01658
32. Cheng R Debroye E Hofkens J Roeffaers MB Efficient photocatalytic CO2 reduction with MIL-100 (Fe)-CsPbBr3 composites Catalysts 2020 10 1352 10.3390/catal10111352
33. Burdyny T Smith WA CO2 reduction on gas-diffusion electrodes and why catalytic performance must be assessed at commercially-relevant conditions Energy Environ. Sci. 2019 12 1442 1453 10.1039/C8EE03134G
34. Protesescu L Yakunin S Bodnarchuk MI Krieg F Caputo R Nanocrystals of cesium lead halide perovskites (CsPbX3, X = Cl, Br, and I): novel optoelectronic materials showing bright emission with wide color gamut Nano Lett. 2015 15 3692 3696 10.1021/nl5048779 25633588
35. Zhang Q Nan H Zhou Y Gu Y Tai M In situ growth of α-CsPbI3 perovskite nanocrystals on the surface of reduced graphene oxide with enhanced stability and carrier transport quality J. Mater. Chem. C 2019 7 6795 6804 10.1039/C9TC01012B
36. Stobinski L Lesiak B Malolepszy A Mazurkiewicz M Mierzwa B Graphene oxide and reduced graphene oxide studied by the XRD, TEM and electron spectroscopy methods J. Electron Spectrosc. Relat. Phenom. 2014 195 145 154 10.1016/j.elspec.2014.07.003
37. Raciti D Wang C Recent advances in CO2 reduction electrocatalysis on copper ACS Energy Lett. 2018 3 1545 1556 10.1021/acsenergylett.8b00553
38. Li S Li Y Liu K Chen M Peng W Laser fabricated carbon quantum dots in anti-solvent for highly efficient carbon-based perovskite solar cells J. Colloid Interface Sci. 2021 600 691 700 10.1016/j.jcis.2021.05.034 34049024
39. Li X Wu Y Zhang S Cai B Gu Y CsPbX3 quantum dots for lighting and displays: room-temperature synthesis, photoluminescence superiorities, underlying origins and white light-emitting diodes Adv. Funct. Mater. 2016 26 2435 2445 10.1002/adfm.201600109
40. E.I. Bîru, H. Iovu, Graphene nanocomposites studied by Raman spectroscopy. J. Raman Spectrosc. 179 (2018). 10.5772/intechopen.73487
41. A. Kaniyoor, S. Ramaprabhu, A Raman spectroscopic investigation of graphite oxide derived graphene. AIP Adv. 2, 032183 (2012). 10.1063/1.4756995
42. Kovalenko MV Protesescu L Bodnarchuk MI Properties and potential optoelectronic applications of lead halide perovskite nanocrystals Science 2017 358 745 750 10.1126/science.aam7093 29123061
43. S. Xu, A. Libanori, G. Luo, J. Chen, Engineering bandgap of CsPbI3 over 1.7 eV with enhanced stability and transport properties. Iscience 24, 102235 (2021). 10.1016/j.isci.2021.102235
44. Crosby GA Demas JN Measurement of photoluminescence quantum yields. Review J. Phys. Chem. 1971 75 991 1024 10.1021/j100678a001
45. Magde D Wong R Seybold PG Fluorescence quantum yields and their relation to lifetimes of rhodamine 6G and fluorescein in nine solvents: Improved absolute standards for quantum yields Photochem. Photobiol. 2002 75 327 334 10.1562/0031-8655(2002)075<0327:FQYATR>2.0.CO;2 12003120
46. Wei Y Cheng Z Lin J An overview on enhancing the stability of lead halide perovskite quantum dots and their applications in phosphor-converted LEDs Chem. Soc. Rev. 2019 48 310 350 10.1039/C8CS00740C 30465675
47. Pols M Vicent-Luna JM Filot I van Duin AC Tao S Atomistic insights into the degradation of halide perovskites: a reactive force field molecular dynamics study J. Phys. Chem. Lett. 2021 12 5519 5525 10.1021/acs.jpclett.1c01192 34096726
48. Chatterjee S Dutta I Lum Y Lai Z Huang K-W Enabling storage and utilization of low-carbon electricity: power to formic acid Energy Environ. Sci. 2021 14 1194 1246 10.1039/D0EE03011B
49. Pander JE Lum JWJ Yeo BS The importance of morphology on the activity of lead cathodes for the reduction of carbon dioxide to formate J. Mater. Chem. A 2019 7 4093 4101 10.1039/C8TA10752A
50. Kumar B Atla V Brian JP Kumari S Nguyen TQ Reduced SnO2 porous nanowires with a high density of grain boundaries as catalysts for efficient electrochemical CO2-into-HCOOH conversion Angew. Chem. Int. Ed. 2017 56 3645 3649 10.1002/anie.201612194
51. Yang Y Robbins JP Ezeonu L Ma Y Sparta N Probing lattice vibrations of stabilized CsPbI3 polymorphs via low-frequency Raman spectroscopy J. Mater. Chem. C 2020 8 8896 8903 10.1039/D0TC02123G