Metal-organic framework/MXene composites: design strategies and applications for energy storage, catalysis, and sensing

In situ growth through chemical bonding

In situ growth is a widely used method to prepare MOF/MXene composites. In this route, MOF crystals nucleate and grow directly on MXene surfaces via coordination or covalent interactions. This method can achieve tight interfacial contact, maximize charge transfer area, and avoid phase separation caused by physical mixing. MXenes possess abundant -O, -OH, and -F terminal groups derived from etching processes. These hydrophilic groups improve dispersion in polar solvents and can anchor metal ions through electrostatic attraction and coordination bonds^[67-69].

The formation of interfacial chemical bonds during in situ growth is considered important for electron transport between insulating MOFs and conductive MXenes. For example, Co²⁺ and Ni²⁺ ions can coordinate with oxygen-containing groups on Ti₃C₂T_x to form Ti-O-Co or Ti-O-Ni bridges, which may act as electron conduction pathways^[32,33]. These bridging bonds are proposed to reduce charge-transfer resistance and influence redox reaction kinetics. Yang et al. found that low-temperature annealing at 300 °C could regulate the content of Ti-O-Co bonds in Co-BDC/Ti₃C₂T_x composites. Optimized interfacial coupling improved electron transfer and enhanced supercapacitor performance^[32]. However, the exact correlation between bond density and electrochemical performance remains incompletely quantified, and the long-term stability of these interfacial bonds under electrochemical polarization requires further investigation.

Solvothermal and hydrothermal methods are commonly adopted to control the size, morphology, and distribution of in situ grown MOFs. MXene nanosheets serve as heterogeneous nucleation sites to form dispersed MOF nanoparticles or nanosheets instead of large aggregated crystals^[70-73] [Figure 2A]. For instance, bimetallic Ni/Co-MOF nanosheets grown on MXene via solvothermal treatment show uniform coverage and adhesion, and remain stable during electrochemical cycling^[74]. Similarly, Co²⁺ can coordinate with 2-methylimidazole on Ti₃C₂T_x surfaces to anchor ZIF-67 polyhedrons. The obtained hierarchical structures combine the pseudocapacitance of MOFs and the conductivity of MXenes^[15,71]. Nevertheless, hydrothermal conditions can also promote MXene oxidation, particularly when the etching-derived terminal groups are predominantly -F, which may hydrolyze under solvothermal conditions and compromise the conductivity of the MXene substrate.

Figure 2. (A) In situ growth of Fe-THBQ@MoSe₂@MXene composite via interfacial Fe-Se coordination. Adapted from Ref.^[70], under CC BY 4.0 license. (B) Electrostatic self-assembly of Zr-MOF/MXene composite. Adapted from Ref.^[75], under CC BY 4.0 license.

In situ growth also enables the fabrication of 2D/2D heterostructures, in which ultrathin MOF nanosheets align on MXene basal planes. This structure can enlarge the exposed active area and shorten ion diffusion paths. Xiao et al. synthesized 2D/2D Ti₃C₂T_x/NiCo MOF heterostructures, in which carboxyl groups on MOF linkers bonded with MXene terminal groups^[14]. The interface maintained structural integrity during lithium-ion battery cycling. The matched dimensions of the two components optimize electrochemical interfaces and provide mechanical strength. However, the uniform synthesis of ultrathin MOF nanosheets remains challenging, and the control of MOF layer thickness on MXene surfaces is often empirical rather than predictive.

The in situ growth method offers direct interfacial bonding, but its applicability depends on the compatibility between MOF synthesis conditions and MXene stability. MOFs synthesized under strongly acidic or basic conditions may degrade MXene nanosheets, and the formation of dense MOF coatings can block the ion-accessible surface of MXenes^[46,47,75]. Therefore, reaction parameters such as pH, temperature, and solvent polarity must be carefully optimized for each MOF/MXene pair.

Electrostatic self-assembly

Different from in situ growth, which relies on chemical reactions, electrostatic self-assembly constructs MOF/MXene heterostructures through physical interactions between oppositely charged materials. MXene nanosheets carry negative zeta potential in aqueous and organic dispersions due to deprotonated -O^- and fluorine terminals, so they can assemble with positively charged MOF particles and derivatives^[77-79]. This method is suitable for combining pre-synthesized MOFs with specific morphologies and for building complex multilayer structures that are difficult to obtain via direct growth.

Layer-by-layer electrostatic assembly is an effective way to prepare 2D/2D heterostructures, and has been used, for example, to couple pre-formed Zr-MOF nanosheets with MXene while preserving their crystallinity^[75] [Figure 2B]. Alternate deposition of MXene and MOF nanosheets forms ordered superlattices with intimate interfacial contact^[40,80]. Such structures provide active sites and electron pathways, while preventing the restacking of 2D materials. As a typical case, ZIF-67-derived CoNi₂S₄ nanocubes can bind with few-layer Ti₃C₂T_x via electrostatic interaction to form interconnected 3D networks, which facilitate ion and electron transport in supercapacitor electrodes^[81]. Electrostatic interaction ensures distribution of active materials on MXene networks and can enhance rate capability and cycling stability. However, the strength of electrostatic forces is sensitive to electrolyte pH and ionic strength, which may limit the stability of assembled structures in some electrochemical environments.

Vacuum-assisted filtration is a scalable technique to convert electrostatically assembled dispersions into free-standing flexible films with controllable thickness and packing density^[82]. During filtration, negatively charged MXene sheets and positively charged MOF particles stack alternately to form a "brick-and-mortar" structure, which combines the mechanical strength of MXenes and the electrochemical activity of MOFs. These flexible films can be directly used as electrodes without extra binders or conductive additives, increasing volumetric energy density and simplifying device fabrication. The mild conditions of electrostatic assembly preserve the crystallinity and porosity of both components and retain their intrinsic electrochemical properties. A limitation is that the mechanical stability of films under repeated bending or stretching depends on the interlayer adhesion strength, which is weaker than covalent bonding and may lead to delamination under mechanical stress.

Post-synthetic derivation

Post-synthetic strategies refer to the chemical or thermal transformation of prefabricated MOF/MXene hybrids into functional derivatives, including metal oxides, sulfides, phosphides, selenides, and nitrogen-doped carbon. Although pristine MOFs have high surface area and regular structures, their low conductivity and limited chemical stability under electrochemical conditions necessitate further modification. The MXene substrate provides mechanical support during transformation, inhibits nanoparticle aggregation, and maintains electrical connectivity throughout the composite^[83,84].

Thermal annealing under controlled atmospheres is a typical post-synthetic treatment. For example, ZIF-67 grown on MXene can be carbonized under inert atmosphere to obtain Co nanoparticle-embedded nitrogen-doped carbon (Co/NC) supported on Ti₃C₂T_x. The composites integrate the catalytic activity of Co, the conductivity of MXenes, and the carbon matrix^[85]. Phosphorization can convert MOF/MXene precursors into metal phosphides such as CoP and Ni₂P, which exhibit activity for hydrogen evolution and improved lithium storage capacity^[86-90] [Figure 3]. The MXene skeleton maintains structural integrity during high-temperature treatment, prevents porous framework collapse, and maintains contact between active phases and conductive substrates. However, high-temperature processing often induces partial MXene oxidation, particularly for Ti-based MXenes, and the resulting TiO₂ surface layer can increase interfacial contact resistance. The balance between MOF carbonization efficiency and MXene preservation requires precise temperature control.

Figure 3. Post-synthetic derivation of CoP@C/Ti₃C₂T_x composite. Adapted from Ref.^[90], under CC BY 4.0 license.

Sulfidation and selenation are also used to convert MOF/MXene precursors into metal chalcogenides with modified electrochemical performance. For example, thermal treatment of ZIF-67/Ti₃C₂T_x hybrids with sulfur sources yields Co₉S₈/MXene and Ni₃S₄/CuS/MXene sandwich structures, which deliver higher specific capacity and rate performance for supercapacitors and batteries^[83,91,92]. These derivatives usually have hierarchical porous structures to buffer volume expansion during ion insertion, while MXenes provide continuous electron pathways and mechanical reinforcement. It should be noted that the sulfidation process often requires sulfur vapor or thiourea at elevated temperatures, which may etch the MXene surface and alter its terminal group chemistry. In general, post-synthetic derivation bridges the gap between the high surface area of MOFs and the practical requirements for conductivity and stability in electrochemical devices, but the structural complexity of the final product makes it difficult to establish precise structure-performance correlations.

Interface engineering

Beyond conventional growth and assembly methods, interface engineering is developed to regulate the interaction between MOFs and MXenes, so as to optimize electronic structures and electrochemical kinetics. Surface modification of MXenes before MOF loading is an approach to improve interfacial compatibility and charge transfer efficiency. For example, ammonia annealing induces nitrogen doping on Ti₃C₂T_x, which enlarges interlayer spacing and improves surface wettability^[93,94]. These changes promote uniform MOF growth and enhance ion accessibility.

Introducing functional interlayers between MOFs and MXenes is another strategy to modulate interfacial properties. Singha et al. inserted ultrathin hexagonal boron nitride (h-BN) between ZIF-67-derived carbon and MXene to construct MOF-MXene-BN heterostructures^[38]. The dielectric interlayer optimizes charge injection and avoids direct chemical reaction between MOF derivatives and MXenes, thus improving the cycling stability of supercapacitors. Similarly, amination treatment endows MXene surfaces with positive or neutral charges, which improves the wettability and adhesion of MOF coatings. This method has been applied to fabricate Ni/Co-MOF@TCT-NH₂ composites with enhanced electrochemical performance^[74]. However, the addition of interlayers inevitably increases the overall internal resistance of the electrode and may reduce the volumetric energy density.

Other chemical functionalization methods include creating defects and vacancies on MXene surfaces to serve as MOF nucleation sites and pre-anchoring metal ions on MXene edges to guide oriented MOF crystallization. These refined engineering methods enable regulation of interfacial electronic states, induce charge redistribution, and can boost the intrinsic activity of catalytic sites^[34,58]. Rational interface design can make MOF/MXene composites exceed the simple performance superposition of individual components and approach the theoretical limit of synergistic interaction. Nevertheless, many interface engineering strategies involve multi-step processing with low yields, which complicates scale-up. The long-term stability of engineered interfaces under continuous electrochemical operation remains insufficiently documented.

The role of interfacial bonding in charge transfer

Interfacial M-O-Ti bonds (M = Co, Ni, Zn, Zr) are frequently invoked to explain the faster charge transfer observed in MOF/MXene composites. The experimental basis for bond formation is comparatively strong. X-ray photoelectron spectroscopy (XPS) consistently reveals shifts in the Ti 2p, O 1s, and metal 2p binding energies that point to Ti-O-Co, Ti-O-Ni, Ti-N, and Zr-O coordination environments, as documented for Co-BDC/Ti₃C₂T_x^[32], NiCo hydroxide grown on MXene^[13], and the 2D/2D CoBDC/Ti₃C₂T_x heterostructure^[33]. Density functional theory (DFT) calculations complement these observations by showing charge-density redistribution at MOF/MXene junctions and modified adsorption energetics of reaction intermediates such as H*, OH*, Li⁺, Na⁺, and Zn^{2+[19,42,43,58]}. Hierarchical electron-coupling engineering in F-doped NiFe layered double hydroxide (LDH)/MOF-74 grown on Ti₃C₂T_x has been shown by combined XPS and DFT analysis to redistribute interfacial charge and optimize the adsorption of oxygen-containing intermediates, directly correlating interfacial coupling with enhanced overall water splitting^[95]; NiCoP/Ti₃C₂T_x interfaces were further calculated to lower the kinetic energy barrier of alkaline hydrogen evolution^[96]. These electronic effects are accompanied by a measurable reduction in charge-transfer resistance relative to physical mixtures or bare MOFs^[19,32,33].

At the same time, the causal weight assigned to interfacial bonds is less securely established. A quantitative relationship between bond density and electrochemical performance has not been demonstrated: most studies report one or a few annealing or growth conditions together with a performance optimum, without independently quantifying the number or density of M-O-Ti linkages. The frequently repeated claim that a higher density of M-O-Ti bonds accelerates electron transfer and thereby improves capacity or activity therefore remains a plausible hypothesis rather than an established mechanism. Even the XPS evidence requires caution, because peak deconvolution in the Ti 2p, C 1s, and O 1s regions of Ti₃C₂T_x is prone to misassignment, and published spectra frequently disagree on component attribution^[97]. These considerations indicate that interfacial M-O-Ti bonding is best regarded as a well-supported contributor to charge transfer whose quantitative structure-performance relationship and long-term stability under polarization remain open questions.

Taken together, the four strategies offer complementary trade-offs. In situ growth gives the strongest interfacial bonding and the most uniform MOF dispersion, but is sensitive to solvothermal conditions, risks MXene oxidation, and is batch-limited. Electrostatic self-assembly is mild and preserves the crystallinity of both components, yet yields weaker physical adhesion that is sensitive to electrolyte pH and ionic strength. Post-synthetic derivation raises intrinsic conductivity and activity most effectively, but sacrifices MOF crystallinity and porosity, risks MXene oxidation, and complicates identification of the true active phase. Interface engineering offers the finest electronic tuning and can exceed simple performance superposition, but is multi-step, low-yield, and costly. In practice the four routes are often combined to balance interfacial strength, activity, and manufacturability^[35]. Table 1 summarizes the key advantages, main drawbacks, and typical applications of these strategies.

Batteries

Before discussing individual battery types, several common design principles emerge from the literature. First, the conductive MXene network is indispensable for compensating the poor electronic conductivity of MOF-derived active materials, ensuring efficient charge transfer during redox reactions. Second, interfacial engineering, whether through chemical bonding between MXene surface terminations and MOF linkers or through heterojunction creation, plays a critical role in stabilizing electrode structures during volume changes accompanying ion insertion. Third, structural hierarchy spanning from nanoscale void spaces to microscale electrode configurations must be carefully orchestrated to accommodate ion size while maintaining mechanical integrity. These principles manifest differently across lithium, sodium, potassium, and zinc-based systems, as detailed below.

Lithium-ion batteries (LIBs)

The limited performance of conventional graphite anodes—theoretical capacity of 372 mAh g^-1, sluggish Li⁺ redox kinetics, and irreversible structural degradation—remains a core bottleneck for LIB advancement. MOF/MXene hybrid nanostructures address these limitations by integrating the high surface area, customizable pore channels, and redox-active metal sites of MOFs with the metallic conductivity, hydrophilic terminated surfaces, and mechanical ductility of MXenes [Figure 4].

Figure 4. Synthesis strategies of MOF/MXene composites for lithium-ion battery anodes. (A) In situ growth of ZnS nanodots/Ti₃C₂T_x MXene hybrids via ZIF-8/MXene precursor followed by sulfidation. Adapted from Ref.^[62], under CC BY 4.0 license. (B) Hydrogen-bonded self-assembly of 3D porous Ti₃C₂/NiCo-MOF composite. Adapted from Ref.^[105], under CC BY 4.0 license.

In situ growth is the dominant preparation technique for LIB anodes, establishing chemical interfacial coupling between MXene substrates and MOF-based active phases. Gao et al. proposed an in situ conversion strategy derived from LDH precursor growth on MXene sheets, transforming surface-confined LDH into pristine Ni_0.8Mn_0.2 Prussian blue analog (PBA) nanoparticles anchored on Ti₃C₂T_x^[49]. This in situ nucleation mechanism ensures contact between electroactive PBA and the conductive MXene network, delivering a reversible capacity of 442 mAh g^-1 with high-rate tolerance. Similarly, Xiao et al. fabricated 2D/2D Ti₃C₂T_x/NiCo MOF heterostructures via solvothermal in situ growth, where carboxyl groups on MOF organic linkers form covalent bonds with -OH/-O terminals of MXene, generating heterointerfaces^[14]. This interfacial interaction accelerates charge transfer and sustains a capacity of 637 mAh g^-1 over 200 cycles. However, these studies are typically conducted at low active mass loadings (< 1 mg cm^-2), and performance at industrially relevant loadings (> 3 mg cm^-2) remains largely unreported.

A critical challenge for alloy-type anode materials (Si, Sn, Sb) lies in extreme volume fluctuation (> 300%) during Li⁺ uptake/release, which can be mitigated via spatial confinement enabled by post-synthetic-derived carbon frameworks combined with MXene scaffolds. Zhang et al. developed a dual-confinement architecture via MOF carbonization post-treatment: ZIF-8 was first grown in situ on MXene surfaces, followed by high-temperature pyrolysis to form nitrogen-doped carbon matrices encapsulating Si nanoparticles, yielding MOF-derived Si/MXene@C hybrids^[44]. The MOF-derived carbon buffer layer together with flexible MXene sheets jointly dissipates mechanical strain while retaining conductive pathways, achieving a stable capacity of 1,006.1 mAh g^-1 over 150 cycles. Similar strategies have been applied to the pristine Sn-MOF/MXene anode (1,009 mAh g^-1)^[43] and the MOF-derived Sb@C/MXene anode (625 mAh g^-1 over 2,000 cycles)^[17], demonstrating the generalizability of this dual-confinement approach. A related conversion-type design, MOF-derived ZnS nanodots anchored on Ti₃C₂T_x, likewise achieved stable lithium storage through MXene-confined conversion^[62]. Notably, these reversible capacities closely approach—and in the Sn case slightly exceed—the theoretical alloying limits of the metals alone (≈994 mAh g^-1 for Li_4.4Sn and ≈660 mAh g^-1 for Li₃Sb)^[106,107]. The apparent excess for the Sn-based anode is attributable to additional Li⁺ storage in the pristine MOF framework (redox-active organic linkers and unsaturated metal nodes), the MXene substrate, and interfacial/pseudocapacitive contributions, and should not be interpreted as exceeding the thermodynamic limit of the full composite.

Beyond the three primary composite strategies, macroscopic structural modulation further optimizes electrode performance. Hydrogen-bonded self-assembly of NiCo-MOF with Ti₃C₂ into a three-dimensional porous network represents another macroscopic strategy for enhancing Li⁺ storage^[105]. Electrospinning constructs one-dimensional fibrous conductive networks; Chen et al. fabricated Ti₃C₂T_x@Co fibrous composites that retain a capacity of 422 mAh g^-1 after 500 cycles^[48]. Utilizing intrinsically conductive MOFs provides another pathway: Sun et al. built conductive MOF@MXene heterostructures with modulated interfacial electron redistribution, achieving 80% capacity retention after 5,000 cycles^[108]. However, the library of intrinsically conductive MOFs remains limited primarily to Cu-, Ni-, and Co-based 2D frameworks, and their chemical stability in carbonate electrolytes is a concern.

The multiple fabrication routes endow MOF/MXene anodes with four key merits: (i) enhanced electronic conductivity via metallic MXene scaffolds; (ii) improved mechanical tolerance toward volume expansion through spatial confinement; (iii) robust interfacial stability from chemical bonding formed during in situ growth; and (iv) tunable Li⁺ storage kinetics via interface engineering and conductive MOF integration. A critical gap is the lack of standardized testing protocols: most studies report gravimetric capacity at low mass loadings, whereas practical cells require areal capacity > 3 mAh cm^-2 with electrode thicknesses > 50 μm.

Sodium-ion batteries (SIBs)

SIBs have garnered interest as a cost-effective alternative to lithium-ion technologies, driven by abundant sodium reserves. However, the larger ionic radius of Na⁺ (1.02 vs. 0.76 Å for Li⁺) induces structural stress, sluggish reaction kinetics, and substantial volume expansion during sodiation/desodiation^[109] [Figure 5].

Figure 5. Synthesis strategies of MOF/MXene composites for sodium-ion battery anodes. (A) In situ growth of CoSe₂@CNTs-MXene via ZIF-67/MXene precursor followed by carbonization and selenization. Adapted from Ref.^[110], under CC BY 4.0 license. (B) In situ growth of nitrogen-doped carbon nanotubes on MXene nanosheets (MXene@CNTs) via ZIF-67/MXene precursor and uric acid-assisted carbonization. Adapted from Ref.^[111], under CC BY 4.0 license.

A critical advancement in MOF/MXene composite design for SIBs involves hollow hierarchical architectures that buffer mechanical stress. Li et al. fabricated MOF-derived hollow MXene@CoS₂/NC composites using MOF/MXene templates followed by nitrogen-doped carbon encapsulation^[16]. The hollow interior accommodates volume expansion of CoS₂, while the MXene skeleton ensures continuous electron transport, enabling a reversible capacity of 620 mAh g^-1 at 0.2 A g^-1 over 200 cycles. However, the initial Coulombic efficiency of such conversion-type anodes is typically below 80%, and the origin of this irreversibility warrants further investigation. Transition metal phosphides encapsulated within MOF-derived carbon matrices and MXene networks offer alternative high-capacity anodes. Liu et al. developed MOF-derived MXene@CoP@NPC composites through phosphorization of ZIF-67 precursors anchored on MXene surfaces^[87]. This architecture delivered a stable capacity of 198 mAh g^-1 at 5 A g^-1, demonstrating rate capability essential for high-power applications. Nevertheless, phosphorization processes typically require toxic phosphorus sources, raising safety and environmental concerns for scale-up. Huang et al. fabricated MOF-derived Ti₃C₂/CoS₂@NPC composites, achieving a capacity of 282.6 mAh g^-1 even at 10 A g^-1, underscoring the role of pore architecture in facilitating rapid Na⁺ diffusion^[78]. A dimensional-gradient CoSe₂@CNTs-MXene anode prepared from MOF precursors further demonstrated high-capacity, stable sodium storage in ether-based electrolytes^[110]. Shi et al. introduced a mechanochemical synthesis approach for preparing MOF/MXene composites, representing a sustainable pathway for large-scale manufacturing^[112]. However, the uniformity of product distribution on MXene surfaces remains less controlled compared to liquid-phase methods. In situ growth of nitrogen-doped CNTs on MXene from ZIF-67 has extended the same conductive-framework concept to dual sodium/potassium storage^[111].

A recurring limitation across SIB studies is the use of ether-based electrolytes, which form stable solid electrolyte interphase layers but exhibit lower ionic conductivity than carbonate electrolytes, potentially masking true rate capability^[113]. Furthermore, MOF/MXene anodes must be evaluated in full-cell configurations against suitable cathodes to assess true practical viability.

Zinc-based batteries and potassium-ion batteries

MOF/MXene heterostructures address the inherent drawbacks of single-component materials for multivalent and large-radius alkali metal-ion batteries^[9]. Pristine MOFs suffer from poor conductivity and fragile frameworks, while pure MXene sheets are limited by insufficient electroactive sites and restacking. The heterointerface shortens ion migration pathways, buffers volumetric variation, and suppresses parasitic side reactions—core bottlenecks restricting cycling lifespan [Figure 6].

Figure 6. Synthesis strategies of MOF/MXene composites for potassium-ion batteries and zinc-based batteries. (A) In situ growth of CoHCC@MX composite via hydroxylated MXene-assisted synthesis for potassium-ion batteries. Adapted from Ref.^[63], under CC BY 4.0 license. (B) In situ growth of bi-ligand ZIF-67 on MXene followed by pyrolysis to form b-CZIF67/MXene composite for zinc-air batteries. Adapted from Ref.^[114], under CC BY 4.0 license.

Zinc-ion batteries. Aqueous zinc-ion batteries face three obstacles: uncontrolled zinc dendrite nucleation, competitive hydrogen evolution, and irreversible dissolution of cathode active species. For anode protection, Watwiangkham et al. combined DFT and molecular dynamics simulations to design Ti₃C₂O₂/CuHHB heterointerfaces with strong Zn adsorption affinity and HER suppression. Theoretical calculations verified that the Ti₃C₂O₂/CuHHB heterointerface possesses stronger Zn adsorption affinity (0.74 eV higher than water adsorption) and lifts the HER limiting potential to -2.41 V, which theoretically inhibits dendrite growth and parasitic hydrogen evolution^[42]. Nevertheless, this work only relies on computational prediction, lacking electrochemical validation to confirm long-term interfacial stability under practical cycling conditions. Different from pure theoretical design, microfluidic 3D printing offers a controllable manufacturing method to construct functional protective layers. Lu et al. deposited MXene/Cu-THBQ composite coatings on zinc powder anodes via microfluidic-assisted shaping. This uniform covering homogenizes Zn²⁺ flux distribution and suppresses HER activity, allowing symmetric cells to operate stably for 1,800 h and matching well with organic cathode materials^[115]. Despite precise control over coating thickness and component distribution, microfluidic printing suffers from high equipment cost and low batch output, which hinders its large-scale industrial promotion.

In terms of zinc-ion battery (ZIB) cathode materials, sandwich-type MOF/MXene heterostructures and MOF-derived carbon-supported metal oxides are two mainstream architectures. Wang et al. synthesized a 2D layered pristine Cu-HHTP/MX composite with inherent one-dimensional ion transport tunnels^[99]. The MXene substrate remedies the low conductivity of pristine Cu-HHTP MOF, enabling reversible Zn²⁺ intercalation redox reactions. The electrode delivers a specific capacity of 260.1 mAh g^-1 at 0.1 A g^-1 and retains 92.5% initial capacity after 1,000 cycles. To further optimize full-cell performance, Sun et al. exploited post-synthetic derivation to fabricate MOF-derived Zn/Co co-doped MnO/C microspheres from bimetallic MOF precursors, which were assembled with Ti₃C₂Cl₂-modified zinc anodes to construct complete ZIB devices. Dual metal doping relieves Jahn-Teller structural distortion of Mn-based active phases, while the MXene coating on anodes restrains dendrite proliferation. The full cell achieves a capacity retention rate of 98.7% over 3,000 charge-discharge cycles^[100]. Hierarchically porous manganese-based ternary composites point to the same conclusion—engineered porosity is central to reversible Zn²⁺ storage—while sharing the same high-loading challenges^[116]. A universal deficiency of these MOF-derived cathodes is the capacity decay under high active mass loading; additionally, the high-temperature pyrolysis procedure required for post-synthetic derivation may partially destroy the original porous MOF framework and reduce the number of accessible active sites.

Potassium-ion batteries. Compared with aqueous ZIBs, non-aqueous potassium-ion batteries suffer from more severe volume expansion because K⁺ possesses a larger ionic radius than Zn²⁺ and Li^+[117]. MOF/MXene composites fabricated via modified in situ growth and dual-confinement strategies have become competitive potassium-ion battery (PIB) anode candidates. For instance, hydroxylated-MXene-assisted in situ growth of ultrafine CoHCC nanoparticles produced a high-performance potassium-ion anode^[63]. Sun et al. proposed a "node-unlocking" interface engineering strategy based on MXene-supported MOF precursors^[50]. Selective removal of partial organic linkers exposes abundant metal node sites on NMD-MOF composites, which elevates K⁺ adsorption energy and facilitates ion capture. When matched with MD-KVO cathodes, the NMD-MOF anode maintains a reversible capacity of 250 mAh g^-1. For conversion-type anode materials with large volumetric strain, Hu et al. adopted an in situ anchoring route to disperse MOF-derived ZnTe@C 0D nanoparticles over Ti₃C₂T_x MXene nanosheets, constructing 0D/2D dual-confinement heterostructures^[45]. The MXene matrix and carbon buffer shell jointly accommodate the structural deformation generated during the reversible conversion reaction (ZnTe ↔ Zn + K_xTe_γ). The composite reserves 230.2 mAh g^-1 after 3,500 continuous cycles. However, all conversion-type MOF/MXene anodes face two common drawbacks: low initial Coulombic efficiency and voltage hysteresis, originating from the slow kinetics of multi-electron redox conversion processes.

Collectively, MOF/MXene heterostructures bring performance improvements to both zinc-based and potassium-ion batteries, yet multiple practical bottlenecks remain unresolved. The intrinsic microporous structure of MOFs blocks the rapid diffusion of large-sized K⁺ and hydrated Zn²⁺; meanwhile, spontaneous restacking of MXene nanosheets during electrode rolling reduces the exposure of heterointerface active sites. Most existing synthesis routes are limited by low productivity and high raw material cost. A further concern is the electrolyte compatibility: non-aqueous PIBs typically use KPF₆ in carbonate solvents, which can react with the surface terminations of MXenes and form resistive surface layers, while aqueous ZIBs operate in mildly acidic electrolytes that may promote MXene dissolution.

Supercapacitors

Market demand for fast-charging, long-lifespan energy storage devices has spurred research on advanced supercapacitor electrode materials. MOF/MXene heterostructures have emerged as a promising direction^[118-120]. Conductive MXene sheets act as interconnected current collectors establishing long-range electron transport pathways, offsetting the insulating characteristic of MOF matrices and inhibiting MOF particle aggregation^[121-125]. Meanwhile, rigid porous MOF crystals serve as nanoscale spacers between MXene layers, reserving interlayer space for electrolyte permeation and rapid ion intercalation. Interfacial chemical bonds such as Ti-O-Co and Ti-O-Ni are proposed to trigger charge redistribution across heterojunctions, lowering ion desolvation energy and enhancing structural robustness^[13,32] [Figure 7].

Figure 7. Synthesis strategies of MOF/MXene composites for supercapacitors. (A) Schiff reaction-induced self-assembly of NH₂-Ni/Co-BDC MOF@Ti₃C₂T_x MXene heterostructure. Adapted from Ref.^[126], under CC BY 4.0 license. (B) In situ growth of Ni-MOF nanorod arrays on Ti₃C₂T_x nanosheets. Adapted from Ref.^[127], under CC BY 4.0 license.

In situ hydrothermal growth represents the most widely adopted preparation strategy. This route leverages abundant -O, -OH, and -F functional groups on etched MXene surfaces as natural heterogeneous nucleation sites, triggering direct crystallization of MOF nanoparticles or nanosheets on MXene substrates and forming stable covalent bridging bonds at heterointerfaces^[128]. The chemical adhesion eliminates loose phase separation frequently observed in mechanically mixed MOF/MXene powders and guarantees uniform dispersion of redox-active MOF phases over conductive MXene networks. Guo et al. proposed a controllable in situ anchoring synthesis followed by alkaline etching treatment to fabricate MOF-derived MXene-supported NiCo LDH composites derived from Ni-doped ZIF-67 precursors^[13]. XPS-identified Ti-O-Ni interfacial bonds anchor the hydroxides onto negatively charged MXene sheets; the composite exhibits fast cross-interface electron transfer and stable structural buffering during redox cycling, consistent with—though not proving—a bond-mediated pathway. Benefiting from this optimized interfacial coupling, the MXene/NiCoZDH electrode delivers a specific capacitance of 877 F g^-1 at 1 A g^-1 and retains 90.9% of initial capacitance after 30,000 continuous charge-discharge cycles^[13]. Ni-MOF nanorod arrays grown directly on Ti₃C₂T_x provide freestanding electrodes whose oriented one-dimensional morphology shortens ion-diffusion paths^[127]. To systematically reveal metal cation effects on capacitive performance, Ji et al. prepared a series of pristine H₃BTC-linked transition metal MOFs (Fe-, Cu-, Co-, Ni-MOF) directly grown on MXene surfaces via identical hydrothermal parameters^[71]. Comparative electrochemical characterization combined with DFT theoretical calculations verified that Ni-MOF@MX-2 exhibits the strongest interfacial electronic coupling effect, which accelerates charge transfer kinetics and suppresses structural pulverization. This composite achieves a high capacitance of 1,160.5 F g^-1 at 1 A g^-1 with 94% capacity retention after 10,000 cycles. Further interfacial regulation based on hard-soft acid-base theory was realized by Zheng et al., who constructed pristine pillared Ni-MOF/Ti₃C₂ composites^[15]. These studies demonstrate that in situ hydrothermal growth enables strong interfacial bonding and uniform MOF distribution, yet the process is inherently batch-limited and the MOF loading density is often difficult to control independently of crystal size. Additionally, the hydrothermal environment can promote partial oxidation of MXene surfaces, particularly for Ti₃C₂T_x etched with HF, which introduces TiO₂ domains that increase interfacial contact resistance^[74,129]. To circumvent this, Schiff-reaction-induced self-assembly has been used to grow NH₂-functionalized Ni/Co-BDC MOFs on Ti₃C₂T_x under non-hydrothermal conditions^[126].

To overcome the poor structural ordering and low batch productivity of traditional hydrothermal batch synthesis, microfluidic assembly and coaxial electrospinning have emerged as advanced continuous manufacturing techniques, providing precise morphological control over fiber-shaped and freestanding flexible supercapacitor electrodes. These processing methods expand the practical application scope of MOF/MXene composites toward wearable and textile energy storage devices. Zhang et al. pioneered microfluidic channel synthesis to fabricate core-shell ZIF-67@Ti₃C₂T_x fibrous electrodes^[128]. Within confined microfluidic flow channels, MXene nanosheets self-assemble into continuous conductive fiber skeletons, while ZIF-67 polyhedrons uniformly anchor on fiber surfaces through in situ coordination reactions. The regularly arranged pore channels inside fibers shorten ion diffusion distances and realize rapid cross-component electron transmission. The resultant fiber electrode reaches an ultrahigh volumetric capacitance of 972 F cm^-3 and maintains 90.8% performance after 20,000 bending cycles, far outperforming conventional slurry-cast rigid electrodes. Building upon this microfluidic framework, Wu et al. designed core-sheath ZIF-L(Zn)@Ti₃C₂T_x fibers with Ti-O-Zn and Ti-F-Zn dual interfacial bonds^[39]. The flexible fibrous network can be woven into stretchable capacitive fabrics, delivering a volumetric capacitance of 1,700 F cm^-3 with exceptional cyclic endurance over 20,000 runs. Coaxial electrospinning serves as another complementary scalable route to construct self-supporting electrode films without extra polymer binders. Pathak et al. prepared hollow MXene-containing carbon nanofibers via coaxial spinning, sequentially modifying fiber surfaces with polypyrrole conductive layers and immobilized ZnCoMOF nanoparticles^[118]. The interconnected three-dimensional conductive network eliminates MXene restacking and establishes dual transport channels for ions and electrons. However, the mass loading of active MOF materials in microfluidic and electrospun fibers is typically lower than in slurry-cast electrodes, which limits the overall areal capacitance. The trade-off between fabrication scalability and energy density remains a key engineering challenge for these advanced processing methods.

Post-synthetic thermal and chemical treatment converts as-prepared MOF/MXene precursors into high-conductivity metal compounds including oxides, hydroxides, sulfides, and phosphides, while inheriting the original hierarchical porous morphology of MOF templates. This strategy fundamentally addresses the intrinsic low conductivity of pristine MOF frameworks. Yang et al. adopted mild low-temperature thermal annealing to modulate Ti-O-Co bridging bond density within Co-BDC/Ti₃C₂T_x 2D/2D heterostructures^[32]. Controlled heating optimizes interfacial electronic coupling without destroying the original layered heterostructure, generating substantially improved capacitive performance compared with unannealed raw composites. Multi-step bottom-up self-assembly and top-down chemical etching enable precise construction of porous nickel phosphate supported on MXene substrates, as reported by Zhang et al.^[84]. MOF-derived nickel phosphate possesses high theoretical pseudocapacitance, yet its poor bulk conductivity limits practical performance; the underlying MXene conductive network solves this bottleneck and reserves hierarchical pores formed during etching, delivering a charge capacity of 639 C g^-1 with 85% retention after 10,000 cycles. In situ sulfurization of MOF/MXene hybrids is widely used to synthesize metal sulfide active phases with abundant redox states. Ruan et al. converted MXene-supported ZIF-67 into dendritic CoNi₂S₄ wrapped by continuous MXene sheets via high-temperature sulfur vapor treatment^[92]. The dendritic nanostructure maximizes exposed redox-active surfaces, while MXene wrapping guarantees rapid charge collection and mechanical protection, reaching a specific capacity of 933 C g^-1. CoS/Ti₃C₂ heterostructures were further shown to accelerate redox kinetics for high-rate supercapacitor electrodes^[65]. Ion exchange combined with sulfidation further creates advanced yolk-shell structured NiCo₂S₄@Co₃S₄ nanocages anchored on Ti₃C₂T_x substrates for asymmetric supercapacitor positive electrodes^[79]. These derivatives demonstrate that post-synthetic conversion can significantly boost conductivity, yet the trade-off is a loss of crystalline MOF porosity and a potential increase in MXene oxidation. The structural complexity of the final product also makes it difficult to establish precise structure-performance correlations, as the electrochemical response may originate from multiple components (residual MOF, derived metal sulfide, oxidized MXene, and carbon matrix) rather than a well-defined interface^[130].

Unlike traditional insulating MOFs that demand carbonization or conductive additives, π-conjugated conductive MOFs deliver intrinsic in-plane electron transport capacity, simplifying electrode composition and elevating volumetric capacitance. Zhu et al. fabricated MXene@Ni-HHTP composite electrodes by assembling two-dimensional conductive Ni-hexahydroxytriphenylene MOF nanosheets with Ti₃C₂T_x substrates^[18]. The fully conjugated ligand framework of Ni-HHTP offers continuous intramolecular conductive channels, while MXene layers suppress mutual restacking of conductive MOF sheets and enhance overall mechanical toughness. This hybrid material realizes a capacitance of 416.6 F g^-1 when matched with capacitive negative electrodes in asymmetric devices and exhibits rate tolerance under high current loads. To further regulate interfacial charge injection dynamics, Singha et al. inserted ultrathin h-BN dielectric nanosheets as intermediate layers between MOF-derived carbon and MXene phases, constructing ternary MOF-MXene-BN heterostructures^[38]. Distribution of relaxation times characterization confirms that the h-BN interlayer optimizes charge migration and inhibits direct chemical corrosion between the two functional components. The sophisticated ternary heterostructure achieves a high charge capacity of 798 C g^-1 and an energy density of 81 Wh kg^-1, retaining 81% capacitance after 5,000 cycles^[38]. Recent work further extends these designs to multifunctional electrodes combining supercapacitive storage with electrocatalytic CO₂RR and to bimetallic Ni-Mo-MOF electrodes serving both supercapacitors and the OER^[55,56]. Device-level demonstrations are also emerging: MXene-integrated MnCo bimetallic MOF hybrids retain capacitive performance in solid-state configurations, and template-assisted NiCo-MOF/MXene nanocomposites operate in both symmetric and asymmetric hybrid devices^[131,132]. However, the practical popularization of conductive MOF/MXene composites is restrained by two key barriers: the limited library of available conductive MOF species (primarily Ni-HHTP, Cu-HHTP, and a few Fe-based frameworks) and complicated ligand synthesis procedures that increase material preparation difficulty and raw material cost. Furthermore, the chemical stability of conductive MOFs in aqueous acidic or alkaline electrolytes is generally inferior to that of conventional ZIFs or carboxylate MOFs.

Several technical bottlenecks still hinder the commercial translation of MOF/MXene supercapacitor electrodes. First, the dominant microporous structure (< 2 nm) of most MOF building blocks impedes rapid diffusion of large-sized solvated ions in organic and ionic liquid electrolytes, generating high interfacial charge transfer resistance and deteriorating high-rate performance. This issue is particularly acute for asymmetric devices using ionic liquid electrolytes, where the solvated ion size can exceed 1 nm, approaching the micropore limit of typical MOFs. Second, even with MOF nanospacers inserted between MXene layers, weak van der Waals attraction still drives partial restacking of MXene nanosheets during electrode rolling and long-term cyclic operation, gradually reducing the number of available heterojunction active sites. Moreover, most reported high-performance electrodes are tested at low active mass loadings (< 2 mg cm^-2) in laboratory conditions; performance attenuation becomes obvious when scaling up electrode thickness to meet practical device energy demands^[133]. A systematic evaluation of MOF/MXene supercapacitor electrodes at loadings > 5 mg cm^-2 and in pouch-cell configurations is largely absent from the current literature. Fourth, the cost of MOF precursors and the complexity of MXene etching and delamination remain significant barriers to industrial-scale manufacturing. High-purity MXene production currently requires multi-step liquid-phase exfoliation with hazardous HF or LiF/HCl mixtures, which raises environmental and safety concerns for factory-scale implementation.

Electrochemical catalysis

Driven by the demand for low-cost, durable electrocatalysts to advance renewable energy conversion technologies, MOF/MXene heterostructures have emerged as a material platform to substitute scarce noble metal catalysts such as Pt, RuO₂, and IrO₂^[134]. Single-component electrocatalysts suffer inherent performance bottlenecks that can hardly be overcome independently. Rational hybridization of MOFs and MXenes constructs complementary heterointerfaces and unlocks synergistic catalytic effects unavailable to single-phase materials^{[19,33,58,135]}. MXene nanosheets act as interconnected conductive skeletons to establish uninterrupted electron highways and suppress MOF particle agglomeration, while porous MOF frameworks provide exposed catalytic centers and channels for electrolyte penetration and reactant mass transfer. Interfacial interactions ranging from electrostatic attraction to covalent bonding modulate the electronic structure of active sites and can lower the activation energy for rate-determining steps.

As summarized in Section "STRATEGIES OF MOF/MXene COMPOSITING", each synthesis route delivers distinct interfacial bonding states, morphological characteristics, and catalytic advantages for MOF/MXene composites. In situ liquid-phase growth directly induces MOF nucleation on MXene surfaces relying on surface -O, -OH, -F anchoring groups, achieving tight solid-liquid derived heterointerfaces^[33,67,69]. Zhao's group adopted an interfacial diffusion-controlled strategy to grow 2D CoBDC nanosheets uniformly on Ti₃C₂T_x substrates, realizing homogeneous heterostructure construction without phase separation^[33]. Electrostatic self-assembly takes advantage of negatively charged MXene sheets and positively charged pre-synthesized MOF nanosheets to build ordered 2D/2D superlattices with maximized contact area^[80]. Post-synthetic thermal or chemical conversion represents the most widely adopted method to boost intrinsic catalytic activity: MOF/MXene precursors are transformed into metal phosphides, sulfides, or metal-carbon composites while retaining the original hierarchical framework^[24,136]. Zong et al. converted ZIF-67/Ti₂NT_x hybrids into CoP/MXene electrocatalysts via low-temperature phosphorization, while Murtaza's team realized CoP/C@Nb₂CT_x heterostructures through sequential carbonization and phosphorization treatments^[23,89]. Supplementary advanced interfacial tuning tactics include electrochemical deposition for ultrathin MOF film coating and dual-ligand modification to strengthen MOF structural flexibility after compositing with MXene substrates^[137]. However, a critical limitation of all four strategies is that the resulting catalytic sites are often heterogeneous and poorly characterized: the active phase may include residual MOF fragments, derived metal compounds, MXene edge sites, and interfacial coordination complexes, making mechanistic attribution difficult [Figure 8].

Figure 8. Synthesis strategies of MOF/MXene composites for electrocatalysis. (A) Post-synthetic sulfidation of ZIF-67/MXene precursor to form CoS@C/MXene nanocomposites at different temperatures. Adapted from Ref.^[138], under CC BY 4.0 license. (B) In situ growth of CoNi-ZIF-67 rhombic dodecahedrons on Ti₃C₂T_x MXene via a coprecipitation reaction. Adapted from Ref.^[139], under CC BY 4.0 license. (C) Electrostatic adsorption of MXene on nickel foam (NF) followed by hydrothermal growth of Ni-doped Co-MOF-74 to form CoNi-MOF-74/MXene/NF. Adapted from Ref.^[64], under CC BY 4.0 license.

MOF/MXene heterostructures exhibit catalytic efficiency toward diverse electrocatalytic reactions, starting with the ORR critical to fuel cells and rechargeable metal-air batteries. Commercial Pt-based ORR catalysts face high cost and methanol crossover poisoning issues, whereas MOF/MXene hybrids with atomically dispersed M-N_x active sites deliver comparable catalytic activity and superior stability. Wang et al. fabricated MOF-derived Fe-N-C@Ti₃C₂T_x composites via stepwise pyrolysis of Fe-doped ZIF-8 grown on MXene surfaces^[19]. Strong heterojunction coupling reduces charge transfer resistance, affording half-wave potentials of 0.887 V in alkaline media and 0.777 V in acidic electrolytes, with negligible activity decay after over 10,000 continuous potential cycles^[19]. Gu and coworkers anchored NH₂-MIL-53(Fe)-derived Fe-N_x single atoms on Ti₃C₂ MXene; the negatively charged MXene surface restrains thermal aggregation of Fe species, modulates the d-band electronic configuration of isolated Fe sites, and strengthens tolerance to methanol cross-contamination relative to commercial Pt/C^[135]. Such systems validate that MXene conductive supports stabilize atomic catalytic centers and accelerate the four-electron ORR pathway. However, the turnover frequency (TOF) of these Fe-N_x sites and their site density remain incompletely quantified, making it difficult to assess whether the observed activity originates from true single-atom sites or from small clusters. Furthermore, the stability of M-N_x motifs under acidic ORR conditions is questionable, as protonation and demetalation can occur during extended operation, a phenomenon that is rarely monitored in laboratory studies.

The OER constitutes the anodic half-reaction of water splitting and metal-air batteries, which requires catalysts stable under high oxidative potentials and capable of lowering four-electron transfer barriers. OER is also the application in which the pristine/derived distinction is least meaningful at the operating stage, because both types of precursors converge on the same working phase. Pristine MOF/MXene systems—including the diffusion-grown CoBDC/Ti₃C₂T_x 2D/2D heterostructure (1.64 V vs. reversible hydrogen electrode at 10 mA cm^-2 in 0.1 M KOH, Tafel slope 48.2 mV dec^-1)^[33], the Fe_2.1Ni_0.7Co_0.2-MIL-88A/Ti₃C₂T_x composite (overpotential 231 mV at 10 mA cm^-2, TOF 2.33 s^-1)^[57], electrodeposited CoFe bimetallic MOF nanosheets on monolayer MXene (overpotential 191 mV, Tafel slope 31 mV dec^-1)^[58], electrostatically assembled Co₂Ni-MOF/Ti₃C₂T_x^[80], and dual-ligand modified Co-MOF/MXene frameworks^[137]—are crystalline in the as-prepared state, but their metal centers are converted in situ to metal (oxy)hydroxides under anodic polarization. The same reconstruction occurs for MOF-derived systems such as NiCoS/MXene obtained by post-synthetic sulfidation and CoNi-ZIF-67/Ti₃C₂T_x composites^[85,139]. The MOF thus acts only as a sacrificial precursor that delivers dispersed metal sites and a porous scaffold, while the MXene supplies the conductive network and stabilizes the reconstructed layer. For this application the materials are therefore classified by their working phase rather than by the crystallinity of their MOF precursor, and performance is attributed to the (oxy)hydroxide/MXene interface, not to the MOF. A critical issue is that the reported current densities are universally below 100 mA cm^-2, and the durability of these reconstructed catalysts at higher current densities remains essentially unknown.

For the cathodic HER of water electrolysis, MOF/MXene heterojunctions and post-converted metal phosphide/sulfide derivatives achieve catalytic performance across both acidic and alkaline environments. Tiwari constructed a pristine covalently bonded Ni₃(HITP)₂-MOF/Ti₃C₂T_x via Ti-N linkages; strong interfacial electron transfer optimizes hydrogen adsorption free energy on metal sites, achieving overpotentials of 180 mV in alkaline media (1.0 M KOH) and 240 mV in acidic conditions (0.5 M H₂SO₄)^[140]. MXene-supported MoNiS derived from MOF precursors balances intermediate adsorption and avoids MXene oxidative corrosion, reaching an overpotential of 153 mV with a Tafel slope of 92 mV dec^-1 in 1 M KOH^[136]. Similarly, MOF-derived cobalt sulfide/carbon nanocomposites on MXene have been reported as bifunctional OER/HER catalysts^[138]. Murtaza's CoP/C@Nb₂CT_x obtained by sequential carbonization and phosphorization is bifunctional, realizing HER and OER overpotentials of 50 and 149 mV, respectively, at 10 mA cm^-2 in 0.1 M KOH, standing out among non-noble metal phosphide electrocatalysts^[89]. For translation to practice, the more informative benchmark is the full-cell demonstration recently provided by scalable electrodeposited NiMoP, which follows catalyst performance through to alkaline electrolyzer operation^[141]. It should be noted that direct comparison of overpotentials is strictly valid only within the same electrolyte system, as different pH values and electrolyte compositions significantly affect reaction kinetics. Additionally, oxygen-vacancy-enriched Ce-MOF/MXene delivers bifunctional HER/OER activity and stable cycling, expanding the application boundary of single-component hybrid catalysts^[142]. However, the MXene surface can undergo oxidation and dissolution under prolonged cathodic polarization in acidic electrolytes, which is rarely addressed at present. The stability data for HER catalysts are typically limited to < 100 h of chronopotentiometry, which remains far below the stack lifetimes of commercial alkaline (50,000-96,000 h) and proton-exchange-membrane (70,000-100,000 h) electrolyzers^[143]. For reference, alkaline electrolyzers operate at current densities of 0.25-0.45 A cm^-2 and proton-exchange-membrane electrolyzers at 1.0-2.0 A cm^-2[144].

Developing bifunctional and trifunctional electrocatalysts capable of driving multiple reactions simultaneously simplifies electrolyzer and battery device assembly and cuts manufacturing costs. Zong's Ti₂NT_x-wrapped ZIF-67-derived CoP composite serves as a bifunctional catalyst for overall water splitting, with HER overpotential of 112 mV, OER overpotential of 241 mV, and a full-cell voltage of 1.61 V at 10 mA cm^-2, exceeding the simple performance superposition of individual catalysts^[23]. Trimetallic BTC-derived NiCoCuO/MXene further reduces overall water splitting voltage to 1.38 V due to multi-valence electronic modulation, outperforming commercial Pt/C||IrO₂ paired electrodes^[101]. Ni-doped Co-MOF-74 grown on MXene-coated nickel foam provided a self-supported electrode for overall water splitting^[64]. Multi-functional catalysts integrating HER, OER, and ORR are suitable for regenerative zinc-air batteries^[114]. He's CoO_x-N-C/TiO₂-C/MXene trifunctional hybrid achieves a water splitting voltage of 1.45 V and a narrow potential gap between ORR and OER, showing value for integrated energy storage-conversion devices^[145]. An MXene-supported MOF-derived CoP nanocomposite has subsequently been shown to combine OER, HER, and ORR activity in a single catalyst^[90]. It is important to note that bifunctional and trifunctional metrics are typically evaluated in separate half-cell configurations using different electrolytes; the true simultaneous catalytic performance in a single electrolyte environment (e.g., 1 M KOH for both HER and OER) often yields lower performance than the sum of individual half-cell values due to competing surface reconstruction processes. The mechanical stability of MOF/MXene catalyst layers during gas evolution (bubble detachment stress) is also rarely evaluated, yet it is a critical factor for membrane electrode assembly durability.

Beyond water electrolysis and air battery catalysis, MOF/MXene composites display performance in emerging electroreduction and small-molecule conversion fields. Electrochemical CO₂RR relies on MOF porous channels to concentrate CO₂ molecules and MXene conductive networks to deliver electrons, yet material selectivity and long-term stability remain major obstacles^[7,10]. For ambient NRR, ZIF-67/Ti₃C₂T_x composites supply abundant Lewis acidic Co sites to activate N≡N triple bonds, attaining an ammonia yield of 6.52 μmol h^-1 cm^-2 and a Faradaic efficiency of 20.2%^[102]. It should be noted that although this value is reported by the cited study, the electrocatalytic NRR community widely acknowledges the risk of false-positive NH₃ signals due to ambient nitrogen contamination. Therefore, rigorous control experiments, such as ¹⁵N₂ isotope-labeled tests, are highly recommended to confirm the origin of the produced ammonia and to verify the reliability of such high Faradaic efficiency values. When applied to fuel cell anodic MOR, ZIF/Ti₃C₂T_x supported ultra-fine Pt nanoparticles deliver a mass activity of 1,900.1 mA mg^-1 and anti-CO poisoning capacity^[146]. MXene substrates also effectively protect HKUST-1 MOF frameworks from high-temperature hydrothermal decomposition and endow MOF-derived Ni nanoparticles with catalytic activity for solid-state hydrogen storage^[20]. However, the TOF and mass activity values for Pt-based catalysts in MOR are often obtained under specific experimental conditions (e.g., scan rate, Pt loading) that are not standardized across studies, making direct comparison problematic^[147].

Future research can break through the limitations of MOF/MXene composites, including the ease of oxidation of MXene and collapse of the MOF skeleton, through five targeted directions. First, researchers should expand the MXene material library beyond conventional Ti₃C₂T_x to Nb₂CT_x, Mo₂TiC₂T_x, and Mo₂Ti₂C₃ series with enhanced anti-oxidation performance and tunable electronic structures. Second, machine learning and high-throughput computational screening can be adopted to predict optimal MOF-MXene matching pairs and interfacial charge transfer behaviors, reducing experimental trial-and-error cycles. Third, operando synchrotron X-ray absorption spectroscopy, in situ Raman, and TEM characterization must be widely utilized to capture dynamic structural reconstruction and active site evolution during real electrocatalytic processes, clarifying reaction mechanisms. Fourth, scalable green manufacturing technologies including continuous flow synthesis and electrochemical exfoliation should be developed to realize large-area electrode mass production. Finally, catalyst performance must be systematically validated in practical membrane electrode assemblies and flow electrolytic cells, narrowing the gap between lab testing indicators and industrial application standards. Systematic breakthroughs in these aspects will exploit the multi-functional catalytic potential of MOF/MXene heterostructures and accelerate the commercialization of low-cost electrocatalytic energy conversion technologies.

Sensing applications

The hybridization of MOFs and two-dimensional MXenes has been widely accepted for sensing technology, overcoming the performance limits of individual sensing materials and generating synergistic signal transduction effects^{[103,148-150]}. Chemiresistive gas sensors represent a mature application direction of MOF/MXene composites, which can operate at room temperature and avoid the high energy consumption of traditional high-temperature metal oxide gas detectors. A series of heterostructures fabricated via etching, self-assembly, and MOF template conversion have been developed for trace detection of toxic, volatile gaseous pollutants. For low-concentration NO₂ monitoring, Liu et al. prepared Cu-BTC/MXene conductive aerogels via solvent-assisted selective pore etching followed by hydrogen-bond-driven self-assembly^[103]. Defect-rich open micropores derived from controlled etching supply gas adsorption sites, and continuous hydrogen-bond-connected MXene conductive pathways boost resistance variation signals upon NO₂ adsorption. The composite sensor delivers a response value of 52.47 toward 50 ppm NO₂ with response/recovery durations of 0.9 and 4.5 s, respectively^[103]. To further utilize MOF as a structural sacrificial template, Peng et al. adopted hydrothermal growth and aging treatment to deposit CuO nanoparticles on MXene substrates, retaining the high porosity inherited from MOF precursors while generating heterojunction interfaces to enhance gas response. This device realizes NO₂ detection with a 30 ppb limit of detection (LOD) and stable signal output over 10 consecutive weeks^[151]. For ammonia sensing vulnerable to humidity interference, ZIF-67 modified via ligand exchange was self-assembled with Ti₃C₂T_x to construct hydrophobic H-MOF/MXene composites. Optimized surface hydrophobicity suppresses water molecule competitive adsorption; even under 76% and 93% relative humidity, signal attenuation only reaches 0.22% and 0.27%, with a theoretical ammonia detection limit as low as 12.8 ppb^[152]. In addition, porphyrin Co-TCPP(Fe) MOF was combined with MXene to form Schottky heterointerfaces for trace NO identification, restoring Fe-N₄ active sites to capture nitric oxide and reaching a 200 ppb detection threshold with response/recovery times of 95 and 15 s^[104]. Sandwich layered conductive Cu-HHTP/MXene heterostructures further expand gas detection targets to triethylamine, featuring an ultra-rapid 4 s response at 50 ppm and stable performance under 80% humidity environments^[66]. These studies demonstrate that MOF/MXene gas sensors achieve detection limits in the ppb range, yet most evaluations are conducted in dry synthetic air. Real-world applications involve complex gas mixtures with varying humidity and temperature, conditions under which MOF frameworks may undergo structural rearrangement or competitive adsorption that degrades selectivity. The long-term drift of baseline resistance in MOF/MXene sensors (> 6 months) is also rarely reported, which is a critical parameter for industrial deployment [Figure 9].

Figure 9. Synthesis strategies of MOF/MXene composites for sensing applications. (A) In situ growth of Cu-HHTP MOF within MXene interlayers to form MXene@Cu-HHTP heterostructure for triethylamine sensing. Adapted from Ref.^[66], under CC BY 4.0 license. (B) In situ growth of Ni-MOF on MXene for glucose sensing. Adapted from Ref.^[153], under CC BY 4.0 license.

Electrochemical sensing based on MOF/MXene hybrid electrodes covers environmental heavy metal detection, food residue analysis, and biomedical small molecule quantification, combining the electrocatalytic activity of MOF metal sites and fast electron transfer networks of MXenes. For neurotransmitter dopamine testing, Zn-MOF was directly compounded with Ti₃C₂ MXene to overcome the insulating limitation of pure MOFs, enabling dopamine quantification with a detection limit of 110 nM^[148]. By switching the MXene substrate to Nb₄C₃T_x and adopting full in situ MOF growth, Boruah et al. further strengthened target molecule affinity on electrode surfaces, realizing an LOD of 0.2 nM and a linear detection window ranging from 1 to 100 nM^[154]. In the field of food safety antibiotic monitoring, one-pot hydrothermal in situ growth of CoFe-MOF on MXene-modified hydrophobic carbon paper provides catalytic sites for chloramphenicol redox reactions, achieving picomolar-level 33 pM detection sensitivity^[155]. ZIF-67-derived bimetallic LDH electrostatically assembled with MXene creates a wide linear range (0.079-521 μM) for ciprofloxacin identification with a 0.013 μM detection limit^[156]. Mycotoxin aflatoxin B1 detection relies on covalent antibody immobilization on Mo-MOF/Ti₃C₂ composites; combined with electrochemical impedance spectroscopy measurement, the sensor achieves an LOD of 8 pg/mL within a linear interval of 0.06-50 ng/mL^[157]. For environmental heavy metal pollution control, one-step hydrothermal synthesized Fe-MOF/MXene forms strong hydroxyl-As binding sites to capture As(III), reaching a detection limit of 0.58 ng L^-1[158]. Flower-shaped Cu-MOF/MXene and calcined Ni-MOF-derived NiO/MXene both serve as non-enzymatic hydrogen peroxide sensing electrodes: the former utilizes Cu-based catalytic centers to attain a 0.35 μM LOD, while the three-dimensional porous framework of the latter effectively prevents MXene sheet stacking during repeated operation^[150,159]. The conductive MOF-on-MXene architecture has also been applied to non-enzymatic glucose detection^[153]. Despite these low detection limits, most electrochemical sensors are validated in buffer solutions containing only the target analyte. The performance in real matrices such as blood serum, food extracts, or wastewater is rarely reported, and electrode fouling by proteins, lipids, or surfactants can severely degrade the practical utility of these devices. Furthermore, the reproducibility of sensor fabrication-particularly the uniformity of MOF coating on MXene surfaces-remains a significant challenge, with batch-to-batch variation in sensitivity often exceeding 15% in reported studies.

Benefiting from the intrinsic flexibility and good biocompatibility of MXene substrates and tunable biological ligand environments of MOFs, MOF/MXene composites have achieved progress in biological analysis and wearable real-time health monitoring. For ultrasensitive extracellular vesicle detection, CuNi bimetallic MOF was hybridized with MXene quantum dots to construct label-free aptasensors^[160]. The large specific surface of bimetallic MOFs immobilizes abundant aptamer recognition fragments, while MXene quantum dots accelerate interfacial electron exchange, supporting trace exosome quantification down to 5 particles μL^-1. For wearable point-of-care melatonin monitoring, biocompatible Zn-glutamate bio-MOF was grown on Nb₂CT_x-coated carbon yarn electrodes, which can be fabricated into adhesive bandage-style wearable sensors with a detection limit of 215 nM for real-time body fluid analysis^[161]. Beyond conventional biomedical testing, this composite sensing system has also been applied to cultural heritage protection research: MXene/CNT/MOF hybrid electrodes realize ultrasensitive collagen detection for archaeological leather relic identification, with a detection limit as low as 83 pg/mL, opening a sensing application direction rarely reported in previous literature^[162]. However, the long-term stability of wearable sensors under continuous skin contact, sweat exposure, and mechanical deformation remains largely uncharacterized. The biodegradation of MOF frameworks in physiological fluids and the potential cytotoxicity of MXene nanosheets are safety concerns that must be addressed before clinical translation. The issue of non-specific binding in complex biological fluids, which can generate false-positive signals, is also insufficiently addressed in most reported studies.

Apart from single-target single-mode sensors, MOF/MXene heterostructures can be designed into integrated multifunctional systems that couple sensing performance with energy harvesting or structural monitoring. MXene/MOF-derived carbon foam integrates piezoresistive structural health monitoring^[163]. The material delivers a piezoresistive sensitivity of 1.98 kPa^-1, serving as a real-time deformation monitoring sensor. Semiconductor Ni₃(HITP) conductive MOF composited with MXene forms flexible temperature sensing electrodes^[164]. The composite possesses a temperature coefficient of resistance of -3.1% °C ^-1, supporting continuous human body temperature and respiration monitoring with 0.1 °C measurement accuracy. Self-powered gas sensing platforms eliminate external power supply by combining triboelectric nanogenerators with Ti₃C₂T_x/MOF-derived CuO heterostructures. Triboelectric energy generated by mechanical deformation drives ammonia detection, and the sensor reaches a response value of 24.8 under 100 ppm ammonia atmosphere without extra power input^[165]. These multifunctional systems represent creative extensions of MOF/MXene composites, though the integration of multiple functions often comes at the expense of sensing selectivity and response speed, as the non-sensing components may introduce additional background noise or interference. The energy conversion efficiency of triboelectric nanogenerators integrated with MOF/MXene sensors is typically below 10%, which may be insufficient for powering data transmission modules in practical IoT applications.

Photochemical catalysis

Photocatalytic water splitting for hydrogen evolution stands as the most extensively studied application of MOF/MXene photocatalysts, where controlled interfacial bonding significantly boosts hydrogen output. Shi constructed UiO-66-NH2(Zr/Ti)/MXene heterostructures via coordination assembly; carboxyl groups strengthen interfacial contact and promote water dissociation, achieving a 20-fold higher H₂ generation rate (2,187 μmol g^-1 h^-1) relative to pristine MOF^[61]. Li further realized Ti-N-linked MIL-NH₂/Ti₃C₂ composites through in situ MXene-derived growth, which shortens electron migration paths and delivers a hydrogen evolution rate of 4,383.1 μmol h^-1 g^-1[166]. Multi-interfacial Ti₃C₂/TiO₂/UiO-66 ternary heterostructures fabricated by thermal annealing create cascade charge transfer channels, yielding a hydrogen production rate of 1,980 μmol h^-1 g^-1[167]. Exfoliation-assembled 2D/2D Cu₂[CuTCPP]/Ti₃C₂ Schottky junctions maximize built-in electric fields, realizing a H₂ yield over 5,000 μmol g^-1 after 3 h of irradiation and enabling unassisted overall water splitting without sacrificial reagents^[60]. These studies demonstrate that interfacial engineering is critical for photocatalytic performance, yet the hydrogen evolution rates are still orders of magnitude lower than those of benchmark semiconductor photocatalysts such as CdS and g-C₃N₄ when normalized to surface area^[168]. Furthermore, the stability of MXene under continuous illumination is questionable, as photoexcited carriers can catalyze the oxidation of Ti₃C₂T_x to TiO₂, leading to progressive degradation of the conductive scaffold [Figure 10]. Where apparent quantum yield (AQY) values are reported for MOF/MXene photocatalysts, they are measured under monochromatic irradiation and remain modest: 4.25% at 350 nm dropping sharply to 0.19% at 400 nm for Cu₂[CuTCPP]/Ti₃C₂^[60], and 3.14% at 420 nm for Ti₃C₂@MIL-NH₂^[166]; the other cited systems do not report quantum efficiency. These values—especially the steep fall-off toward the visible range—remain well below the 10% solar-to-hydrogen conversion efficiency regarded as the minimum for competitively priced, large-scale photoelectrochemical systems^[169].

Figure 10. Synthesis strategies of MOF/MXene composites for photocatalysis. (A) Carboxyl group-assisted coordination route for constructing UiO-66-NH₂(Zr/Ti)/carboxyl-functionalized MXene (UZT/CFMX) heterostructure. Adapted from Ref.^[61], under CC BY 4.0 license. (B) Assembly of 2D Cu₂[CuTCPP] MOF with exfoliated Ti₃C₂ MXene to form a 2D/2D Schottky junction for photocatalytic hydrogen evolution. Adapted from Ref.^[60], under CC BY 4.0 license.

Beyond hydrogen evolution, MOF/MXene hybrids exhibit photocatalytic performance spanning CO₂ photoreduction, ambient nitrogen fixation, and photocatalytic uranium(VI) reduction. For CO₂ conversion, in situ grown Co-Co LDH/Ti₃C₂ nanoarrays derived from MOF precursors deliver a CO generation rate of 1.25 × 10⁴ μmol h^-1 g^-1 with an AQY of 0.92%^[170]. Zn/Fe-ZIF/Ti₃C₂T_x composites realize 96% epichlorohydrin conversion under simulated sunlight via photo-thermal synergism and extended near-infrared light response^[59]. MXene quantum dot/Ni-MOF Type-II heterostructures activate N≡N triple bonds, attaining an ammonia production rate of 88.79 μmol g^-1 h^-1 under ambient light irradiation^[171]. Similarly, MnO_x/UiO-66/Ti₃C₂ dual-cocatalyst systems realize 98.4% U(VI) photoreduction for radioactive wastewater treatment, extending the same interfacial design principles to actinide removal^[172]. While these metrics appear promising, it is important to note that most photocatalytic tests are conducted in small-batch reactors under idealized conditions (pure water, controlled pH, sacrificial reagents for H₂ evolution). The performance under realistic solar illumination and continuous-flow conditions is rarely reported, and the scalability of photocatalytic reactor design remains unexplored.