Introduction

Since graphene was first prepared by mechanical exfoliation in 2004 [1], various two-dimensional (2D) materials have attracted extensive attention on account of their unique physical and chemical properties [2]. These materials consist of atomically thin sheets with inherently large surface areas; they can be used extensively in various areas, such as electrocatalysis [3], photocatalysis [4], energy storage [5], membrane separation [6, 7], and biotherapy [8]. Besides graphene, a wide range of atomically thin 2D materials have also been successfully prepared, including transition metal dichalcogenides [9,10,11], phosphorenes [12,13,14], silicenes [15, 16], germanene antimonenes [17, 18], boron nitrides [19,20,21], and layered double hydroxides [22].

Transition metal carbides, carbonitrides, and nitrides (MXenes) are a new addition to the family of 2D materials [23]. The common form of MXene is Mn+1XnTx (n = 1, 2, 3), where M represents an early transition metal, X represents carbon and/or nitrogen, and Tx denotes surface functional groups, such as −O, −OH, or −F [43,

Structural and Electronic Properties

Structural Properties

The overall crystal geometry of MXene presents a hexagonal close-packed structure, which is analogous to its MAX-phase precursor. Here, M atoms are arranged in a close-packed structure, and octahedral sites are occupied by X atoms. The adjacent layered units are connected via van der Waals forces, similar to other 2D materials [53].

MXenes are usually prepared in aqueous solutions, including acidic fluorides. Therefore, the surface of MXenes is occupied by a mixture of −OH, −O, and −F terminations. For brevity, these molecules are denoted Mn+1XnTx, where T represents the surface termination. Non-terminated MXenes have never been obtained [23, 76]. Recent computational studies demonstrate that the surface termination exerts significant impacts on the properties of MXenes. For example, Hu et al. [77] systematically studied the chemical origin of termination-functionalized MXenes by Bader charge analysis and thermodynamic calculations; the materials revealed stability in the order of Ti3C2O2 > Ti3C2F2 > Ti3C2(OH)2 > Ti3C2H2 > Ti3C2, which was attributed to the splitting of the highly degenerated 3d orbitals of surface Ti. In another study, Fu and co-workers [78] systematically explored the effects of several functional groups (i.e., −Cl, −F, −H, −O, and −OH) on the stabilization, mechanical properties, and electronic structures of a representative MXene (Ti3C2); the authors found that oxygen-functionalized Ti3C2 shows better thermodynamic stabilization and strength than their other counterparts due to significant charge transfers from inner bonds to the outer surface of the material. While MXenes with specific terminations may be gained by a post-synthesis method, very few studies on this topic have been reported. For example, Meng et al. [79] predicted that S-functionalized Ti3C2 displays metallic behavior, a stable structure, a low diffusion barrier, and outstanding storage capacity for Na-ion batteries.

Besides theoretical explorations, surface termination of MXenes such as Ti3C2Tx and V2CTx has also been investigated by using experimental methods. For instance, Wang et al. [80] revealed the surface atomic scale of Ti3C2Tx through aberration-corrected scanning transmission electron microscopy (STEM); the group found that surface functional groups (e.g., −OH, −F, and −O) are randomly distributed on the MXene surfaces and prefer to occupy the top sites of the central Ti atom. Karlsson’s group [81] observed individual and double sheets of Ti3C2 by aberration-corrected STEM-EELS and revealed sheet coverage and intrinsic defects and TiOx adatom complexes. In another study, Sang and co-workers [103] presented hierarchical MoS2/Ti3C2-MXene@C nanohybrids by coupling MoS2 nanosheets on carbon-stabilized Ti3C2 MXene. The obtained catalyst exhibited excellent performance with a low overpotential of 135 mV at 10 mA/cm2 and a low Tafel slope of 45 mV/dec; these values are smaller than those of other counterpart catalysts (Fig. 5g). Indeed, our group presented Co-MoS2/Mo2CTx nanohybrids by engineering Co-doped MoS2 coupled with Mo2CTx MXene [98]. The resulting hybrids exhibited a low overpotential of 112 mV at 10 mA/cm2 and good stability in 1 mol/L KOH aqueous solution.

Fig. 5
figure 5

a Work functions (denoted by dots) of different MXenes with O terminations compared with the ionization energies (denoted by dashed lines) of monolayer MoS2, WS2, MoSe2, WSe2, and MoTe2. b Schematic of the MoS2-catalyzed HER and metallic energy-band feature of MoS2 induced by p-type Schottky barrier-free contact. c Hydrogen adsorption energies of 2H-MoS2, 1T-MoS2, and 2H-MoS2/MXene heterostructures. The numbers in brackets represent the corresponding different H coverages. Reproduced with permission [126]. Copyright 2019, American Chemical Society. d Typical TEM and SEM images of a nanoroll-like MoS2/Ti3C2Tx hybrid and schematic of the MoS2/Ti3C2Tx-catalyzed HER process. e Polarization curves of a MoS2/Ti3C2Tx hybrid, pure MoS2, Ti3C2Tx nanosheets, and Pt/C. Reproduced with permission [42]. Copyright 2019, Elsevier Ltd. f Polarization curves of interlayer expanded-MoS2/Ti3C2 at various temperatures. Reproduced with permission [100]. Copyright 2018, Royal Society of Chemistry. g Polarization curves of MoS2/Ti3C2-MXene@C, MoS2/oxidized MXene, MoS2/rGO@C, Ti3C2 MXene, and Pt/C catalysts. Reproduced with permission [103]. Copyright 2011, Wiley-VCH. h Reaction free energy (ΔGH*) of HER on the most active sites of different graphene/MXene heterostructures and on the Pt (111) surface. i Changes in ΔGH* during HER on N-doped graphene over a V2C MXene monolayer. Reproduced with permission [128]. Copyright 2018, Royal Society of Chemistry

Full size image

Du et al. [102] reported the in situ growth of the Ni-based bimetal phosphorus trisulfide (Ni1−xFexPS3) on the surface of Ti3C2Tx MXene nanosheets by a simple self-assembly and subsequent solid-state reaction process. The optimized hybrids (Ni0.7Fe0.3PS3@MXene) exhibited a low overpotential of 196 mV for HER in 1 mol/L KOH solution. Zhou et al. [128] theoretically designed several heterostructures of N-doped graphene/MXenes (Ti2C, Nb2C, V2C, and Mo2C) as catalysts for HER. DFT calculations suggested that N-doped graphene/heterostructures possess the lowest reaction free energies (close to 0 eV) and a low Tafel reaction barrier (1.3 eV) for HER (Fig. 5g, h) owing to the strong electronic coupling between the MXene and N-doped graphene.

Recent studies indicate that the HER performance of MXenes could be improved by do** with metal atoms. Li et al. [129] studied the HER properties of modified M2XO2-type MXenes bearing transition metal atoms by high-throughput computational methods. Addition of transition metal atoms to several combinations, such as Os-Ta2CO2, Ir-Sc2CO2, Ag-Nb2NO2, Re-Nb2NO2, and W-Nb2NO2, could change the relevant reaction mechanism (from Volmer–Heyrovsky to Volmer–Tafel), induce electron redistributions on the surface of the MXene, and, ultimately, result in distinct enhancements in HER activity. Du et al. [43] reported an MXene (Ti3C2Tx)-based hybrid with simultaneous Nb do** and surface Ni/Co alloy modification. DFT calculations indicated that Nb do** could shift the Fermi energy level toward the conduction band, leading to improved conductivity. Moreover, the surface M–H affinity was modified by the Ni/Co alloy, and the optimized catalyst showed the lowest Gibbs free energy for adsorbed H* (Fig. 6a, b). The resultant Ni0.9Co0.1@NTM (Nb-doped Ti3C2Tx) hybrids delivered excellent HER performance, only requiring a small overpotential of 43.4 mV to deliver a current density of 10 mA/cm2 in 1 mol/L KOH solution (Fig. 6c, d), and exhibited long-term stability. Using in situ co-reduction, Li et al. [96] prepared Pt/Ti3C2Tx via alloying Pt with Ti from the surface of Ti3C2Tx. In situ X-ray absorption spectroscopy revealed that Pt transforms from a single atom into intermetallic compounds with increasing temperature (Fig. 6e). The as-prepared Pt/Ti3C2Tx-550 showed outstanding HER performance and only needed a low overpotential of 32.7 mV at 10 mA/cm2 (Fig. 6f); it also demonstrated a small Tafel slope of 32.3 mV/dec. HER current normalization processing revealed that the respective mass activity and specific activity of Pt/Ti3C2Tx-550 are 4.4 and 13 times higher than those of Pt/Vulcan at an overpotential of 70 mV (Fig. 6g, h). As shown in Fig. 6i, DFT calculations demonstrated that (100)- and (111)-terminated Pt3Ti nanoparticles show H* binding comparable with Pt (111). However, (110)-termination showed that H* adsorption was excessively exergonic, leading to poisoning of the relative overpotential.

Fig. 6
figure 6

a Atomistic configuration of pristine monolayer Ti3C2O2 with H* adsorption, Nb doped on pristine monolayer Ti3C2O2 with H* adsorption, Co/Ni replacement of Ti atoms on Nb-doped pristine monolayer Ti3C2O2, and the three different H* adsorption O sites. b Gibbs free energies for H* adsorbed on active sites shown in a M-doped Ti3C2O2. c Polarization curves of a series of NiCo@Nb-doped Ti3C2Tx MXene nanohybrids, Ni@Nb-doped Ti3C2Tx MXene nanohybrid, Nb-doped Ti3C2Tx MXene, and Pt/C in 1 mol/L KOH. d Corresponding Tafel plots of a series of NiCo@Nb-doped Ti3C2Tx MXene nanohybrids, Ni@Nb-doped Ti3C2Tx MXene nanohybrid, Nb-doped Ti3C2Tx MXene, and Pt/C in 1 mol/L KOH. Reproduced with permission [43]. Copyright 2019, Wiley-VCH. e Magnitude of the Fourier transform of the k2 weighted Pt LIII edge in situ EXAFS of Pt/Ti3C2Tx reduced at different temperatures compared with that of Pt/SiO2. f Polarization curves of Pt/Vulcan, Pt/Ti3C2Tx at different temperatures, and Ti3C2Tx. g Mass activity of Pt/Vulcan and Pt/Ti3C2Tx catalysts with different treatments. h Specific activity of Pt/Vulcan and Pt/Ti3C2Tx catalysts. i DFT-calculated free energy diagrams of hydrogen evolution at the Pt (111), Pt3Ti (111), Pt (100), Pt3Ti (100), Pt (110), and Pt3Ti (110) surfaces. Reproduced with permission [96]. Copyright 2019, American Chemical Society

Full size image

In another work, Zhang et al. [104] reported a novel electrochemical exfoliation method to prepare Mo2TiC2Tx MXene nanosheets for HER. The obtained nanosheets possessed an abundance of exposed basal planes and Mo vacancies providing numerous active sites on which to immobilize single atoms and improve the HER catalytic property of the MXenes (Fig. 7a). Pt atoms anchored onto the Mo2TiC2Tx nanosheets showed excellent catalytic performance. The obtained Mo2TiC2Tx–PtSA catalysts only needed low overpotentials of 30 and 77 mV to deliver current densities of 10 and 100 mA/cm2, respectively. The as-prepared catalyst showed an outstanding mass activity of 8.3 A/mg, which is around 40 times greater than that of commercial Pt/C (0.21 A/mg; Fig. 7b, c). Strong covalent bonding between Mo2TiC2Tx and positively charged Pt atoms endowed the Mo2TiC2Tx–PtSA catalyst with outstanding long-term stability. DFT calculations suggested that single-atom Pt could lead to the redistribution of the electronic structure of Mo2TiC2Tx and move up the d orbitals-electron domination close to the Fermi level (Fig. 7d, e), resulting in improved catalytic activity. As presented in Fig. 7f, the obtained Mo2TiC2Tx–PtSA catalyst showed a low adsorption energy of − 0.08 eV, which is significantly lower than those of Mo2TiC2O2 (− 0.19 eV) and Pt/C (− 0.10 eV). Other Pt-modified MXenes nanohybrids have been achieved by different methods, such as photo-induced reduction [97], NaBH4 reduction [99], and solution plasma modification [101], and the resulting hybrids generally showed remarkably improved performance for HER compared with pristine MXenes. ** low-cost and high-efficiency catalysts for ORR remains a crucial endeavor.

Liu and Li [132] simulated a series of Pt/v-Tin+1CnTx (n = 1–3, T = O and/or F) heterostructures by DFT calculations. As displayed in Fig. 10a, F-terminated MXenes were predicted to display better performance in ORR than their O-terminated counterparts; however, F-terminated MXenes may demonstrate lower stability on account of their weaker chemical bonding. A variety of MXene-based materials have been explored to enhance ORR performance. For instance, Li et al. [48] prepared FePc/Ti3C2Tx hybrids by a facile self-assembly method in dimethylformamide solution. Owing to the presence of Ti3C2Tx, obvious Fe 3d electron delocalization and spin-state transition of Fe(II) ions were confirmed by a series of characterization analysis, such as ESR and Mössbauer spectroscopy, as presented in Fig. 10b–d. More importantly, changes in electron configuration led to lower local electron densities and higher spin states in the Fe(II) centers, which promoted oxygen adsorption and reduction in active FeN4 sites. As shown in Fig. 10e, f, the optimized hybrids showed lower half-wave potentials (− 0.886 vs. RHE) compared with pure FePc (− 0.886 vs. RHE) and commercial Pt/C (− 0.84 V vs. RHE). The catalysts also, respectively, showed two- and fivefold higher specific ORR activity than pure FePc and commercial Pt/C in 0.1 mol/L KOH solution.

Fig. 10
figure 10

a Free energy diagram of ORR intermediates on Pt/v-Tin+1CnT2 (n = 1–3, T = O or F) surfaces. Reproduced with permission [132]. Copyright 2019, American Chemical Society. b Fe Mössbauer transmission spectra and c deconvolution of pristine FePc and FePc/Ti3C2Tx. d X-band ESR spectra of pristine Ti3C2Tx, FePc, and FePc/Ti3C2Tx. e Polarization curves and f the corresponding Tafel plots of pristine FePc, FePc/Ti3C2Tx, and Pt/C. Reproduced with permission [48]. Copyright 2018, Wiley-VCH. g Preparation of Co/N-CNTs@Ti3C2Tx composites. h C K-edge XANES spectra of Ti3C2Tx MXene, Co/N-CNTs@Ti3C2Tx, and Co/N-CNTs. i N K-edge XANES spectra of Co/N-CNTs and Co/N-CNTs@Ti3C2Tx. j Ti L-edge XANES spectra of Ti3C2Tx MXene and Co/N-CNTs@Ti3C2Tx. k Polarization curves of Ti3C2Tx, Co/N-CNTs, Co/N-CNTs@Ti3C2Tx, and Pt/C. Reproduced with permission [111]. Copyright 2018, Wiley-VCH

Full size image

Zhang et al. [111] presented a new type of Co/N-CNTs@Ti3C2Tx hybrid synthesized by an in situ growth strategy (Fig. 10g). The resulting catalyst showed superior ORR catalytic performance with a low onset potential of 0.936 V versus RHE and a half-wave potential of 0.815 V versus RHE in 0.1 mol/L KOH aqueous solution (Fig. 10k); such performance was attributed to strong interfacial coupling and electron transfers in the composite, which were well verified by XANES (Fig. 10h–j). A series of nanohybrids, such as Mn3O4/Ti3C2Tx nanocomposites [114], C3N4/Ti3C2 heterostructures [47], FeNC/MXene nanohybrids [112], urchin-like MXene-Ag0.9Ti0.1 nanowire composites [113], and FeCo (3:1)-N-d-Ti3C2 MXene hybrids [115], have also been proven to display outstanding ORR performance.

Nitrogen Reduction Reaction

NH3 is considered a promising alternative energy carrier on account of its high energy density. At present, large-scale NH3 production is primarily conducted via the Haber–Bosch method at high-pressure and high-temperature conditions using H2 and nitrogen N2 as the virgin gas. However, this process consumes large amounts of energy and generates massive amounts of CO2. Thus, develo** sustainable and economical N2-fixation methods is urgently needed. Electrocatalytic NRR has attracted much attention due to its innate advantages, including reaction under ambient conditions and water as the hydrogen source [133]. However, NRR processes remain at the infant stages of development, and designs of efficient and low-cost electrocatalysts continue to challenge researchers.

MXene-based materials have recently been studied as catalysts for NRR. For example, Azofra et al. [50] predicted the N2-capture behaviors of M3C2 MXenes using DFT calculations and found that V3C2 and Nb3C2 are excellent candidates as NRR catalysts due to their low reaction energies of 0.32 and 0.39 eV (vs. a standard hydrogen electrode), respectively (Fig. 11a). V3C2 showed a low activation barrier of 0.64 eV, which is smaller than that of Nb3C2 (0.85 eV), for the first proton–electron transfer (rate-determining step). Gao et al. [134] predicted the catalytic activity of a series of single atoms (i.e., Sc, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Mo, Ru, Rh, Pd, Ag, Cd, and Au) anchored onto Ti3C2O2 by calculating their Gibbs free energies. The authors suggested that end-on N2 adsorption is energetically advantageous and that negative free energies represent outstanding N2 activation properties. Hydrogenations of N2 into *NNH and of *NH2 into NH3 were considered possible potential-limiting steps. In another study, Cheng et al. [135] carried out DFT calculations to investigate the catalytic activity of single transition metal atom (Mo, Mn, Fe, Co, Ni, or Cu)-decorated M2NO2-type MXenes (M = Ti, V, and Cr) for NRR. Mo/Ti2NO2 was screened as a very promising candidate catalyst with a low overpotential of 0.16 eV. This result could be ascribed to the strong bonding strength between Mo and Ti2NO2. Moreover, Mo/Ti2NO2 showed a low Gibbs free energy (0.12 eV) for NH3 desorption, which promotes NH3 release, and exhibited excellent metallic characteristics, which could effectively promote electron transfer between Mo and Ti2NO2. Zheng and co-workers [136] studied the NRR performance of single-atom B-decorated MXenes using DFT calculations. Here, B-doped Mo2CO2 and W2CO2 MXenes showed excellent catalytic activity and selectivity with limiting potentials of − 0.20 and − 0.24 V, respectively (Fig. 11b–d). Hydrogenation of *N2 into *N2H could be facilitated by the high tendency of B-to-adsorbate electron donation. However, conversion of *NH2 into *NH3 was seriously hindered by strong B–N bonding.

Fig. 11
figure 11

a Minimum energy path for N2 conversion into NH3 catalyzed by V3C2 (top) and Nb3C2 (bottom) MXenes. Reproduced with permission [50]. Copyright 2016, Royal Society of Chemistry. Free energy profiles for the NRR catalyzed by group b IV (Ti, Zr, Hf), c V (V, Nb, Ta), and d VI (Cr, Mo, W) MXenes with B centers. Reproduced with permission [136]. Copyright 2019, American Chemical Society

Full size image

Some experiments have been performed to investigate the NRR activity of MXene-based materials. For example, Luo et al. [49] first verified that the central Ti atom in the MXene Ti3C2Tx is the most active site for N2 adsorption (1.34 eV) by comparison with C (− 0.16 eV), O (− 1.21 eV), and lateral Ti (− 0.95 eV) atoms. In addition, the basal plane of MXene is inert relative to edge planes owing to the former’s lower exposure of Ti sites, as shown in Fig. 12a, b. When smaller Ti3C2Tx MXenes were dispersed on vertically aligned metal FeOOH nanosheets, a faradaic efficiency of 5.78% under − 0.2 V versus RHE was obtained; this value is 1.25 times higher than the maximum value obtained from an MXene/stainless steel mesh (4.62%) under − 0.1 V versus RHE, as presented in Fig. 12c, d. Li and co-workers [118] directly applied small-sized (~ 50–100 nm) F-free Ti3C2Tx nanosheets for NRR. The obtained catalyst showed an NH3 yield of 36.9 μg/(h mgcat) and faradaic efficiency of 9.1% at − 0.3 V versus RHE in 0.1 mol/L HCl (Fig. 12e). These values are, once again, much larger than those of F-based MXenes due to the unique size effect and fluorine-free characteristics to the novel catalysts. In another study, Zhao et al. [119] reported that Ti3C2Tx MXene nanosheets could serve as catalysts for NRR. The catalysts achieved an NH3 yield of 20.4 μg/(h mgcat) and a faradaic efficiency of 9.3% at − 0.4 V versus RHE. DFT results demonstrated that the distal NRR mechanism was more favorable, and the related *NH2/NH3 reaction was the rate-determining step. Zhang et al. [117] prepared TiO2/Ti3C2Tx hybrids by using a simple hydrothermal method and studied their catalytic activity for NRR. The obtained hybrids were tested in 0.1 mol/L HCl and showed good catalytic performance with an NH3 yield of 26.32 μg/(h mgcat) and faradaic efficiency of 8.42% at − 0.60 V versus RHE (Fig. 12f, g); these results are believed to originate from the synergistic effect between TiO2 nanoparticles and Ti3C2Tx nanosheets. Kong and co-workers [120] reported that an MnO2-decorated Ti3C2Tx MXene nanohybrid could serve as an electrocatalyst for NRR with excellent durability and outstanding selectivity. This nanohybrid showed a large NH3 yield of 34.12 μg/(h mgcat) and high faradaic efficiency of 11.39% under 0.55 V versus RHE in 0.1 mol/L HCl (Fig. 12h). As shown in Fig. 12i, DFT calculations indicated that unsaturated surface Mn atoms could serve as active sites for adsorption and activation of N2. The first hydrogenation process in this strategy was identified as the rate-determining step.

Fig. 12
figure 12

a Optimized structures of Ti3C2Tx MXenes and the corresponding adsorption energies for N2 on various atomic sites and H2O on the middle Ti atomic site. b–d Faradic efficiencies of a Ti3C2Tx MXene/stainless steel mesh and Ti3C2Tx MXene/FeOOH at different potentials, respectively. Reproduced with permission [49]. Copyright 2018, Elsevier Ltd. e NH3 yields and faradaic efficiencies of F-free Ti3C2Tx nanosheets and Ti3C2Tx/carbon paper at various potentials. Reproduced with permission [118]. Copyright 2019, Royal Society of Chemistry. f NH3 yields and faradaic efficiencies of TiO2/Ti3C2Tx at various potentials. g Amounts of NH3 obtained from carbon paper (CP), TiO2/CP, Ti3C2Tx/CP, and TiO2/Ti3C2Tx/CP at − 0.6 V after 2 h of electrolysis. Reproduced with permission [117]. Copyright 2019, American Chemical Society. h NH3 yields and faradaic efficiencies of MnO2–Ti3C2Tx at various potentials. i Gibbs free energy profiles for NRR over MnO2 (110)–MXene surfaces through the traditional distal pathway. Reproduced with permission [120]. Copyright 2019, Royal Society of Chemistry

Full size image

CO2 Reduction Reaction

Large-scale anthropogenic CO2 emissions cause serious environmental issues, including global warming and extinction of species, among others. Converting CO2 by CO2RR into value-added chemicals and fuels has attracted extensive research attention due to the environment-friendly characteristics of this technology [137, 138].

The electrocatalytic CO2RR activity of MXenes has been explored by using theoretical DFT calculations. For example, Chen et al. [139] studied different −OH terminated MXenes for CO2RR by theoretical calculation and found that Sc2C(OH)2 is a highly promising candidate for catalyzing the CO2RR of CO2 into CH4 with a limiting potential of − 0.53 V. This excellent performance could be attributed to the high reactivity of H atoms in the −OH termination groups of the MXene, which is conducive to the formation of stable structures with intermediates and lowering of the necessary overpotential. MXene catalysts with low charge migration during the potential-limiting step have also been suggested to demonstrate good CO2RR performance. Li et al. [140] predicted that IV–VI series MXenes show excellent performance for CO2 capture. Cr3C2 and Mo3C2 MXenes have been considered highly promising candidates for the selective conversion of CO2 into CH4. The authors also found that the formation process of OCHO· and HOCO· radicals occurs as a spontaneous reaction in the early hydrogenation steps, which was the rate-determining step of CO2 into CH4 conversion process. According to the calculated minimum energy path results, the CO2 → CH4 conversion process over bare Cr3C2 and Mo3C2 required overpotentials of 1.05 and 1.31 eV, respectively (Fig. 13a, b). However, functional group (e.g., −O or −OH)-terminated MXenes (Mo3C2) required very low energy inputs (Fig. 13c, d). In another study, Handoko and co-workers [51] reported that W2CO2 and Ti2CO2 are highly promising M2XO2 MXene candidates for CO2RR owing to their low overpotential and good selectivity. This excellent performance could be attributed to the accessibility of the *HCOOH pathway, which is energetically more favorable compared with *CO pathway. In addition, O termination groups on MXenes help stabilize the reaction intermediates. Thus far, however, no experimental study on MXene-based catalysts for CO2RR has yet been reported.

Fig. 13
figure 13

a Side view of the minimum energy path for CO2 conversion into *CH4 and **H2O catalyzed by Mo3C2. b Minimum energy path for CO2 conversion into *CH4 and **H2O catalyzed by Cr3C2. Minimum energy path for CO2 conversion into CH4 and H2O over c Mo3C2(OH)2 and d Mo3C2O2. Reproduced with permission [140]. Copyright 2019, American Chemical Society

Full size image

Summary and Outlook

As an emerging class of 2D materials, MXenes show tremendous potential in electrochemical energy conversion. In this review, we systematically summarized recent advances in MXenes-based materials in electrocatalysis, including HER, OER, ORR, NRR, and CO2RR. Many high-performance MXenes-based catalysts featuring distinct inherent properties, such as excellent metallic conductivity, rich surface chemistry, and unique morphology, have been prepared. We outlined two common strategies for improving the electrocatalytic property of MXene-based catalysts. First, surface functional groups (e.g., −O, −OH, and –F) and exposed terminal metal sites (e.g., Ti, Mo, Nb, and V) can serve as catalytic activity sites, as verified by theoretical calculations and experiments. Thus, regulating the surface chemistry of these molecules is a promising strategy to enhance the electrocatalytic property of MXenes. Second, constructing nanohybrids with other active components (e.g., nanoparticles, monoatomics, and other 2D materials) is another effective strategy to improve the electrocatalytic performance of MXene-based materials. The surface functional groups of MXenes endow them with the ability to easily form strong interactions with different components. Many metallic MXenes show enhanced charge-carrier transfer properties, and their 2D structure can prevent the active materials from aggregating.

Despite the initial successes obtained from MXene-based electrocatalysts, however, many challenges remain to be solved. For example, more novel MXenes must have been predicted and synthesized by theoretical calculations and experimental methods. The electrocatalytic performance of these materials should also be systematically investigated. The electrocatalytic performance of MXenes-based materials for some applications (e.g., CO2RR) remains mostly theoretical. Thus, experimental studies should be performed to verify the results of theoretical calculations. Moreover, great efforts have been exerted to develop MXene-based catalysts for electrocatalysis, but elucidating the relevant catalytic mechanism has proven to be difficult. Therefore, more advanced characterizations (e.g., in situ microscopy and spectroscopy) and theoretical calculations must be conducted to promote the rational design of MXene-based catalysts.