Stabilizing indium sulfide for CO2 electroreduction to formate at high rate by zinc incorporation | Nature Communications

Nature Communications volume 12, Article number: 5835 (2021) Cite this article

16k Accesses

121 Citations

31 Altmetric

Metrics details

An Author Correction to this article was published on 15 October 2021

This article has been updated

Recently developed solid-state catalysts can mediate carbon dioxide (CO2) electroreduction to valuable products at high rates and selectivities. However, under commercially relevant current densities of > 200 milliamperes per square centimeter (mA cm−2), catalysts often undergo particle agglomeration, active-phase change, and/or element dissolution, making the long-term operational stability a considerable challenge. Here we report an indium sulfide catalyst that is stabilized by adding zinc in the structure and shows dramatically improved stability. The obtained ZnIn2S4 catalyst can reduce CO2 to formate with 99.3% Faradaic efficiency at 300 mA cm−2 over 60 h of continuous operation without decay. By contrast, similarly synthesized indium sulfide without zinc participation deteriorates quickly under the same conditions. Combining experimental and theoretical studies, we unveil that the introduction of zinc largely enhances the covalency of In-S bonds, which “locks” sulfur—a catalytic site that can activate H2O to react with CO2, yielding HCOO* intermediates—from being dissolved during high-rate electrolysis.

Electrosynthesis of value-added fuels using carbon dioxide (CO2) as a feedstock provides an appealing route to reduce climate-changing CO2 emission and a solid stepping-stone towards carbon neutrality1,2. Over the past decade, the development of catalysts that are active and selective for CO2 reduction reaction (CO2RR), and meanwhile, suppress the competing hydrogen evolution, has been the subject of intensive study. This has resulted in a variety of carbon-based products to be synthesized from CO2, such as carbon monoxide (CO)3, formate (HCOO−)4, methane5, and higher hydrocarbons and oxygenates (e.g., ethylene6, ethanol7, and n-propanol8). Despite remarkable advances, recent techno-economic analyses showed that CO and formate are probably the only products that can achieve the industrialization trend of CO2RR in the near future9,10. Regarding the formate, its profitable production requires current densities of ≥ 200 mA cm−2, Faradic efficiency (FE) of > 90%, and power conversion efficiencies of > 50% (refs. 9,10). Further, adopting solid-electrolyte electrolyzers permits the continuous production of formic acid without the separation process, making it even more economically viable4.

Early research on CO2RR from Hori and co-workers revealed that a number of metals, such as lead, mercury, indium (In), bismuth (Bi), cadmium, and tin, could convert CO2 to formate, but many of these metals suffer from unsatisfactory selectivity or toxic issue11. Improvements in efficiency and selectivity have been achieved on nontoxic metallic catalysts via controlling of catalyst morphologies and dimensionalities12,13, creation of vacancies14, and introduction of other elements (e.g., O, S, and P) to form new phases15,16,17,18. When immobilized on the gas diffusion electrodes that surmount CO2 mass transport limitation, commercially relevant rates (> 200 mA cm−2) and FE (> 90%) were observed to be reached on Bi nanosheets19, Bi2O3 nanotubes20, Bi metal-organic layers21, and InP quantum dots17. However, the prospect on the potential of these catalysts for long-term operational stability is elusive. At high current densities, catalyst stability perhaps becomes a very important challenge. Often, CO2RR activity deteriorates rapidly during high-rate electrolysis, owing to reasons like catalysis agglomeration22, active-phase change12,23, and element dissolution15,24. Unfortunately, previous research effort on catalyst stability, especially working at commercially relevant current densities, has remained rather rare. To make renewably powered formate electrosynthesis from CO2 to be practical, it is critically necessary to develop catalysts that are not only active but also stable, and to gain insights on mechanisms of mediating the intrinsic stability.

Here, we report that incorporation of zinc (Zn) into indium sulfide (In2S3) synthesis enables tuning over its phase and structure, which dramatically improves the long-term stability of the resultant catalyst (ZnIn2S4) although the catalyst morphology remains almost unchanged. Comprehensive experiments coupled with computational studies reveal an enhanced covalency of In−S bonds mediated by Zn, which overstabilizes sulfur—a catalytic site that can activate H2O to react with CO2, leading to the formation of HCOO* intermediates—in the catalyst structure. Consequently, we achieved nearly 100% CO2-to-formate conversion at a current density of 300 mA cm−2 over 60 h without degradation, corresponding to a high production rate of 8,894 μmol cm−2 h−1.

We had an interest in indium sulfide as a catalyst because S-doped In was shown by Wang and co-workers to be effective for catalyzing CO2RR to formate. The presence of S enables facile activation of H2O to form adsorbed H*, which consequently reacts with absorbed CO2 to yield HCOO* intermediates16. However, the stability of S-doped In was only assessed under ~60 mA cm−2 during a 10 h period; the prospect of such catalyst for durable high-rate CO2-to-formate conversion is unclear. Very recently, Xia et al. reported that exfoliated ultrathin ZnIn2S4 nanosheets with rich Zn vacancies show improved CO2RR ability to formate14. Although interesting, its long-term stability at current densities relevant to commercial operation (> 200 mA cm−2) was not evaluated. These results motivated us to examine the ability of indium sulfide instead of S-doped In for mediating CO2 to formate. We synthesized indium sulfide hydrothermally by the reaction of InCl3·4H2O and C2H5NS in deionized water (DIW) at 160 °C (Supplementary Fig. 1). Cubic In2S3 (JCPDS 65-0459; Fig. 1i) was produced after 6 h, exhibiting flower-like morphology composed of hierarchically organized nanosheets (Supplementary Fig. 2). Indeed, we observed good formate selectivity on In2S3, but the performance degraded quickly at high current densities owing to the dissolution of S2− ions (discussion later).

a, b SEM images of the ZnIn2S4 catalyst. The right panel in b shows the crystal structure of ZnIn2S4. Scale bars, 5 μm (a) and 1 μm (b). c STEM-EDX elemental mapping of ZnIn2S4, exhibiting a uniform spatial distribution of Zn (red), In (green), and S (yellow), respectively. Scale bar, 1 μm. d, e Atomic-resolution Z-contrast images of ZnIn2S4 along [001] zone axis. Scale bars, 1 nm (d) and 0.5 nm (e). f The corresponding FFT pattern of (d). g The line intensity profile acquired along the yellow arrow in (d). h Atomic model of ZnIn2S4 along [001] zone axis. i–k XRD patterns (i), UPS spectra (j), and BET surface area analysis (k) of ZnIn2S4 and In2S3, respectively.

Previous experimental studies revealed that adding Zn in some transition metal chalcogenides (e.g., Co3S4)25 can enhance the structure robustness. Thus we sought to improve the stability of high-rate CO2RR by incorporating Zn into indium sulfide. We used the same hydrothermal method for preparing the desired product except the addition of ZnCl2 during the synthesis (Supplementary Fig. 1). Intriguingly, we obtained hexagonal-structured ZnIn2S4 (JCPDS 65-2023; Fig. 1i) microflowers that consist of hierarchically organized nanosheets (Fig. 1a, b), which closely resemble In2S3 described above. The thicknesses of nanosheets were determined to be ~8.69 nm for ZnIn2S4 and 9.32 nm for In2S3 through atomic force microscopy (AFM) measurements (Supplementary Fig. 3). We note that the synthesis of flower-like ZnIn2S4 was previously reported26,27,28, whereas the analogous morphologies of ZnIn2S4 and In2S3 that synthesized by the same protocol here will underpin a fair performance comparison. Energy-dispersive X-ray (EDX) spectrum elemental mapping exhibits a uniform spatial distribution of Zn, In, and S (Fig. 1c). This simple synthetic strategy enables the production of high-yield ZnIn2S4 material with good fidelity for potential large-scale adoption (Supplementary Fig. 4).

We studied the detailed atomic structure of the ZnIn2S4 by high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM). The atomic-resolution Z-contrast images in Fig. 1d, e clearly reveal hexagonal lattice, where In atoms exhibit higher image intensity than the overlapped Zn and S atoms (Supplementary Fig. 5). The fast Fourier transform (FFT) result exhibits the (100) and (010) reflections (Fig. 1f). Using image contrast, the In and Zn(S) atoms can be further identified by the line intensity profile (Fig. 1g) acquired along the yellow arrow in Fig. 1d. The corresponding atomic model depicts that all the overlapped Zn and S atoms are located at the centers of honeycomb (Fig. 1h). Without the addition of Zn leads to the crystallization of cubic In2S3 by the same synthetic protocol (Fig. 1i and Supplementary Fig. 2). Structurally, ZnIn2S4 belongs to (ZnS)mIn2S3 (m = 1−3) system29, which bears an orderly alternation of S and Zn(In) (Fig. 1b, right). The sequence of atoms along the [001] direction is S−ZnT−S−InO−S−InT−S, where ZnT and InT occupy the tetrahedral coordination and InO occupies the octahedral site, respectively29,30. By comparison, one-third of the tetrahedral sites in In2S3 is unoccupied31. The incorporation of Zn alters the coordination environment of indium sulfide and thus might tailor favorably the electronic structure and catalytic properties.

To probe the electronic structures of ZnIn2S4 and In2S3, we measured the work function by ultraviolet photoelectron spectroscopy (UPS) (Fig. 1j). Our results show a lower work function of ZnIn2S4 (5.12 eV) compared to In2S3 (5.29 eV), revealing a superior electronic property by the incorporation of Zn element, consistent with electrochemical impedance spectroscopy (EIS) results (Supplementary Fig. 6). We speculate, on the basis of the above results, that CO2RR may be highly favored on such ternary In-based sulfide owing to the modulated coordination environment and electronic structure. Moreover, we determined the Brunauer−Emmett−Teller (BET) surface areas of ZnIn2S4 and In2S3 to be 71.3 and 70.0 m3 g−1 (Fig. 1k), respectively.

We examined CO2RR properties of ZnIn2S4 and In2S3 catalysts in a flow cell (Supplementary Fig. 7) using recirculated 1 M KHCO3 (pH 8.4) as electrolyte. CO2 gas was fed at the cathode with a flow rate of 24 mL min−1; the outlet gas flow rate was also measured for accurate product analysis (see “Methods”; Supplementary Fig. 8). We quantified the solution-phase and gas-phase products by using nuclear magnetic resonance (NMR) spectroscopy and on-line gas chromatography (Supplementary Fig. 9), respectively. The linear sweep voltammetry curves in Fig. 2a show sharp reduction peaks for ZnIn2S4 and In2S3 catalysts in a CO2 environment. In a N2 environment, however, the two catalysts exhibit a slight current−voltage response. In comparison with In2S3, the onset potential for CO2RR on ZnIn2S4 catalyst shifted to a more positive value, implying enhanced CO2RR kinetics (Fig. 2a). Figure 2b shows that the Faradaic efficiency (FE) for formate on ZnIn2S4 catalyst was always greater than on In2S3 at all potentials examined (Supplementary Figs. 10, 11). Notably, the ZnIn2S4 catalyst yields peak FE of 99.3% for formate at −1.18 V versus a reversible hydrogen electrode (RHE), while the competing hydrogen evolution reaction (HER) on this catalyst was substantially suppressed (Fig. 2b and Supplementary Fig. 12). With this FE, we achieved a CO2RR to formate partial current density of ~298 mA cm−2 (Fig. 2c), representing the highest value reported to date under KHCO3 environments (Fig. 2e). We also performed reference measurements of hexagonal ZnS (JCPDS 39-1363) that synthesized by the identical route for comparison, which, however, overwhelmingly produces H2 (Supplementary Fig. 13). Additionally, our series of control experiments disclosed that the optimum CO2RR performance was gained on ZnIn2S4 catalyst that hydrothermally synthesized at 160 °C for 6 h with a ZnCl2:InCl3·4H2O ratio of 1:2 (Supplementary Figs. 14–19).

a, b The linear sweep voltammetry curves (a) and potential-dependent Faradaic efficiencies for products (b) on ZnIn2S4 and In2S3. c, d Partial current density (c) and half-cell PCE (d) for CO2-to-formate conversion on ZnIn2S4 and In2S3. e, f Comparison of formate partial current densities and FEs (e), and formate production rates (f) for various catalysts reported under KHCO3 environments (see Supplementary Table 1 for details). g Stability test of the ZnIn2S4 and In2S3 at 300 mA cm−2. The electrolyte was occasionally replaced by new 1 M KHCO3 solution (red arrows) to recover the ionic concentration and conductivity of the anolyte. The error bars represent the standard deviation of three independent measurements.

Figure 2d presents the half-cell power conversion efficiency (PCE) for CO2-to-formate conversion under various applied potentials. At −1.18 V versus RHE, our full-cell (CO2 + H2O → formate + O2) device shows a half-cell formate PCE exceeding 50% on ZnIn2S4 catalyst. A comprehensive review of recent literature revealed that our ZnIn2S4 catalyst exhibits superb selectivity and partial current density (Fig. 2e), which result in a formate production rate of up to 8,894 μmol cm−2 h−1, outperforming all previous results4,14,16,19,20,21,24,32,33,34,35,36,37,38,39,40,41,42,43 that have been reported under KHCO3 environments (Fig. 2f).

We used density functional theory (DFT) to obtain insights into the CO2RR properties of the studied catalyst (see “Methods” for details). We compared the Gibbs free energies (△G) for the formation of formate intermediate (HCOO*) on the surfaces of ZnIn2S4, In2S3, and In models (Supplementary Figs. 20–22). The computed barrier of HCOO* formation is 117 meV for ZnIn2S4 and 120 meV for In2S3, smaller than that of 270 meV for In, implying that S sites favorably mediate the HCOO* formation. Our calculations further reveal the lowest barrier of HCOOH* formation on ZnIn2S4, leading to its superior CO2-to-formate ability. These results indicate that formate preferentially generates on ZnIn2S4 catalyst. By contrast, *COOH (intermediate of CO product) formation is highly endergonic on the three In-based catalysts (Supplementary Figs. 20–22), causing the production of CO to be virtually prohibited. Although HER process is largely hampered, our DFT results reveal that, as compared to In2S3 and In, the S sites of ZnIn2S4 enable much smaller hydrogen adsorption free energy of 370 meV. Early works15,16 have reported that S acts as a promotor to enhance CO2-to-formate conversion, we thus reasonably surmise that S sites on ZnIn2S4 surface permit easier H2O dissociation to adsorbed H* species, which then react with CO2 to yield HCOO* intermediates.

Aside from activity, long-term stability—especially operating at high current densities (> 200 mA cm−2)—is another critical metric for CO2 electrolysis technique to be practical10. Figure 2g shows the key finding that we wish to report in this work: that is, the CO2RR stability of indium sulfide can be remarkably improved by the incorporation of Zn. We tested the stability of ZnIn2S4 catalyst at a profitable current density as large as 300 mA cm−2, during this process portions of the electrolyte were frequently taken out for NMR analysis. The formate FE could be held at > 97% over 60 h of continuous electrolysis without the need of additional overpotentials (Fig. 2g). By contrast, In2S3 reference exhibited a rapid drop in formate selectivity, whereas the FE toward H2 climbed up to ~90% within 8 h. We hypothesize that such severe performance drop might be caused by the structure degradation during high-rate electrolysis. Notably, the exceptional stability of ZnIn2S4 catalyst enables us to produce ~327 mmol formate after 60 h (Fig. 2g).

We combined multiple characterization techniques to track the structural evolution of ZnIn2S4 and In2S3 catalysts during CO2 electrolysis under various current densities and operating times (Fig. 3). X-ray diffraction (XRD) patterns and scanning electron microscopy (SEM) studies reveal that the phase and morphology of ZnIn2S4 catalyst were well retained when progressively increasing the current density even up to 500 mA cm−2 (Fig. 3a, c). By contrast, In2S3 catalyst undergoes a complete phase transition to metallic In (JCPDS 65-9292; Fig. 3b) at a current density of mere 50 mA cm−2, accompanied by a dramatic morphology change (Fig. 3c, below images) owing to the loss of S that leads to structure collapse. Our Raman spectroscopy measurements on ZnIn2S4 show that two characteristic peaks at 248 (LO1 mode) and 340 cm−1 (LO2 mode)44,45 were retained after 60 h of operation at 300 mA cm−2 (Fig. 3d). However, the characteristic Raman modes (A1g and Eg)46 of In2S3 disappeared while Raman signals from metallic In (ref. 47) were detected within mere 1 min (Fig. 3e and Supplementary Figs. 23a, 24). The Raman results are consistent well with our post-mortem SEM analyses (Supplementary Fig. 25) and XRD results (Supplementary Fig. 23b).

a, b XRD patterns of ZnIn2S4 (a) and In2S3 (b) after CO2 /Subscript> electrolysis under various current densities for 10 min. c Corresponding SEM images of ZnIn2S4 (above) and In2S3 (bottom). Scale bars, 1 μm (above) and 500 nm (bottom). d–g Raman spectra of ZnIn2S4 (d) and In2S3 (e), and S 2p XPS spectra of ZnIn2S4 (f) and In2S3 (g) after CO2 electrolysis for various times at 300 mA cm−2. h STEM-EDX elemental mappings of ZnIn2S4 (scale bar: 1 μm) and In2S3 (scale bar: 600 nm) after running for 60 h and 8 h at 300 mA cm−2, respectively. i SEM-EDX measurements of the remained sulfur in catalysts after running for various times at 300 mA cm−2. The error bars represent the standard deviation of three independent measurements. j TEM (above, scale bars: 50 nm) and SAED patterns (down, scale bars: 5 1/nm) of ZnIn2S4 catalyst after CO2 electrolysis for various times at 300 mA cm−2.

Of note that the severe loss of S for In2S3 catalyst was further verified by X-ray photoelectron spectroscopy (XPS; Fig. 3g), STEM-EDX elemental mapping (Fig. 3h) and SEM-EDX (Fig. 3i). This is starkly contrasted with ZnIn2S4 whose chemical state and content of each elements (i.e., Zn, In, and S) were nearly unaltered after 60 h of high-rate CO2 electrolysis (Fig. 3f, h and Supplementary Figs. 26, 27). We quantified the amount of S remained in ZnIn2S4 and In2S3 catalysts by using SEM-EDX (Supplementary Figs. 28–30), which permits a quantitative compositional analysis at a relatively large scale. As shown in Fig. 3i, the amount of S in In2S3 drops to 2.13 wt% from its original value (23.6 wt%) within the first 1 h, followed by a slow drop to almost zero over the next 2 h. This result is consistent with the significantly increased S amount in an electrolyte that measured by inductively coupled plasma atomic emission spectroscopy (ICP-AES; Supplementary Fig. 31). By contrast, the ZnIn2S4 catalyst shows negligible loss of S after 60 h. Moreover, selected-area electron diffraction (SAED) analyses of the used ZnIn2S4 catalyst reveal that the single-crystalline hexagonal phase well maintains after the aggressive long-term stability test (Fig. 3j). We further note that ZnS catalyst also performs very stable at high current densities (Supplementary Fig. 32) owing to the strong interaction between Zn and S (refs. 16,48,49), although it mainly produces H2 (Supplementary Fig. 13).

Our results above conclusively demonstrate that the stability degradation of In2S3 can be attributable to S leaching, and show the primacy of Zn as a stabilizer in indium sulfide that prevents S to be leached out. Besides indium sulfide, S leaching was also widely observed in other metal sulfides, while the dissolution mechanism is rather complex6,18,50,51. We turned to use DFT calculations to study the cause of the enhanced stability of indium sulfide after incorporating Zn. Compared with In2S3 having tetrahedral vacancies31, the tetrahedral and octahedral sites in ZnIn2S4 are fully occupied30 after Zn incorporation. Notably, in ZnIn2S4, all Zn atoms bind with S through tetrahedral coordination, which implies the formation of strong ZnT−S bonds considering that tetrahedral structures commonly give covalent feature52. Our computed differential charge density and its projection on the (110) plane map clearly reveal an enhanced electron cloud between Zn and S atoms (Fig. 4a, b), revealing electron donation from Zn to S due to the strong reducibility of Zn atoms. The transfer of electrons from In to S can also be seen more pronounced in ZnIn2S4 (Fig. 4a, b) than that in In2S3 (Fig. 4d, e), which leads to charge accumulation around In−S bonds and correspondingly higher covalency53.

a, b Differential charge density (a) and projection on the (110) plane (b). c ELF of ZnIn2S4. d, e Differential charge density (d) and projection on the (011) plane (e). f ELF of In2S3. The azure and yellow clouds represent electron density depressions and accumulations, respectively. g–i COHPs for In−S bonding (g) and Zn−S bonding (h) of ZnIn2S4, as well as In−S bonding (i) of In2S3.

The calculated electronic localization function (ELF) of the tetrahedral InT−S and octahedral InO−S bonds in ZnIn2S4 are 0.84 and 0.79 (Fig. 4c), which compare larger than that of 0.71 and 0.76 in In2S3 (Fig. 4f), indicating a greater localization of S−InO−S−InT−S (ref. 54). Likewise, the ELF of tetrahedral ZnT−S bond in ZnIn2S4 was calculated to be 0.81 (Fig. 4c), pointing to its localized covalent feature. The interatomic bond strengths were further quantitatively analyzed by the projected crystal orbital Hamilton population (pCOHP) method (Fig. 4g−i). We found that the anti-bonding orbitals of In−S and Zn−S for ZnIn2S4 are less occupied. Moreover, our calculations yield integrated pCOHP values below the Fermi level of −0.763 and −0.737 for In−S bonds in ZnIn2S4 and In2S3, respectively, again demonstrating greater bond strengths in ZnIn2S4 (ref. 55). These results, therefore, indicate that the bond breaking between In(Zn) and S in ZnIn2S4 is kinetically cumbersome, which explains the negligible S dissolution and thus exceptional long-term stability of the ZnIn2S4 catalyst (Supplementary Fig. 33).

In closing, we have shown long-term formate electrosynthesis from CO2 at a high current density of 300 mA cm−2 on a cost-effective indium sulfide catalyst modulated by Zn. The extraordinary catalyst stability can be explained by the increase of In−S covalency, which substantially prevents sulfur dissolution during CO2RR. We achieved selective and fast CO2-to-formate conversion with a formate FE of 99.3% and a notable formate production rate of 8,894 μmol cm−2 h−1. These findings will advance the development of efficient and durable catalysts for commercial-scale electrosynthesis of formate.

All chemicals were used as received without further purification. Indium chloride tetrahydrate (InCl3·4H2O), thioacetamide (C2H5NS), and Zinc dichloride (ZnCl2), were purchased from Sinopharm Chemical Reagent Co., Ltd (Shanghai, China). In a typical experiment, 3.0 mmol InCl3·4H2O was dissolved in 150 mL deionized water (DIW), and then 6.1 mmol C2H5NS was added with vigorous stirring for 20 min. The 20 mL resultant solution was transferred into a 50 mL Teflon-lined autoclave, sealed, and heated at 160 °C for 6 h. After the reaction, the obtained In2S3 powders were washed with excess DIW and absolute ethanol for at least three times, and then dried at room temperature in an oven under vacuum for further characterization. The synthesis of ZnIn2S4 was the same with the synthesis of In2S3, except the addition of 1.5 mmol ZnCl2 during the first step. For the synthesis of ZnS, it needs to replace 3.0 mmol InCl3·4H2O with 1.5 mmol ZnCl2.

XRD was performed on a Japan Rigaku DMax-γA X-ray diffractometer with Cu Kα radiation (λ = 1.54178 Å). The morphology of the samples was investigated by SEM (Zersss Supra 40) and TEM (JEOL 2010F(s)). The STEM and HRTEM images, SAED and EDX elemental mapping were taken on JEMARM 200 F Atomic Resolution Analytical Microscope with an acceleration voltage of 200 kV. SEM-EDX was determined by GeminiSEM 500 with an Oxford Aztec series X-ray energy spectrum. Raman spectra was measured on a Raman microscope (HORIBA) with a 785 nm excitation laser. ICP-AES data were obtained by an Optima 7300 DV instrument. N2 adsorption/desorption isotherms were recorded on an ASAP 2020 accelerated surface area and a porosimetry instrument (Mictromeritics), equipped with an automated surface area, at 77 K by using Barrett−Emmett−Teller calculations. XPS was taken on an X-ray photoelectron spectrometer (ESCALab MKII) with an X-ray source (Mg Kα hυ = 1253.6 eV).

The catalyst ink was prepared by ultrasonic dispersion of 10 mg catalyst powders in 1 ml isopropanol, which was mixed with 50 μL of 5 wt% Nafion. The resulted ink was uniformly spread on the gas diffusion layer (GDL, Sigracet 29 BC) of 3 × 3 cm2 in the area by using an airbrush, yielding the prepared electrode with a catalyst loading of ~1.0 mg cm−2.

All electrochemical measurements were performed in a flow cell with VSP-300 Potentiostat (Bio-Logic, France). For experiments in flow cells, gaseous CO2 (99.999%) was passed through the gas chamber at the back side of the gas diffusion electrodes. Both catholyte and anolyte (1 M KHCO3) were continuously circulated through the cathode and anode chambers separated by the cation exchange membrane (Nafion™ 117), which was used to avoid the crossover issues of formate56. The cathode is the prepared gas diffusion electrode (GDE, 1 × 1 cm2), and the anode is a piece of nickel foam (1 × 1 cm2). The CO2 inlet flowrates were kept constant at 24 mL min−1 by a mass flow controller (C100L, Sierra). KHCO3 electrolyte flowrates were maintained constant at 20 mL min−1 controlled by a peristaltic pump (BT100-2J, Longer Pump). The CO2 electrolysis lasted for 10 min unless otherwise specified. The linear sweep voltammetry (LSV) curves of ZnIn2S4 and In2S3 were performed in CO2-fed and Ar-fed 1 M KHCO3 solution. All potentials were measured against an Ag/AgCl (saturated KCl) reference electrode and converted to the RHE reference scale with iR correction on account of the equation:

Where the solution resistance Rs was determined by EIS over a frequency range from 100 KHz to 10 mHz.

The gas products were analyzed by gas chromatography (GC-2014, Shimadzu) equipped with thermal conductivity detector (TCD) to quantify H2 concentration and flame ionization detector (FID) to analyze the content of CO. Considering CO2 consumption, the outlet flow rate was monitored by a mass flowmeter (AST10-HLC, Asert Instruments) before flowing to the on-line GC. The Faradaic efficiency for gas products (FEx) was calculated by the following formula:

where F is the Faraday constant (96485 C mol−1), I is the total current density, nx is electrons transferred for reduction to product x, Cx is volume fraction of the product x detected by GC, u is outlet gas flowrate and VM is molar volume (22.4 L mol−1).

The formate products were quantified by 1H NMR spectra measured with a Bruker 400 MHz spectrometer. Typically, 400 µL of the electrolyte after CO2RR electrolysis was mixed with 200 µL of D2O containing 50 ppm (m/m) dimethyl sulphoxide (DMSO) as the internal standard. The area ratio of the formate peak to the DMSO peak was compared to the standard curve to quantify the concentration of formate. The molar quantity of formate (nformate) was calculated via multiplying the concentration of formate with the volume of the catholyte. The Faradaic efficiency of the formate (FEformate) can be calculated by the following equation:

where t is the CO2 electrolysis time.

The half cell (cathodic) power conversion efficiency (PCE, assuming the overpotential of the oxygen evolution reaction is zero) of the formate products was calculated using:

where E is the applied potential vs RHE, Eformate is thermodynamic potential (−0.02 V vs RHE) of CO2RR to formate57.

The production rate for formate was calculated using the following equation:

where Q is the total charge passed and S is the geometric area of the electrode (1 cm2).

The DFT calculations were performed by Vienna ab initio simulation package (VASP)58 program with projector augmented wave (PAW)59 method and the kinetic energy cut off was set to be 500 eV. The convergence criterion for electronic self-consistent iteration was set to be 10−4 eV. The atomic positions were fully relaxed until the force on each atom is less than 0.02 eV Å−1. The Perdew−Burke−Ernzerhof60 generalized gradient approximation exchange-correlation functional was used throughout. The slab model of In (101), In2S3 (311), and ZnIn2S4 (102) surface were constructed from the optimized In, In2S3, and ZnIn2S4 crystal structure. At the same time, a vacuum layer of 15 Å is established in the c-axis direction to ensure the separation between slabs. In addition, the surface formate species takes a unit negative charge, and, the present DFT calculation is not so great as to describe this kind of system carrying a neat charge. Thus, HCOOH was considered as the final product to describe this reaction rather than formate, in line with the DFT calculation in many researches18,61,62. The COHPs were computed using the developed lobster program63,64,65.

Here, The Gibbs free energies were calculated at 25 °C and 1 atm:

where ΔEads, ΔZPE, T, ΔS, U, and e are the binding energy, zero-point energies changes, temperature, entropy changes, applied potential at the electrode, and charge transferred, respectively.

All experimental data within the article and its Supplementary Information are available from the corresponding author upon reasonable request.

A Correction to this paper has been published: https://doi.org/10.1038/s41467-021-26430-5

Montoya, J. H. et al. Materials for solar fuels and chemicals. Nat. Mater. 16, 70–81 (2017).

Article ADS CAS Google Scholar

Mistry, H., Varela, A. S., Kühl, S., Strasser, P. & Cuenya, B. R. Nanostructured electrocatalysts with tunable activity and selectivity. Nat. Rev. Mater. 1, 16009 (2016).

Article ADS CAS Google Scholar

Gao, F.-Y. et al. High-curvature transition-metal chalcogenide nanostructures with a pronounced proximity effect enable fast and selective CO2 electroreduction. Angew. Chem. Int. Ed. 59, 8706–8712 (2020).

Article CAS Google Scholar

Fan, L., Xia, C., Zhu, P., Lu, Y. & Wang, H. Electrochemical CO2 reduction to high-concentration pure formic acid solutions in an all-solid-state reactor. Nat. Commun. 11, 3633 (2020).

Article ADS CAS PubMed PubMed Central Google Scholar

Chen, Z. et al. Grain-boundary-rich copper for efficient solar-driven electrochemical CO2 reduction to ethylene and ethanol. J. Am. Chem. Soc. 142, 6878–6883 (2020).

Article CAS PubMed Google Scholar

Zhuang, T.-T. et al. Steering post-C–C coupling selectivity enables high efficiency electroreduction of carbon dioxide to multi-carbon alcohols. Nat. Catal. 1, 421–428 (2018).

Article CAS Google Scholar

Niu, Z.-Z. et al. Hierarchical copper with inherent hydrophobicity mitigates electrode flooding for high-rate CO2 electroreduction to multicarbon products. J. Am. Chem. Soc. 143, 8011–8021 (2021).

Article CAS PubMed Google Scholar

Jiang, K. et al. Metal ion cycling of Cu foil for selective C–C coupling in electrochemical CO2 reduction. Nat. Catal. 1, 111–119 (2018).

Article CAS Google Scholar

Jouny, M., Luc, W. & Jiao, F. General techno-economic analysis of CO2 electrolysis systems. Ind. Eng. Chem. Res. 57, 2165–2177 (2018).

Article CAS Google Scholar

Bushuyev, O. S. et al. What should we make with CO2 and how can we make it? Joule 2, 825–832 (2018).

Article CAS Google Scholar

Hori, Y., Wakebe, H., Tsukamoto, T. & Koga, O. Electrocatalytic process of CO selectivity in electrochemical reduction of CO2 at metal electrodes in aqueous media. Electrochim. Acta 39, 1833–1839 (1994).

Article CAS Google Scholar

Ye, K. et al. In situ reconstruction of a hierarchical Sn−Cu/SnOx core/shell catalyst for high-performance CO2 electroreduction. Angew. Chem. Int. Ed. 59, 4814–4821 (2020).

Article CAS Google Scholar

Han, N. et al. Ultrathin bismuth nanosheets from in situ topotactic transformation for selective electrocatalytic CO2 reduction to formate. Nat. Commun. 9, 1320 (2018).

Article ADS PubMed PubMed Central CAS Google Scholar

Wang, Z. et al. Exfoliated ultrathin ZnIn2S4 nanosheets with abundant zinc vacancies for enhanced CO2 electroreduction to formate. ChemSusChem 14, 852–859 (2020).

Article CAS Google Scholar

Zhou, J.-H. et al. Boosting electrochemical reduction of CO2 at a low overpotential by amorphous Ag−Bi−S−O decorated Bi0 nanocrystals. Angew. Chem. Int. Ed. 58, 14197–14201 (2019).

Article CAS Google Scholar

Ma, W. et al. Promoting electrocatalytic CO2 reduction to formate via sulfur-boosting water activation on indium surfaces. Nat. Commun. 10, 892 (2019).

Article ADS PubMed PubMed Central CAS Google Scholar

Grigioni, I. et al. CO2 electroreduction to formate at a partial current density of 930 mA cm−2 with InP colloidal quantum dot derived catalysts. ACS Energy Lett. 6, 79–84 (2020).

Article CAS Google Scholar

Zheng, X. et al. Sulfur-modulated tin sites enable highly selective electrochemical reduction of CO2 to formate. Joule 1, 794–805 (2017).

Article CAS Google Scholar

Yang, J. et al. Bi-based metal-organic framework derived leafy bismuth nanosheets for carbon dioxide electroreduction. Adv. Energy Mater. 10, 2001709 (2020).

Article CAS Google Scholar

Gong, Q. et al. Structural defects on converted bismuth oxide nanotubes enable highly active electrocatalysis of carbon dioxide reduction. Nat. Commun. 10, 2807 (2019).

Article ADS PubMed PubMed Central CAS Google Scholar

Cao, C. et al. Metal-organic layers leading to atomically thin bismuthene for efficient carbon dioxide electroreduction to liquid fuel. Angew. Chem. Int. Ed. 59, 15014–15020 (2020).

Article CAS Google Scholar

Wu, J., Sun, S.-G. & Zhou, X.-D. Origin of the performance degradation and implementation of stable tin electrodes for the conversion of CO2 to fuels. Nano Energy 27, 225–229 (2016).

Article CAS Google Scholar

Shi, Y. et al. Unveiling hydrocerussite as an electrochemically stable active phase for efficient carbon dioxide electroreduction to formate. Nat. Commun. 11, 3415 (2020).

Article ADS CAS PubMed PubMed Central Google Scholar

Cheng, H. et al. Surface nitrogen-injection engineering for high formation rate of CO2 reduction to formate. Nano Lett. 20, 6097–6103 (2020).

Article ADS CAS PubMed Google Scholar

Huang, Z.-F. et al. Hollow cobalt-based bimetallic sulfide polyhedra for efficient all-pH-value electrochemical and photocatalytic hydrogen evolution. J. Am. Chem. Soc. 138, 1359–1365 (2016).

Article CAS PubMed Google Scholar

Chai, B., Peng, T., Zeng, P., Zhang, X. & Liu, X. Template-free hydrothermal synthesis of ZnIn2S4 floriated microsphere as an efficient photocatalyst for H2 production under visible-light irradiation. J. Phys. Chem. C. 115, 6149–6155 (2011).

Article CAS Google Scholar

Chen, Z. et al. Photocatalytic degradation of dyes by ZnIn2S4 microspheres under visible light irradiation. J. Phys. Chem. C. 113, 4433–4440 (2009).

Article ADS CAS Google Scholar

Gou, X. et al. Shape-controlled synthesis of ternary chalcogenide ZnIn2S4 and CuIn(S,Se)2 nano-/microstructures via facile solution route. J. Am. Chem. Soc. 128, 7222–7229 (2006).

Article CAS PubMed Google Scholar

Otsuka, R., Seo, W. S. & Koumoto, K. Thermoelectric properties of layer-structured (ZnS)mIn2S3. Key Eng. Mater. 181−182, 59–62 (2000).

Article Google Scholar

Aymerich, F., Meloni, F. & Mula, G. Electron band structure of α-ZnIn2S4 and related polytypes. Solid State Commun. 29, 235–238 (1979).

Article ADS CAS Google Scholar

Fu, X. et al. Photocatalytic performance of tetragonal and cubic β-In2S3 for the water splitting under visible light irradiation. Appl. Catal. B 95, 393–399 (2010).

Article CAS Google Scholar

Ma, W. et al. Monoclinic Scheelite Bismuth Vanadate derived Bismuthene nanosheets with rapid kinetics for electrochemically reducing carbon dioxide to formate. Adv. Funct. Mater. 31, 2006704 (2021).

Article CAS Google Scholar

García de Arquer, F. P. et al. 2D metal oxyhalide-derived catalysts for efficient CO2 electroreduction. Adv. Mater. 30, 1802858 (2018).

Article CAS Google Scholar

Xing, Y. et al. Bi@Sn core–shell structure with compressive strain boosts the electroreduction of CO2 into formic acid. Adv. Sci. 7, 1902989 (2020).

Article CAS Google Scholar

Liang, C. et al. High efficiency electrochemical reduction of CO2 beyond the two-electron transfer pathway on grain boundary rich ultra-small SnO2 nanoparticles. J. Mater. Chem. A 6, 10313–10319 (2018).

Article CAS Google Scholar

Shang, H. et al. Design of a single-atom indiumδ+-N4 interface for efficient electroreduction of CO2 to formate. Angew. Chem. Int. Ed. 59, 22465–22469 (2020).

Article CAS Google Scholar

Kwon, I. S. et al. Selective electrochemical reduction of carbon dioxide to formic acid using indium–zinc bimetallic nanocrystals. J. Mater. Chem. A 7, 22879–22883 (2019).

Article CAS Google Scholar

Zhang, X. et al. Defects and conductive nitrogen−carbon framework regulated ZnInOx nanosheets for boosting CO2 electrocatalytic reduction. Appl. Catal. B 279, 119383 (2020).

Article CAS Google Scholar

Luo, W., Xie, W., Li, M., Zhang, J. & Züttel, A. 3D hierarchical porous indium catalyst for highly efficient electroreduction of CO2. J. Mater. Chem. A 7, 4505–4515 (2019).

Article CAS Google Scholar

Zhang, A. et al. Harmonizing the electronic structures of the adsorbate and catalysts for efficient CO2 reduction. Nano Lett. 19, 6547–6553 (2019).

Article ADS CAS PubMed Google Scholar

Wang, J. et al. Heterostructured intermetallic CuSn catalysts: high performance towards the electrochemical reduction of CO2 to formate. J. Mater. Chem. A 7, 27514–27521 (2019).

Article CAS Google Scholar

Zhang, A. et al. Nickel doping in atomically thin tin disulfide nanosheets enables highly efficient CO2 reduction. Angew. Chem. Int. Ed. 57, 10954–10958 (2018).

Article CAS Google Scholar

Yuan, X. et al. Decoration of In nanoparticles on In2S3 nanosheets enables efficient electrochemical reduction of CO2. Chem. Commun. 56, 4212–4215 (2020).

Article CAS Google Scholar

Shen, S., Chen, J., Wang, X., Zhao, L. & Guo, L. Microwave-assisted hydrothermal synthesis of transition-metal doped ZnIn2S4 and its photocatalytic activity for hydrogen evolution under visible light. J. Power Sources 196, 10112–10119 (2011).

Article ADS CAS Google Scholar

Lopez-Rivera, S. A., Martinez, L., Fontal, B., Giriat, W. & Medina, F. Raman study of a ZnIn2S4 layered compound. Semicond. Sci. Technol. 10, 645–652 (1995).

Article ADS CAS Google Scholar

Tao, H., Mao, S., Dong, G., Xiao, H. & Zhao, X. Raman scattering studies of the Ge–In sulfide glasses. Solid State Commun. 137, 408–412 (2006).

Article ADS CAS Google Scholar

RRUFFTM Project https://rruff.info/Indium.

Wu, Y. et al. Converting copper sulfide to copper with surface sulfur for electrocatalytic alkyne semi-hydrogenation with water. Nat. Commun. 12, 3881 (2021).

Article ADS CAS PubMed PubMed Central Google Scholar

Hu, C. et al. In situ electrochemical production of ultrathin nickel nanosheets for hydrogen evolution electrocatalysis. Chem 3, 122–133 (2017).

Article CAS Google Scholar

Luc, W. et al. SO2 induced selectivity change in CO2 electroreduction. J. Am. Chem. Soc. 141, 9902–9909 (2019).

Article CAS PubMed Google Scholar

Liu, D., Liu, Y. & Li, M. Understanding how atomic sulfur controls the selectivity of the electroreduction of CO2 to formic acid on metallic Cu surfaces. J. Phys. Chem. C. 124, 6145–6153 (2020).

Article CAS Google Scholar

Aymerich, F., Meloni, F. & Mula, G. Electronic properties of the layer compound ZnIn2S4. Phys. B+C. 99, 314–317 (1980).

Article ADS CAS Google Scholar

Tao, W. et al. Electronic structure and oxidation mechanism of nickel–copper converter matte from first-principles calculations. ACS Omega 5, 20090–20099 (2020).

Article CAS PubMed PubMed Central Google Scholar

Wei, T.-R. et al. Exceptional plasticity in the bulk single-crystalline van der Waals semiconductor InSe. Science 369, 542–545 (2020).

Article ADS CAS PubMed Google Scholar

Hao, Y.-C. et al. Promoting nitrogen electroreduction to ammonia with bismuth nanocrystals and potassium cations in water. Nat. Catal. 2, 448–456 (2019).

Article CAS Google Scholar

Ma, M., Kim, S., Chorkendorff, I. & Seger, B. Role of ion-selective membranes in the carbon balance for CO2 electroreduction via gas diffusion electrode reactor designs. Chem. Sci. 11, 8854–8861 (2020).

Article CAS PubMed PubMed Central Google Scholar

Kuhl, K. P., Cave, E. R., Abram, D. N. & Jaramillo, T. F. New insights into the electrochemical reduction of carbon dioxide on metallic copper surfaces. Energy Environ. Sci. 5, 7050–7059 (2012).

Article CAS Google Scholar

Kresse, G. & Hafner, J. Ab initio molecular dynamics for liquid metals. Phys. Rev. B 47, 558–561 (1993).

Article ADS CAS Google Scholar

Blöchl, P. E. Projector augmented-wave method. Phys. Rev. B 50, 17953–17979 (1994).

Article ADS Google Scholar

Perdew, J. P., Burke, K. & Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 77, 3865–3868 (1996).

Article ADS CAS PubMed Google Scholar

Yoo, J. S., Christensen, R., Vegge, T., Nørskov, J. K. & Studt, F. Theoretical insight into the trends that guide the electrochemical reduction of carbon dioxide to formic acid. ChemSusChem 9, 358–363 (2016).

Article CAS PubMed Google Scholar

Jiang, B., Zhang, X.-G., Jiang, K., Wu, D.-Y. & Cai, W.-B. Boosting formate production in electrocatalytic CO2 reduction over wide potential window on Pd surfaces. J. Am. Chem. Soc. 140, 2880–2889 (2018).

Article CAS PubMed Google Scholar

Maintz, S., Deringer, V. L., Tchougréeff, A. L. & Dronskowski, R. LOBSTER: a tool to extract chemical bonding from plane-wave based DFT. J. Comput. Chem. 37, 1030–1035 (2016).

Article CAS PubMed PubMed Central Google Scholar

Dronskowski, R. & Bloechl, P. E. Crystal orbital Hamilton populations (COHP): energy-resolved visualization of chemical bonding in solids based on density-functional calculations. J. Phys. Chem. 97, 8617–8624 (1993).

Article CAS Google Scholar

Deringer, V. L., Tchougréeff, A. L. & Dronskowski, R. Crystal orbital Hamilton population (COHP) analysis as projected from plane-wave basis sets. J. Phys. Chem. A 115, 5461–5466 (2011).

Article CAS PubMed Google Scholar

Download references

This work is supported by the National Basic Research Program of China (Grant 2018YFA0702001), the National Natural Science Foundation of China (Grants 21975237 and 51702312), the Anhui Provincial Research and Development Program (Grant 202004a05020073), the USTC Research Funds of the Double First-Class Initiative (Grant YD2340002007), the Fundamental Research Funds for the Central Universities (WK2340000101), and the Recruitment Program of Global Youth Experts.

These authors contributed equally: Li-Ping Chi, Zhuang-Zhuang Niu, and Xiao-Long Zhang.

Division of Nanomaterials & Chemistry, Hefei National Laboratory for Physical Sciences at the Microscale, University of Science and Technology of China, Hefei, 230026, China

Li-Ping Chi, Zhuang-Zhuang Niu, Xiao-Long Zhang, Peng-Peng Yang, Jie Liao, Fei-Yue Gao, Zhi-Zheng Wu, Kai-Bin Tang & Min-Rui Gao

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

You can also search for this author in PubMed Google Scholar

M.R.G. and K.B.T. conceived and supervised the project. L.P.C. performed the experiments, collected and analyzed the data. Z.Z.N. helped with the test and analysis of NMR. X.L.Z. carried out the DFT calculations. P.P.Y., J.L., F.Y.G., and Z.Z.W. helped with electrochemical data collection and analysis. M.R.G., K.B.T., and L.P.C. co-wrote the paper. All authors discussed the results and commented on the paper.

Correspondence to Kai-Bin Tang or Min-Rui Gao.

The authors declare no competing interests.

Peer review information Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work. Peer reviewer reports are available.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

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 license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license 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 license, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

Chi, LP., Niu, ZZ., Zhang, XL. et al. Stabilizing indium sulfide for CO2 electroreduction to formate at high rate by zinc incorporation. Nat Commun 12, 5835 (2021). https://doi.org/10.1038/s41467-021-26124-y

Download citation

Received: 21 June 2021

Accepted: 14 September 2021

Published: 05 October 2021

DOI: https://doi.org/10.1038/s41467-021-26124-y

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Nature Communications (2024)

Catalysis Letters (2024)

Science China Materials (2024)

Rare Metals (2024)

Nano Research (2023)