Structural modulation of metal-organic framework glasses for alkali metal ion transport and conduction

Disordered coordination networks in MOF glasses

Disordered coordination networks are regarded as one of the defining structural characteristics of MOF glasses and are fundamentally distinct from the highly periodic frameworks observed in crystalline MOFs. Unlike crystalline counterparts that rely on long-range periodic ordering, MOF glasses generally preserve only short-range ordering around the metal coordination environment while exhibiting substantial long-range disorder throughout the extended framework. For example, in ZIF-derived glasses, Zn-N coordination and tetrahedral Zn-imidazolate local structures are often preserved after glass formation, whereas longer-range Zn-Zn or framework correlation peaks become broadened or weakened^[31]. This suggests that the periodic framework structure is transformed into an isotropic disordered network.

Figure 1 further summarizes this local preservation-long-range disorder structural feature. Single-crystal X-ray diffraction (XRD) indicates that certain coordination bonds in crystalline coordination frameworks become weakened before melting and may act as initiation sites for crystal-structure collapse and melting^[32] [Figure 1A]. The differential scanning calorimetry (DSC) and dynamic mechanical analysis (DMA) results show that MOF glass formation is accompanied by clear melting, glass transition, and temperature-dependent viscosity changes, suggesting that these materials are not simply amorphized products but undergo an identifiable crystal-liquid-glass phase transition [Figure 1B]. Variable-temperature solid-state nuclear magnetic resonance (NMR) and molecular simulations reveal the presence of dynamically heterogeneous coordination environments during glass formation [Figure 1C]. Furthermore, reverse Monte Carlo (RMC) modeling, combined with pair distribution function (PDF), extended X-ray absorption fine structure (EXAFS), and magic-angle spinning nuclear magnetic resonance (MAS NMR) data, reconstructs the atomistic glass structure and further demonstrates that short-range coordination motifs can be retained within the glassy network, while medium- and long-range structures become significantly disordered [Figure 1D].

Figure 1. Selected characterizations of structures and phase transitions in MOFs. (A) Single-crystal XRD can determine the crystal structure, where mean atomic displacement retrieved from variable temperature measurements helps clarify the melting mechanism through Lindemann’s rule; (B) Frequency-dependent dynamic mechanical analysis and DSC are used to determine thermal behavior, such as T_g; (C) Variable-temperature solid-state NMR and molecular simulation are examples of characterization methods that provide insight into structural dynamics; (D) RMC can provide atomistic modeling of the glass structure through the fitting of experimental data from PDF, EXAFS, and MAS NMR. DSC: Differential scanning calorimetry; DMA: dynamic mechanical analysis; MAS NMR: magic-angle spinning nuclear magnetic resonance; EXAFS: extended X-ray absorption fine structure; MOFs: metal-organic frameworks; XRD: X-ray diffraction; T_g: glass transition temperature; RMC: reverse Monte Carlo; PDF: pair distribution function.

Compared with crystalline MOFs, such structural heterogeneity is particularly important because it introduces locally flexible coordination sites, that are difficult to achieve in ordered crystalline frameworks. The resulting disordered coordination networks therefore provide a structurally adaptive platform for ion transport, where local coordination fluctuations and dynamically rearrangeable environments may facilitate ion hopping and reduce migration barriers^[33]. These characteristics increasingly suggest that the functional advantages of MOF glasses are not derived from preserving crystalline order, but rather from exploiting controlled disorder within coordination networks to enable emergent transport behaviors inaccessible in rigid crystalline MOFs.

Thermodynamic characteristics of MOF glasses

MOF glasses are the product of certain crystalline MOFs that undergo melting, quenching, and finally vitrify^[34-36]. The thermodynamic characteristics of MOF glasses are fundamentally governed by the interplay between framework mobility, coordination dynamics, and temperature-dependent structural rearrangements during melting and vitrification. Unlike conventional inorganic glasses, MOF glasses originate from coordination networks in which metal-ligand bonds remain partially dynamic even in the liquid state, enabling a unique form of thermodynamic softening prior to complete structural collapse.

Among them, zeolitic imidazolate frameworks (ZIFs) family, especially ZIF-62, has received considerable interest due to its excellent glass-forming performance. Figure 2 illustrates the glass-forming process of ZIF-62^[37]. Adopting similar structures to zeolites, the ZnN₄ tetrahedra in ZIF-62 replace SiO₄ tetrahedra and are linked by rigid organic ligands rather than less-constrained bridging oxygens [Figure 2A]. This zeolite-like tetrahedral topology reduces the angular constraints, and enables continuous framework reorganization upon heating, thereby fundamentally underpinning its ability to undergo vitrification prior to decomposition. The dominant prerequisite for MOF glass formation is the existence of a well-defined melting regime prior to framework decomposition, namely that the melting temperature (T_m) lies below the decomposition temperature (T_d), resulting in a sufficiently wide melting window^[31] (ΔT = T_d - T_m). Upon rapid cooling from the molten state, the disordered coordination network becomes kinetically arrested, suppressing crystal nucleation and growth, and yielding an amorphous glassy solid^[37] [Figure 2B and C]. The melt-quench nature of MOF glasses allows shaping and device fabrication (e.g., hot imprinting and membrane processing). Compared with crystalline MOFs, bulk MOF glasses have been shown to possess mechanical robustness, which enables membrane fabrication and thermal processability, thereby enhancing their potential for device integration^[22,38-41].

Figure 2. Structure and melt-quench glass formation of ZIF-62. (A) Zeolite-like tetrahedral framework of crystalline ZIF-62, consisting of tetrahedrally coordinated Zn²⁺ centers connected by imidazolate-based ligands; (B) Temperature-time pathway illustrating the melting, cooling and annealing process used to produce a_gZIF-62; (C) Flow chart of float glass fabrication method. ZIF-62 and the derived materials are shown in white. (A-C) Reproduced from reference^[40], licensed under CC BY 4.0.

In the case of ZIF-4, molecular dynamics simulations provide mechanistic insight into the microscopic origin of MOF melting. As shown in Figure 3A, crystalline ZIF-4 transforms into a disordered melt at 1,500 K and can subsequently be quenched into a glassy state. The corresponding PDFs in Figure 3B show that the structural peaks broaden upon melting, indicating the loss of long-range order while short-range correlations remain partially preserved. Figure 3C further shows that the N-Zn-N bond-angle distribution becomes significantly broader in the molten state, suggesting distortion of the local ZnN₄ tetrahedral coordination environment. These changes indicate that ZIF-4 melting is associated with Zn-N coordination lability and local network rearrangement rather than complete collapse of the metal-ligand framework. Meanwhile, Figure 3D shows that the pore-size distribution becomes broader after melt-quenching, reflecting the formation of a more disordered glassy network. Overall, ZIF-4 melting produces a liquid-like disordered state in which local Zn-N connectivity and partial framework memory are retained, while medium- and long-range order is progressively lost^[42].

Figure 3. Melt-quenching simulation of ZIF-4 at zero pressure using deep learning force field. (A) Atomic configurations of ZIF-4 at different stages of the melt-quenching process, i.e., ZIF-4 crystal, molten ZIF-4 at 1,500 K, and quenched ZIF-4 glass at 300 K; (B-D) Calculated (B) pair distribution functions, (C) N-Zn-N bond angle distribution functions, and (D) pore size distribution of the ZIF-4 crystal, melt (at 1,500 K), and glass. (A-D) Reproduced from reference^[42], licensed under CC BY 4.0.

Given the melting behavior of certain MOFs, the possible role of framework mobility in MOF-glass ion transport can be discussed with reference to soft matter electrolytes based on this broader principle: structural mobility and transient free-volume formation can influence ion migration in disordered solid electrolytes. Figure 4A shows the tensile deformation of polyethylene oxide (PEO)/LiClO₄ polymer electrolyte films, while both the in-plane and out-of-plane ionic conductivities increase with deformation/strain [Figure 4B]. This enhancement is attributed to cavitation-induced tiny voids, particularly in the amorphous regions of the semicrystalline polymer matrix [Figure 4C]. These voids increase free volume and may facilitate cooperative rearrangements of polymer chains, thereby improving ionic conductivity. In MOF-derived glass electrolytes, an analogous broad principle may apply, although through a distinct structural mechanism: glass-state framework mobility may promote ion migration through transient free-volume generation, local bond-angle fluctuations, and cooperative rearrangement of coordination motifs, rather than through polymer-chain segmental motion or a purely static hopping pathway.

Figure 4. Framework mobility of soft matter electrolytes. (A) Photo images of PEO samples subjected to tensile deformation; (B) In-plane and out-of-plane ionic conductivities of PEO electrolyte (soft matter electrolyte) with respect to tensile deformation (in the direction of the red arrow). (i) Out-of-plane ionic conductivity vs. tensile deformation of PEO/Li salt film; (ii) Out-of-plane enhancement in ionic conductivity vs. tensile strain; (iii) In-plane ionic conductivity vs. tensile deformation; (iv) In-plane enhancement in ionic conductivity vs. tensile strain; (C) Depiction of semi-crystalline polymer microstructure at various stages of tensile deformation. (A-C) Reproduced from reference^[43], licensed under CC BY 4.0. PEO: Polyethylene oxide.

In addition, it is important to note that MOF-glass conductivity is not governed by free volume or viscosity alone both of which are influenced by the vitrification process. Figure 5 further shows that, in Zn-H_nPO₄-based MOF glasses, changes in microdomain size and coordination-network connectivity alter bulk-phase viscosity but have only limited influence on ion conductivity. This suggests that local coordination chemistry, especially metal coordination number and ligand-based hopping-site density, can be more decisive than macroscopic viscosity. Therefore, MOF glasses should not be viewed simply as ordered crystals transformed into static amorphous solids, but as dynamically disordered yet partially connected coordination networks, in which framework mobility, free-volume evolution, thermodynamic relaxation, and local coordination motifs collectively regulate both vitrification and ion-transport properties.

Figure 5. Comparable ion conductivities with change in average size of continuous networks. Zn-H_nPO₄ frameworks showed little change in ion conductivity with a change in the connectivity of the coordination networks. The gray areas provide visual aids to distinguish the increasingly disconnected coordination network. This figure is reproduced with permission from reference^[44]. Copyright 2022, American Chemical Society.

Porosity and pore behavior in MOF glasses

Porosity and pore behavior in MOF glasses are fundamentally governed by the competition between structural densification during vitrification and the retention of accessible free volume inherited from the parent crystalline framework. Early studies on melt-quenched ZIF glasses showed that vitrification is often accompanied by substantial densification and loss of gas-accessible porosity^[45]. For example, melt-quenched ZIF-4 was reported to form a nearly non-porous glass with a Brunauer-Emmett-Teller surface area below 5 m²·g^-1, indicating severe collapse or contraction of the originally porous framework^[35]. Subsequent positron annihilation lifetime spectroscopy studies further showed that a_gZIF-4 retains internal free volume, but much of this porosity is inaccessible to conventional gas sorption probes, highlighting the distinction between intrinsic free volume and gas-accessible porosity^[46,47]. Similar reductions in accessible pore volume and surface area have also been observed in pressure-induced and mechanically amorphized ZIFs, suggesting that framework collapse, densification, and loss of long-range order commonly reduce the accessible porosity of amorphous MOF phases^[48].

However, subsequent studies showed that porosity in MOF glasses is not completely lost, but can evolve into a partially retained and dynamically heterogeneous microporous network. As Figure 6 shows, porous glasses derived from ZIF-76 and ZIF-76-mbIm demonstrate that bulky benzimidazolate-based linkers can inhibit dense packing in the liquid state and help preserve free volume after vitrification. The PDFs of crystalline and glassy samples remain nearly identical below ~6 Å, indicating that local C-C/C-N and Zn-N correlations are preserved, while the remaining features beyond 6 Å, especially around 7.5 and 10.8 Å, further suggest retained medium-range Zn-imidazolate-Zn-benzimidazolate connectivity in the glass state [Figure 6A]. This retained structure is confirmed to lead to functional pore accessibility, as a_gZIF-76-mbIm reversibly adsorbs CO₂ and CH₄ [Figure 6B]. Meanwhile, vitrification contracts the pore-size distribution but still preserves microporous voids, consistent with positron annihilation lifetime spectroscopy (PALS)-detected pores of ~4.8 and 7.2 Å [Figure 6C]. But the slower CO₂ uptake in the glass indicates there are more tortuous and constricted diffusion pathways in glass state [Figure 6D]. Together, these results suggest that transport accessibility in MOF glasses arises from interconnected disordered cavities rather than conventional crystalline pore channels.

Figure 6. Structural retention and gas-accessible microporosity in ZIF-derived glasses. (A) ZIF-76 pair distribution function D(r). Inset shows medium-range order. Zn light blue, Cl green, C grey, N dark blue, H omitted for clarity; (B) Adsorption isotherms of a_gZIF-76-mbIm (filled symbols = adsorption, empty symbols = desorption); (C) Pore size distributions as determined by a NLDFT method from CO₂ adsorption isotherms at 273 K; (D) Time-dependent CO₂ uptake profiles at 273 K at a pressure of 5 Torr. (A-D) Reproduced from reference^[49], licensed under CC BY 4.0. NLDFT: Non-local density functional theory.

Recent studies have therefore increasingly explored pore engineering strategies to overcome vitrification-induced densification, including linker steric modulation, mixed-linker design, crystal-glass composites, and porogen-assisted approaches inspired by traditional porous glass fabrication^[50]. In particular, porogen strategies involving thermally decomposable salts, polymers, or sacrificial fillers have been proposed as promising routes for introducing hierarchical porosity and tunable pore architectures into MOF glasses. The introduction of hierarchical porosity may be especially important because ion and gas transport in dense glassy networks is frequently limited by diffusion bottlenecks and restricted pore interconnectivity. Moreover, the macroscopically grain-boundary-free nature of MOF glasses provides an additional advantage over crystalline MOFs, as continuous isotropic transport pathways may emerge when pore connectivity is successfully retained. Consequently, pore behavior in MOF glasses should not be interpreted simply as a loss of crystallinity-induced porosity, but rather as the evolution of adaptive disordered free-volume networks whose accessibility, connectivity, and transport functionality can potentially be engineered through controlled liquid-state structural design and vitrification processes.

Li⁺ and Na⁺ transport behaviors and mechanisms in MOF glasses

For alkali metal cations, transport in MOF glasses is not simply governed by rigid pore diffusion^[52]. Instead, it is strongly affected by the coupled effects of free volume, coordination disorder, linker dynamics, and local ion-framework interactions.

The recent ZIF-62 glass electrolyte study showed that Li⁺ migration in glassy ZIF-62 occurs through a “relay-race” mechanism, where Li⁺ is dynamically delivered between N sites from imidazole and benzimidazole ligands. Compared with crystalline ZIF-62, the glass phase provides a larger Li⁺ migration space, higher calculated Li⁺ diffusivity, and more continuous homogeneous transport channels. In crystalline ZIF-62, Li⁺ motion is confined to localized regions, whereas glassy ZIF-62 provides a larger spatial distribution for Li⁺ migration [Figure 7A and B]. The mean square displacement (MSD) analysis further shows that the calculated Li⁺ diffusivity in glassy ZIF-62 is 3.97 × 10^-9 m²·s^-1, which is higher than that in crystalline ZIF-62 (1.43 × 10^-9 m²·s^-1) [Figure 7C]. Within the same simulation time frame, the Li⁺ migration distance is approximately 1.49 Å in crystalline ZIF-62, but increases to approximately 3.92 Å in glassy ZIF-62, indicating faster Li⁺ migration and a more open transport space in the glassy phase [Figure 7D and E]. In addition, Li⁺ trajectories in crystalline ZIF-62 are relatively localized and mainly distributed around N species, whereas in glassy ZIF-62, the Li⁺ trajectories become more continuous along N atoms [Figure 7F and G]. The radial distribution function further reveals that the Li-N distance increases from approximately 2.14 Å in the crystalline phase to 2.22 Å in the glassy phase. This longer Li-N distance suggests weaker interactions between Li⁺ and N sites in glassy ZIF-62, making Li⁺ easier to detach from one N site and migrate to another [Figure 7H]. Consistently, the Li⁺ migration activation energy calculated from the Arrhenius relationship was 0.17 eV for glassy ZIF-62 quasi-solid-state electrolyte (GZ-62-QSSE), lower than that of crystalline ZIF-62 quasi-solid-state electrolyte (CZ-62-QSSE), 0.20 eV. In contrast, the binding energy between Li⁺ and GZ-62-QSSE was only 0.21 eV, markedly lower than that between Li⁺ and CZ-62-QSSE, 2.91 eV^[53]. These results demonstrate that incompletely coordinated Zn²⁺ sites and distorted coordination environments generated by vitrification can immobilize anions, weaken Li⁺ binding, and thereby promote Li⁺ migration.

Figure 7. Li⁺ migration mechanism in crystalline and glassy ZIF-62. Atom position distribution of Li in (A) ZIF-62 crystal and (B) ZIF-62 glass; (C) Li MSD in crystalline and glassy ZIF-62 during 5 ps; (D and E) Structural snapshots of Li⁺ migration in ZIF-62 crystal and ZIF-62 glass from AIMD simulations. Movement trajectories of Li element in (F) ZIF-62 crystal and (G) ZIF-62 glass; (H) RDF of Li-N in ZIF-62 crystal and ZIF-62 glass. The yellow curves in A, B, F and G represent the movement trajectories of Li ions. This figure is quoted with permission from reference^[53]. Copyright 2025, Wiley-VCH GmbH. RDF: Radial distribution function; AIMD: ab initio molecular dynamics.

Another important Li⁺-related mechanism is modifier-induced network depolymerization. In lithium benzimidazolate-modified ZIF-62 glasses, Li(bIm) acts as a compatible glass modifier. Rather than merely occupying the residual pores, Li(bIm) becomes incorporated into the disordered glass network and disrupts the original Zn-Im/bIm-Zn connectivity, generating non-Zn-bridging linker environments and partially depolymerizing the framework. This structural modification lowers the glass-transition temperature and increases the configurational freedom of the network. These results suggest that Li⁺ can influence ion transport not only as a potential mobile charge carrier but also by modifying the local coordination network and increasing structural flexibility^[54].

Porosity and free volume also play a central role. MOF glass formation often causes partial pore collapse, which may impede Li⁺ transport by generating dead-end pathways. Therefore, alkali-ion transport in MOF glasses should be understood as a coupled structure-dynamics process. The most important mechanisms include: N-site-assisted hopping, free-volume-mediated diffusion, modifier-induced network depolymerization, coordination disorder, and glass-dynamics-assisted ion migration. Future MOF-glass-based SSEs may be rationally designed by combining alkali salt incorporation, defect engineering, pore accessibility control, and thermodynamic softening to create continuous, isotropic, and dynamically adaptive ion-transport networks^[26].

Na⁺ and K⁺ transport in MOF glasses: an emerging but underdeveloped direction

In disordered or amorphous solid electrolytes, Na⁺ transport is strongly affected by the topology of the local free-volume network, the distribution of transient coordination sites, and the connectivity of ion-hopping pathways. Recent studies on amorphous and nanocrystalline Na-Y-Zr-Cl halide electrolytes showed that the introduction of amorphous domains can enhance Na⁺ conductivity by reducing grain-size-related transport barriers and generating more continuous ion-transport pathways^[55]. Similarly, molecular-dynamics studies on amorphous alkali oxyhalides indicate that alkali-ion migration is governed by local anion coordination, residence time near framework-forming units, and ion-ion correlation effects rather than by rigid crystallographic channels alone^[56]. These findings suggest that, in MOF glasses, Na⁺ migration should be understood as transport through a dynamically connected disordered free-volume network, where both bottleneck size and the flexibility of local coordination environments are critical.

MOF-based Na⁺ solid electrolytes further highlight the importance of pore size and channel chemistry. Nozari et al. introduced the sodium-salt-containing ionic liquid (Na_0.1EMIM_0.9) (Na0.1[1-ethyl-3-methylimidazolium]0.9) bis(trifluoromethanesulfonyl)imide (TFSI) into ZIF-8 and demonstrated that MOF-ionic liquid host-guest interactions can promote high Na⁺ conductivity, while structural disorder improves the stability of the composite electrolyte^[57]. In MOF-74 systems, the one-dimensional hexagonal channels and open metal sites provide larger ion-conducting pathways and sites for anion immobilization. For example, Na/Mg-MOF-74 exhibited an ionic conductivity of 8.53 × 10^-4 S·cm^-1 at 70 °C, which was attributed to its stable ion-conducting channels and exposed Mg²⁺ sites^[58]. Similarly, Na[EMIm][TFSI]@Zn-MOF-74 reached 5 × 10^-4 S·cm^-1 at room temperature and enabled stable sodium plating for over 100 h^[59].

Therefore, the design principles established for Li⁺ conducting MOF glasses cannot be directly transferred to Na⁺ systems. Because Na⁺ has a larger ionic radius and different coordination preferences, Na⁺ conductors generally require wider bottlenecks, larger accessible free volume, weaker ion-framework binding, and more continuous percolated pathways. However, current studies remain limited: most Na⁺ MOF electrolytes rely on crystalline MOFs or ionic-liquid-filled composites, while direct evidence for Na⁺ migration in genuine MOF glass networks is still scarce. Future work should therefore combine structural characterization, ion-dynamics simulations, and electrochemical measurements to clarify how glass topology, free-volume distribution, and anion immobilization jointly regulate Na⁺ transport.

Compared with Li⁺ transport, K⁺ conduction in MOF glasses remains largely unexplored. To date, most reported alkali-metal-ion-conducting MOF glasses and MOF-glass-containing electrolytes have focused on Li⁺ transport, as exemplified by glassy ZIF-based quasi-solid-state electrolytes for lithium-metal batteries^[53]. In contrast, K⁺ conducting MOF systems reported so far are mainly based on crystalline single-ion-conducting frameworks or MOF/polymer composite electrolytes rather than melt-quenched MOF glasses^[60,61]. For example, MOF-808-SO₃K has been reported as a single potassium-ion conducting crystalline MOF, in which immobilized sulfonate groups provide fixed anionic sites and K⁺ serves as the mobile charge carrier^[61]. More recently, MOF-containing composite polymer electrolytes have also been developed for solid-state potassium batteries, where UiO-66 is incorporated into a poly(vinylidene fluoride-co-hexafluoropropylene) (PVDF-HFP)-based polymer matrix to enhance mechanical stability, increase polymer disorder, and provide K⁺ migration channels^[60]. These examples demonstrate the feasibility of K⁺ transport in MOF-based electrolyte systems, but they also highlight that the structure-transport relationships established in crystalline MOFs or MOF/polymer composites cannot be directly transferred to MOF glasses.

The scarcity of K⁺ conducting MOF glasses likely arises from several coupled challenges. K⁺ has a larger ionic radius and lower charge density than Li⁺, making its transport more sensitive to pore aperture, free-volume size, and the availability of weakly coordinating sites^[62,63]. Besides, many glass-forming MOFs are charge-neutral frameworks without intrinsic fixed anionic groups; therefore, K⁺ transport generally requires the incorporation of external potassium salts, ionic liquids, or framework-bound anionic functionalities^[54]. The preparation of homogeneous K⁺ containing MOF glasses is also synthetically challenging, as the potassium salts may crystallize, phase-separate, or decompose during high-temperature melting and quenching. This issue is particularly relevant because MOF glasses are generally produced by heating/melting followed by rapid quenching^[64-66]. These factors explain why K⁺ conduction in MOF glasses remains an important research gap and suggest that future studies should focus on designing glass-forming frameworks with sufficiently large and connected free-volume elements, introducing fixed anionic sites or weakly coordinating ligands, and suppressing salt crystallization to enable continuous K⁺ transport.

Additive modulation strategies in developing MOF glass ionic conductor

Thermodynamic modulation toward accessible MOF vitrification

From a thermodynamic perspective, the key challenge in accessing MOF glasses lies in enabling framework reorganization to occur prior to decomposition. Introducing additives into MOF glasses can be achieved by chemical substitution, guest incorporation, and topology engineering. They provide an intrinsic means of thermodynamic modulation by directly altering local coordination strength and network rigidity. The additives refer to the introduction of foreign atoms, ions or molecules (impurities) into crystals or glass, which have a relatively low concentration but can alter the chemical environment and electronic structure of the material.

In porous carbon synthesis, the dissolution-rearrangement of ZIF-8 in a MgCl₂·6H₂O molten salt medium below 300 °C demonstrates that a salt environment can substantially lower the energetic threshold for structural transformation, compared to conventional solid-state pyrolysis of ZIF-8, which typically requires more than 700 °C to induce framework collapse and carbonization^[67]. This observation highlights the general capability of molten salts to unlock low-temperature structural reorganization pathways and provides a conceptual basis for extending such strategies to MOF systems more broadly. Together, these results highlight the general role of molten salts in lowering both energetic and kinetic barriers for MOF melting and vitrification. Recent work by Cao et al. directly demonstrated that co-melting heterocycle-based benzimidazolium halide salts (Cl^-, Br^-, I^-) with ZIF crystals enables framework melting at temperatures below 300 °C and yields exceptionally low glass transition temperatures (T_g < 150 °C)^[68]. Figure 8 shows that, during thermal treatment, halide anions coordinate with Zn centers while proton exchange from the cation modifies the local coordination environment, leading to the formation of mixed-linker configurations. This process enables continuous structural reconstruction of the framework. Upon further heating, partial evaporation of linker species may occur, followed by re-polymerization into a modified network with increased Zn-halide connectivity.

Figure 8. Mechanism of modification for ZIF glasses using benzimidazolium halide salts. This figure is quoted with permission from reference^[68], licensed under CC BY 4.0.

Molten inorganic salts can directly interact with metal-ligand coordination environments during thermal processing, enhancing mass transport, thereby enabling controlled framework reconstruction and pore evolution^[69]. Beyond this, they also act as intrinsic sources of mobile charge carriers, introducing ions such as Li⁺ into the evolving network and thereby enabling ionic conduction within the resulting glass.

Guest modulating strategies demonstrate that thermodynamic accessibility of MOF glasses can be rationally engineered. Nevertheless, these approaches primarily regulate when and how melting occurs, rather than the functional properties of the resulting glasses. Consequently, thermodynamic modulation alone remains insufficient for achieving high-performance MOF glass electrolytes, motivating parallel efforts in transport regulation discussed in the following section.

Conductivity enhancement of MOF glasses

Beyond thermodynamic accessibility, achieving practically relevant ionic conductivity constitutes an equally critical challenge for MOF-based solid electrolytes, including both crystalline MOFs and MOF glasses. This section focuses specifically on strategies for enhancing ionic transport in MOFs and MOF glasses, where the framework serves as a mechanically rigid host for mobile charge carriers introduced via guests^[13,70,71]. Unlike thermodynamic modulation, these approaches target carrier concentration, mobility, and percolation pathways, and are often developed independently of melting or vitrification considerations.

Early studies predominantly focused on intrinsic ionic conduction, especially proton transport mediated by hydrogen-bonded networks, which are usually formed by acidic ligands, coordinated water molecules, or guest acids^[72,73]. In such proton-conducting MOFs, conductivities in the range of 10^-4-10^-2 S·cm^-1 have been reported under high humidity (RH > 90%) and moderate temperatures (60-90 °C)^[74-76]. However, the strong dependence on water content and the instability of hydrated structures significantly limit long-term operation and high-temperature applicability^[77].

To address these challenges, increasing attention has been directed toward extrinsic ionic conduction, in which MOFs serve as rigid scaffolds hosting mobile ionic species introduced via guest salts or ionic liquids^[78]. Ionic liquid-infiltrated MOFs (IL@MOFs) often show enhanced ion transport, with conductivities reaching 10^-4 S·cm^-1 at room temperature, while offering favorable thermal robustness compared with conventional polymer-based electrolytes^[57,79,80]. Table 1 summarizes previous research on conducting MOFs and MOFs-based electrolytes systems.

Table 1

Representative MOF-glass, non-ZIF MOF, and IL@MOF ion-conducting systems with material-level conductive phase, performance, and limitations

Category	Specific material	Conductive phase/transport pathway	Ion	Performance	Comparative strength	Limitations/trade-offs	Suggested reference
ZIF-derived MOF electrolyte	Glassy ZIF-4 QSSE	Glassy ZIF-4 network with lean Li salt/solvent phase confined in the glass-derived framework	Li+	1.61 × 10-4 S·cm-1 at 30 °C	Relatively high room-temperature ionic conductivity, stable glassy state improves processability and ion transport	Quasi-solid-state rather than fully dry single-phase glass; residual liquid/solvent contribution should be stated clearly	[27]
	ZIF-62 glass in PEO-LiTFSI/IL electrolyte	Disordered ZIF-62 glass as isotropic filler; Li+ mainly transported through IL-rich phase and improved filler/electrolyte wetting	Li+	2.41 × 10-4 S·cm-1 at 30 °C; 9.46 × 10-4 S·cm-1 at 70 °C		Composite effect, not intrinsic dry MOF-glass conduction; performance depends on polymer, IL and filler loading	[81]
	Vitrified MOF composite (glassy ZIF-62) electrolyte (PGZ)	Vitrified MOF-derived composite framework with polymer/gel electrolyte phase enabling Li+ migration	Li+	~6.3 × 10-4 S·cm-1		Composite rather than single-component glass; the exact conductive phase must be separated from the MOF-glass scaffold	[82]
	Porous ZIF-62 glass gelled polymer electrolyte (PMG-GPE)	Top-down carved porous MOF glass with 3D interconnected graded-aperture channels; gel/polymer phase supplies Li salt and continuous ion medium	Li+	Up to ~1.9 × 10-3 S·cm-1 at RT		Gelled polymer/composite electrolyte; conductivity cannot be attributed solely to dry MOF glass	[79]
	Boundary-free ZIF-62 glass SSE for Li-O2 batteries	Continuous MOF-glass phase designed to remove grain boundaries and support Li+ transport/interface stability	Li+	~5 ×10-4 S·cm-1 at 20 °C		Cell chemistry is Li-O2-specific; synthesis and oxygen-electrolyte interface may not directly transfer to Li-ion or Na-ion cells	[25]
	NaTFSI/[EMIM][TFSI]@ZIF-8	Superionic sodium IL phase confined in/disordered by ZIF-8; structural disorder improves stability and transport	Na+	~6 × 10-3 S·cm-1 at RT	Very high Na+ conductivity; key benchmark for IL-impregnated MOF sodium conductors	Non-glassy phase, Main conduction comes from IL-rich phase; mixed-ion contribution and liquid containment must be discussed	[57]
Crystalline carboxylate MOF	MOF-808-based solid-state electrolyte	MOF-808 host/framework combined with Li salt phase; open Zr-cluster chemistry may assist salt dissociation and Li+ migration	Li+	3.08 × 10-3 S·cm-1 at 25 °C	High room-temperature ionic conductivity, combined with a high lithium-ion transference number and good cycling stability	MOF-808 itself is not an intrinsic fast Li+ conductor, and its performance mainly depends on confined liquid/quasi-liquid electrolyte species within its pores	[83]
	MOF-808-SO3K wetted with propylene carbonate	Single-ion K+ conductor: tethered sulfonate anions immobilize countercharge; PC wets pellet and assists K+ mobility	K+	3.1 × 10-5 S·cm-1 at 303 K; K+ transference number 0.76	Immobilized sulfonate anions enable single K+ conduction with a high K+ transference number	Lower conductivity than many IL@MOF Li/Na systems; solvent-wetted pellet rather than dry glass; particulate crystalline nature may limit practical solid-state battery applications	[61]
Crystalline carboxylate MOF + IL	[BMP][TFSI]@UiO-66	Confined ionic liquid in UiO-66 pores; ionic liquid phase dominates conduction	Li+	3.3 × 10-4 S·cm-1 at RT	Competitive structural design; the higher packing density enhances ion transport	Liquid-containing system; possible mixed cation/anion transport and leakage/plasticization issues	[84]
	[EMIM][TFSI]@UiO-67	Confined EMIM-TFSI phase in larger-pore UiO-67; Li+ transport assisted by IL mobility	Li+	1.0 × 10-4 S·cm-1 at RT	Shows pore-size/linker extension effect relative to UiO-66-type hosts	IL phase can dominate conductivity; larger pores may reduce confinement and ion selectivity	[85]
	[EMIM][TFSI]@MOF-525(Cu)	Ionic liquid confined in crystalline MOF-525(Cu) pores, with nanowetted interfaces improving electrode contact	Li+	3.0 × 10-4 S·cm-1 at RT	High RT conductivity for an IL@MOF electrolyte; nanowetted interfaces improve Li-metal contact and reduce interfacial resistance	Not MOF glass; conductivity mainly arises from IL phase rather than intrinsic MOF framework; apparent conductivity may be affected by excess surface IL; Li+ transference and IL retention should be considered	[86]
	[BMP][TFSI]@UiO-66	Li-containing ionic liquid confined in UiO-66 pores	Li+	3.3 × 10-4 S·cm-1 at RT	High RT conductivity for crystalline IL@MOF electrolyte; nanostructured UiO-66/Li-IL improves bulk Li+ transport and electrode/electrolyte interfacial contact	Not intrinsic MOF or MOF-glass conduction; conductivity mainly depends on confined Li-IL phase; possible mixed-ion transport from IL; long-term IL retention and surface/excess IL effects should be considered	[87]
	[EMIM][TFSI]+ UiO-66@UiO-67	Li-containing ionic liquid hosted in UiO-66@UiO-67 nanopores	Li+	2.1 × 10-4 S·cm-1	Core-shell MOF-in-MOF combines high IL uptake from UiO-67 shell and ion-selective confinement from UiO-66 core; higher conductivity and Li+ transport than homogeneous UiO-66/Li-IL or UiO-67/Li-IL hosts	Not intrinsic MOF or MOF-glass conduction; conductivity mainly arises from Li-IL phase; mixed-ion transport and IL retention/excess IL effects should be considered; core-shell synthesis is more complex than single-MOF hosts	[88]
	MIL-121/Li + [EMIM][TFSI]	Apparent conduction mainly from EMIM-TFSI-rich external/interparticle phase; Li+ in MIL-121/Li remains largely immobile	Li+	5.0 × 10-4 S·cm-1 at 30 °C for MIL-121/Li + 23 wt% EMIM-TFSI	Stable confinement of ionic liquid to enhance Li+ transport	Not intrinsic MOF conduction; dry MIL-121/Li is nearly insulating; IL may not enter pores effectively; apparent conductivity can be dominated by external IL	[89]
	NaTFSI in Bmpyr-TFSI@UiO-66	Na salt dissolved in pyrrolidinium-TFSI IL confined by sulfonated/UiO-66-type MOF host	Na+	3.6 ×10-4 S·cm-1 at RT	Promising for Na-metal batteries	Quasi-solid/IL-laden system; ionic liquid content and cation transference need careful reporting	[90]
	NaFSI/[EMIM][FSI]@UiO-66	FSI-based Na ionic liquid confined in UiO-66 pores	Na+	10-4-10-3 S·cm-1 at RT	FSI chemistry can enhance ion mobility relative to bulkier TFSI systems	FSI-based systems may raise interfacial/corrosion concerns; still liquid-containing	[91]
	Na[EMIM][TFSI]@Zn-MOF-74	Na-containing EMIM/TFSI ionic phase in 1D MOF-74 channels with open metal sites	Na+	5 × 10-4 S·cm-1 at RT	Synergistic ion transport through MOF-74 one-dimensional channels and ionic liquid; Promising for Na-metal batteries	Poor moisture stability and open metal site interactions can trade off conductivity and long-term stability	[59]
Carboxylate MOF glass	agZW-UiO-67·MSA	Zwitterion-grafted UiO-67 plus Brønsted acid; MIMS-MSA acid-base subsystem provides proton transport in melt-quenched glass	H+	1.57 × 10-2 S·cm-1 at 100 °C	Breakthrough in glass formation of carboxylate MOFs	Proton conductor, not Li+/Na+/K+ electrolyte; requires acid/zwitterion modification and elevated temperature	[92]
	ag ZW-UiO-68·MSA and agZW-DUT-5·MSA	Zwitterion/acid-enabled glass formation in Zr/Al carboxylate frameworks	H+/potential platform	Glass formation shown; Tm ~156 °C for ZW-UiO-68·MSA and ~152 °C for ZW-DUT-5·MSA	Low-melting carboxylate MOF glasses	Alkali-ion battery conductivity not yet benchmarked; should be framed as material-platform evidence, not Li/Na/K electrolyte performance	[92]
	Mg/Mn adipate-based carboxylate MOF glasses	Flexible aliphatic carboxylate network; melt-quenched Mg2+/Mn2+ adipate glasses	No battery ion reported	Tm ~284 °C (Mg) and ~238 °C (Mn); improved hardness/elastic modulus compared with typical ZIF glasses	Low-melting carboxylate MOF glasses	No Li+/Na+/K+ conductivity yet; mainly structural/mechanical evidence	[66]
Cu-based MOF + IL	[DEME][TFSI]@HKUST-1	IL confined in HKUST-1 pores; Cu paddlewheel framework provides polar host environment	Li+	4.0 × 10-4 S·cm-1 at RT	High RT conductivity among IL@MOF systems	HKUST-1 has moisture sensitivity; Cu sites/framework stability should be considered for battery compatibility	[93]
	[EMIM][TFSI]@ HKUST-1	Ionic liquid-based ionogel provides the main ion-conducting phase; LLZO ceramic core contributes Li+ conduction pathways; MOF shell offers porous channels and metal sites for ion regulation/selectivity	Li+	3.87 × 10-4 S·cm-1 at RT	High ionic conductivity, wide electrochemical window, high thermal stability, improved Li dendrite suppression and stable Li metal cycling	Composite system rather than intrinsic MOF or MOF-glass conductor; Li+ transference number of ionogel electrolytes remains a key concern; performance depends on IL content, interfacial wetting, and core-shell filler dispersion	[94]
Phosphate-based coordination-polymer glass	[Zn3(H2PO4)6(H2O)3] (1,2,3-benzotriazole) glass	Metal phosphate/azole coordination polymer glass; phosphate groups and azole assist anhydrous proton hopping	H+	8.0 × 10-3 S·cm-1 at 120 °C	Anhydrous proton conduction combined with glass processability	Proton conductor at elevated temperature, not alkali-ion battery electrolyte; moisture/temperature dependence must be distinguished from Li/Na/K systems	[95]

LiTFSI: Lithium bis(trifluoromethanesulfonyl)imide; NaTFSI: sodium bis(trifluoromethanesulfonyl)imide; [EMIM][TFSI]:1-ethyl-3-methylimidazolium bis(trifluoromethanesulfonyl)imide; [BMP][TFSI]:1-butyl-1-methylpyrrolidinium bis(trifluoromethanesulfonyl)imide; [EMIM][FSI]: 1-ethyl-3-methylimidazolium bis(fluorosulfonyl)imide; NaFSI: sodium bis(fluorosulfonyl)imide; MAS: methanesulfonic acid; [DEME][TFSI]: N,N-diethyl-N-methyl-N-(2-methoxyethyl) ammonium bis (trifluoromethanesulfonyl)imide; MOF: metal-organic framework; ZIF: zeolitic imidazolate framework; IL: ionic liquid; QSSE: quasi-solid-state electrolyte; PEO: polyethylene oxide; PGZ: poly(1,3-dioxolane) (PDOL)/glassy ZIF-62; PMG-GPE: porous MOF glass gelled polymer electrolyte; SSE: solid-state electrolyte; RT: room temperature; EMIM: 1-ethyl-3-methylimidazolium; TFSI: bis(trifluoromethanesulfonyl)imide; FSI: bis(fluorosulfonyl)imide; MIMS: 1-(1-methyl-3-imidazolio)propane-3-sulfonate; LLZO: lithium lanthanum zirconium oxide.

Ionic conduction in IL@MOF systems originates from the synergistic interplay between confined ionic liquids and the host framework^[96]. As Figure 9 shows, the ionic liquid provides a reservoir of mobile charge carriers, while nanoscale confinement within MOF pores alters ion coordination and suppresses ion pairing, thereby increasing the population of free charge carriers. Meanwhile, the MOF framework actively modulates ion transport through interfacial interactions and steric constraints. Specifically, framework sites can immobilize or slow down anions via coordination or adsorption, resulting in enhanced cation transference. In addition, the well-defined pore architecture establishes continuous ion-conduction pathways and reduces tortuosity for ion migration. As a result, ionic transport in IL@MOFs is governed by a combination of confinement-enhanced ion dissociation, framework-mediated ion selectivity, and percolated transport pathways within the confined liquid phase.

Figure 9. Ion-transport mechanism in IL@MOF electrolytes. Alkali-metal-ion-containing ionic liquids are loaded into MOF nanopores to form IL@MOF hybrid electrolytes. Within the confined MOF channels, Li⁺ or Na⁺ cations can migrate along the pore network, whereas bulky anions interact with the MOF framework and exhibit restricted mobility. The confinement effect and framework-anion interactions can facilitate cation-dominated ion transport. MOF: Metal-organic framework; IL: ionic liquid.

While substantial progress has been made in enhancing ionic conductivity through guest incorporation and framework disorder, these strategies frequently rely on elevated operating temperatures or liquid-like transport behavior to achieve high conductivities. Consequently, ionic conductivity enhancement has remained largely decoupled from thermodynamic accessibility and structural stability under device-relevant operating conditions, underscoring the need for integrated approaches that simultaneously address transport and thermodynamic constraints.

Transient flux strategies in developing MOF glass ionic conductor

Different from additive modulation, transient flux-assisted processing relies on the presence of an external molten medium during heating, which facilitates mass transport and framework rearrangement without necessarily becoming a permanent component of the final glass^[97]. In MOF systems, meltable frameworks phases can act as transient fluxes, enabling the co-melting of otherwise non-meltable MOFs and expanding the accessible vitrification window. By lowering kinetic barriers associated with solid-state diffusion and coordination rearrangement, flux-assisted pathways allow melting to occur at temperatures below the intrinsic decomposition threshold. Previous research used meltable ZIF-62 to dissolve non-fusible ZIF-8 in a liquid phase effectively lowering the overall melting threshold, and the whole composite achieved co-melting at 430-450 °C^[98] [Figure 10A and B]. Flux melting also provides a promising pathway for constructing ion-conductive MOF glasses through the formation of newly accessible disordered pore networks. Although the original crystalline pore structure of ZIF-8 collapses during vitrification, the incorporation of ZIF-8 into the liquid ZIF-62 matrix disrupts dense liquid-state packing and generates a less compact disordered framework with enlarged pore-limiting apertures and enhanced free volume [Figure 10C-E]. The resulting flux-melted glass exhibits significantly improved H₂ accessibility and larger transport apertures compared with pure a_gZIF-62, indicating the emergence of more continuous and accessible diffusion pathways within the disordered coordination network [Figure 10F]. Such dynamically disordered pore environments may be particularly beneficial for ionic transport because they can reduce steric constraints and facilitate ion hopping through interconnected free-volume domains. In contrast to rigid crystalline pore channels, these isotropic disordered networks may also provide macroscopically grain-boundary-free transport pathways and adaptable coordination environments capable of accommodating alkali ions during migration. Therefore, flux-mediated pore restructuring not only expands the vitrification accessibility of non-melting MOFs, but also suggests a viable strategy for engineering transport-accessible MOF glasses with potential ionic conduction functionality.

Figure 10. Dissolving ZIF-8 by ZIF-62. (A) Atomic configuration of a high-temperature liquid ZIF; (B) C_p and mass as a function of temperature for (ZIF-8)(ZIF-62) (20/80); (C) Structure factors S(q) of (ZIF-8)(ZIF-62) (20/80), a_g[(ZIF-8)_0.2(ZIF-62)_0.8] and ZIF-8, alongside that of a_gZIF-62; (D) Corresponding X-ray pair distribution functions D(r); (E) Temperature resolved SAXS profile of (ZIF-8)(ZIF-62) (20/80); (F) H₂ at 77K gas isotherms for the zinc-based crystalline mixtures and glasses. (A-F) adapted with permission from reference^[98]. Copyright 2019, Royal Society of Chemistry. SAXS: Small-angle X-ray scattering; ZIF: zeolitic imidazolate framework.

Coordination-environment engineering strategies

At a higher structural level, topology and network engineering offer a powerful means of thermodynamic regulation by modifying the global constraint landscape of MOFs. Reductions in framework connectivity, changes in network topology, or cooperative interactions with confined guests can increase configurational freedom and defect density, thereby lowering the energetic barriers associated with melting and vitrification. These topology-driven strategies highlight that glass formation in MOFs is governed not only by local bond chemistry but also by collective network constraints. In ZIFs, works have shown that partial substitution or mixing of metal nodes (e.g., Zn/Co) enables continuous modulation of metal-ligand bond strength thereby systematically lowering the melting temperature and expanding the glass-forming window^[99,100]. There have been studies clearly demonstrating that simply changing the metal center can transform an otherwise non-meltable framework into one exhibiting reversible melting behavior^[101]. At the microscopic level, metal-node engineering alters local coordination geometry, electronic structure, and metal-organic bond energetics.

Early work by Bennett et al. targeted linker substitution in ZIFs - where a fraction of imidazolate linkers in ZIF-4 (Zn(Im)₂) is replaced during synthesis by bulkier benzimidazolate or methylbenzimidazolate ligands to form TIF-4 and ZIF-62 - systematically lowers the melting temperature from ~863 K (ZIF-4) to ~740 K (TIF-4) and ~710 K (ZIF-62), suppresses reconstructive recrystallization, and enables direct melt-quenching into a glass while preserving local Zn-N coordination^[31]. Building on this foundation, recent studies by Xue et al. demonstrated that even highly stable carboxylate-based frameworks, such as UiO-67 and DUT-5, can be rendered meltable through reductions in network connectivity, or cooperative interactions with pore-confined guests^[92]. From a thermodynamic perspective, topology-driven strategies effectively reduce the rigidity of the coordination network, increase structural degrees of freedom, and enhance defect tolerance, thereby lowering the energetic barriers associated with melting or vitrification^[12].

A promising strategy: molten salt

Building on the diverse strategies developed for ion-conducting MOF glasses, we further propose that molten-salt chemistry may offer a useful conceptual framework for designing and understanding ion-conducting MOF glasses. This perspective is inspired by molten-salt strategies widely used in carbon and oxide chemistry, where molten salts can act not only as ion reservoirs, but also as reaction media, fluxes, structure-directing agents, and thermal stabilizers. When translated to MOF-glass systems, molten salts may play multiple roles: they can introduce mobile alkali-metal carriers, weaken or screen metal-ligand interactions, lower the effective melting or softening temperature, suppress premature framework decomposition, and facilitate vitrification during quenching. In our view, four coupled mechanisms are particularly relevant when translating these approaches to salt-assisted MOF systems, as demonstrated in Figure 11.

Figure 11. Coupled mechanism for molten-salt-assisted ionic conductive MOF glass.

Halide or mixed chloride salts behave as low-viscosity ionic fluxes. They provide a continuous liquid environment that dramatically accelerates mass transport and lowers kinetic barriers for bond exchange and framework rearrangement. When a crystalline MOF is immersed in such a molten-salt medium, the framework no longer needs to reach its intrinsic melting point to undergo large-scale reorganization; instead, depolymerization-repolymerization events can be triggered at significantly reduced temperatures within the salt flux. This is mechanistically analogous to flux-melting using a low-T_m MOF glass, but the role of the liquid phase is now played by an inorganic salt.

Second, molten salts are not inert solvents. Their highly polarizable anions (e.g., Cl^-) and small, strongly coordinating cations can directly interact with metal-ligand bonds in the MOF^[102]. At the microscopic level, these ions compete with the original linkers for coordination to the metal nodes and screen metal-ligand electrostatics. This weakens metal-ligand coordination, promotes linker cleavage, and locally depolymerizes the network, effectively softening the framework. In ZIFs, for example, partial replacement of Zn-imidazolate coordination by halide or alkali-ion coordination can facilitate Zn-N bond rearrangement, thereby lowering the thermal transition temperatures and enabling access to a viscous, melt-like state before decomposition^[54,68,97].

Besides, the molten-salt environment also exerts a spatial confinement effect. In a salt-filled pore space, the surrounding ionic medium acts as a liquid matrix that limits crystallite growth, suppresses sintering, and stabilizes highly defective or partially depolymerized states that would be unstable in the absence of the salt. Once incorporated into the softened coordination network, ionic species such as Li⁺, Na⁺, and Cl^- can act as network modifiers and become integrated into the local coordination structure rather than simply occupying residual pore space^[54]. Their incorporation creates chemically heterogeneous coordination environments, including undercoordinated alkali-ion sites and mixed metal-linker/halide coordination around the metal nodes^[68]. Meanwhile, the dynamic disruption and reformation of metal-linker-metal bridges during vitrification increases the population of non-bridging or dangling linkers and undercoordinated metal sites, resulting in a more depolymerized and structurally heterogeneous glass network^[103].

Once vitrification occurs, the same ions that acted as flux now play a second role: they form nanoconfined ionic domains within the amorphous framework. In pores, interstitial regions, and defect-rich zones, the molten salt behaves as a highly dissociated ionic medium. Under nanoconfinement, its structure and dynamics differ from the bulk, but the effective ion mobility can remain high^[104,105]. The result is a percolating network of salt-rich pathways threaded through a mechanically robust MOF glass matrix. From the perspective of ionic conductivity, this hybrid state combines (i) the isotropic, macroscopically grain-boundary-free transport of a glass with (ii) continuous, liquid-like ion highways provided by the nanoconfined salt.

Conductivity characterization

Structure characterization

X-ray absorption spectroscopy: detecting atomic structure

X-ray absorption spectroscopy (XAS) is an effective technique for investigating the local electronic structure of atoms and their coordination environments within MOF glass structures^[31]. X-ray absorption near-edge structure (XANES) can reveal the oxidation state and local electronic structure of atoms, while EXAFS can probe the bond distances to neighboring atoms and their coordination numbers^[115,116]. Xue et al. used Zr K-edge XANES and EXAFS to analyze the local structure of a family of carboxylate MOFs^[92]. XAS provides element-specific insight into the local structure of MOF glasses. The Zr K-edge XANES spectra of ZW-UiO-67·MSA before and after vitrification show only minor changes, indicating that the local electronic state and coordination environment of Zr are largely preserved during glass formation [Figure 12A]. A similar result is observed for ZW-UiO-67·TFSA, further supporting the retention of the local Zr environment after vitrification [Figure 12B]. The corresponding Fourier-transformed extended X-ray absorption fine structure (FT-EXAFS) spectra of ZW-UiO-67·MSA exhibit closely related first- and higher-shell features before and after vitrification, suggesting that the local Zr-O coordination shell and Zr-based node structure remain substantially intact in the glassy phase [Figure 12C]. Likewise, the FT-EXAFS spectra of ZW-UiO-67·TFSA confirm that the local coordination motifs are largely maintained despite the loss of crystallographic long-range order [Figure 12D]. Together, these results show that MOF vitrification primarily disrupts long-range periodicity while preserving the essential local coordination framework.

Figure 12. Zr K-edge X-ray absorption spectroscopy analysis of ZW-UiO-67-based glasses. (A) XANES spectra at the K-edge of the Zr atoms for derivatives of ZW-UiO-67·MSA and references; (C) Corresponding radial distance χ(R) space spectra from k³-weighted Fourier transform of χ(k)-function from the Zr K-edge EXAFS. Comparison spectra for ZW-UiO-67·TFSA and derivatives are shown in (B) and (D). Insets in (C) and (D) show the octahedral Zr₆O₄(OH)₄ node. Color codes: Zr, purple; O, red; H, light gray. (A-D) Reproduced from reference^[92], licensed under CC BY 4.0. MOF: Metal-organic framework; XANES: X-ray absorption near-edge structure.

THz/far-IR: probing metal bond and inorganic

Terahertz/far-infrared (THz/far-IR) spectroscopy is an effective technique for probing the quasi-localized properties and bonding characteristics of MOF composites. Through in situ THz/far-IR measurements, further information on the evolution of these composites during MOF formation can be collected. As Figure 13A shows, our team has applied this technique for the first time to reveal the dynamic behavior of glassy MOF-perovskite composites during the sintering process^[122].

Figure 13. (A) In situ THz/far-IR spectroscopy for monitoring temperature-dependent bonding evolution during sintering; (B) ZIF-62(Co)Fe before and after sintering. (i) Larger Co-N peak after vitrification; (ii) Fe-N formation after heating. (A) adapted with permission from reference^[122]. Copyright 2021, American Association for the Advancement of Science; (B) is adapted with permission from reference^[123]. Copyright 2022, Wiley-VCH GmbH.

In another study [Figure 13B], our team employed this technique to monitor the metal-N bonding environment in modified ZIF-62(Co)^[123]. In a_gZIF-62(Co)-Fe, the broad Co-N band at approximately 334 cm^-1 is consistent with structural amorphization, while the newly emerged peak at around 344 cm^-1 indicates the formation of Fe-N stretching vibrations^[123].

PALS: revealing free-volume defects and nanoscale voids

PALS has emerged as a particularly informative probe for characterizing nanoscale free volume, inaccessible microporosity, and defect-related cavities in MOF glasses. PALS can measure the lifetime of positronium species localized within low-electron-density regions, allowing characteristic cavity sizes and pore-size distributions to be inferred from positronium annihilation behavior. This capability has been demonstrated across several MOF-glass systems, including ZIF-62 glasses with exceptional glass-forming ability, ZIF-62-derived glass membranes, and ZIF-76-based glasses with permanent accessible porosity^{[37,40,49,124]}. These studies collectively establish PALS as a valuable complementary technique for evaluating how vitrification reshapes the intrinsic free volume and porosity of MOF-derived glassy networks.

A representative example is provided by Thornton et al. in their study of porosity in MOF glasses^[46]. In this work, PALS was used to compare the pore-size distributions of crystalline ZIF-4, dense crystalline ZIF-zni, and melt-quenched a_gZIF-4. As shown in Figure 14, the PALS-derived pore-size distributions reveal that a_gZIF-4 possesses a pore structure intermediate between the open framework ZIF-4 and the denser ZIF-zni phase, while also exhibiting a broader distribution characteristic of a structurally disordered glassy network^[46]. This example illustrates the specific value of PALS for identifying nanoscale vacancies, voids, and defect-like free-volume elements that may persist after framework collapse or vitrification, including pore environments that may be inaccessible to gas adsorption measurements.

Figure 14. Pore size distributions measured with PALS and simulated with Zeo++ for ZIF-4, ZIF-zni and a_gZIF-4. PALS results for the third and fourth components (τ3 and τ4) are represented as Gaussian distributions which correspond to the small and large cavities, respectively. This figure is quoted with permission from reference^[46], licensed under CC BY 3.0. PALS: Positron annihilation lifetime spectroscopy.

For ion-conducting MOF glasses, such information is particularly significant. Ionic transport in disordered hybrid glasses may be influenced not only by the chemical coordination environment of mobile ions, but also by the size, distribution, and possible connectivity of free-volume elements or defect-rich regions within the glassy framework. PALS therefore has strong potential to serve as a structural probe for assessing whether enhanced ionic conductivity correlates with increased free volume, nanoscale void formation, or defect-mediated transport pathways. However, this interpretation should be made cautiously: PALS does not directly measure ion mobility or prove an ion-conduction mechanism. Rather, it provides critical structural evidence that should be correlated with electrochemical impedance spectroscopy, ion transference measurements, solid-state NMR, XPS/XAS, and local-structure analyses to establish a more complete relationship between free volume, defect structure, and ion transport in MOF glass electrolytes.

Future applications

The structural programmability of MOF glasses provides a promising platform for developing next-generation solid-state ion conductors. MOF glasses exhibit features including disorder in long-range yet remain chemically tunable coordination networks, sub-nanometer free volume, dynamic metal-ligand environments, and macroscopically grain-boundary-free monolithic architectures, which create opportunities to design ion-transport pathways that are not limited to intrinsic crystallographic pores, but can instead be regulated through glass composition, coordination chemistry, defect concentration, free-volume distribution, and thermal history.

One particularly attractive direction is the development of salt-assisted MOF-glass electrolytes. The incorporation of light-metal halide salts, such as LiCl or NaCl, may serve a dual function: reducing the effective vitrification barrier of rigid coordination frameworks and simultaneously introducing mobile ionic carriers. This approach may expand MOF-derived glasses beyond intrinsically meltable ZIF systems to include frameworks that are traditionally difficult to vitrify. If the salt-framework interaction is properly controlled, such modified glasses have the potential to exhibit even higher ion conductivity than pristine MOF glass under near-room-temperature or moderately elevated-temperature conditions, thereby occupying an important materials space between polymer electrolytes and ceramic solid electrolytes.

Hybrid MOF glasses are also promising for ion conduction and may be deployed in intermediate-temperature solid-state batteries, particularly sodium- and lithium-based stationary energy-storage systems. Their amorphous and isotropic nature can suppress grain-boundary resistance, improve interfacial conformity with electrodes, and provide mechanically adaptable electrolyte layers. At temperatures approximating the glass transition point, partial softening of the coordination network may enhance local structural dynamics, promote ion mobility, and improve electrode reaction kinetics, while avoiding the severe thermal-management requirements associated with conventional high-temperature ceramic electrolytes. These advantages are especially relevant for stationary batteries, where long-term durability, processability, and interfacial stability are often more important than maximum gravimetric energy density.

Moreover, as demonstrated by Ogawa et al., beyond alkali-metal-ion conduction, MOF glasses offer considerable potential as anhydrous or low-humidity proton conductors^[39]. The combination of coordinatively unsaturated metal sites, hydrogen-bonding ligands, phosphate- or sulfonate-containing units, and nanoconfined ionic domains could generate proton-transfer pathways that remain active under conditions where hydrated polymer membranes lose performance. This makes MOF-glass electrolytes attractive for intermediate-temperature proton-conducting fuel cells, membrane reactors, and electrochemical hydrogen-separation devices. Compared with conventional polymer membranes, MOF glasses may provide improved thermal robustness, reduced dehydration sensitivity, and more chemically programmable proton-transfer environments.

Another important application area is ion-selective membranes and electrochemical separation devices. Wang et al. have initiated investigations into MOF glass composites for lithium enrichment^[125]. Because the ionic species, coordination sites, and free-volume architecture can be independently tuned, MOF glasses could be engineered for selective transport of Li⁺, Na⁺, K⁺, Ag⁺, H⁺, or multivalent ions. The absence of grain boundaries may also reduce non-selective leakage pathways, which is a major limitation in many polycrystalline ion-conducting membranes.

MOF-glass ion conductors may further contribute to solid-state sensors and harsh-environment ionic devices. Their compositional flexibility, thermal stability, and resistance to grain-boundary degradation make them suitable for gas sensors, humidity sensors, electrochemical probes, and high-temperature ionic devices. By selecting appropriate metal nodes, ligands, and incorporated salts, the transport response could be coupled with specific chemical stimuli, enabling ion-conducting glasses with sensing functions.

More broadly, the key opportunity lies in transforming meltability from a rare intrinsic property into a programmable design parameter. Through flux-assisted vitrification, coordination-environment modulation, and compositing or interface-mediated transport design, MOF glasses could evolve from a small family of meltable ZIF-derived materials into a broader class of processable ionic glasses.

Challenges

Despite these opportunities, several fundamental and technological challenges must be addressed. The first challenge is to establish reliable structure-transport relationships. Ion conduction in MOF glasses is expected to involve multiple coupled factors, including free-volume size, defect concentration, metal-ligand bond dynamics, salt distribution, local coordination disorder, and segmental motion near the glass transition temperature. However, these factors are difficult to decouple experimentally. A high apparent conductivity alone is insufficient to prove efficient ionic transport; it must be supported by activation-energy analysis, isotope-effect measurements, DC polarization, impedance spectroscopy using appropriate blocking and non-blocking electrodes, and structural probes such as PDF, EXAFS, PALS, and variable-temperature spectroscopy as discussed in Section "CHARACTERIZATION OF MOF-GLASS-BASED SSEs".

Long-term stability is another central concern. Salt-assisted vitrification and ion incorporation may create metastable structures that are beneficial for transport but vulnerable to structural relaxation, salt redistribution, phase segregation, or crystallization during extended operation. Repeated thermal cycling, operation near or above T_g, electrochemical bias, and contact with reactive electrodes may progressively alter the defect population, free-volume distribution, and local coordination environment. Understanding how ionic mobility couples with glass relaxation will therefore be essential for predicting lifetime performance.

Device integration also remains nontrivial. Although the amorphous and isotropic nature of MOF glasses may improve physical contact with electrodes compared with rigid crystalline ceramics, the formation of interphases at electrode-electrolyte boundaries remains poorly understood. Interfacial resistance, electrode compatibility, redox stability, dendrite suppression in metal batteries, and mechanical integrity under cycling must be systematically evaluated. The electrochemical stability window of each MOF-glass electrolyte should also be carefully determined, since organic ligands, redox-active metal nodes, and incorporated salts may introduce decomposition pathways under bias.

Scalable synthesis and processing are also key barriers. Electrolytes typically require homogeneous composition and compatibility with established workflows. Achieving these requirements for fabricating MOF glass ion conductor at large scale is challenging because small variations in precursor crystallinity, particle size, salt loading, heating rate, pressure, or quenching conditions may strongly influence glass structure and ionic conductivity. Future work should therefore move beyond small bulk pellets toward thin films, supported membranes, laminated electrolytes, and prototype cells.

For solid-state battery applications, the practical viability of MOF glass electrolytes should be evaluated beyond ionic conductivity. Key issues include the electrochemical stability window, chemical compatibility with metal anodes, oxidative stability against high-voltage cathodes, interfacial decomposition, and dendrite suppression. Previous studies have shown that many solid electrolytes are not thermodynamically stable against alkali-metal electrodes and may form resistive decomposition interphases during cycling^[126,127]. Moreover, interfacial instability can increase cell resistance, promote uneven metal deposition, and accelerate dendrite growth^[128]. Therefore, future MOF glass electrolytes should be designed with stable electrode interfaces, sufficient mechanical robustness, and suppressed parasitic reactions.

Finally, the field requires clearer performance benchmarks and standardized testing protocols. Conductivity values should be reported together with activation energy, temperature and humidity conditions, sample geometry, electrode configuration, electronic transference number, cycling stability, and post-mortem structural analysis. For battery-related applications, symmetric-cell cycling, full-cell testing, critical current density, and interfacial impedance evolution should be included.

Conclusion

Overall, salt-assisted and structurally modulated MOF glasses offer a promising route to modulate glass formation, coordination chemistry, and ion transport within a single design framework. Their greatest promise lies not simply in achieving ion conductivity, but in enabling processable, chemically programmable, and interface-adaptable ion conductors. If the challenges of mechanistic verification, long-term stability, interfacial compatibility, and scalable processing can be addressed, MOF-glass ion conductors could become a versatile platform for solid-state batteries, proton-conducting devices, ion-selective membranes, and even integrated electrochemical energy systems.