# Fileset

[Small - 2026 - Ma - Engineering Proton Conductive Metal Organic Glasses Through Secondary Network Formers.pdf](https://mdr.nims.go.jp/filesets/879c3475-6f27-4520-8e70-60808aec7de3/download)

## Creator

[Nattapol Ma](https://orcid.org/0000-0002-6162-1834), [Hideka Ando](https://orcid.org/0009-0004-1487-4478), [Renzhi Ma](https://orcid.org/0000-0001-7126-2006), [Takashi Nakanishi](https://orcid.org/0000-0002-8744-782X)

## Rights

[Creative Commons BY-NC Attribution-NonCommercial 4.0 International](https://creativecommons.org/licenses/by-nc/4.0/)

## Other metadata

[Engineering Proton Conductive Metal–Organic Glasses Through Secondary Network Formers](https://mdr.nims.go.jp/datasets/d11d0c24-f48d-4fc0-9356-8f45e77883bb)

## Fulltext

Engineering Proton Conductive Metal–Organic Glasses Through Secondary Network FormersSmallwww.small-journal.comRESEARCH ARTICLEEngineering Proton Conductive Metal–Organic GlassesThrough Secondary Network FormersNattapol Ma1 Hideka Ando1 Renzhi Ma2 Takashi Nakanishi21International Center for Young Scientists (ICYS), National Institute for Materials Science, Tsukuba, Ibaraki, Japan 2Research Center for MaterialsNanoarchitectonics (MANA), National Institute for Materials Science, Tsukuba, Ibaraki, JapanCorrespondence: Nattapol Ma (ma.nattapol@nims.go.jp)Received: 21 November 2025 Revised: 8 February 2026 Accepted: 9 February 2026Keywords: amorphous materials | coordination polymers | metal–organic frameworks | proton conductivities | secondary network formersABSTRACTCrystal–liquid–glass phase transitions in coordination polymers (CPs) and metal–organic frameworks (MOFs) have openednew opportunities for materials processing and for accessing novel or enhanced functionalities inherited from their crystallineprecursors. However, strategies to modulate the properties of the resulting glassy states, collectively referred to as metal–organicglasses (MOGs), have primarily relied on crystal engineering. Such approaches face intrinsic limitations, as the rare occurrenceof melting behavior in CPs/MOFs and the narrow compositional windows that sustain a stable liquid phase restrict access tonew structures and properties. Inspired by the compositional tunability of conventional oxide glass, this work explores a strategyto modulate MOG properties by incorporating inorganic zirconium hydrogen phosphate as a secondary network former. Wehypothesize that the mismatch between tetrahedrally coordinated Zn2+ in the parent MOG and octahedrally coordinated Zr4+in the additive induces distinct structural and functional modifications. By systematically varying the content of the zirconiumhydrogen phosphate, we demonstrate a linear increase in the glass transition temperature, viscosity, and anhydrous protonconductivity, reaching 2.6 mS cm−1 at 150◦C. These results highlight the potential of translating design principles from inorganicglass science to fine-tune the properties of MOGs.1Ti(atsndsgicTw©ShIntroductionhe recent discovery of crystal–liquid–glass phase transitionsn coordination polymers (CPs) and metal–organic frameworksMOFs) has opened new opportunities for materials processingnd for accessing novel or enhanced functionalities derived fromheir crystalline precursors [1–3]. This concept has attractedignificant attention, motivating a growing effort to identifyew meltable CPs/MOFs with unique properties [4, 5]. Toate, however, strategies to modify the properties of the glassytates of CPs/MOFs, collectively referred to as metal–organiclasses (MOGs), have primarily relied on crystal engineer-ng. In this approach, CPs/MOFs are first designed in theirrystalline form with the expectation that targeted function-his is an open access article under the terms of the Creative Commons Attribution-NonCommercial Liork is properly cited and is not used for commercial purposes.2026 The Author(s). Small published by Wiley-VCH GmbHmall, 2026; 22:e14459ttps://doi.org/10.1002/smll.202514459alities will persist after melt-quenching [6, 7]. Yet, the rarityof melting behavior in CPs/MOFs, observed in less than 1%of known CPs/MOFs [5, 8], and the narrow compositionalwindows that support a stable liquid state impose intrinsiclimitations on the accessible properties of MOGs. For instance,among the series M(H2PO4)2(1,2,4-triazole)2 (M = Zn2+, Cd2+,Mn2+, Co2+, Fe2+, Cr2+), only the Zn2+ analogue melts uponheating, while the others remain crystalline until thermaldecomposition [9–12]. Developing approaches that allow directtuning of MOG properties, without requiring the synthesisof new meltable crystals, would therefore greatly expand thedesign space and unlock broader applications, particularly forsystems in which crystal-to-glass transitions are difficult toachieve.cense, which permits use, distribution and reproduction in any medium, provided the original1 of 10http://www.small-journal.comhttps://doi.org/10.1002/smll.202514459https://orcid.org/0000-0002-6162-1834https://orcid.org/0009-0004-1487-4478https://orcid.org/0000-0001-7126-2006https://orcid.org/0000-0002-8744-782Xmailto:ma.nattapol@nims.go.jphttp://creativecommons.org/licenses/by-nc/4.0/https://doi.org/10.1002/smll.202514459http://crossmark.crossref.org/dialog/?doi=10.1002%2Fsmll.202514459&domain=pdf&date_stamp=2026-02-13FIGURE 1 Schematic illustration of (A) conventional approach in metal–organic glass properties modifications via crystal engineering and (B)the proposed structural modification process via secondary network former addition in this work.TtpboEttwagmetoeigs“cva[pIe(dcosr2 16136829, 2026, 22, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202514459 by Nattapol Ma - National Institute For , Wiley Online Library on [19/04/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creo date, controlling MOG properties, including viscosity [13],hermal behavior [7], mass and ion transport [14], mechanicalroperties [15], optics [16] and magnetics [17] responses, andeyond, remains a central challenge, as only a small numberf CPs/MOFs can form stable liquids upon heating (Figure 1A).ven slightmodifications in crystal composition potentially causehe material to retain its crystalline state up to the decompositionemperature, rather thanmelting [9, 11, 18]. This contrasts sharplyith conventional oxide glasses, where property control is rarelychieved by designing new crystalline precursors. Instead, oxidelass networks are continuously tuned through compositionalodification with network modifiers, secondary network form-rs (intermediates), and other additives [19, 20]. Such composi-ional flexibility enables the formation of an enormous diversityf oxide glasses with tunable atomic structures and physical prop-rties, including transparency, chemical durability, and mechan-cal strength, supporting applications ranging from architecturallass [21] to nuclear waste immobilization [22]. Inspired by theuccess in oxide glasses, recent studies have explored the use ofnetwork modifiers” in MOGs to induce depolymerization andhain fragmentation through ligand exchanges, thereby reducingiscosity or altering thermal behavior in selected systems suchs zinc–phosphate–azole CPs [23, 24], metal-bis(acetamides)25], and zeolite imidazolate frameworks [13]. However, com-osition tuning in glasses is not limited to depolymerization.t can also involve incorporating “secondary network form-rs” or “intermediates” that participate in network formationpolymerization), thereby enhancing structural stability or intro-ucing new physical properties through cross-linking or chargeompensation of modifier ions [26]. Yet, to date, no examplesf MOG property modulation through the incorporation ofuch secondary network formers or intermediates have beeneported.of 10aIn this article, we present a strategy to modulate the ther-mal properties, viscosity, and anhydrous proton conductivityof a representative zinc–phosphate–azole-based MOG throughthe systematic incorporation of zirconium hydrogen phosphate,which potentially acts as a secondary network former (Figure 1B).This role is analogous to that of Al2O3 as a network-formingoxide in conventional silica-based glasses. We hypothesize thatintroducing octahedrally coordinated Zr4+ into the tetrahedrallycoordinated Zn2+ chains of the parentMOG introduces additionalphosphate sites, which facilitate polymerization and strengthenhydrogen-bonding connectivity (Figure 2A). These structuralmodifications are expected to influence the macroscopic prop-erties of the resulting glasses strongly. We demonstrate that theglass transition temperature, viscosity, and anhydrous protonconductivity can all be tuned through compositional control.Synchrotron X-ray total scattering, pair distribution function(PDF) analysis, thermal and rheological measurements, andpositron annihilation lifetime spectroscopy (PALS) reveal prop-erty trends consistent with the behavior typically associated withsecondary network formers in conventional oxide glasses, therebyenabling direct compositional tuning of the glasses‘ macroscopicproperties.2 Results and Discussion2.1 Synthesis, Melting Behavior, and GlassFormationA representative CP, [Zn(HPO4)(H2PO4)2](H2Im)2 (HIm = imi-dazole, Figure S1), was selected because it forms a stable liquidstate and is broadly compatible with functional dopants [23, 27].The compound, hereafter referred to as ZnPIm, was synthesizedSmall, 2026tive Commons LicenseFIGURE 2 (A) Schematic illustration of the hypothesized structure tuning in metal–organic glasses through the incorporation of secondarynetwork formers. (B) First and (C) second heating DSC profiles of all samples under N2 atmosphere measured at a heating rate of 10◦C min−1.(D) Variation of melting temperature (Tm), glass transition temperature (Tg), enthalpy change (∆H), and the Tg/Tm ratio as a function of the molefraction (x) of ZrP in ZnPIm1−xZrPx.odsbHtZs((TtpratveSdAZeTfTZftS 16136829, 2026, 22, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202514459 by Nattapol Ma - National Institute For , Wiley Online Library on [19/04/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Cren the gram scale following previously reported procedures (seeetailed method in Supporting Information) [23, 27]. ZnPIm con-ists of 1D chains of tetrahedrally coordinated Zn2+ ions bridgedy orthophosphate (HPO42−). Two additional monocoordinated2PO4− anions complete the coordination sphere. Charge neu-rality is maintained by two interchain imidazolium cations pern2+. The powder X-ray diffraction (PXRD) pattern of the as-ynthesized ZnPIm matched well with the simulated patternFigure S2). Heating ZnPIm above its onset melting temperatureTm = 155.7◦C) resulted in crystalmelting (Figure 2B,C; Figure S3).he liquid phase remained stable up to ca. 195◦C, beyond whichhermal decomposition began. Cooling the melts prior to decom-osition yielded a vitrified glassy state, denoted ZnPIm-g. Uponeheating, differential scanning calorimetry (DSC) confirmedn onset glass transition temperature (Tg) of 11.0◦C. At roomemperature (25◦C), ZnPIm-g remains a practically stable, highlyiscous supercooled liquid, showing no signs of crystallizationven at a slow DSC scan rate of 0.5◦C min−1 up to 160◦C (Figure4) [28]. Further details of temperature-dependent viscosity areiscussed in the viscoelasticity section.morphous zirconium hydrogen phosphate with the formular(HPO4)2∙nH2O (ZrP) was selected as an additive and isxpected to act as a secondary network former (Figure S5).he amorphous nature of ZrP is expected to diminish inter-acial mismatch in comparison to its crystalline counterparts.he short-range structure of ZrP closely resembles that of γ-r(H2PO4)(PO4)⋅2H2O (γ-ZrP), as confirmed by pair distributionunction analysis (Figures S6–S8) [29]. Further details of thisechnique will be discussed in a later section. The crystalline γ-mall, 2026phase consists of octahedrally coordinated Zr4+ ions, each boundto two H2PO4− and four distinct PO43− groups. Upon heating to1200◦C under N2 atmosphere during thermogravimetric analysis(TGA), ZrP exhibits continuous weight loss, reaching a total of∼14.95% (see details in Figure S10). This weight loss is attributedto the removal of one water molecule per Zr(HPO4)2 unit viaphosphate condensation [30, 31], while the remaining loss isassigned to residual water in the ZrP. Based on these observations,the composition of ZrP was estimated to be Zr(HPO4)2∙1.6H2O.Preparation of modified MOGs began by homogenizing as-synthesized ZnPIm and ZrP at different mole ratios using a mixermill operated at 15 Hz for 10 min (Figure 1B). The resultingmixtures are referred to as ZnPIm1−xZrPx, where x represents themole fraction of ZrP (x= 0.1, 0.2, 0.3, 0.4, 0.5, and 0.7) (Figure S12).PXRD patterns of all mixtures closely resembled that of ZnPIm,with no evidence of ZrP crystallization (Figure S13). The actual Zrto Zn mole fractions of all samples were analyzed by inductivelycoupled plasma optical emission spectroscopy (ICP-OES, TableS1). The thermal stability of the mixtures was comparable to thatof ZnPIm. However, samples with higher ZrP content exhibitedgreater weight loss below 200◦C, attributed to the release of ZrP’swater molecules (Figures S14–S20).Upon heating at 10◦C min−1, DSC profiles of all partiallydehydrated samples (see the detailed method in the SupportingInformation) exhibited a single endothermic melting peak, withonset Tm ranging from 148.7◦C to 156.3◦C (Figure 2B,D; Table S2).This variation in Tm is considerably smaller than that typicallyobserved in systems displaying eutectic behavior [23] or flux3 of 10ative Commons LicensemCmfTZffbTsh2saidZStmgii(oo[fcwccchcscoAlccrTTaoeaiwtayZ(mt4 16136829, 2026, 22, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202514459 by Nattapol Ma - National Institute For , Wiley Online Library on [19/04/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creaelting [24], as previously reported for the Zn-azole-phosphateP family. The total enthalpy change (ΔH) associated withelting decreased progressively with increasing ZrP content,rom 45.5 kJ mol−1 for ZnPIm-g to 5.6 kJ mol−1 for ZnPIm0.3ZrP0.7.his trend is consistent with the reduced fraction of meltablenPIm in the mixtures. When the ΔH is normalized to the moleraction of ZnPIm in each sample, the value initially increasesrom 45.5 kJ mol−1 for ZnPIm to 50.2 kJ mol−1 for ZnPIm0.9ZrP0.1efore gradually decreasing to 44.8 kJ mol−1 for ZnPIm0.5ZrP0.5.he increase in the normalized ΔH relative to pristine ZnPImuggests the dissolution event of ZrP into the ZnPIm melts. Atigher ZrP contents, the normalized ΔH further decreases to0.3 kJ mol−1 for ZnPIm0.3ZrP0.7 (Table S2). This reduction likelytems from differences in the extent of mechanically inducedmorphization of ZnPIm during the milling process, particularlyn ZrP-rich systems [32–34]. This observation is supported by theecreasing crystallinity levels derived from the PXRD data of thenPIm1−xZrPx samples (Figure S21).ubsequent DSC heating measurements revealed a single glassransition for all samples (Figure 2C; Figure S22–S28) [35]. Theelt-quenchedmaterials are hereafter denoted as ZnPIm1−xZrPx-, and Tg values are reported as onset temperatures. A linearncrease in Tg was observed with higher ZrP content, increas-ng from 11.0◦C for ZnPIm-g to 57.4◦C for ZnPIm0.3ZrP0.7-gFigure 2D; Table S2). Correspondingly, the Tg/Tm ratio (basedn absolute temperature), an indicator that reflects the tendencyf a liquid to form a glass upon cooling rather than crystallizing36, 37], also increases linearly with ZrP content, from 0.66or ZnPIm-g to 0.78 for ZnPIm0.3ZrP0.7-g. This increase in Tgontrasts with all previous examples in other MOG systems,here additives typically act as network modifiers, promotinghain fragmentation and thereby lowering the Tg of the parentompounds [13, 24, 25]. Binary glass systems composed of MOGonstituents generally exhibit either a weighted-average Tg inomogeneously mixed glasses [38, 39], or distinct Tg valuesorresponding to the individual parent components in phase-eparated systems [40]. The effect of ZrP observed here morelosely resembles that of Al3+ as a secondary network formerr intermediate in Na2O–Al2O3–SiO2 glasses, where increasingl3+ content promotes the degree of polymerization (or cross-inking) and, consequently, raises Tg [26]. In addition, the heatapacity change at Tg (∆Cp) decreases with increasing ZrPontent (Figures S29 and S30), suggesting enhanced networkigidity and a reduced configurational degree of freedom aboveg [26, 41].o verify that the observed increase in Tg arises from ZrPnd parallels the effect typically observed when adding sec-ndary network formers in conventional oxide glasses, a controlxperiment was conducted using Al2O3 as a chemically inertdditive. Under the relatively mild melting conditions employedn this work (< 175◦C), Al2O3 remains unreactive. This contrastsith aluminosilicate glass systems, which require processingemperatures above 1100◦C [26]. An equimolar mixture of ZnPImnd Al2O3 was prepared following the same synthetic procedure,ielding ZnPIm0.5[Al2O3]0.5 (Figure S31). The DSC profile ofnPIm0.5[Al2O3]0.5 exhibits a single endothermic peak of meltingTm) at 159.2◦C (Figure S32). More importantly, the Tg of theelt-quenched sample (ZnPIm0.5[Al2O3]0.5-g) remained essen-ially unchanged at 10.5◦C, compared to 11.0◦C of the ZnPIm-g.of 10This value is significantly lower than the 36.3◦C observed forZnPIm0.5ZrP0.5-g with comparable additive contents, suggestingthat the higher Tg potentially arises from the contribution of ZrP,rather than from the mere presence of an inert additive.2.2 Morphology and Free Volume of MOGsModified MOGs were prepared on a larger scale by heatingZnPIm1−xZrPx under an inert Ar atmosphere at 175◦C for 10 min(see the detailed method in the Supporting Information), thenallowing it to cool naturally to room temperature (Figure S33).This process yielded amorphous solids that exhibited broaddiffuse scattering features in their XRD patterns, confirming theabsence of long-range order (Figure S34). Thermogravimetricanalysis results for all modified MOGs prepared by meltingat 170◦C under N2 atmosphere show less than 1% weight lossup to at least 190◦C (Figures S35–S41). A lower overall weightloss at 500◦C was also observed with increasing ZrP content.Additionally, TGA results to 1200◦C of all modified MOGssuggest that any residual water after melt-quenching is belowthe detection limit of TGA-based analysis (Figure S42 and TableS3). Fourier transform infrared (FTIR, Figure S43) spectra ofrepresentative ZnPIm0.8ZrP0.2-g are largely identical to thoseof ZnPIm-g, indicating that the imidazole molecules remainprotonated and do not coordinate to either Zn2+ or Zr4+ ions inthe melt-quenched MOGs. The observed water-related signalsare attributed to the hygroscopic nature of the samples, sinceFTIR measurements were performed under ambient conditions.Scanning electron microscopy (SEM, Figures S44–S51) revealedsmooth, grain-boundary-free surfaces with no evidence of phasesegregation from ZnPIm-g up to ZnPIm0.6ZrP0.4-g in micrometerlength scale. Within this composition range, energy-dispersiveX-ray (EDX, Figures S52–S57) mapping further confirmed thatboth constituents were homogeneously distributed throughoutthe observed regions. At higher ZrP contents, specifically inZnPIm0.5ZrP0.5-g and ZnPIm0.3ZrP0.7-g, distinct Zn- and Zr-richdomains become clearly visible, with the distinction betweenthem becoming increasingly pronounced as the ZrP concentra-tion increases. The melt-quenched samples remain amorphousfor at least 2 weeks under ambient conditions, as confirmed bythe absence of Bragg peaks in the PXRD patterns (Figure S58).We further utilized positron annihilation lifetime spectroscopy(PALS, Figure 3) to monitor changes in the pore radius of thefree volume within the modified MOGs, providing an indirectassessment of phase segregation and local structures. The lifetime(τ3) and relative intensity (I3) of the longest-lived component,orthopositronium (o-Ps), are correlated with the average pore(cavity) radius and the relative number of cavities, respectively[42]. Consequently, these parameters reflect the distribution offree spaces within the samples. Mismatched interfaces betweensegregated phases are expected to generate interfacial cavities,thereby increasing the overall cavity size in the samples [12, 43].The estimated pore radii (and τ3) for pure ZnPIm-g and ZrP are0.241 nm (1.56 ns) and 0.286 nm (2.00 ns), respectively. In themodified MOGs, the free volume decreases below that of pristineZnPIm-g for compositions of ZnPIm1−xZrPx-g with x < 0.5, sug-gesting enhanced miscibility between the two components [44].When x> 0.5, the free volume begins to increase again, consistentwith the phase segregation observed in the SEM images ofSmall, 2026tive Commons LicenseFIGURE 3 Pore radius as a function of composition ofZnPIm1−xZrPx-g and PALS parameter (τ33I3). These cavity sizeswere derived from the orthopositronium (o-Ps) lifetime component(τ3) obtained via Positron Annihilation Lifetime Spectroscopy (PALS).The black line and blue dashed line represent the pore radius valueof ZnPIm-g and linear additive fractional free volume reference lines,respectively. See Figure S59 for the PALS spectra and Table S4 for thecomplete parameters.ZdcTaribd2TrXyiphcTP(PoispdfS 16136829, 2026, 22, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202514459 by Nattapol Ma - National Institute For , Wiley Online Library on [19/04/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable CreanPIm0.5ZrP0.5-g and ZnPIm0.3ZrP0.7-g. Furthermore, the PALS-erived free volume parameter (τ33I3) as a function of MOGomposition displays a sub-additive trend across all compositions.his behavior indicates that the reduced free volume limits over-ll molecular mobility, thereby restricting local conformationalearrangements [28, 45]. Consequently, the observed increasen Tg with increasing ZrP content, as determined by DSC, cane attributed to the restricted molecular mobility arising fromecreased free volume..3 Short-to-Intermediate Range Structurehe influence of ZrP incorporation on the short- to intermediate-ange structure of the modified MOGs was investigated using-ray total scattering and pair distribution function (PDF) anal-ses (Figure 4). This technique provides real-space structuralnformation by quantifying the probability of finding atomicairs at specific interatomic distances, represented as weightedistograms of atom–atom correlations [46]. All samples wereharacterized both before and after the melt-quenching process.he peak assignments here were confirmed by calculated partialDFs derived from single-crystal structures of ZnPIm and γ-ZrPFigures S7 and S60) [29, 47, 48]. Prior to melt-quenching, theDF profiles of the mixtures closely resembled a superpositionf the parent ZnPIm and ZrP components, with relative peakntensities varying according to the composition of each con-tituent (Figure 4; Figure S61). In the mixed samples, a neweak was observed at approximately 5.6 Å (marked by the redashed line), located between the characteristic 5.2 and 5.8 Åeatures of ZrP and ZnPIm, respectively, corresponding to themall, 2026Zr⋯Zr1 and Zn⋯Zn correlations (metal–ligand–metal connectiv-ity). However, the origin of this feature cannot be unambiguouslyassigned based on the present data.After melt-quenching, the reciprocal-space total scattering struc-ture factor, S(q), showed the disappearance of Bragg peaks,confirming the transformation to an amorphous phase (FiguresS62–S68). The PDF features of the resulting modified MOGsremain largely consistent with those of their parent mixturesat pair distances below approximately 6.5 Å, but with reducedamplitudes. This observation indicates partial preservation ofshort-range metal–ligand–metal connectivity, including Zr⋯Zrand Zn⋯Zn, consistent with the retention of local coordinationnetworks. At shorter pair distances (below 3.5 Å), the PDFfeatures are nearly identical to those observed before melt-quenching, confirming the retention of the local tetrahedraland octahedral coordination geometries around Zn2+ and Zr4+centers [24]. At longer pair distances (above 7 Å), the pronouncedbroadening of the PDF peaks signifies the loss of long-rangestructural order accompanying vitrification [49]. 91Zr solid-statenuclear magnetic resonance (NMR) reveals subtle differencesbetween pristine ZrP and melt-quenched ZnPIm0.5ZrP0.5-g, sug-gesting modest changes in the local environment around theZr4+ centers (Figure S77) [50]. However, further interpretation islimited by the relatively broad NMR signals. Based on structuralcharacterization, ZnPIm-g and ZnPIm0.8ZrP0.2-g were chosen asrepresentative samples for further characterization to assess howZrP incorporation affects the viscoelastic and proton-conductingbehaviors of the samples.2.4 ViscoelasticityThe influence of ZrP on mechanical response was studied bycomparing ZnPIm-g and ZnPIm0.8ZrP0.2-g samples using rhe-ological analysis. Temperature-dependent dynamic mechanicalanalysis (DMA) and viscosity measurements were conductedto examine the softening behavior upon heating. Two distinctviscoelastic regions were identified, as indicated by the dashedlines in Figure 5A,B [51]. At lower temperatures, both sam-ples exhibit solid-like elasticity, with a nearly constant storagemodulus (G″) exceeding the loss modulus (G′). Upon heat-ing, softening begins at approximately 12.3◦C for ZnPIm-g and36.4◦C for ZnPIm0.8ZrP0.2-g. Further heating to above the mainrelaxation temperature (Tα), at 30.4◦C for ZnPIm-g and 60.8◦Cfor ZnPIm0.8ZrP0.2-g, leads to a transition into viscous, liquid-like behavior. The pronounced increase in both softening andrelaxation temperatures in ZnPIm0.8ZrP0.2-g demonstrates thatZrP incorporation enhances network rigidity, thereby stabilizingsolid-like behavior at elevated temperatures. Note that the Tαvalue is frequency dependent [52].Temperature-dependent viscosity profiles of both ZnPIm-gand ZnPIm0.8ZrP0.2-g follow the Mauro-Yue-Ellison-Gupta-Allan(MYEGA) equation (Figure 5C; Table S5) [58]. The glass transi-tion temperatures derived from the viscosity curves (Tg,η) wereestimated by extrapolating the viscosity to 1012 Pa∙s. [58, 59] Adistinct increase in Tg,η was observed, from 6.5◦C for ZnPIm to31.5◦C for ZnPIm0.8ZrP0.2-g, consistent with the trend observedin DSC experiments of the modified MOGs. Throughout themeasurement range, the viscosity profile of ZnPIm0.8ZrP0.2-g5 of 10tive Commons LicenseFIGURE 4 Local structures and pair distribution functions (PDFs) of modified samples. (A) Local coordination environment and pair-distancelabeling around (A) Zn2+ in ZnPIm and (B) Zr4+ in γ-ZrP. Zn and Zr are shown as polyhedra, while P, O, C, and N atoms are represented in orange, red,gray, and light blue, respectively. (C) Experimental PDFs of ZnPIm, ZrP, and a series of ZnPIm1−xZrPx samples and their melt-quenched counterparts,ZnPIm-g and ZnPIm1−xZrPx-g. Peak labels are assigned based on partial PDFs (Figures S7 and S60) simulated from crystal structures of γ-ZrP [29] andZnPIm [47]. Additional S(q) data and extended PDF data up to 30 Å are provided in Figures S8 and S62–S76.stpivaUZIuZrina[rfbceccbaTZm6 16136829, 2026, 22, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202514459 by Nattapol Ma - National Institute For , Wiley Online Library on [19/04/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Crehifts markedly toward higher temperatures. For example, theemperature at which the viscosity reaches the standard workingoint (η = 103 Pa∙s) increases from 73◦C for ZnPIm-g to approx-mately 109◦C for ZnPIm0.8ZrP0.2-g. At these temperatures, theiscosity of both samples is comparable to molten soda-lime glassbove 1100◦C and is suitable for industrial forming processes [60].pon further heating to approximately 150◦C, the viscosity ofnPIm-g decreases below the practicalmelting point (η< 10 Pa∙s).n contrast, ZnPIm0.8ZrP0.2-g maintains viscosities above 10 Pa∙sp to at least 160◦C. For comparison, at 73◦C, theworking point ofnPIm-g (η = 103 Pa∙s), the viscosity of ZnPIm0.8ZrP0.2-g alreadyeaches 2.2 × 105 Pa∙s. This substantial enhancement in viscositys consistent with the hypothesis that ZrP may act as a secondaryetwork former, potentially promoting polymerization throughdditional phosphate coordination associated with Zr4+ species14, 26]. However, this interpretation is based on macroscopicheological behavior and does not constitute direct evidenceor specific atomic-scale cross-linking. Comparable behavior haseen reported in synthetically controlled MOGs with varyingoordination-network sizes, where materials possessing morextended coordination networks exhibit significantly higher vis-osity (4.0 × 104 Pa∙s at 120◦C) than their highly fragmentedounterparts (0.68 Pa∙s at 120◦C) [14]. Similar trends have alsoeen reported for Na2O–Al2O3–SiO2 glasses incorporating Al3+ assecondary network former [26].he supercooled liquid characteristics of ZnPIm-g andnPIm0.8ZrP0.2-g were compared with those of various referenceaterials using a fragility diagram (Figure 5D) [6, 41, 53–56].of 10aThe fragility indices (m) of ZnPIm-g and ZnPIm0.8ZrP0.2-gwere estimated to be 83.0 ± 16.1 and 79.9 ± 16.2, respectively,with the values overlapping within experimental uncertainty.Compared with other systems, both materials exhibit higherfragility than inorganic ZnCl2, with intermediate fragility (m =23) [61], and show fragility comparable to that of the metallicglass Pt60Ni15P25 (m = 67.2) [55]. Relative to previously reportedMOGs, the fragility indices of ZnPIm-g and ZnPIm0.8ZrP0.2-gfall between Zn(imidazolate)1.75(benzimidazolate)0.25 (ZIF-62)[6], with an exceptionally low m of 23, and the relatively fragile[Zn3(H2PO4)6(H2O)3](1,2,3-benzotriazole) (m = 124) [24].2.5 Anhydrous Proton ConductivityThe proton conductivities of ZnPIm-g, ZnPIm0.8ZrP0.2-g, andZnPIm0.5ZrP0.5-g were compared under anhydrous conditions,representing the pristine MOG, a modified sample with a homo-geneous ZrP distribution, and a modified sample exhibiting clearphase segregation, respectively. All samples, in their moltenstates, were loaded into a conductivity cell of fixed geometry. Con-ductivity measurements were carried out by alternating current(AC) impedance spectroscopy (Figures S78 and S79). Each samplewas equilibrated at the target temperature for at least 3 h prior tomeasurement, and all experiments were performed under a dryAr atmosphere to maintain anhydrous conditions. Note that thetemperature range used for proton conductivity measurementsalso covers temperatures above Tg for both materials, where theybehave as highly viscous supercooled liquids.Small, 2026tive Commons LicenseFIGURE 5 Viscoelasticity and anhydrous proton conductivity of melt-quenched samples. Temperature-dependent storage modulus (G’) and lossmodulus (G”) of (A) ZnPIm-g and (B) ZnPIm0.8ZrP0.2-g measured at 1 Hz, 0.02% strain, and a cooling rate of 2◦C min−1. (C) Temperature-dependentviscosity of ZnPIm-g and ZnPIm0.8ZrP0.2-g. (D) Fragility plot of ZnPIm-g and ZnPIm0.8ZrP0.2-g in comparison with reference compounds. Data arereproduced from ref. [6, 41, 53–56]. (E) Variable-temperature proton conductivity of ZnPIm-g, ZnPIm0.8ZrP0.2-g, and ZnPIm0.5ZrP0.5-g under a dry Aratmosphere. Data for ZnPIm-g are reproduced from ref. [23]. Fitting parameters are available in Tables S6 and S7. Conductivity values and errors areavailable in Table S8. (F) Walden plots of ZnPIm-g and ZnPIm0.8ZrP0.2-g compared with 85% and 98% H3PO4, and representative protic ionic liquids.Reference data of H3PO4, protic ionic liquids, and the calibrated ideal KCl line are taken from Ref. [54, 57].TZ[SmZsacrc×l5cadnce1ppciiieS 16136829, 2026, 22, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202514459 by Nattapol Ma - National Institute For , Wiley Online Library on [19/04/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Crehe temperature-dependent proton conductivity of pristinenPIm-g follows Vogel–Fulcher–Tammann (VFT)-type behavior62–64], exhibiting a maximum proton conductivity of 1.1 × 10−4cm−1 at 110◦C (Figure 5E) [23]. Above this temperature, reliableeasurements could not be obtained because the viscosity ofnPIm-g decreases below its physical melting point, leading toample leakage from the measurement setup. Incorporation ofsmall amount of ZrP in ZnPIm0.8ZrP0.2-g increases the protononductivity by nearly five-fold compared to ZnPIm-g at 110◦C,eaching 5.1 × 10−4 S cm−1. Upon further heating to 150◦C, theonductivity value of ZnPIm0.8ZrP0.2-g reaches a maximum at 2.610−3 S cm−1. In contrast, ZnPIm0.5ZrP0.5-g exhibits significantlyower proton conductivity, 3.9 × 10−6 S cm−1 at 110◦C and.1 × 10−5 S cm−1 at 150◦C. This reduction in conductivity isonsistent with the phase segregation observed in SEM/EDXnalyses, which reveal the formation of ZrP-rich domains. Theseomains likely disrupt the continuity of the proton-conductingetwork, particularly given the relatively low intrinsic protononductivity of ZrP. For example, Zr(PO4)(H2PO4)⋅2H2O (γ-ZrP)xhibits anhydrous proton conductivity of 8 × 10−8 S cm−1 at80◦C [65]. Although the maximum conductivity achieved in theresent work remains below the benchmark value required forractical applications (10−2 S cm−1) [66], the enhanced protononductivity of ZnPIm0.8ZrP0.2-g is achieved without a decreasen Tg or viscosity relative to pristine ZnPIm-g (Table S9). Futuremprovements in proton conductivity may be realized by replac-ng the parent MOG and/or the additive with components thatxhibit higher intrinsic proton conductivity.mall, 2026All three examples display VFT-type temperature dependence,indicating that proton transport is coupled to structural dynamicsand facilitated by structural relaxation, a behavior characteristicof glassy or polymeric ion conductors [59]. This differs fromArrhenius-type behavior, typically associated with thermallyactivated hopping of charge carriers in crystalline conductors.The VFT fitting parameters are summarized in Table S10. Theactivation factor (B) decreases in the order ZnPIm0.5ZrP0.5-g(5423 K) > ZnPIm-g (2626 K) > ZnPIm0.8ZrP0.2-g (1725 K),indicating progressively weaker temperature dependence andlower apparent activation energies for proton conduction [59, 67].The relationship between viscosity (η) and equivalentconductivity (Λm) of ZnPIm-g and ZnPIm0.8ZrP0.2-g wasexamined using a Walden plot to elucidate the underlyingproton-conductivity mechanisms (Figure 5F) [68, 69]. AlongWalden’s ideal line, conductivity is directly proportional tofluidity, indicating that ion migration is fully coupled to theviscous flow of the medium, a characteristic of dilute fluidsystems where ion transport follows the vehicle mechanism[59]. In our study, both samples fall within the superionicregion, deviating substantially from Walden’s ideal line. Suchdeviations indicate that charge carriers are partially decoupledfrom the matrix dynamics and that proton transport is governedpredominantly by the Grotthuss mechanism rather thanviscous diffusion. For reference, phosphoric acid, which islocated slightly above Walden’s ideal line, exhibits ∼97% ofits proton conductivity via the Grotthuss mechanism [59, 70].7 of 10ative Commons LicenseIfppZlvcZcvηttdt3TtetptclwavrswgrniiglrwhasANfNbTBe2maKa8 16136829, 2026, 22, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202514459 by Nattapol Ma - National Institute For , Wiley Online Library on [19/04/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Crean our systems, the coordination networks within the MOGramework likely restrict counter-ion mobility while facilitatingroton hopping along interconnected hydrogen-bondathways [14, 71]. Compared with the pristine ZnPIm-g,nPIm0.8ZrP0.2-g shows a larger deviation from Walden’s idealine toward the superionic region. This shift reflects its higheriscosity and enhanced proton conductivity, implying weakeroupling between structural relaxation and proton transport.nPIm0.8ZrP0.2-g maintains a broader viscoelastic proton-onducting regime than ZnPIm-g before transforming into aiscous liquid upon surpassing the working-point threshold (log−1 =−4 Poise−1) [56]. These observations support our hypothesishat ZrP incorporation introduces additional phosphate siteshat improve hydrogen-bond connectivity, thereby stabilizingynamic proton-conducting pathways and promoting long-rangeransport.Conclusionhis study presents a strategy tomodulate the properties ofMOGshrough the incorporation of amorphous ZrP, which is hypoth-sized to act as a secondary network former. We demonstratehat ZrP significantly influences the glassy-state properties ofroton-conductive ZnPIm, enabling controlled tuning of the glassransition temperature, viscoelasticity, and anhydrous protononductivity. By systematically varying the amount of ZrP, ainear increase in Tg from 11.0◦C to 57.4◦C is achieved, togetherith anhydrous proton conductivities up to 2.6 × 10−3 S cm−1t 150◦C and over two orders of magnitude enhancement iniscosity at 73◦C relative to the pristine MOG. Synchrotron X-ay total scattering and PDF analyses indicate preservation ofhort-range metal–ligand–metal connectivity in the glassy state,hile both metal centers retain their characteristic coordinationeometries. On the basis of these macroscopic thermal andheological trends, we envisage that ZrP functions as a secondaryetwork-forming component that promotes increased polymer-zation within the coordination network, analogous to the role ofntermediate or network-forming oxides in conventional silicatelasses. Direct microscopic experimental evidence for this cross-inking mechanism is not provided in the present work andemains an important topic for future investigation. Overall, thisork establishes a foundation for the compositional design ofybrid glasses, bridging the principles from oxide glass sciencend coordination chemistry to enable new routes for tailoring thetructure, thermal behavior, and functional properties of MOGs.cknowledgements.M. acknowledges the support from ICYS for a research fellowship,rom the JST PRESTO grant JPMJPR25MB, from JSPS KAKENHI Grantumbers JP24K23109 and JP25K18055, from the Sumitomo Foundationasic science grant number 2402150, and from the Iketani Science andechnology Foundation grant number 0371207-A. We acknowledge theL04B2 beamlines at SPring-8 for the synchrotron X-ray total scatteringxperiments with the approval of JASRI (Proposal Numbers 2024B1167,025A1067, and 2025B1248). This work was supported by the World Pre-ier International Research Center Initiative (WPI), MEXT, Japan. Wecknowledge support fromDr. Hiroki Yamada (SPring-8) and Dr. Sorachaosasang (NIMS) during the X-ray total scattering measurement. Wecknowledge Dr. Soracha Kosasang (NIMS) for high-temperature TGAof 10measurements. We thank Dr. Shinsuke Ishihara (NIMS) for providingaccess to FTIR.We acknowledge support from theNIMS Surface and BulkAnalysis Unit and NIMS Nanofabrication Facilities.Conflicts of InterestThe authors declare no conflict of interest.Data Availability StatementThe data that support the findings of this study are available from thecorresponding author upon reasonable request.References1. D. Umeyama, S. Horike, M. Inukai, T. Itakura, and S. Kitagawa,“Reversible Solid-to-Liquid Phase Transition of Coordination PolymerCrystals,” Journal of the American Chemical Society 137 (2015): 864–870,https://doi.org/10.1021/ja511019u.2. T. D. Bennett, J.-C. Tan, Y. Yue, et al., “Hybrid Glasses From Strongand FragileMetal-Organic Framework Liquids,”Nature Communications6 (2015): 8079, https://doi.org/10.1038/ncomms9079.3. Y. Hirai, T. Nakanishi, Y. Kitagawa, et al., “Luminescent Coordina-tion Glass: Remarkable Morphological Strategy for Assembled Eu(III)Complexes,” Inorganic Chemistry 54 (2015): 4364–4370.4. N. Ma and S. Horike, “Metal–Organic Network-Forming Glasses,”Chemical Reviews 122 (2022): 4163–4203.5. N. Ma, S. Kosasang, E. K. Berdichevsky, T. Nishiguchi, and S. Horike,“Functional Metal–organic Liquids,” Chemical Science 15 (2024): 7474–7501.6. A. Qiao, T. D. Bennett, H. Tao, et al., “A Metal-Organic Frame-work With Ultrahigh Glass-Forming Ability,” Science Advances 4 (2018):aao6827, https://doi.org/10.1126/sciadv.aao6827.7. L. Frentzel-Beyme, M. Kloß, P. Kolodzeiski, R. Pallach, and S. Henke,“Meltable Mixed-Linker Zeolitic Imidazolate Frameworks and TheirMicroporous Glasses: From Melting Point Engineering to SelectiveHydrocarbon Sorption,” Journal of the American Chemical Society 141(2019): 12362–12371, https://doi.org/10.1021/jacs.9b05558.8. T. D. Bennett and S. Horike, “Liquid, Glass andAmorphous Solid Statesof Coordination Polymers and Metal–Organic Frameworks,” NatureReviews Materials 3 (2018): 431–440, https://doi.org/10.1038/s41578-018-0054-3.9. D. Umeyama, N. P. Funnell, M. J. Cliffe, et al., “Glass Formation viaStructural Fragmentation of a 2DCoordinationNetwork,”Chemical Com-munications 51 (2015): 12728–12731, https://doi.org/10.1039/C5CC04626B.10. W. Chen, S. Horike, D. Umeyama, et al., “Glass Formation of aCoordination Polymer Crystal for Enhanced Proton Conductivity andMaterial Flexibility,” Angewandte Chemie International Edition 55 (2016):5195–5200, https://doi.org/10.1002/anie.201600123.11. Y. Ohara, A. Hinokimoto, W. Chen, et al., “Formation of CoordinationPolymer Glass by Mechanical Milling: Dependence on Metal Ions andMolecular Doping for H+ Conductivity,” Chemical Communications 54(2018): 6859–6862.12. S. Kosasang, N. Ma, D. L. Christensen, M. M. Smedskjaer, H. Yamada,and S. Horike, “Metal–Organic Crystallized Glasses: MicrostructureFormation and Their Properties,” Journal of the American ChemicalSociety 147 (2025): 33870–33878.13. F. Cao, S. S. Sørensen, A. K. R. Christensen, et al., “ContinuousStructure Modification of Metal-organic Framework Glasses via HalideSalts,” Nature Communications 16 (2025): 7001, https://doi.org/10.1038/s41467-025-62143-9.14. T. Ogawa, K. Takahashi, T. Kurihara, et al., “Network Size Controlin Coordination Polymer Glasses and Its Impact on Viscosity and H+Conductivity,” Chemistry of Materials 34 (2022): 5832–5841.Small, 2026tive Commons Licensehttps://doi.org/10.1021/ja511019uhttps://doi.org/10.1038/ncomms9079https://doi.org/10.1126/sciadv.aao6827https://doi.org/10.1021/jacs.9b05558https://doi.org/10.1038/s41578-018-0054-3https://doi.org/10.1039/C5CC04626Bhttps://doi.org/10.1002/anie.201600123https://doi.org/10.1038/s41467-025-62143-91MN1H(1SGs1ML41od2OTC2Ra2tJA2PA2PI22C2“aA2HP92fS(2LPmC3TSt3mS2S 16136829, 2026, 22, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202514459 by Nattapol Ma - National Institute For , Wiley Online Library on [19/04/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applic5. M. Kim, H.-S. Lee, D.-H. Seo, S. J. Cho, E. Jeon, and H. R.oon, “Melt-quenchedCarboxylateMetal–Organic FrameworkGlasses,”ature Communications 15 (2024): 1174.6. J. Hou, P. Chen, A. Shukla, et al., “Liquid-Phase Sintering of Leadalide Perovskites and Metal-organic Framework Glasses,” Science 3742021): 621–625, https://doi.org/10.1126/science.abf4460.7. L. León-Alcaide, L. Martínez-Goyeneche, M. Sessolo, et al., “Directynthesis of an Iron Metal-Organic Framework Antiferromagneticlass,” Nature Communications 16 (2025): 8783, https://doi.org/10.1038/41467-025-63837-w.8. W.-L. Xue, P. Kolodzeiski, H. Aucharova, S. Vasa, et al., “Highly Porousetal-organic Framework Liquids and Glasses via a Solvent-Assistedinker Exchange Strategy of ZIF-8,” Nature Communications 15 (2024):420, https://doi.org/10.1038/s41467-024-48703-5.9. S. Kohara, K. Suzuya, K. Takeuchi, et al., “Glass Formation at the Limitf Insufficient Network Formers,” Science 303 (2004): 1649–1652, https://oi.org/10.1126/science.1095047.0. Y. Onodera, S. Kohara, H. Masai, A. Koreeda, S. Okamura, and T.hkubo, “Formation of Metallic Cation-oxygen Network for Anomaloushermal Expansion Coefficients in Binary Phosphate Glass,” Natureommunications 8 (2017): 15449, https://doi.org/10.1038/ncomms15449.1. S. M. Wiederhorn and D. R. Clarke, “Architectural Glass,” Annualeview of Materials Research 52 (2022): 561–592, https://doi.org/10.1146/nnurev-matsci-101321-014417.2. I. W. Donald, B. L. Metcalfe, and R. N. J. Taylor, “The Immobiliza-ion of High Level Radioactive Wastes Using Ceramics and Glasses,”ournal of Materials Science 32 (1997): 5851–5887, https://doi.org/10.1023/:1018646507438.3. N.Ma,N.Horike, L. Lombardo, et al., “Eutectic CsHSO4-Coordinationolymer Glasses With Superprotonic Conductivity′,” Journal of themerican Chemical Society 144 (2022): 18619–18628.4. N. Ma, S. Kosasang, S. Horike, and H. Yamada, “Liquid Coordinationolymers With Anhydrous Proton Conductivity,” Angewandte Chemienternational Edition 64 (2025): 202504618, https://doi.org/10.1002/anie.02504618.5. S. S. Sørensen, X. Ren, T. Du, et al., “Water as a Modifier in a Hybridoordination Network Glass,” Small 19 (2023): 2205988.6. C. Le Losq, D. R. Neuville, P. Florian, G. S. Henderson, and D.Massiot,The Role of Al3+ on Rheology and Structural Changes in Sodium Silicatend Aluminosilicate Glasses and Melts,” Geochimica et Cosmochimicacta 126 (2014): 495–517.7. N. Ma, S. Impeng, S. Bureekaew, N. Morozumi, M. Haga, and S.orike, “Photoexcited Anhydrous Proton Conductivity in Coordinationolymer Glass,” Journal of the American Chemical Society 145 (2023):808–9814, https://doi.org/10.1021/jacs.3c01821.8. X. Zheng, K. Nagura, T. Takaya, K. Hashi, and T. Nakanishi, “Questor a Rational Molecular Design of Alkyl-Distyrylbenzene Liquid byubstitution Pattern Modulation,” Chemistry – A European Journal 292023): 202203775.9. M. A. Salvadó, P. Pertierra, S. García-Granda, L. M. Barcina, R.lavona, and J. Rodríguez, “Hydrogen Bond Network of the Layeredhosphates γ-Zr(H2PO4)(PO4) 2H2O and γ-Hf(H2PO4)(PO4) 2H2ODeter-ined by Neutron Powder Diffraction,” Zeitschrift für Kristallographie-rystalline Materials 216 (2001): 326–330.0. D. E. Harrison, H. A. McKinstry, and F. A. Hummel, “High-emperature Zirconium Phosphates,” Journal of the American Ceramicociety 37 (1954): 277–288, https://doi.org/10.1111/j.1151-2916.1954.b14037.x.1. G. Wallez, J.-P. Souron, and M. Quarton, “Crystal Structure and Ther-al Evolution of Inedite α-Zr2O(PO4)2 and α-Hf2O(PO4)2,” Solid Stateciences 8 (2006): 1061–1066, https://doi.org/10.1016/j.solidstatesciences.006.02.054.mall, 2026abl32. J. F. Willart, A. De Gusseme, S. Hemon, G. Odou, F. Danede, and M.Descamps, “Direct Crystal to Glass Transformation of Trehalose Inducedby Ball Milling,” Solid State Communications 119 (2001): 501–505, https://doi.org/10.1016/S0038-1098(01)00283-6.33. E. Ma, “Amorphization in Mechanically Driven Material Systems,”Scripta Materialia 49 (2003): 941–946, https://doi.org/10.1016/S1359-6462(03)00477-9.34. J. F.Willart andM. Descamps, “Solid State Amorphization of Pharma-ceuticals,”Molecular Pharmaceutics 5 (2008): 905–920, https://doi.org/10.1021/mp800092t.35. Z. Guo, C. Pan, A. Shinohara, and T. Nakanishi, “Merging π-molecular Functions Achieved Through Homogeneous Liquid-LiquidBlending of Solvent-free Alkyl-π Liquids,” Science and Technology ofAdvanced Materials 26 (2025): 2515007, https://doi.org/10.1080/14686996.2025.2515007.36. K. Ito, C. T. Moynihan, and C. A. Angell, “Thermodynamic Determi-nation of Fragility in Liquids and a Fragile-to-Strong Liquid Transition inWater,” Nature 398 (1999): 492–495, https://doi.org/10.1038/19042.37. G. N. Greaves and S. Sen, “Inorganic Glasses, Glass-forming Liquidsand Amorphizing Solids,” Advances in Physics 56 (2007): 1–166, https://doi.org/10.1080/00018730601147426.38. L. Longley, S. M. Collins, C. Zhou, et al., “Liquid Phase Blendingof Metal-Organic Frameworks,” Nature Communications 9 (2018): 2135,https://doi.org/10.1038/s41467-018-04553-6.39. C. Thanaphatkosol, N.Ma, K. Kageyama, et al., “Modulation of ProtonConductivity in Coordination Polymer Mixed Glasses,” Chemical Com-munications 58 (2022): 6064–6067, https://doi.org/10.1039/D2CC01266A.40. L. Longley, C. Calahoo, R. Limbach, et al., “Metal-organic Frameworkand Inorganic Glass Composites,” Nature Communications 11 (2020):5800, https://doi.org/10.1038/s41467-020-19598-9.41. C. A. Angell, “Formation of Glasses From Liquids and Biopolymers,”Science 267 (1995): 1924–1935, https://doi.org/10.1126/science.267.5206.1924.42. S. J. Tao, “Positronium Annihilation in Molecular Substances,” TheJournal of Chemical Physics 56 (1972): 5499–5510, https://doi.org/10.1063/1.1677067.43. C. Hugenschmidt, U. Holzwarth, M. Jansen, S. Kohn, and K. Maier,“Crystallization of Silica Studied by Positron Annihilation,” Journal ofNon-Crystalline Solids 217 (1997): 72–78, https://doi.org/10.1016/S0022-3093(97)00104-X.44. J. C. Machado, G. G. Silva, and L. S. Soares, “PositronAnnihilation and Differential Scanning Calorimetry Investigationsin Poly(methylmethacrylate)/Low Molecular Weight Poly(ethyleneoxide) Polymer Blends,” Journal of Polymer Science Part B: PolymerPhysics 38 (2000): 1045–1052, https://doi.org/10.1002/(SICI)1099-0488(20000415)38:8%3c1045::AID-POLB5%3e3.0.CO;2-N.45. A. J. Hill, M. D. Zipper, M. R. Tant, G. M. Stack, T. C. Jordan, andA. R. Shultz, “A Free Volume Approach to the Mechanical Behaviour ofMiscible Polycarbonate Blends,” Journal of Physics-Condensed Matter 8(1996): 3811.46. M. W. Terban and S. J. L. Billinge, “Structural Analysis of MolecularMaterials Using the Pair Distribution Function,” Chemical Reviews 122(2022): 1208–1272, https://doi.org/10.1021/acs.chemrev.1c00237.47. S. Horike, D. Umeyama, M. Inukai, T. Itakura, and S. Kitagawa,“Coordination-Network-Based Ionic PlasticCrystal forAnhydrous ProtonConductivity,” Journal of the American Chemical Society 134 (2012):7612–7615, https://doi.org/10.1021/ja301875x.48. C. L. Farrow, P. Juhas, J. W. Liu, et al., “PDFfit2 and PDFgui:Computer Programs for Studying Nanostructure in Crystals,” Journal ofPhysics-Condensed Matter 19 (2007): 335219.49. T. L. Christiansen, S. R. Cooper, and K. M. Ø. Jensen, “There’s noPlace Like Real-Space: Elucidating Size-Dependent Atomic Structure of9 of 10e Creative Commons Licensehttps://doi.org/10.1126/science.abf4460https://doi.org/10.1038/s41467-025-63837-whttps://doi.org/10.1038/s41467-024-48703-5https://doi.org/10.1126/science.1095047https://doi.org/10.1038/ncomms15449https://doi.org/10.1146/annurev-matsci-101321-014417https://doi.org/10.1023/A:1018646507438https://doi.org/10.1002/anie.202504618https://doi.org/10.1021/jacs.3c01821https://doi.org/10.1111/j.1151-2916.1954.tb14037.xhttps://doi.org/10.1016/j.solidstatesciences.2006.02.054https://doi.org/10.1016/S0038-1098(01)00283-6https://doi.org/10.1016/S1359-6462(03)00477-9https://doi.org/10.1021/mp800092thttps://doi.org/10.1080/14686996.2025.2515007https://doi.org/10.1038/19042https://doi.org/10.1080/00018730601147426https://doi.org/10.1038/s41467-018-04553-6https://doi.org/10.1039/D2CC01266Ahttps://doi.org/10.1038/s41467-020-19598-9https://doi.org/10.1126/science.267.5206.1924https://doi.org/10.1063/1.1677067https://doi.org/10.1016/S0022-3093(97)00104-Xhttps://doi.org/10.1002/(SICI)1099-0488(20000415)38:8%3c1045::AID-POLB5%3e3.0.CO;2-Nhttps://doi.org/10.1021/acs.chemrev.1c00237https://doi.org/10.1021/ja301875xNA5Cc15GC5m(5‘Ph5CJ15fILj5H5Wao15cS5ST16S6ZT16a6oh6Tu16zaR96N1 16136829, 2026, 22, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/smll.202514459 by Nattapol Ma - National Institute For , Wiley Online Library on [19/04/2026]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applanomaterials Using Pair Distribution Function Analysis,” Nanoscaledvances 2 (2020): 2234–2254, https://doi.org/10.1039/D0NA00120A.0. A. Sutrisno, L. Liu, J. Dong, and Y. Huang, “Solid-State 91Zr NMRharacterization of Layered and Three-Dimensional Framework Zir-onium Phosphates′,” The Journal of Physical Chemistry 116 (2012):7070–17081.1. J. Qiao, J.-M. Pelletier, and R. Casalini, “Relaxation of Bulk Metalliclasses Studied by Mechanical Spectroscopy,” The Journal of Physicalhemistry B 117 (2013): 13658–13666, https://doi.org/10.1021/jp4067179.2. R. H. Colby, L. J. Fetters, and W. W. Graessley, “The Melt Viscosity-olecular Weight Relationship for Linear Polymers,”Macromolecules 201987): 2226–2237, https://doi.org/10.1021/ma00175a030.3. C. A. Angell, C. T. Moynihan, and M. Hemmati, “‘Strong’ andSuperstrong’ Liquids, and an Approach to the Perfect Glass State viahase Transition,” Journal of Non-Crystalline Solids 274 (2000): 319–331,ttps://doi.org/10.1016/S0022-3093(00)00222-2.4. J.-P. Belieres and C. A. Angell, “Protic Ionic Liquids: Preparation,haracterization, and Proton Free Energy Level Representation,” Theournal of Physical Chemistry B 111 (2007): 4926–4937, https://doi.org/10.021/jp067589u.5. A. Takeuchi, H. Kato, and A. Inoue, “Vogel–Fulcher–Tammann Plotor Viscosity Scaled With Temperature Interval Between Actual anddeal Glass Transitions for Metallic Glasses in Liquid and Supercoolediquid States,” Intermetallics 18 (2010): 406–411, https://doi.org/10.1016/.intermet.2009.08.015.6. U. Fotheringham, Springer Handbook of Glass, ed. J.D. Musgraves, J.u, and L. Calvez, (Springer International Publishing, 2019): 79–112.7. C. Schreiner, S. Zugmann, R. Hartl, and H. J. Gores, “FractionalaldenRule for Ionic Liquids: Examples FromRecentMeasurements andCritique of the So-Called Ideal KCl Line for the Walden Plot,” Journalf Chemical & Engineering Data 55 (2010): 1784–1788, https://doi.org/10.021/je900878j.8. J. C. Mauro, Y. Yue, A. J. Ellison, P. K. Gupta, and D. C. Allan, “Vis-osity of Glass-Forming Liquids,” Proceedings of the National Academy ofciences 106 (2009): 19780–19784, https://doi.org/10.1073/pnas.0911705106.9. Y. Wang, N. A. Lane, C.-N. Sun, F. Fan, T. A. Zawodzinski, and A. P.okolov, “Ionic Conductivity and Glass Transition of Phosphoric Acids,”he Journal of Physical Chemistry B 117 (2013): 8003–8009, https://doi.org/0.1021/jp403867a.0. J. E. Shelby, Introduction to Glass Science and Technology (Royalociety Of Chemistry, 2005), https://doi.org/10.1039/9781847551160.1. P. Lucas, G. J. Coleman, M. Venkateswara Rao, et al., “Structure ofnCl 2Melt. Part II: Fragile-to-Strong Transition in a Tetrahedral Liquid,”he Journal of PhysicalChemistry B 121 (2017): 11210–11218, https://doi.org/0.1021/acs.jpcb.7b10857.2. H. Vogel, “The Law of the Relation Between the Viscosity of Liquidsnd the Temperature,” J Phys Z 22 (1921): 645–646.3. G. S. Fulcher, “Analysis of Recent Measurements of the Viscosityf Glasses,” Journal of the American Ceramic Society 8 (1925): 339–355,ttps://doi.org/10.1111/j.1151-2916.1925.tb16731.x.4. G. Tammann and W. Hesse, “Die Abhängigkeit der Viscosität von deremperatur bie unterkühlten Flüssigkeiten,” Zeitschrift für anorganischend allgemeine Chemie 156 (1926): 245–257, https://doi.org/10.1002/zaac.9261560121.5. G. Alberti, M. G. Bernasconi, and M. Casciola, “Preparation of γ-irconium Phosphate Microcrystals With High Degree of Crystallinitynd Proton Conductivity of Their Hydrogen and Ammonium Forms,”eactive Polymers 11 (1989): 245–252, https://doi.org/10.1016/0923-1137(89)0110-3.6. B. C. H. Steele and A. Heinzel, “Materials for Fuel-cell Technologies,”ature 414 (2001): 345–352, https://doi.org/10.1038/35104620.0 of 1067. M. A. B. H. Susan, T. Kaneko, A. Noda, and M. Watanabe, “Ion GelsPrepared by in Situ Radical Polymerization of VinylMonomers in an IonicLiquid and Their Characterization as Polymer Electrolytes,” Journal of theAmerican Chemical Society 127 (2005): 4976–4983, https://doi.org/10.1021/ja045155b.68. C. Austen Angell, Y. Ansari, and Z. Zhao, “Ionic Liquids: Past, Presentand Future,” Faraday Discussions 154 (2012): 9–27, https://doi.org/10.1039/C1FD00112D.69. W. Xu and C. A. Angell, “Solvent-Free Electrolytes With AqueousSolution-Like Conductivities,” Science 302 (2003): 422–425, https://doi.org/10.1126/science.1090287.70. J.-P. Melchior, K.-D. Kreuer, and J. Maier, “Proton Conduction Mech-anisms in the Phosphoric Acid–Water System (H4P2O7–H3PO4⋅2H2O): a1H, 31P and 17O PFG-NMR and Conductivity Study,”Physical ChemistryChemical Physics 19 (2017): 587–600.71. T. Ogawa, K. Takahashi, S. S. Nagarkar, et al., “Coordination PolymerGlass From a Protic Ionic Liquid: Proton Conductivity and MechanicalProperties as an Electrolyte,” Chemical Science 11 (2020): 5175–5181,https://doi.org/10.1039/D0SC01737J.Supporting InformationAdditional supporting information can be found online in the SupportingInformation section.SupportingFile: smll72859-sup-0001-SuppMat.pdf.Small, 2026icable Creative Commons Licensehttps://doi.org/10.1039/D0NA00120Ahttps://doi.org/10.1021/jp4067179https://doi.org/10.1021/ma00175a030https://doi.org/10.1016/S0022-3093(00)00222-2https://doi.org/10.1021/jp067589uhttps://doi.org/10.1016/j.intermet.2009.08.015https://doi.org/10.1021/je900878jhttps://doi.org/10.1073/pnas.0911705106https://doi.org/10.1021/jp403867ahttps://doi.org/10.1039/9781847551160https://doi.org/10.1021/acs.jpcb.7b10857https://doi.org/10.1111/j.1151-2916.1925.tb16731.xhttps://doi.org/10.1002/zaac.19261560121https://doi.org/10.1016/0923-1137(89)90110-3https://doi.org/10.1038/35104620https://doi.org/10.1021/ja045155bhttps://doi.org/10.1039/C1FD00112Dhttps://doi.org/10.1126/science.1090287https://doi.org/10.1039/D0SC01737J Engineering Proton Conductive Metal-Organic Glasses Through Secondary Network Formers 1 | Introduction 2 | Results and Discussion 2.1 | Synthesis, Melting Behavior, and Glass Formation 2.2 | Morphology and Free Volume of MOGs 2.3 | Short-to-Intermediate Range Structure 2.4 | Viscoelasticity 2.5 | Anhydrous Proton Conductivity 3 | Conclusion Acknowledgements Conflicts of Interest Data Availability Statement References Supporting Information