Transparent and High-porosity Aluminum Alkoxide Network-forming Glasses

doi:10.21203/rs.3.rs-3935946/v1

Download PDF

Article

Transparent and High-porosity Aluminum Alkoxide Network-forming Glasses

https://doi.org/10.21203/rs.3.rs-3935946/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 26 Aug, 2024

Read the published version in Nature Communications →

Version 1

posted

You are reading this latest preprint version

Metal-organic network-forming glasses are an emerging type of material capable of combining the modular design and high porosity of metal-organic frameworks and the high processability and optical transparency of glasses. However, a generalizable strategy for achieving both high porosity and high glass forming ability in modularly designed metal-organic networks has yet to be developed. Herein, we developed a new series of metal-organic network-forming glasses, aluminum alkoxide glasses, by linking aluminum-oxo clusters with alcohol linkers in the presence of a modulator template. These glasses exhibit well-defined glass transitions and high surface areas up to 500 m²/g, making them one of the most porous glassy materials. The aluminum alkoxide glasses also have optical transparency and fluorescent properties, and their structures were elucidated by pair-distribution functions and compositional analysis. A systematic glass transition study suggested that progressive increase in network connectivity during the evaporation of a coordinatively competitive solvent is key to the bottom-up glass synthesis. Aluminum alkoxide glass can also encapsulate crystalline MOFs to yield composite materials with higher porosities. These findings could significantly expand the library of microporous metal-organic network-forming glasses and enable their future applications.

Physical sciences/Chemistry/Materials chemistry/Metal–organic frameworks

Physical sciences/Materials science/Soft materials/Glasses

Physical sciences/Chemistry/Inorganic chemistry/Solid-state chemistry

Physical sciences/Chemistry/Materials chemistry/Soft materials/Glasses

A new series of glasses made from aluminum cluster linked with alcohols that are both transparent and highly porous.

Metal-organic network-forming glasses are an emerging type of glassy material that can combine high porosity, optical transparency, chemical robustness, and defect-free interfaces.^1,2 Conceptually, a glassy network with micropores can be created from an oxide glass structure by replacing metal and oxygen atoms with metal clusters and organic linkers, which results in a coordinative network. However, synthetically preparing such porous glassy networks is fundamentally challenging: obtaining glasses of coordinative networks is challenging due to their high melting point and low thermal stability. Retaining microporosity in glasses is even more challenging because glasses tend to have structures similar to those of their corresponding supercooled liquids and are usually nonporous. Consequently, the number of known glassy coordinative networks is very limited compared to the vast library of crystalline metal-organic frameworks (MOFs), and even smaller fractions of these networks have gas accessible microporosity.^3–10 Although perturbative and other methods have allowed additional MOFs to be vitrified,^11–14 there has been no systematic strategy for synthesizing coordination network glasses with high internal surface areas.

The bottom-up synthesis of glassy coordinative networks represents an important opportunity to develop new microporous glassy coordinative networks. It was discovered that porous titanium phenolate networks can be directly synthesized using a solvent evaporation approach,¹⁵ which can also be applied to amorphous titanium-carboxylate networks.¹⁶ However, for both cases, the glass transition was not directly observed, and their liquid precursor tended to have poor glass-forming ability when rigid linkers were employed to increase porosity. In this report, we developed a series of microporous aluminum alkoxide glasses with well-defined glass transitions that can combine high porosity and glass forming ability as well as light weight, optical transparency and fluorescent properties (Fig. 1). These aluminum alkoxide glasses are synthesized through evaporating coordinatively competitive solvents that dissolve the aluminum precursor and organic linker (Fig. 1a-c), during which the organic linker gradually links the aluminum-oxo clusters (AlOCs) to form a coordinative network and cause vitrification (Fig. 1d). A bulky, nonvolatile monodentate alcohol is introduced as a modulator template during synthesis to prevent pore collapse, after which the alcohol is removed by solvent exchange to open pores (Fig. 1e). The modulator template can also provide additional configurational degrees of freedom, thus increasing the glass-forming ability. Consequently, the aluminum alkoxide glasses with modulator templates exhibit well-defined glass transitions in their as-synthesized form, which retain their backbone structure after the activation process. Using the modulator-templating approach, we achieved a BET surface area of 500 m²/g for aluminum alkoxide glasses that incorporate a highly flexible linker and have high glass-forming ability; these glasses are arguably the most porous glass materials with well-defined glass transitions. We believe our findings provide a generalizable strategy to achieve a large library of highly porous glassy networks available for various applications.

Synthetic design of aluminum alkoxide glasses

It is known that simple aluminum alkoxides can form AlOCs in solution, especially in the presence of carboxylic acid ligands.^17–19 In our synthesis, we harnessed the chemistry of AlOC and aimed to use multidentate alcohol linkers to link these AlOCs to form a glassy coordinative network. We demonstrate this chemistry using two different linkers, bis(2-hydroxyethyl) terephthalate (BHET) and methanetetrayltetrakis(benzene-4,1-diyl))tetramethanol (MTBT) (Fig. 1, Figs. S1-2). The BHET has highly flexible terminal segments and good solubility, whereas the MTBT is rather rigid and can support porous structures by itself. In a typical synthesis, 310 mg of BHET is dissolved in 2 mL of 1-butanol, 500 µL of glacial acetic acid, and 4 mL of ethanol at elevated temperature, after which the mixture is added to 300 mg of aluminum sec-butoxide in a glass vial and stirred overnight on a 120°C hotplate. The resulting clear and colorless solution was then evaporated in a petri dish heated to 125°C, after which the solution became increasingly viscous and then vitrified to give the Al-BHET glass (Fig. 1d). The as-synthesized aluminum alkoxide glasses were then washed with THF, a THF/ethanol mixture and acetone through solvent exchange and then activated with supercritical CO₂ and heated under vacuum. Al-BHET was characterized by various spectroscopic techniques (Figs. S3-S8), and infrared spectroscopy revealed the deprotonation of the linker hydroxyl groups. In the synthesis of Al-BHET, the alcohol solvents have the same hydroxy group as the linker, which functions as a network modulator by coordinatively competing with the linker and reducing the connectivity of the network. Thus, during solvent evaporation, the network connectivity gradually increases, resulting in increased viscosity and relaxation time, causing vitrification.

The Al-MTBT glass was synthesized and characterized in a similar manner but with a lower concentration and an ethylene glycol methyl ether/THF/ethanol solvent mixture (Section S4, Figs. S9-S13). In contrast to the synthesis of Al-BHET, which is highly robust and tolerant to various synthetic conditions, Al-MTBT glass requires specific solvent mixtures that balance MTBT solubility and modulator concentration. The synthesis of Al-MTBT also needs to be carried out with rapid evaporation; otherwise, precipitation and crystallization occur. Both optical images and scanning electron microscopy images of Al-BHET and Al-MTBT show smooth surfaces and homogeneous features consistent with their glassy nature (Fig. 1d, Fig. S2). According to the comparison of Al-bearing materials between BHET and MTBT, flexible linkers clearly reliably provide monolithic glassy coordinative networks, whereas rigid linkers such as MTBT pose challenges for successful vitrification. However, although flexible linkers could endow higher glass forming ability, they also tend to give rise to lower porosity, as shown in the next section, thus presenting a trade-off that needs to be mitigated with more elaborate synthetic design.

Gas accessible porosity and pore expansion with modulator template

To study the porosity of the aluminum alkoxide glasses, their gas and vapor adsorption isotherms were measured (Fig. 2). For Al-BHET, its small pores do not result in nitrogen uptake at 77 K. CO₂ uptake at 195 K shows that Al-BHET is still porous, with a pore volume of 0.11 cm³/g (Fig. 2a). The methanol uptake at room temperature also confirmed the porosity of the material (Fig. S3). The skeleton density of Al-BHET was determined by a He pycnometer to be 1.41 g/cm³, which is approximately half of the value of titanium phenolate glass (Table S2) ¹⁵. From the pore volume measured by CO₂ uptake and skeleton density, the volumetric porosity of Al-BHET was determined to be 13%. For Al-MTBT, its more rigid linker gives larger pores accessible to N₂ with a BET surface area of 363 m²/g, a skeleton density of 1.33 g/cm³ and a pore fraction of 23% (Fig. 2b, Figs. S9-10, Table S3). The porosity of Al-MTBT was also characterized by CO₂ uptake (Figs. S10-11). Apart from porosity, both aluminum alkoxide networks can also have optical transparency and fluorescent properties. The Al-BHET is colorless and has no measurable absorption in the visible range of 400–750 nm (Fig. 2c). MTBT is slightly yellow in color, as the linker has adsorption features near 400 nm. Remarkably, Al-MTBT emits blue fluorescence originating from the linker (Fig. 2d), which could have future applications in smart coating and lighting applications.

The comparison between Al-BHET and Al-MTBT revealed that flexible linkers such as BHET can provide high glass forming ability and a uniform morphology at the cost of porosity. Efforts to increase porosity by choosing rigid linkers such as MTBTs face fundamental limitations in forming uniform glassy materials. To overcome such limitations and mitigate the trade-off between porosity and glass forming ability, we developed a modulator templating approach that can achieve high porosity even with flexible linkers. In this approach, a nonvolatile bulky monodentate alcohol is introduced as a modulator that can coordinate to AlOC during synthesis and prevent pore collapse during solvent evaporation. After vitrification, the modulator can be removed by solvent exchange with short alcohol and provide micropores. We demonstrate such an approach by using 1,1,1-triphenyl-2,5,8,11-tetraoxatridecan-13-ol (TPTO) as a modulator, and we denote the resulting network-forming glass as Al-BHET-TPTO (section S5, Figs. S14-S20). When 23% TPTO (molar ratio compared to BHET) was initially added to the synthesis, Al-BHET-TPTO had a surface area of 500 m²/g, as measured by nitrogen uptake at 77 K (Fig. 3a) with a void fraction of 30%. The Al-BHET-TPTO is also transparent (Fig. S16) and shows well-defined glass transition behavior in its as-synthesized form (Fig. S17), which makes the Al-BHET-TPTO arguably one of the most porous glassy materials determined by the BET surface area measured with N₂ uptake. In comparison, Al-BHET without a modulator template has no measurable nitrogen uptake or a 13% void fraction when measured by 195 K CO₂, which clearly shows the effect of introducing TPTO on increasing microporosity (Fig. 3b). The removal of the TPTO template during solvent exchange can be confirmed by digested NMR and infrared spectroscopy (Figs. S19-20), which show that only a very small amount of TPTO is left in the activated glass compared to the amount initially added. Notably, the glass transition behaviors of Al-BHET and Al-BHET-TPTO are different. Al-BHET does not have a well-defined glass transition temperature, and its glassy nature is deduced from analyzing its vitrification process, which will be elaborated upon in the next paragraph. On the other hand, the as-synthesized Al-BHET-TPTO has well-defined glass transitions due to the increased configurational degree of freedom associated with the modulator effect of TPTO. Thus, the modulator template not only increases the porosity but also increases the glass-forming ability and provides a handle for studying the glass transition behavior of these aluminum alkoxide glasses.

The chemical space of Al/BHET/TPTO and the glass transition of aluminum alkoxide glasses

The vitrification of aluminum alkoxide glass involves fixing the amount of aluminum and BHET and decreasing the amount of coordinatively competitive solvent. The volatility of regular solvents used in synthesis makes it difficult to track such processes by differential scanning calorimetry (DSC). In this regard, nonvolatile TPTO can be used as a modulator to study the glass transition behavior of metal-linker-modulator tertiary systems when the amount of modulator is reduced. Gradually reducing the amount of TPTO in the Al/BHET/TPTO tertiary system increases the glass transition temperature, accompanied by a less pronounced change in heat capacitance corresponding to the glass transition. Moreover, with decreasing TPTO, the Al-BHET-TPTO mixture becomes mechanically hardened. These observations indicate that reducing the amount of modulator results in greater network connectivity: increased network connectivity means a greater degree of cross-linking and a greater energy barrier for configurational motion, which accounts for the greater Tg and mechanical strength; increased network connectivity also means that the Al/BHET/TPTO would resemble covalently linked networks such as dry silica,²⁰ which often behave as strong liquids with small changes in heat capacitance during the glass transition. Such reasoning from physical measurements is consistent with chemical intuition that the removal of a coordinatively competitive modulator would allow the BHET linker to link up AlOCs. These studies on the Al-BHET-TPTO system can be readily generalized to all other Al-linker-solvent/modulator systems and rationalize the synthesis and vitrification of aluminum alkoxide glasses, even though their Tg cannot be measured directly due to the high volatility or low melting point of the solvent.

We further studied another example of the Al/BHET/TPTO tertiary system to show the general features of aluminum alkoxide glasses with templating modulators, which have a composition of TPTO:BHET = 1:1 and Al:OH = 1:2 (termed Al-BHET-TPTO-2; Figs. S21-27). The as-synthesized Al-BHET-TPTO-2 shows a clear glass transition at 20°C (Fig. S23). After the solvent exchange and activation processes, Al-BHET-TPTO-2 exhibited a pore volume of 0.24 cm³/g, as characterized by CO₂ uptake at 195 K, which was 83% of that of the Al-BHET-TPTO membrane discussed above. Infrared spectra and digested NMR data show that a large portion of TPTO is removed during solvent exchange and activation, whereas the prominent features of the pair-distribution functions remain largely unchanged for the as-synthesized, solvent-exchanged and activated Al-BHET-TPTO-2 glass. Thus, we could conclude that Al-BHET-TPTO-2, like Al-BHET-TPTO, is a glassy material with a well-defined glass transition determined by DSC measurements and high porosity, as shown by gas adsorption measurements, where its structural backbone remains unchanged during the solvent exchange and activation processes, as evidenced by the pair distribution function. It is worth noting that the surface area of the Al/BHET/TPTO glass does not show monotonic dependence on the BHET:TPTO ratio, as the modulator affects the network structure in a rather complex manner.

Structural elucidation of aluminum alkoxide network-forming glasses

To provide a reasonable structural model for aluminum alkoxide glasses, we combined compositional analysis, spectroscopy, and total scattering techniques to study Al-BHET as a model compound. In this process, we focus on describing short- and medium-range orders, which are essentially dictated by the structure of the AlOCs. As AlOCs are formed in situ, a variety of AlOCs are certainly present, and we aim to find a representative AlOC that embodies the most prominent features of these glasses. With the synthesis carried out in the presence of alcohol and carboxylic acid, the resulting AlOCs would have mixed acid and alcohol linkers. X-ray total scattering measurements produce the pair-distribution function of Al-BHET, where the first two peaks at 1.4 and 1.9 Å are attributed to C-C/O and Al-O bonds, respectively (Fig. 5a, b). The peak at 2.83 Å corresponds to the Al-Al pair, which clearly shows the presence of AlOC. The absence of a peak at 3.3 Å excluded the presence of wheel-shaped AlOC because their carboxylate groups with the µ₂-η¹:η¹ coordination mode would results in an Al-O pair at 3.3 Å. In addition, these wheeled-shaped clusters typically have low solubility and can precipitate from the solution. On the other hand, AlOCs such as AlOC-41 could be more plausible candidates for representing the AlOCs in Al-BHET, as they contain only a small amount of bridging carboxylic acids. The AlOC-41 cluster also has a higher solubility and can be more readily formed under the synthesis conditions used to prepare Al-BHET. We compared the pair-distribution function simulated using the core of AlOC-41 attached to the BHET linker with the measured data, which showed good consistency. Notably, most of the peaks in the pair-distribution function are associated with various atom pairs, and the allocation in Fig. 5a does not suggest that the corresponding peaks originated exclusively from these pairs. In addition, the 2.36 Å peak does not heavily involve Al and thus would primarily originate from the linker; this peak is not shown in Fig. 5a for simplicity. Considering the relatively high acid-to-alcohol ratio of 2.25 found via digested NMR, part of the carboxylic acid would be a terminal ligand that replaces BHET. This provides the empirical formula of Al₈O₂(OH)₃(BHET)_4.2(CH₃COO)_9.6∙2H₂O, with a calculated composition of C 43.0% H 4.4% O 41.5% Al 11.1%, which is consistent with the measured elemental analysis result of C 43.1% H 4.9% O 42.8% Al 10.3%. The presence of AlOC was also confirmed by X-ray absorption spectroscopy, where the near edge feature and EXAFS of the Al-BHET were distinctly different from those of alumina (Figs. S6-7).^21,22 The consistency of the above analysis shows that it is reasonable to speculate that the aluminum alkoxide glasses are composed of AlOCs linked by alcohol linkers and reinforced by carboxylic acid ligands, where the AlOCs would have mixed ligands and similar types and sizes of AlOC-41. Considering the similarity in their synthetic conditions, the AlOCs and their connectivity in the Al-BHET-TPTO glass would be similar to those in the Al-BHET glass.

High-porosity MOF@glass composites

The Al-BHET network-forming glass can also serve as a host for crystalline MOFs and produce glass ceramics with larger surface areas,^23,24 which we demonstrated by incorporating UiO-66 in the Al-BHET glass (Fig. 6). The UiO-66@Al-BHET glass ceramics were synthesized by adding UiO-66 nanocrystals to a solution for Al-BHET synthesis, where the UiO-66 nanocrystals retained their crystallinity, as evidenced by SEM and XRD (Fig. 6a, c). The UiO-66@Al-BHET glass ceramic composite exhibited high uptake, with a surface area of 680 m²/g (Fig. 6b). The synthetic compatibility of aluminum alkoxide glass, which allows the formation of MOF@glass composites, could lead to further applications.

In summary, we developed a new class of metal-organic network-forming glass made by linking AlOCs with alcohol linkers. Its synthesis can be carried out via a bottom-up approach, where evaporation of coordinatively competitive alcohol solvents leads to increased network connectivity and subsequent vitrification. The introduction of a modulator template provides both a high surface area and glass-forming ability, making Al-BHET-TPTO one of the most porous glassy materials in terms of its BET surface area. The glass transition behavior of the metal-linker-modulator tertiary material system indicated that the relaxation mechanism near the fictive temperature was modulator-induced coherent framework dynamics. The development of aluminum alkoxide glasses and modulator templating strategies represents an important advance in the field of metal-organic network-forming glasses, which have expanded the structural diversity of these materials, provided viable solutions for achieving high porosity, and paved the way for future applications in smart coatings, membranes, and composite materials.

Acknowledgment

The authors thank Ms. L. Long at ShanghaiTech University for assistance in gas adsorption measurements, Dr. R. Gao and Mr. C. Qi for assistance in fluorescence measurements and Ms. R.-F. Ma at Guangxi Normal University for X-ray total scattering measurements. X-ray absorption spectroscopy at the Al K-edge was carried out at BL02B02 of the Shanghai Synchrotron Radiation Facility, which was supported by the ME2 project under contract from the National Natural Science Foundation of China (11227902). Y.Z. acknowledges support from the Science and Technology Commission of Shanghai Municipality (22QC1401500).

Data and materials availability

The data that support the findings of this study are available from the corresponding authors upon request. All data are available in the main text or the supplementary materials.

Competing interests

The authors declare no competing interests.

Ma N, Horike S (2022) Metal–Organic Network-Forming Glasses. Chem Rev 122:4163–4203
Bennett TD, Horike S (2018) Liquid, glass and amorphous solid states of coordination polymers and metal–organic frameworks. Nat Rev Mater 3:431–440
McHugh LN, Bennett TD (2022) Introducing porosity into metal–organic framework glasses. J Mater Chem A 10:19552–19559
Fan Z et al (2023) Creating glassy states of dicarboxylate-bridged coordination polymers. Chem Commun 59:14317–14320
Frentzel-Beyme L et al (2019) Porous purple glass – a cobalt imidazolate glass with accessible porosity from a meltable cobalt imidazolate framework. J Mater Chem A 7:985–990
Zhou C et al (2018) Metal-organic framework glasses with permanent accessible porosity. Nat Commun 9:5042
Frentzel-Beyme L, Kolodzeiski P, Weiß J-B, Schneemann A, Henke S (2022) Quantification of gas-accessible microporosity in metal-organic framework glasses. Nat Commun 13:7750
Christensen R et al (2023) Medium-Range Order Structure Controls Thermal Stability of Pores in Zeolitic Imidazolate Frameworks. J Phys Chem Lett 14:7469–7476
Fan Z et al (2023) Formation of Porosity toward Acetylene upon Vitrification of Non-porous Photochromic Coordination Polymer Crystals. Chem Mater 35:3859–3866
Hou J et al (2020) Halogenated Metal–Organic Framework Glasses and Liquids. J Am Chem Soc 142:3880–3890
Yin Z et al (2022) Synergistic Stimulation of Metal–Organic Frameworks for Stable Super-cooled Liquid and Quenched Glass. J Am Chem Soc 144:13021–13025
Wei Y-S, Fan Z, Luo C, Horike S (2023) Desolvation of metal complexes to construct metal–organic framework glasses. Nat Synth. 10.1038/s44160-023-00412-5
Widmer RN et al (2019) Pressure promoted low-temperature melting of metal–organic frameworks. Nat Mater 18:370–376
Vaidya S et al (2020) Transparent and luminescent glasses of gold thiolate coordination polymers. Chem Sci 11:6815–6823
Zhao Y, Lee S-Y, Becknell N, Yaghi OM, Angell CA (2016) Nanoporous Transparent MOF Glasses with Accessible Internal Surface. J Am Chem Soc 138:10818–10821
Xu W et al (2023) High-Porosity Metal‐Organic Framework Glasses. Angew Chem Int Ed 62:e202300003
Wang ST, Liu YJ, Feng CC, Fang WH, Zhang J (2023) The largest aluminum molecular rings: Phenol-thermal synthesis, photoluminescence, and optical limiting. Aggregate 4:e264
Li Y et al (2022) Record Aluminum Molecular Rings for Optical Limiting and Nonlinear Optics. Angew Chem Int Ed 61:e202116563
Zhang X-Z, Wang X-F, Fang W-H, Zhang J (2021) Synthesis, Structures, and Fluorescence Properties of Dimeric Aluminum Oxo Clusters. Inorg Chem 60:7089–7093
Yue Y (2015) Anomalous Enthalpy Relaxation in Vitreous Silica. Front Mater 2:54
Fontaine A, Lagarde P, Raoux D, Esteva JM (1979) Soft X-ray absorption and EXAFS on the K edge of aluminium. J Phys F: Met Phys 9:2143
Fulton JL et al (2015) Electronic and Chemical State of Aluminum from the Single- (K) and Double-Electron Excitation (KL_II&III, KL_I) X-ray Absorption Near-Edge Spectra of α-Alumina, Sodium Aluminate, Aqueous Al₃⁺·(H₂O)₆, and Aqueous Al(OH)₄^–. J Phys Chem B 119:8380–8388
Hou J et al (2019) Metal-organic framework crystal-glass composites. Nat Commun 10:2580
Lin R et al (2023) Metal–organic framework glass composites. Chem Soc Rev 52:4149–4172

There is NO Competing Interest.

SIZhangZhao.docx

Download PDF

Journal Publication

published 26 Aug, 2024

Read the published version in Nature Communications →

Version 1

posted

You are reading this latest preprint version

Transparent and High-porosity Aluminum Alkoxide Network-forming Glasses

Status:

Journal Publication

Version 1

Abstract

Figures

One-Sentence Summary

Introduction

Results and Discussion

Synthetic design of aluminum alkoxide glasses

Gas accessible porosity and pore expansion with modulator template

The chemical space of Al/BHET/TPTO and the glass transition of aluminum alkoxide glasses

Structural elucidation of aluminum alkoxide network-forming glasses

High-porosity MOF@glass composites

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1