Boosting the reduction of CO₂ and dimethylamine for C–N bonding to synthesize DMF via modulating the electronic structures of indium single atoms

Abstract

The electrocatalytic reduction of CO₂ and dimethylamine (HN(CH₃)₂) for C–N coupling is a promising strategy for synthesizing N,N-dimethylformamide (DMF). However, the generation of suitable coupling intermediates via the dehydrogenation of HN(CH₃)₂ and reduction hydrogenation of CO₂ is the key challenge for achieving C–N bonding to synthesize DMF. We optimized the electronic structure of HN(CH₃)₂ and CO₂ for adsorption on indium single atoms by adjusting the coordination structure, thereby promoting the hydrogen transfer from nitrogen of dimethylamine to CO₂, which generated the intermediates of *N(CH₃)₂ and *COOH for C–N bonding to synthesize DMF. The yield of DMF synthesized on InN₃ reached 41.3 μmol L⁻¹ h⁻¹, which was about 12 times greater than that of InN₄ at −0.8 V. In situ technology and DFT calculations jointly demonstrated that compared with InN₄, InN₃ optimized the electron distribution of adsorbed CO₂ and HN(CH₃)₂. The electron density of hydrogen on HN(CH₃)₂ decreased, exhibiting its electrophilic properties. In addition, oxygen of CO₂ accumulated electrons near the dimethylamine end and exhibited strong electron-rich properties, which led to hydrogen transfer from dimethylamine to CO₂, generating the species *N(CH₃)₂ and *COOH that are conducive to C–N coupling to synthesize DMF on InN₃. This work provides important theoretical guidance for the C–N coupling of CO₂ and amines.

---

Broader context

Environmental pollution and energy crisis are considered two major problems that threaten the survival and development of human beings. CO₂, which is a typical greenhouse gas, and dimethylamine, which is a water pollutant, need to be removed urgently. The simultaneous conversion of CO₂ and dimethylamine via C–N coupling is a promising strategy for synthesizing N,N-dimethylformamide (DMF), which has a wide range of industrial and other applications. However, generation of suitable coupling intermediates via the dehydrogenation of HN(CH₃)₂ and reduction hydrogenation of CO₂ is the key challenge in C–N coupling for synthesizing DMF. In this work, we optimized the electronic structure of HN(CH₃)₂ and CO₂ for adsorption on indium single atoms by adjusting the coordination structure, thereby promoting the hydrogen transfer from dimethylamine to CO₂, which generated the intermediates *N(CH₃)₂ and *COOH for C–N coupling to synthesize DMF. This work provides important theoretical guidance for the C–N coupling of CO₂ and secondary amines.

Introduction

N,N-Dimethylformamide (DMF), which is a type of organic nitride and a universal solvent, is an important chemical intermediate and platform compound, with its annual production reaching millions of tons. Its extensive industrial applications underscore its significant economic and technological values.^1–3 Traditional industrial synthesis of DMF is carried out using CO as the carbon source and dimethylamine as the nitrogen source under high-temperature (120 °C) and high-pressure (1.8 MPa) conditions, which has the disadvantages of harsh reaction conditions and high energy consumption. In this regard, green synthesis of DMF driven by clean electricity via electrocatalytic C–N coupling using greenhouse gas CO₂ as the carbon source and dimethylamine has important environmental and energy significance (CO₂ + HN(CH₃)₂ + H₂O + 2e⁻ → HCON(CH₃)₂ + 2OH⁻). However, realizing electrocatalytic reduction of CO₂ and dimethylamine for C–N coupling to form DMF is very difficult, and the most significant challenge is the generation of intermediates that are conducive to C–N coupling to synthesize DMF via dehydrogenation of dimethylamine and the reduction hydrogenation of carbon dioxide.

The electronic structure of the catalyst directly affects the C–N coupling^4,5 between ammonia dehydrogenation and CO₂ hydrogenation. This is because coordination environment of interface atoms affects the adsorption structure of reactants on the catalyst, thereby altering the electronic structure of adsorbed reactants and improving the adsorption behaviour of the catalyst towards reactants or reaction intermediates,^6,7 which optimize the reaction pathway. Research has found that optimizing the electronic structure of catalysts by adjusting the coordination structure can promote the dehydrogenation of ammonia. Zan et al. constructed asymmetric Ru₁N₂O₁ by changing the coordination environment of Ru single-atom catalysts, which not only enhanced its adsorption of ammonia but also optimized the electronic configuration of Ru.⁸ It led to direct activation of N–H bonds without the involvement of surface oxygen, thereby facilitating dehydrogenation. Moreover, adjusting the electronic structure of the catalyst through the coordination structure can promote the activation of reduction of carbon dioxide. Zhong et al. reported that the electronic structure of nickel was regulated by adjusting the coordination environment around the bimetallic nickel sites. The high electrocatalytic activity of Ni₂–N₃C₄ towards reduction of CO₂ could be attributed to the modulation of its electronic structure on the Ni sites and the resulting appropriate binding energy for *COOH and *CO intermediates.⁹ Besides, owing to the unique electronic properties of indium species, it can be highly selectively reduced to CO₂ as HCOOH. It can be seen that formic acid contains HCO groups, and the reduction of CO₂ to HCO groups is also an important step in the N-formylation of dimethylamine and CO₂ to synthesize DMF.

Therefore, we constructed an indium single-atom catalyst with a low coordination structure, which optimized its electronic structure via adjusting the coordination environment for the reduction of CO₂ and dimethylamine to synthesize DMF. InN₃ with low N-coordinated indium single-atom catalyst led to easy dehydrogenation of dimethylamine, which was then coupled with the CO₂ reduction intermediate for C–N-coupled N-formylation to synthesize DMF. The low coordination structure of indium single atoms significantly affects the electronic properties of the catalyst and reactants. Differential charge analysis revealed that the indium single-atom catalyst of InN₃ affects the electron distribution of its adsorbed reactants dimethylamine and CO₂. The electron density of hydrogen of nitrogen on dimethylamine decreases, exhibiting electrophilic properties. In addition, the oxygen atom of electrons on CO₂ accumulates near the dimethylamine end, exhibiting strong electron-rich properties, which weakens the hydrogen atom binding ability of dimethylamine on its nitrogen, thus facilitating dehydrogenation transfer to the oxygen of CO₂, achieving hydrogen transfer from dimethylamine to CO₂, generating *N(CH₃)₂ and *COOH intermediates that are conducive to coupling, thereby efficiently forming C–N bonds, and the free energy of N-formylation on InN₃ is lower, making it easier to generate DMF. We designed and prepared a low N-coordinated indium single-atom catalyst of InN₃ to promote efficient C–N coupling between CO₂ and dimethylamine for the electrochemical synthesis of DMF. The yield of DMF synthesized by InN₃, a low N-coordinated indium single-atom catalyst, reached 41.3 μmol L⁻¹ h⁻¹ at −0.8 V, with a corresponding faradaic efficiency of 22.4%. Synchrotron radiation, in situ infrared spectroscopy, in situ Raman spectroscopy, and DFT theoretical calculations jointly demonstrate that by regulating the coordination structure of indium single atoms, changes in electronic structure are induced, which promote hydrogen transfer from dimethylamine to CO₂. It is more conducive to CO₂ reduction and hydrogenation. This efficiently generates intermediates of *N(CH₃)₂ and *COOH that facilitate C–N coupling to synthesise DMF. This work provides important theoretical guidance for the C–N coupled N-formylation of CO₂ and amines, thereby paving the way for advancing electrochemical CO₂ conversion technologies.

Results and discussion

DFT calculations predicted the possibility of synthesizing DMF from CO₂ and dimethylamine. In order to gain a deeper understanding of the feasibility and reaction pathway of the synthesis of N,N-dimethylformamide using CO₂ and dimethylamine via C–N coupling, Indium single-atom catalysts with different coordination numbers of InN₃ and InN₄ were used as models to illustrate which type of coordination structure promotes the C–N coupling of dimethylamine and CO₂ to form DMF. The free energy for the C–N coupled N-formylation of CO₂ and dimethylamine to generate DMF in both InN₃ and InN₄ configurations is shown in Fig. 1a: *(CH₃)₂NH + CO₂ → *(CH₃)₂N + *COOH → *(CH₃)₂NCOOH → *(CH₃)₂NCO → *(CH₃)₂NCHO → (CH₃)₂NCHO. The detailed adsorption configuration is shown in Fig. 1b. Initially, low nitrogen coordination is beneficial for the adsorption of nitrogen-containing substances, and in this process, dimethylamine is advantageous for adsorption on InN₃ with low nitrogen coordination. The adsorption energy of InN₃ for dimethylamine is −0.09 eV, and the change in Gibbs free energy is less than zero, indicating that InN₃ is prone to adsorbing dimethylamine. The free energy of InN₄ adsorption of dimethylamine is 0.64 eV, and the change in Gibbs free energy is greater than zero, indicating that it is an endothermic reaction. InN₄ is not easily adsorbed with dimethylamine. Then, the CO₂ and adsorbed dimethylamine undergo hydrogen transfer. The hydrogen on nitrogen of dimethylamine is transferred to carbon dioxide, achieving the reduction hydrogenation of carbon dioxide to generate a *COOH intermediate, while dimethylamine generates a *N(CH₃)₂ intermediate. In the process of hydrogen transfer between CO₂ and dimethylamine, InN₃ is more prone to this process. The free energy for hydrogen transfer to generate *N(CH₃)₂ + *COOH intermediate is 0.37 eV, while the free energy for this step on InN₄ is twice that of InN₃, reaching 0.74 eV. In the process of C–N coupling of CO₂ and dimethylamine to synthesise DMF, the most difficult to occur is the C–N coupling of *N(CH₃)₂ and *COOH intermediates, as the highest free energy for C–N coupling occurs throughout the entire reaction pathway. The free energy of the intermediate *(CH₃)₂NCOOH generated on the InN₄ catalyst reached 1.23 eV. The catalyst with a low-coordination structure, InN₃, can greatly reduce the reaction free energy of C–N coupling compared to InN₄, which is only 0.60 eV, making it more prone to C–N coupling. The process of *(CH₃)₂NCO → *(CH₃)₂NCHO is an also crucial step in the formylation after C–N coupling. From the configuration of InN₃ adsorbing (CH₃)₂NCO groups (Fig. 1b), it can be seen that the metal indium site adsorbs the carbon atom on the carbonyl group of *(CH₃)₂NCO, which is beneficial for its direct catalytic hydrogenation. The oxygen atoms of *(CH₃)₂NCO was adsorbed on the metal indium site of InN₄, which belongs to indirect catalytic hydrogenation. In addition, it is unfavourable for the hydrogenation acylation on carbon. The free energy for the reduction hydrogenation acylation of carbon atom on the carbonyl group of *(CH₃)₂NCO occurring on InN₃ is even lower at 0.24 eV. However, the free energy for generating *(CH₃)₂NCHO on InN₄ is as high as 1.06 eV, making it very difficult to acylate. In summary, low coordination of InN₃ can reduce the free energy of the key reaction step *(CH₃)₂N + *COOH in the synthesis of DMF from dimethylamine and carbon dioxide, and is more conducive to C–N coupling and N-formylation synthesis of DMF ((CH₃)₂NCHO).

Fig. 1 Gibbs free energy and pathway diagram (a) and molecular adsorption configuration diagram (b) of different nitrogen-coordinated indium single-atom catalysts (InN₄, InN₃) for the reduction of CO₂ and dimethylamine to synthesize DMF via C–N bonding. Differential charge density of indium single-atom catalysts (InN₄, InN₃) with different nitrogen coordination for co-adsorption of dimethylamine and CO₂ (c), where yellow represents electron accumulation and green represents electron depletion.

We used differential charge density to analyse the reason why InN₃ achieves catalytic reduction of CO₂ and dimethylamine to synthesize DMF. We compared the differential charge density of indium single-atom catalysts (InN₄, InN₃) with different nitrogen coordination environments, and explained their catalytic mechanism from the aspects of electron distribution. The most critical step in the catalytic reduction of CO₂ and dimethylamine to synthesize DMF is the hydrogen transfer between dimethylamine and CO₂, generating reaction intermediates *N(CH₃) and *COOH that are favourable for C–N coupling. This step involves the dehydrogenation of dimethylamine, which is very difficult to occur in the reduction reaction. We optimized the adsorption configurations of InN₄ and InN₃ for both CO₂ and dimethylamine via differential charge density. The analysis results showed that InN₃ adsorbs dimethylamine, and CO₂ molecules are perpendicular to the dimethylamine molecules adsorbed on InN₃ (Fig. S1, ESI†). Charge transfer occurs on the CO₂ molecules, and the electron density of oxygen on CO₂ near dimethylamine increases in InN₃, exhibiting electron-rich properties. However, the electron density of the hydrogen atom on nitrogen of dimethylamine is missing, resulting in strong electrophilicity (Fig. 1c), which easily causes hydrogen on dimethylamine to detach from dimethylamine and form a bond with the electron-rich oxygen on CO₂, leading to hydrogen transfer and the key step of *(CH₃)₂NH + CO₂ → *N(CH₃)₂ + *COOH. When the InN₄ catalyst simultaneously adsorbs CO₂ and dimethylamine, CO₂ is parallel to the dimethylamine molecules adsorbed on InN₄. From the configuration diagram of differential charge density, it can be seen that the side of CO₂ near the dimethylamine is electron deficient, and the position of hydrogen on the primary amine of the dimethylamine molecule is also electron deficient, which causes charge repulsion. Therefore, hydrogen transfer between dimethylamine molecule and CO₂ molecule is not easy to occur on InN₄, completing the key step of *(CH₃)₂NH + CO₂ → *N(CH₃)₂ + *COOH. Therefore, the free energy of the key reaction intermediate *N(CH₃)₂ + *COOH, which is conducive to coupling, generated on InN₄ is as high as 0.74 eV, which is twice as high as that on InN₃. In summary, the low coordination of InN₃ alters the electron arrangement of adsorbed dimethylamine and carbon dioxide, making dimethylamine more prone to dehydrogenation and CO₂ reduction hydrogenation, thereby causing C–N coupling between *N(CH₃)₂ and *COOH intermediates to form DMF. In order to deeply understand the interaction between the hydrogen atom of the amino group in HN(CH₃)₂ and the oxygen atom of CO₂ on InN₃, the projected density of states (PDOS) was calculated (Fig. S2, ESI†). The results indicated that the overlap between the s orbital of the hydrogen atom on amino group in HN(CH₃)₂ and the p orbital of the oxygen atom in CO₂ was high, indicating that there was electronic interaction and strong binding force between the hydrogen atom of amino group in HN(CH₃)₂ and the oxygen atom of CO₂, facilitating hydrogen transfer of HN(CH₃)₂ and CO₂ to synthesize DMF.

Synthesis and structural characterization of indium single-atom catalysts. The indium single-atom catalysts with different nitrogen coordinations (InN₄ and InN₃) were prepared by controlling different calcination temperatures (900 °C, 1000 °C) (Fig. S3, ESI†). From its XRD pattern (Fig. 2a), it can be seen that the material after high-temperature calcination only has a peak shape of carbon and no crystal shape of metal indium nanoparticles. It is preliminarily determined that the catalyst we prepared is not a nanoparticle, possibly because the prepared catalyst is atomically dispersed and does not have its peak structure.¹⁰ The N vacancies were confirmed through electron paramagnetic resonance (EPR) testing. All samples exhibit a Lorentz line with a corresponding g-value of 2.003,¹¹ which can be attributed to unpaired electrons in the aromatic ring of localized π-state carbon atoms in the calcined MOF material.^12,13 As the calcination temperature increases, the EPR signal of the sample gradually increases, indicating that nitrogen atoms were lost, leaving behind excess electrons. The material of indium-doped ZIF-8 calcined at 1000 °C has the strongest signal, indicating the highest content of N vacancies, possibly InN₃. The weakest and almost non-existent signal was obtained by calcination at 900 °C, which may be InN₄ (Fig. 2c). Then we demonstrate the specific coordination configuration through synchrotron radiation. By scanning electron microscopy, it can be observed that the prepared indium single-atom InN₃ exhibited a regular dodecahedral structure (Fig. S4a and b, ESI†), which is similar in morphology to similar materials prepared in the literature.^14–16 Through EDS mapping, it can be observed that In and N atoms are uniformly distributed (Fig. S4c and d, ESI†). The surface chemical properties of the prepared InN₃ catalyst was analysed by XPS. We conducted a detailed analysis of the valence states of N and In elements in the catalyst, and N 1s can be fitted into four main components, corresponding to pyridine N (398.4 eV), pyrrole N (399.0 eV), graphite N (401.2 eV), and In–N (400.2 eV)¹⁷ (Fig. 2b). We continued to analyze the valence state of indium element in catalysts InN₄ and InN₃. In 3d_5/2 and In 3d_3/2 are located at 445.0 eV and 452.5 eV,^10,18–20 respectively, indicating that the valence state of In in the prepared catalyst InN₃ is +3 (Fig. S5, ESI†). Its valence state is the same as that of the indium species in InN₄. We used synchrotron radiation to confirm that the catalyst is a single atom and determine the coordination environment of catalyst. By processing the R space of InN₃ and InN₄ catalysts prepared by synchrotron radiation detection, the different coordination environments of indium single atoms were confirmed. We used synchrotron X-ray absorption spectroscopy (XAS) to conduct in-depth research on the electronic and coordination structures of the catalyst. By comparing the white line peaks of the k-edge X-ray absorption energy near-edge structure spectra of different indium species (In foil, In₂O₃, InN₃, and InN₄), it can be seen that the white line peak positions between InN₃ and In₂O₃ are almost the same, and the valence state of In on InN₃ is the same as the oxdation state of In₂O₃, which is +3 (Fig. 2d). Single-atom catalysts of InN₃ and InN₄ were confirmed via synchrotron radiation. In Fig. 2e, the k-edge Fourier transform k³-weighted EXAFS spectra (R space) of indium in InN₃ and InN₄ show only one distinct peak, which is an important indicator of single-atom catalysts.^21–23 Compared to In foil, the In₂O₃ material has In–In bonds, while InN₃ and InN₄ only have In–N bonds at 1.53 Å, indicating that the material exists in the form of In with N coordination²⁴ (Fig. 2e). We further demonstrated through wavelet transform of synchrotron radiation that the catalyst we prepared is a single-atom catalyst of indium. By comparing the wavelet transform plots of k-edge EXAFS signals of In foil, In₂O₃, InN₃, and InN₄, it can be found that there are only In–In bonds in the In foil (Fig. 2j), while In₂O₃ has both In–In and In–O bonds (Fig. 2k), but there are no In–In bonds in InN₃ (Fig. 2m) and InN₄ (Fig. 2l), there is only In–N bonds. In addition, the maximum value of k in InN₃ is slightly smaller than that in InN₄, which may be caused by the low coordinated structure of nitrogen. We further analyzed the coordination configuration of indium single atoms using synchrotron radiation and elaborated on the coordination environment of single-atom indium. By analyzing the R space of InN₃ and InN₄ (Fig. 2g), it was found that the decrease in peak intensity of In–N bonds indicates a decrease in their coordination number, which roughly proves the successful preparation of single-atom catalysts for InN₃ and InN₄. In addition, it can be seen from the changes in q-space (Fig. 2f) that our data quality is very good. The coordination number of indium single atoms was determined by fitting indium single-atom catalysts calcined at different temperatures (900 °C, 1000 °C). Through EXAFS fitting of the K-edge of indium, analysis shows that the indium single atom obtained by calcination at 900 °C is connected to four N atoms in the first coordination shell of indium, resulting in indium nitrogen-tetracoordinated indium single-atom InN₄. The configuration diagram is shown in Fig. 2i. The indium single atom obtained by calcination at 1000 °C is connected to three N atoms in the first coordination shell of indium, resulting in indium-nitrogen triple-coordinated indium single-atom InN₃. The configuration diagram is shown in Fig. 2h. Through synchrotron radiation analysis of the catalysts, it was demonstrated that the prepared catalysts InN₃ and InN₄ are single-atom catalysts. In addition, it is consistent with the target catalyst model we designed through DFT calculations.

Fig. 2 XRD (a), XPS spectra of N in InN₃ (b) and EPR signals (c) of indium single-atom catalysts obtained after calcination at different temperatures, the k-edge X-ray absorption energy near-edge structural spectra (XANES) of In foil, In₂O₃, InN₃, and InN₄ (d) and their k-edge Fourier transform k³-weighted EXAFS spectra (R space) (e), k-edge Fourier transform k³-weighted EXAFS spectra (q space) of In InN₃ and InN₄ (f); k-edge Fourier transform k³-weighted EXAFS spectra (R space) of In InN₃ and InN₄ (g); K-edge EXAFS fitting curves of In corresponding to R-space for InN₃ (h) and InN₄ (i); Wavelet transform k³-weighted EXAFS spectra of In foil (j), In₂O₃ (k), InN₃ (l), and InN₄ (m).

Performance evaluation of InN₃ and InN₄ for electrocatalytic synthesis of DMF from CO₂ and dimethylamine. We conducted performance evaluation on the electrocatalytic reduction of CO₂ and dimethylamine to synthesise DMF on InN₄ and InN₃ single-atom catalysts. Through the linear scan curve test (Fig. 3a), it can be seen that the current of the InN₃ single-atom catalyst is higher than that of the InN₄ single-atom catalyst, indicating that the InN₃ single-atom catalyst is beneficial for catalyzing the reduction of CO₂ and dimethylamine. We found that the InN₃ catalyst prepared at a calcination temperature of up to 1000 °C exhibited smaller electrochemical impedance than that of InN₄ in the electrochemical impedance spectra, making it easier to conduct electrons for electrochemical reduction reactions (Fig. 3b). We determined the double-layer capacitance of InN₄ and InN₃ single-atom catalysts by testing their cyclic voltammetry spectra at different scan rates in the Faraday interval (Fig. S6a and b, ESI†), and their electrochemical active area was further determined. The slope analysis indicated that the InN₃ single atom has a higher electrochemical active area (Fig. 3c). The performance of electrochemical synthesis of DMF from CO₂ and dimethylamine was evaluated at different potentials. The yields of N,N-dimethylformamide (DMF) were detected by HPLC (Fig. S7, ESI†). The prepared InN₄ and InN₃ catalysts were used as working electrodes. A platinum plate was used as the counter electrode, and a saturated calomel electrode was used as the reference electrode to test the performance of DMF synthesis at different voltages. It was found that the InN₃ catalyst exhibited the highest electrocatalytic reduction of CO₂ and dimethylamine to DMF at −0.8 V, reaching 41.3 μmol L⁻¹ h⁻¹ (Fig. 3g), with a corresponding faradaic efficiency of up to 22.4% (Fig. 3h). However, it can be found that different nitrogen-coordinated indium single-atom catalysts can electrochemically synthesize DMF, but their performance varies greatly. At −0.8 V, the catalyst of InN₄ only achieved a yield of 3.4 μmol L⁻¹ h⁻¹ for electrocatalytic reduction of CO₂ and dimethylamine to synthesize DMF (Fig. 3d), and the corresponding faradaic efficiency was only 2.4% (Fig. 3e). By comparing the yield of different nitrogen-coordinated indium single-atom catalysts for DMF synthesis at the same optimal potential, it was found that the yield of InN₃ electrocatalytic reduction of CO₂ and dimethylamine to synthesize DMF at the optimal potential of −0.8 V was about 12 times that of InN₄ (Fig. 3f). Comparing the efficiency of reduction of CO₂ and dimethylamine to synthesize DMF on InN₄, the faradaic efficiency of InN₃ in catalyzing DMF synthesis at the optimal potential of −0.8 V is more than 9 times that of InN₄ (Fig. 3i). In summary, the unique low coordination structure of InN₃ single-atom catalysts exhibits superior catalytic activity for the reduction of carbon dioxide and dimethylamine to synthesize DMF. We also compared the performance of different electrolyte solutions in the synthesis of DMF from CO₂ and dimethylamine. The FE for synthesizing DMF in an electrolyte solution of 0.1 M KHCO₃ is 22.4%, while the FE in KOH and Na₂SO₄ is 1.2% and 5.8%, respectively (Fig. S8, ESI†), which are much lower than the efficiency of synthesizing DMF in an electrolyte solution of KHCO₃. For the long-term stability test, electrochemical synthesis of DMF was carried out for 12 hours. There is no significant decrease in current density, which indicated that InN₃ is relatively stable (Fig. S9, ESI†).

Fig. 3 Linear scan curves (a) and EIS (b) of InN₃ and InN₄ during electrocatalytic reduction of CO₂ and dimethylamine to synthesize DMF; relationship between the charge and discharge current density and the cyclic voltammetry scan rate on InN₃ and InN₄ electrodes (c); yield and faradaic efficiency for electrocatalytic reduction of CO₂ and dimethylamine to synthesize DMF at different voltages using InN₄ (d) and (e) and InN₃ (g) and (h) single-atom catalysts, as well as the yield comparison map (f) and faradaic efficiency comparison map (i) of InN₃ and InN₄ for synthesizing DMF at different potentials.

In situ characterization of the mechanism for synthesizing DMF. We investigated the reaction mechanism of different nitrogen-coordinated indium single-atom catalysts (InN₄ and InN₃) for the catalytic reduction of CO₂ and dimethylamine C–N bonding to synthesize DMF via in situ infrared spectroscopy and in situ mass spectrometry. The intermediates involved in the catalytic reduction of CO₂ and dimethylamine using indium single-atom catalysts InN₃ and InN₄ with different nitrogen coordinations were detected by in situ infrared spectroscopy. It combines with intermediate species detected by in situ mass spectrometry to elucidate the reaction mechanism for synthesizing DMF. We characterized the adsorption of CO₂ and dimethylamine by InN₃ single-atom catalysts via in situ infrared spectroscopy. We found that a large absorption peak at a wavelength of 1241 cm⁻¹ was detected on InN₃, and the intensity of the peak gradually increases over time until it no longer changes at 30 minutes, indicating that the catalyst adsorption has reached saturation. The absorption peak at 1241 cm⁻¹ corresponds to the adsorption of dimethylamine²⁵ (Fig. 4a), while InN₄ did not detect any adsorption peak for dimethylamine. The adsorption energies of InN₃ and InN₄ single-atom catalysts for dimethylamine were compared through DFT calculations. Through DFT theoretical calculations, it was found that the adsorption energy of InN₃ for dimethylamine is −0.09 eV, while the adsorption energy of InN₄ for dimethylamine is 0.64 eV (Fig. 4d). The more negative the free energy, the more likely it is to occur. Therefore, InN₃ is more conducive to adsorbing dimethylamine. It has been demonstrated through infrared spectroscopy and DFT calculations that InN₃ is more favourable for adsorbing dimethylamine.

Fig. 4 Detection of adsorption of dimethylamine on InN₃via in situ infrared spectroscopy (a), adsorption energy of dimethylamine on InN₃ and InN₄ (d); two-dimensional in situ infrared spectra and 3D in situ FT-IR spectra for electrocatalytic reduction of CO₂ and dimethylamine to synthesize DMF on InN₃ (b) and (e) and InN₄ (c) and (f). In situ Raman contour diagram (g) and two-dimensional diagram (h) of InN₃ electrocatalytic reduction of CO₂ and dimethylamine to synthesize DMF, and in situ mass spectrometry detection of intermediate species in the process of InN₃ catalyst electrocatalytic reduction of dimethylamine and CO₂ to synthesize DMF (i).

The reaction intermediates in the synthesis of DMF from CO₂ and dimethylamine in InN₃ and InN₄ catalysts through in situ infrared spectroscopy were detected to explore the mechanism of C–N bonding. Initially, we used in situ electrochemical infrared spectroscopy to detect the reaction intermediates of the InN₃ catalyst for electrocatalytic reduction of CO₂ and dimethylamine to synthesize DMF. By applying voltage, we found that in situ electrochemical infrared spectroscopy detected intermediates corresponding to C–O, C–N, and COOH bonds at wavelengths of 1173 cm⁻¹, 1449 cm⁻¹, and 1664 cm⁻¹,²⁶ respectively, indicating that these reaction intermediates are likely to be key reaction intermediates for the synthesis of DMF (Fig. 4b and e). However, the InN₄ single-atom catalyst did not have a peak for C–N bonds during the electrocatalytic reduction of CO₂ and dimethylamine, and only exhibited a peak for the *COOH intermediate at a wavelength of 1654 cm⁻¹ (Fig. 4c and f).²⁶ It suggests that InN₄ is not conducive to C–N bonding. By comparing the in situ infrared spectra of different nitrogen-coordinated indium single-atom catalysts for the catalytic reduction of CO₂ and dimethylamine C–N bonding, it was found that the low nitrogen-coordinated indium single-atom catalyst InN₃ is more conducive to C–N bonding. The reason may be that the low nitrogen-coordinated indium single-atom catalyst of InN₃ is more conducive to the adsorption of nitrogen-containing reactant dimethylamine and thus more conducive to C–N coupling with CO₂ reduction intermediates to form DMF, while InN₄ is not conducive to the adsorption of dimethylamine and only reduction of carbon dioxide, the reduction intermediates of carbon dioxide are not conducive to C–N coupling with dimethylamine. Our experimental result on electrocatalytic synthesis of DMF is consistent with the conclusion obtained from in situ infrared spectroscopy. Compared with InN₄, InN₃ electrocatalytically reduces dimethylamine and CO₂ to synthesize N,N-dimethylformamide in the highest yield.

We further investigated the reaction mechanism of efficient C–N bonding between CO₂ and dimethylamine through in situ Raman spectroscopy using the indium single-atom catalyst of InN₃ for the electrochemical synthesis of DMF. The in situ Raman spectra of InN₃ revealed peaks for the reactant dimethylamine at 890 cm⁻¹ and 1018 cm⁻¹, as shown in Fig. 4h, as well as peaks for the carbon material at 1357 cm⁻¹ and 1602 cm⁻¹. As the voltage increases, a C–N peak²⁷ appears at 1064 cm⁻¹ at –0.8 V (Fig. 4g and h). However, we detected a very small peak of C–N bond and did not observe any other Raman peaks of other reaction intermediates on InN₄ (Fig. S10, ESI†), indicating that the InN₄ catalyst is not easy to catalyze the C–N bonding of dimethylamine and CO₂ to synthesize DMF. This is also consistent with our experimental results of in situ infrared spectroscopy and DFT calculations. At the same time, we also used in situ mass spectrometry detection as a supplementary method to investigate the mechanism of efficient C–N bonding electrochemical synthesis of DMF between CO₂ and dimethylamine catalyzed by InN₃. We found that the species COOH intermediate has a mass-to-charge ratio of m/z of 45, and did not find the species CO with an m/z value of 28 (Fig. 4i). Therefore, we believe that the *COOH intermediate is a key reaction intermediate for its C–N coupling. From in situ infrared spectroscopy and in situ online mass spectrometry, it is inferred that the reaction pathway for the C–N coupling of CO₂ and dimethylamine to generate DMF is as follows: NH(CH₃)₂ + CO₂ → *N(CH₃)₂ + *COOH → *COOHN(CH₃)₂ → *CON(CH₃)₂ → HCON(CH₃)₂.

Conclusions

We constructed indium single-atom catalysts (InN₃ and InN₄) with different nitrogen coordinations to electrocatalyze the synthesis of DMF from CO₂ and dimethylamine. The adjustment of the coordination structure of indium single-atom catalysts changes the electronic structure when adsorbing dimethylamine and CO₂. It made dimethylamine more easily dehydrogenated and carbon dioxide reduced hydrogenated. The low N-coordinated single-atom catalyst of InN₃ can simultaneously catalyze the reduction of CO₂ into a *COOH intermediate and the dehydrogenation of dimethylamine to *N(CH₃)₂ intermediate. The *COOH intermediate and *N(CH₃)₂ intermediate undergo efficient C–N coupled N-formylation to synthesize DMF. The indium single atom of InN₃ affects the electronic properties of the catalyst for the adsorption of CO₂ and dimethylamine. The electron density of hydrogen of nitrogen on dimethylamine decreases, exhibiting electrophilic properties. In addition, oxygen of CO₂ accumulates electrons near the dimethylamine end, exhibiting strong electron-rich properties, which weakens the constraint ability of hydrogen atoms on dimethylamine, thus facilitating dehydrogenation transfer to the oxygen of CO₂, achieving hydrogen transfer from dimethylamine to CO₂, and generating species *N(CH₃)₂ and *COOH that are conducive to coupling. Subsequently, it efficiently forms C–N bonds and ultimately generates DMF. We promote the efficient C–N coupled electrochemical synthesis of DMF by modulating the electronic structures of indium single atoms. The yield of DMF synthesized reached 41.3 μmol L⁻¹ h⁻¹ at −0.8 V on InN₃, with a corresponding faradaic efficiency of 22.4%. In situ infrared spectroscopy, in situ Raman spectroscopy, and DFT calculations have jointly demonstrated that the intermediates *COOH and *N(CH₃)₂, which undergo hydrogen transfer between CO₂ and dimethylamine, are key reaction intermediates for C–N coupling. The indium single atom of InN₃ reduces the Gibbs free energy of this critical step, which is beneficial for C–N bonding N-formylation. This work provides important theoretical guidance for dehydrogenation C–N coupling of CO₂ and amines.

Author contributions

Guohua Zhao conceived the conceptual idea. Jingui Zheng designed the experiments and wrote the original manuscript. Lingzhi Sun carried out materials synthesis, characterization and electrochemical measurements. Shaohan Xu and Qihao Xie contributed to data analysis and improved the paper quality by discussions. Xun Pan supervised the project.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This research was financially supported by the National Natural Science Foundation of China (No. 22476153, 22076140, 21537003). Additionally, we acknowledge the help from Shanghai Synchrotron Radiation Facility (SSRF, No. 2022-SSRF-PT-020439, 2022-SSRF-PT-020432).

Notes and references

1. A. Jutand, Chem. Rev., 2008, 108, 2300–2347.
2. S. Ding and N. Jiao, Angew. Chem., Int. Ed., 2012, 51, 9226–9237.
3. S. Yamaguchi, H. Shinokubo and A. Osuka, Inorg. Chem., 2009, 48, 795–797.
4. H. Li, L. Xu, S. Bo, Y. Wang, H. Xu, C. Chen, R. Miao, D. Chen, K. Zhang, Q. Liu, J. Shen, H. Shao, J. Jia and S. Wang, Nat. Commun., 2024, 15, 8858.
5. X. Zhang, X. Zhu, S. Bo, C. Chen, M. Qiu, X. Wei, N. He, C. Xie, W. Chen, J. Zheng, P. Chen, S. P. Jiang, Y. Li, Q. Liu and S. Wang, Nat. Commun., 2022, 13, 5337.
6. Y. Wan, M. Zheng, W. Yan, J. Zhang and R. Lv, Adv. Energy Mater., 2024, 14, 2303588.
7. J. Huang, S. Ajmal, A. Kumar, J. Guo, M. Mujahid Alam, A. G. Al-Sehemi and G. Yasin, J. Energy Chem., 2025, 101, 132–155.
8. Z. Liu, H. Xu, Y. Fan, W. Huang, F. Yu, Z. Qu and N. Yan, Environ. Sci. Technol., 2024, 58, 10717–10728.
9. Y.-N. Gong, C.-Y. Cao, W.-J. Shi, J.-H. Zhang, J.-H. Deng, T.-B. Lu and D.-C. Zhong, Angew. Chem., Int. Ed., 2022, 61, e202215187.
10. P. Lu, X. Tan, H. Zhao, Q. Xiang, K. Liu, X. Zhao, X. Yin, X. Li, X. Hai, S. Xi, A. T. S. Wee, S. J. Pennycook, X. Yu, M. Yuan, J. Wu, G. Zhang, S. C. Smith and Z. Yin, ACS Nano, 2021, 15, 5671–5678.
11. J. Zhang, X. Yang, G. Xu, B. K. Biswal and R. Balasubramanian, Adv. Mater., 2024, 36, 2309205.
12. X. Zhang, P. Ma, C. Wang, L. Gan, X. Chen, P. Zhang, Y. Wang, H. Li, L. Wang, X. Zhou and K. Zheng, Energy Environ. Sci., 2022, 15, 830–842.
13. Y. Zhang, Y. Jiang, G. Jiang, T. Or, R. Gao, H. Zhang, Z. Bai, N. Chen, Y.-P. Deng and Z. Chen, Nano Energy, 2023, 115, 108672.
14. R. Zeng, L. Lin, H. Gong, Y. Li, J. Xu, L. Huang, W. Wang, S. Lin, D. Tang and S. Guo, Chem. Catal., 2023, 3, 100514.
15. H. Yin, F. Dong, H. Su, Z. Zhuang, Y. Wang, D. Wang, Y. Peng and J. Li, ACS Nano, 2023, 17, 25614–25624.
16. H. Hu, J. Wang, B. Cui, X. Zheng, J. Lin, Y. Deng and X. Han, Angew. Chem., Int. Ed., 2022, 61, e202114441.
17. Y. Guo, Z. Jin, J. Lu, L. Wei, W. Wang, Y. Huang and A. Wang, Energy Environ. Sci., 2023, 16, 5274–5283.
18. C. Lv, C. Lee, L. Zhong, H. Liu, J. Liu, L. Yang, C. Yan, W. Yu, H. H. Hng, Z. Qi, L. Song, S. Li, K. P. Loh, Q. Yan and G. Yu, ACS Nano, 2022, 16, 8213–8222.
19. J. Wu, F. Huang, Q. Hu, D. He, W. Liu, X. Li, W. Yan, J. Hu, J. Zhu, S. Zhu, Q. Chen, X. Jiao and Y. Xie, J. Am. Chem. Soc., 2024, 146, 26478–26484.
20. T. Zhou, K. Wang, Z. Luo, B. Xiao, B. Zi, H. Zheng, Q. Lu, J. Zhao, M. Chen, T. He, Y. Zhang, J. Zhang, H. Sun and Q. Liu, Int. J. Hydrogen Energy, 2024, 82, 367–373.
21. K. Chen, N. Zhang, F. Wang, J. Kang and K. Chu, J. Mater. Chem. A, 2023, 11, 6814–6819.
22. E. Zhang, L. Tao, J. An, J. Zhang, L. Meng, X. Zheng, Y. Wang, N. Li, S. Du, J. Zhang, D. Wang and Y. Li, Angew. Chem., Int. Ed., 2022, 61, e202117347.
23. B. Sun, X. Wang, Z. Li, H. Liu, W. Jiang, K. Song, Z. Wang, P. Wang, Y. Liu, Z. Zheng, Y. Dai, B. Huang and H. Cheng, Chem. Catal., 2024, 4, 100862.
24. H. Shang, T. Wang, J. Pei, Z. Jiang, D. Zhou, Y. Wang, H. Li, J. Dong, Z. Zhuang, W. Chen, D. Wang, J. Zhang and Y. Li, Angew. Chem., Int. Ed., 2020, 59, 22465–22469.
25. M. J. Buttler and D. C. McKean, Spectrochim. Acta, 1965, 21, 465–483.
26. H. Shi, H. Wang, Y. Zhou, J. Li, P. Zhai, X. Li, G. G. Gurzadyan, J. Hou, H. Yang and X. Guo, Angew. Chem., Int. Ed., 2022, 61, e202208904.
27. P. Zhan, J. Zhuang, S. Yang, X. Li, X. Chen, T. Wen, L. Lu, P. Qin and B. Han, Angew. Chem., Int. Ed., 2024, 63, e202409019.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ee05681g

---