Efficient hydrogen generation from NaBH₄ methanolysis using a lithium-enhanced Dawson-type polyoxometalate catalyst

Abstract

In this work, a Dawson-type lithium polyoxometalate Li₆[α-P₂W₁₈O₆₂]·28 H₂O (LiPOM), was synthesized and thoroughly characterized using X-ray diffraction (XRD), phosphorus-31 nuclear magnetic resonance (³¹P NMR), Fourier transform infrared spectroscopy (FT-IR), ultraviolet–visible spectroscopy (UV-Vis), scanning electron microscopy (SEM), transmission electron microscopy (TEM), atomic force microscopy (AFM) techniques and Brunauer–Emmett–Teller (BET) surface area analysis techniques to confirm its structural integrity and morphology. The catalytic performance of LiPOM for hydrogen production via NaBH₄ methanolysis was systematically investigated under varying reaction temperatures, catalyst dosages, and NaBH₄ concentrations. The obtained results demonstrate that LiPOM exhibits enhanced hydrogen evolution efficiency compared with similar catalytic systems reported in the literature. Kinetic analysis based on the Arrhenius model yielded an activation energy of 28.12 kJ mol⁻¹, enthalpy of 25.56 kJ mol⁻¹, and entropy of − 128.98 J mol⁻¹ K⁻¹. The improved catalytic activity is attributed to the preserved Dawson-type framework and the highly rough surface morphology of LiPOM, which together facilitate effective reactant interaction and borohydride activation. These findings indicate that lithium-stabilized Dawson-type POMs are promising catalysts for environmentally benign hydrogen production processes.

Introduction

Fossil fuel resources are limited, and their overuse leads to environmental pollution. Their high carbon footprint and excessive greenhouse gas emissions have necessitated the need for a clean and sustainable energy system. This has increased the interest in production based on this system and solutions for different studies^1–4. Among the systems currently used, there are methods such as storage in tanks in pressurized gas and/or liquid phase at very low temperature, adsorption on activated carbon, carbon nanotubes and metal hydrides, but it is possible to categorize storage in three groups: physical, chemical and hybrid. Both the structure of hydrogen and thermodynamic requirements make it advantageous to store hydrogen in chemical structures. Among these chemical compounds, metal boron hydrides are of great importance. The most preferred metal boron hydride is sodium boron hydride. NaBH₄ is a hydrogen-rich salt with a theoretical gravimetric hydrogen content of 10.8%. High purity and controlled hydrogen can be produced from NaBH₄ and has the advantages of a non-toxic and non-flammable stable solution⁵. The methanolysis reaction of NaBH₄ is as given in Eq. (1)⁶:

1

Noble metals such as Pd, Pt, Ru and Rh are highly preferred catalysts for methanolysis reactions due to their properties such as activity, selectivity and stability. They also have high stability and the capacity to resist physical and chemical transformations that may occur over time^7–9. However, since they are very limited in nature and very costly for hydrogen production applications, alternatives to these metals and other alternatives that can be more easily accessed are being sought without the need to add these metals^10,11. As an alternative to these catalysts, especially non-noble transition metal-containing catalysts that can be used in hydrogen production have very high activities. Due to these advantageous properties and activities, they are a good alternative compared to noble metal-containing catalysts^10,12. Since the methanolysis reaction of NaBH₄ in hydrogen production takes place on the surface of the metal catalyst, the metal catalysts used show activity by contacting the substances with which they will react, thanks to their active ends, and they are resistant to adverse conditions thanks to their porous surface structures and high surface area. Thanks to this advantageous feature, it has a higher activity. Depending on this activity, it increases the catalytic performance for hydrogen production. As a catalyst, nanoparticles containing transition metals in particular and nanocomposites obtained with them are used in nanotechnological applications^13,14.

In the literature, the economic sustainability of borohydride-based hydrogen production systems depends particularly on the ability to reconvert borate derivatives formed after hydrolysis back into NaBH₄ or LiBH₄ with low energy input. In this context, Ouyang et al. demonstrated high-efficiency NaBH₄ recovery from NaBO₂ using a Mg-mediated mechanochemical regeneration approach¹⁵; Zhu et al. presented closed-loop strategies for integrating borohydrides into the hydrogen-energy process chain¹⁶; and Chen et al. showed that low-cost, single-step regeneration of LiBH₄ from hydrolysis products is possible¹⁷. These studies reveal that Li(Na)BH₄ systems can be evaluated not only as hydrogen production reactants but also as cyclical chemical energy carriers. From this perspective, the Li⁺ ion content and redox-active polyoxometalates (POMs) skeleton of the Li-POM structure used in our study offer a functional platform that can both improve catalytic hydrolysis kinetics in borohydride-based systems and potentially work in compatibility with lithium-based closed-loop approaches. Thus, the Li-POM system can be considered as a material architecture that can be integrated into a long-term vision of a regeneration-based hydrogen economy by increasing the hydrogen production rate. In addition, metal oxide nanoparticles are widely used in electro-chemical sensing systems due to their environmentally friendly properties and important features such as n-type semiconductor structure, high photosensitivity, biocompatibility and chemical stability, as well as chemical stability, wide band gap and significant electron transfer rate^18–20. In general, incorporation of these particles into catalytic systems can significantly modify their electronic structure and charge-transfer characteristics. An increase in the bandgap is typically associated with enhanced structural stability and reduced electron mobility, whereas a decrease in the bandgap may facilitate electron transfer and improve catalytic activity. Owing to their tunable electronic properties, redox behavior, and structural versatility, these materials have attracted considerable attention as promising catalysts in energy-related applications, particularly hydrogen generation and other electron-transfer-driven catalytic processes^12,21. Although oxidative stress has adverse effects on various biological systems, reactive oxygen species produced by metal oxides such as molybdenum oxide, vanadium oxide, indium oxide, tungsten oxide and cerium oxide can also be used to remove various pollutants such as dyes, antibiotics and biological wastes from the environment through photocatalytic degradation^22–24. The use of organometallic compounds in electronics, optoelectronics and recently especially in catalytic applications has become a field of interest²⁵. The molecule that absorbs the light travels to the excited state of the energy of the photon and stores the energy in the form of chemical energy for a short while. One such compound is the transition metal with metal oxide group anions, which have now become known as POMs. POMs, particularly those containing transition metals, possess remarkable structural stability, reversible multi-electron redox properties, and tunable electronic characteristics. These features make them attractive candidates for a wide range of catalytic applications, including hydrogen generation reactions. Among the various POM architectures, Dawson-type phosphotungstates have received considerable attention owing to their robust framework and the presence of multiple redox-active tungsten centers capable of facilitating electron-transfer processes. Despite the extensive use of polyoxometalates in catalysis, the hydrogen-generation performance of Dawson-type lithium phosphotungstate, Li₆[α-P₂W₁₈O₆₂]·28 H₂O (LiPOM), during NaBH₄ methanolysis has not been systematically investigated. To the best of our knowledge, this is the first study reporting the catalytic application of LiPOM for hydrogen production from sodium borohydride methanolysis. Unlike conventional noble-metal-based catalysts, LiPOM offers a structurally well-defined and redox-active molecular framework containing multiple tungsten centers capable of facilitating electron-transfer processes. In this work, the catalyst was comprehensively characterized, and its hydrogen-generation performance was evaluated through kinetic and thermodynamic analyses. Furthermore, the relationship between the Dawson structure, W-centered redox activity, and catalytic performance is discussed to provide insight into the origin of its catalytic behavior. These findings demonstrate the potential of Dawson-type lithium polyoxometalates as cost-effective and efficient alternatives for sustainable hydrogen generation.

In this study, a Dawson-type lithium polyoxometalate, Li₆[α-P₂W₁₈O₆₂]·28 H₂O, was synthesized and comprehensively characterized using XRD, ³¹P NMR, FT-IR, UV–Vis, SEM, TEM, AFM and BET techniques to confirm its structural integrity and morphology. The catalytic performance of LiPOM for hydrogen generation via NaBH₄ methanolysis was systematically investigated under varying reaction temperatures, catalyst loadings, and NaBH₄ concentrations. Furthermore, kinetic parameters were determined to evaluate reaction efficiency and gain insight into the catalytic role of the LiPOM framework. This study aims to provide a clearer understanding of lithium-stabilized Dawson-type POMs as promising catalysts for controlled hydrogen production and future sustainable energy applications. The overall experimental approach of the study, including LiPOM synthesis, characterization, and catalytic evaluation process for hydrogen production from NaBH₄ methanollysis, is schematically presented in Scheme 1.

Scheme 1.

A schematic representation of the experimental workflow and research strategy, from the synthesis of the Li₆[α-P₂W₁₈O₆₂]·28 H₂O catalyst to the evaluation of its hydrogen production performance.

Materials and methods

Materials

All chemicals used in this study, including lithium tungstate (Li₂WO₄), phosphoric acid (H₃PO₄), lithium chloride (LiCl), sodium borohydride (NaBH₄), and methanol, were purchased from Sigma-Aldrich. All reagents were of analytical grade with a purity of ≥ 99%, and were used as received without further purification.

Synthesis of Dawson type POM compound

3.82 g of lithium tungstate (Li₂WO₄) was added to 6 ml of boiling water and stirred for one hour. Then, 8 g of 85% phosphoric acid (H₃PO₄) was added dropwise to this mixture and stirred under reflux for two hours. After cooling to 60 °C, 5 g of LiCl was added. After this addition, the mixture was stirred under reflux again for 2 h. The mixture was then filtered through filter paper and recrystallized twice with methanol and then with 15 mL of 3.28 M LiCl solution²⁶. Dawson-type phosphotungstates can exist in two principal isomeric forms, namely the α- and β-isomers. The α-isomer is represented by Li₆[α-P₂W₁₈O₆₂]·14 H₂O, whereas the β-isomer is represented by Li₆[β-P₂W₁₈O₆₂]·19 H₂O. In the present study, the synthesized compound corresponds to the α-Dawson isomer. The formation reaction of Li₆[α-P₂W₁₈O₆₂] is given in Eq. (2)²⁶:

2

Characterization of Dawson type LiPOM

The structure of the synthesized LiPOM compound was characterized by 31P NMR, FT-IR, UV-Vis, SEM, TEM, and AFM methods. 31P NMR spectra were recorded at room temperature in (CD3)2CO solvent and using a Varian ASW400 (121.49 MHz) model spectrometer. A Bruker Vertex 70 spectrophotometer was used to record Fourier transform infrared spectroscopy (FT-IR) spectra of the compound embedded in KBr pellets in transmittance mode (2000 to 400 cm-1 range and 4 cm-1 resolution). The surface morphology was observed by using analytical techniques such as a Zeiss (EVOLS10) model Scanning Electron Microscope (SEM), a Transmission Electron Microscope (TEM), an NT-MDT Ntegra model AFM and Micromeritics 3Flex BET surface area device (BET).

Catalytic activity study of LiPOM

Catalytic activity studies were carried out using the water displacement method based on measuring the volume of gas produced every one minute during the hydrogen production reaction^27–29. In this method, hydrogen generated by a reaction occurring in an inverted gas collection tube completely filled with water was collected in a region. The volume of gas accumulated in this region was measured and recorded every minute. These recorded volumes were used to calculate the instantaneous hydrogen production rate (mL/min). Based on these calculations, the reaction kinetics were calculated based on the relationship between reaction and time. The methanolysis reaction of NaBH₄ was chosen to determine the catalytic activity of LiPOM. All catalytic activity determination studies were conducted in a Schlenk tube system to maintain an inert atmosphere. The effect of parameters such as catalyst amount (2.5 mg, 5 mg, 7.5 mg and 10 mg LiPOM), substrate concentration (300 mM, 350 mM, 400 mM, 450 mM NaBH₄) and reaction temperature (25 °C, 30 °C, 35 °C, 40 °C) were systematically investigated.

Result and discussions

Structural characterization

As shown in Scheme 2, the combination of fragments from two Keggin ions forms the Dawson structure [X₂M₁₈O₆₂]_n⁻. There are three metal groups in both the Kegkin ion structures, and these groups are divided to form the Dawson structure, and the remaining fragments are reassembled. Lacunary polyoxometalate structures are generally formed by the controlled removal of one or more addenda metal–oxo units from a plenary polyoxometalate framework, commonly under basic or hydrolytic conditions. The resulting vacant coordination sites provide reactive positions for further functionalization with metal ions or other inorganic/organic species. The open coordination centers of the structures directly relate with the advantage of being able to bond with metal or other nonmetal atoms. This bonding state is utilized to modify the structures and properties of POMs compounds toward applied usage³⁰.

Scheme 2.

General polyhedral representation of (X₂M₁₈O₆₂]_n⁻) polyoxometalate in Dawson structure (left), symbolic atom representation (ball and stick drawing) of synthesized LiPOM compound structure (right)²⁶.

The FT-IR spectrum confirms the successful formation of the Dawson-type polyoxometalate framework and the presence of hydration water molecules in the LiPOM structure. The broad absorption band at 3573 cm⁻¹ is attributed to O–H stretching vibrations of adsorbed or coordinated water molecules, while the band at 1613 cm⁻¹ corresponds to the H–O–H bending vibration. The characteristic band observed at 1080 cm⁻¹ is assigned to P–O stretching vibrations of the phosphate groups within the Dawson framework. The absorption band at 977 cm⁻¹ is attributed to the stretching vibration of terminal W = O bonds, whereas the band at 775 cm⁻¹ corresponds to bridging W–O–W vibrations within the polyoxometalate skeleton. As shown in Fig. 1a, vibrational bands are consistent with the reported spectral features of Dawson-type phosphotungstate polyoxometalates and confirm the preservation of the heteropolyanion structure after synthesis^31,32. The presence of the characteristic P–O, W = O and W–O–W vibrations confirms the preservation of the Dawson framework, which is essential for maintaining the redox-active tungsten centers responsible for catalytic hydrogen generation³³.

Fig. 1.

(a) FT-IR, (b) UV-Vis, (c) ³¹P NMR spectrum (121.49 MHz, D₂O) and (d) XRD of LiPOM compound.

The LiPOM has also been explored in pure water through absorption spectra. As indicated in Fig. 1b, the electronic transition at 254 nm by oxygen ligand-to-metal charge transfer band (O→W) of the POM anion represents the characteristic charge transfer transition peak of the Dawson-type α-P₂W₁₈O₆₂^34,35. When one electron or more are reduced in the process of transition, a new broad band between 500 and 900 nm is formed. The new band is due to periodic charge transfer transition between the reduced tungsten metal and either of the other d₀ tungsten metals or as a consequence of metal-centered d–d transition, the absorbance is thereby decreased in the UV region). Colour change upon the visible spectrum absorption due to periodic charge transfer in POM complexes results from the LMCT O→metal transition. This absorption of UV radiation is due to this effect. POMs show an increased photocatalytic activity through improved O→M charge transfer upon this absorption^35,36.

³¹P-NMR is frequently used to determine the purity of vacancy structure because it is highly sensitive to structural changes of polyoxoanions such as α-, β- isomers and substitution and demonstrates the chemical shift in the structure. The differences in the values of the chemical shift (δ) and the lacunarity of the structure are measured by the change in the electron density of the oxygen atoms of the central PO₄ tetrahedron³⁷. As can be noticed from Fig. 1c, the 31P-NMR spectrum of LiPOM dissolved in D₂O displays a sharp α-isomer peak at δ = −12.51 ppm. In addition, there are small β-isomer impurity peaks close to −11.01 and − 11.78 ppm, and the peaks are consistent with the literature data^33,38,40.

The XRD analyses in Fig. 1d clearly reveal the structural properties of the Li-POM catalyst. The sharp peak around 2θ ≈ 8° indicates that Li⁺ ions form a regular packing between POM clusters and that the laminar structure is preserved⁴¹. Other diffraction peaks (2θ ≈ 33°, 38°, 48°, and 56°) are clearly visible but not dominant; this indicates the partial amorphous/semi-crystalline character of the catalyst and small crystal sizes⁴². Similarly, the low peak intensity confirms that Li⁺ ions are homogeneously integrated into the structure and that no separate metallic Li phase is formed. The obtained XRD results reveal that the Li-POM catalyst maintains its structural integrity, containing semi-crystalline/amorphous areas alongside regular crystalline regions, providing a nano-sized and accessible active surface area; these features create a favorable basis for catalytic performance⁴³. The absence of impurity peaks and the excellent agreement with the reported diffraction pattern of Dawson-type phosphotungstates indicate the successful formation of a highly crystalline LiPOM phase. The preservation of the Dawson framework is particularly important because it provides the redox-active tungsten centers responsible for catalytic activity.

Morphological characterization of LiPOM compound

As shown in Fig. 2, the morphological changes in the films of the fabricated LiPOM compound containing α-P₂W₁₈O6₂ (Dawson-type) were examined using SEM, TEM, and AFM techniques. When these images are examined together, it is seen that the compound structure has a porous structure and forms an agglomeration structure in some regions. As shown in Fig. 2a, SEM images demonstrate that small POM clusters with an average size of 100–200 nm are individually bonded to form a large porous structure. In Fig. 2b, TEM images clearly show the crystalline structure of the POM clusters. These clusters were determined to have an average diameter ranging from 50 to 100 nm. Surface images and roughness analyses of the Dawson Li-POM compound were performed by AFM and are shown in Fig. 2c and d. This method is useful for investigating the variations in the topography. When the results obtained with this method are examined, the average surface roughness value (R_a) for the compound is 29.9891 nm. The presence of nanostructures and high polarity allow the surface to reach the desired potential value. This advantage allows for greater electron injection. The relatively high surface roughness suggests the presence of numerous accessible surface sites. Such a nanostructured morphology may facilitate reactant adsorption and improve contact between borohydride species and catalytically active tungsten centers, thereby contributing to the observed hydrogen-generation performance.

Fig. 2.

(a) SEM, (b) TEM, (c-d) AFM (2D and 3D), and (e) EDX images of LiPOM compound.

Lithium is an important element for hydrogen production, battery development, and new technologies, and analytical techniques, including SEM, capable of resolving this element spatially and quantitatively have not yet reached their detectable potential. Since the required sensitivity for detecting Li in lithium-containing compounds has not been achieved and an EDX detector optimized for detecting extremely low-energy radiation (< 100 eV) cannot be used, lithium metal cannot be detected in EDX analyses of elements with very low atomic numbers, despite their presence in the basic P-O-W compound structure shown in Fig. 2e^44,45.

Surface properties of the Li-POM structure were examined in detail using the Brunauer–Emmett–Teller (BET) method with N₂ adsorption–desorption isotherms, as shown in Fig. 3a. The analysis determined the specific surface area of the sample to be 1.97 m²/g according to the BET model. This value is highly consistent with the surface areas reported in the literature for Li-POMs, which typically have a crystalline, dense, and limited pore structure^46,47. Furthermore, the calculation of the single-point surface area as 2.28 m²/g in Fig. 3b and the fact that this value is quite close to the BET surface area confirms the suitability of the relative pressure (p/p₀) range selected during the measurement and the reliability of the adsorption data. The fact that the surface area calculated using the Langmuir model yields a significantly higher value of 37.17 m²/g indicates that the assumptions of homogeneous surface and single-layer adsorption in this model are not valid for Li-POM structures. Therefore, the BET model was used as the basis for surface area evaluations, and the Langmuir surface area was considered only as a comparative. Consequently, this compact and crystalline structure offers an advantageous platform for catalytic and electrochemical applications where electron transfer processes are crucial.

Fig. 3.

(a) N₂ adsorption-desorption isotherms, and (b) relative pressure (p/p₀) range graph of Li-POM.

The specific surface area of LiPOM was determined to be 1.97 m² g⁻¹. Although this value is relatively low compared with highly porous heterogeneous catalysts, LiPOM exhibited remarkable hydrogen-generation activity during NaBH₄ methanolysis. This apparent discrepancy indicates that catalytic performance is not governed solely by surface area. Unlike conventional porous catalysts, polyoxometalates derive their catalytic efficiency primarily from their molecular-level redox properties and electron-transfer capabilities. In particular, the Dawson-type framework contains multiple redox-active tungsten centers capable of reversible W(VI)/W(V) transitions, which facilitate borohydride activation and hydrogen evolution. Furthermore, the characteristic O→W ligand-to-metal charge-transfer transition observed in the UV–Vis spectrum confirms the electron-transfer capability of the tungsten-oxo framework. The nanostructured morphology and relatively high surface roughness revealed by AFM and TEM analyses provide accessible active sites and promote effective interaction between the catalyst surface and reactant molecules. Similar behavior has been reported for molecular polyoxometalate catalysts, where catalytic activity is strongly influenced by electronic structure and active-site accessibility rather than by specific surface area alone. Therefore, the high catalytic activity of LiPOM originates mainly from its redox-active molecular architecture and efficient electron-transfer characteristics rather than from a large surface area.

Hydrogen production of LiPOM

Effect of catalyst concentration on catalytic activity

The hydrogen production efficiency of LiPOM catalyst at different concentrations is shown in Fig. 4. During the experimental studies, LiPOM catalyst was taken at different concentrations of 2.5 mg, 5 mg, 7.5 mg and 10 mg control experiments were carried out. During the experimental work, NaBH₄ concentration was kept at 300 mM and temperature was kept at 25 ± 0.5 °C. Hydrogen production graph is shown in Fig. 4a. These results show that the hydrogen release rate increases in direct proportion with the increase in the amount of catalyst. This increase is also explained by the fact that more interaction between NaBH₄ and LiPOM occurs during the methanolysis reaction. Figure 4b shows the first order plot of the hydrogen release rate according to the catalyst concentration. Accordingly, hydrogen release rate depending on the NaBH₄ concentration was calculated as R² = 0.9994. The results indicate that LiPOM catalyst shows a superior catalytic activity.

Fig. 4.

(a) Hydrogen release rate of LiPOM catalyst at different concentrations, (b) first order hydrogen release rate plot (300 mM, 25 ± 0.5 °C).

Effect of NaBH₄ concentration on catalytic activity

The hydrogen release rate activity of LiPOM catalyst at different substrate concentrations is shown in Fig. 5. In order to investigate the effect of substrate concentration on hydrogen release rate, control experiments were carried out with 300 mM, 350 mM, 400 mM, 450 mM NaBH₄. During the methanolysis reaction, 2.5 mg LiPOM catalyst and temperature was kept constant at 25 ± 0.5 °C. Figure 5a shows the effect of hydrogen release rate on NaBH₄. These results show that as the NaBH₄ concentration increases, the hydrogen release rate also increases. Accordingly, the first order graph curve of the hydrogen release rate is given in Fig. 5b. The hydrogen release rate depending on the NaBH₄ concentration was calculated as R² = 0.9916.

Fig. 5.

(a) NaBH₄ hydrogen release rate at different concentrations, (b) first order plot of hydrogen release rate (2.5 mg LiPOM, 25 ± 0.5 °C).

Effect of different temperatures on catalytic activity

The catalytic activity of LiPOM catalyst at different temperatures was investigated depending on the hydrogen release rate as shown in Fig. 6. Accordingly, control experiments were carried out at 25 °C, 30 °C, 35 °C and 40 °C, keeping the catalyst (2.5 mg LiPOM) and substrate concentration (300 mM) constant. The hydrogen release at different temperatures is shown in Fig. 6a. The results show that the increase in hydrogen release rate is directly proportional to the increase in temperature and is consistent with the literature. Similarly, Arrhenius and Eyring plots depending on the hydrogen release rate in Fig. 6a are shown in Fig. 6b and c, respectively. Additionally, the slope of the line obtained from the Arrhenius graph is evaluated as -Eₐ/R, and the point where it intersects the y-axis is evaluated as lnA. According to the results obtained, the activation energy of the reaction was derived using the Arrhenius equation as shown in Eq. (3)^48,49, and the activation energy was calculated as 28.12 kJ mol^− 1.

3

 

Fig. 6.

(a) Hydrogen release performance of LiPOM catalyst at different temperatures, (b) Arrhenius and (c) Eyring plot (2.5 mg LiPOM, 300 mM NaBH₄).

This transformation in Eq. (3) is calculated according to the first-order plot of ln (k) versus T^− 1 to find the activation energy, given the universal gas constant R. The activation energy Ea was calculated as R² = 0.9638 for the line taken from these data points, as shown in Fig. 6b. In addition, enthalpy and anthropy calculations were made based on the Eyring plot of the hydrogen release rate given in Fig. 6c. The Eyring equation was calculated as given in Eq. (4)^50,51. Enthalpy (∆H) and entropy (∆S) values were calculated as 25.56 kJ mol^− 1 and − 128.98 J mol^− 1 K^− 1, respectively.

4

Furthermore, the catalytic behavior of the LiPOM framework was evaluated by determining its kinetic parameters. The results show that the redox properties of W centers in POM structures are well-known in the literature^52,53; in this context, the obtained results support a role that may contribute to electron transfer in the methanollysis process. However, this suggestion is based on kinetic analyses rather than direct mechanistic evidence.

Furthermore, the catalytic behavior of the LiPOM framework was evaluated by determining its kinetic parameters. The observed catalytic activity can be directly correlated with the structural and electronic characteristics of the LiPOM catalyst. XRD analysis confirmed the preservation of the Dawson-type framework, which provides a robust molecular architecture containing multiple redox-active tungsten centers. These W centers are known to undergo reversible WVI/WV transitions and can facilitate electron-transfer processes during borohydride activation. In addition, the UV–vis spectrum exhibited a characteristic O→W ligand-to-metal charge-transfer transition at 254 nm, indicating the electron-accepting nature of the tungsten-oxo framework. Such charge-transfer capability is beneficial for promoting electron redistribution during the catalytic cycle. Morphological analyses further revealed a nanostructured surface composed of interconnected POM clusters with a relatively high surface roughness (Ra = 29.99 nm), providing accessible active sites for reactant adsorption. Although the BET surface area of LiPOM was relatively low (1.97 m² g⁻¹), the catalyst exhibited remarkable hydrogen-generation activity. This finding suggests that the catalytic performance is governed primarily by the redox-active molecular structure and accessibility of active sites rather than by surface area alone. Furthermore, Li⁺ ions contribute to maintaining the structural integrity of the Dawson framework and may facilitate substrate activation through electrostatic interactions with BH₄⁻ species. Therefore, the enhanced catalytic performance of LiPOM can be attributed to the synergistic combination of redox-active W centers, charge-delocalized Dawson architecture, Li⁺-assisted stabilization, and nanostructured surface morphology.

In Table 1, kinetic and thermodynamic parameters of different catalyst systems in methanol conversion reaction are presented as comparisons. In this context, the synthesised LiPOM catalyst stands out with its low activation energy, high conversion frequency and thermodynamic properties. The activation energy of the LiPOM catalyst is 28.12 kJ mol^− 1 and when compared with other catalysts in the literature such as PdNi@VC (54.20 kJ mol^− 1) and ZnSnO₃-SnO₂ (kJ mol^− 1), its ability to initiate the reaction even at low temperatures stands out. The positive activation enthalpy value (ΔH = 25.56 kJ mol⁻¹) indicates that the NaBH₄ methanolysis reaction requires an energy input during the formation of the activated complex. This finding suggests that the reaction follows a kinetically energy-barrier-controlled mechanism. Furthermore, the increase in hydrogen production rate with rising temperature supports the endothermic nature of the activation process. Consequently, the obtained ΔH value indicates that the reaction proceeds more readily at higher temperatures in the presence of the Li-POM catalyst and that the formation of the transition state is thermally supported. Another entropy change (ΔS = −128.98 J mol^− 1 K^− 1) also indicates a more ordered structure in the direction of the products. A negative entropy indicates an increase in order at the molecular level during the reaction, which is usually associated with the conversion of small and mobile molecules such as methanol into products with less freedom. Furthermore, beyond all these thermodynamic data, the TOF value of LiPOM was calculated as 942.61 s-¹, which indicates that the number of conversions per second per unit active surface area of the catalyst is very high. In this context, LiPOM draws a very active catalyst profile not only thermodynamically but also kinetically. Although there are few systems with high TOF values in the literature, this feature strengthens the application potential of LiPOM to an excellent degree. In conclusion, LiPOM catalyst shows a low enthalpy and activation energy, entropy regulating effect for methanol based reactions, but shows a very high kinetic performance with conversion frequency. Current research in hydrogen production and storage indicates that production and storage processes for sustainable energy systems must be addressed in conjunction with materials science-based approaches. In this context, the effects of metal alloys, metal hydrides, and nanostructured materials on hydrogen production kinetics and storage performance have been examined in detail. In recent years, it has been reported that POM-based systems offer significant potential for catalytic hydrogen production due to their multi-electron redox properties and structural tunability. In this context, Li-POM-containing materials stand out as a promising alternative in terms of efficiency and stability in hydrogen production processes^57–59.

Table 1.

Comparative kinetic, thermodynamic and catalytic parameters of representative noble-metal, transition-metal and polyoxometalate catalysts reported for NaBH₄ methanolysis in the literature.

Catalyst System	Reaction Type	Ea (kJ/mol)	ΔH (kJ/mol)	ΔS (J/mol·K)	TOF (s− 1)	Ref.
CO–ZnO NPs	MeOH	17.45	14.87	−194.37	805.17	12
PtRu@SWCNT	MeOH	17.29	15.46	−171.68	136.25	54
ZnSnO3-SnO2	MeOH	43.19	40.65	−178.89	832.38	7
ZnO/Co@f-MWCNT	MeOH	16.36	13.73	−122.80	-	55
PdNi@VC	MeOH	54.20	51.7	−134.5	-	56
LiPOM	MeOH	28.12	25.56	−128.98	942.61	This work

Proposed catalytic mechanism of LiPOM during NaBH₄ methanolysis

A plausible catalytic mechanism for hydrogen generation over Li₆[α-P₂W₁₈O₆₂]·28 H₂O is proposed based on the structural characterization results, UV–vis analysis, and kinetic data obtained in this study. Dawson-type polyoxometalates are well known for their reversible multi-electron redox behavior, particularly through W(VI)/W(V) transitions, which enable efficient electron transfer processes. During NaBH₄ methanolysis, borohydride ions first adsorb onto the catalyst surface. The tungsten centers within the Dawson framework may act as transient electron-accepting sites, facilitating charge redistribution and weakening the B–H bonds of BH₄⁻ species. This process promotes hydride transfer to methanol molecules, resulting in accelerated hydrogen release. The characteristic O→W ligand-to-metal charge-transfer transition observed at 254 nm in the UV–vis spectrum supports the electron-transfer capability of the tungsten-oxo framework. Moreover, the highly delocalized electronic structure of the Dawson anion allows temporary stabilization of reduced tungsten species without significant structural distortion. Li⁺ ions play an additional role by stabilizing the polyoxometalate framework and maintaining the integrity of the catalyst structure during the reaction. Furthermore, electrostatic interactions between Li⁺ ions and borohydride species may facilitate substrate activation and improve accessibility to catalytic sites. The nanostructured morphology and relatively high surface roughness revealed by AFM analyses provide numerous accessible active sites for reactant adsorption. Consequently, the combined effects of redox-active W centers, the electronically delocalized Dawson framework, and Li⁺-assisted stabilization contribute to the enhanced hydrogen-generation performance of the LiPOM catalyst. To provide a clearer understanding of the catalytic pathway and the possible roles of W centers and Li⁺ ions during hydrogen evolution, a mechanistic illustration is presented in Scheme 3.

Scheme 3.

Proposed catalytic mechanism of LiPOM during NaBH₄ methanolysis.

As illustrated in Scheme 1, the catalytic cycle begins with the adsorption of BH₄⁻ species onto the LiPOM surface through interactions with terminal W = O groups and electrostatically accessible Li⁺ sites. Subsequent electron transfer from BH₄⁻ to redox-active W(VI) centers facilitates B–H bond activation and promotes hydride transfer to methanol molecules. Hydrogen evolution occurs through successive proton–hydride interactions, while borate-containing species are formed as by-products. During the catalytic process, the reduced W(V) centers are reoxidized to W(VI), regenerating the active sites and maintaining catalytic turnover. The proposed mechanism is consistent with the characteristic O→W charge-transfer transition observed in the UV–vis spectrum, the preserved Dawson framework evidenced by XRD analysis, and the favorable kinetic parameters obtained from Arrhenius and Eyring evaluations.

Conclusions

Dawson-type lithium polyoxometalate, Li₆[α-P₂W₁₈O₆₂]·28 H₂O, was successfully synthesized and comprehensively characterized to verify its structural and physicochemical features. UV–Vis spectroscopy confirmed the presence of ligand-to-metal charge transfer transitions characteristic of tungsten-based Dawson frameworks, while FT–IR spectra displayed the diagnostic vibrational modes of the [P₂W₁₈O₆₂]⁶⁻ anion, confirming preservation of the polyoxometalate skeleton. The ³¹P NMR spectrum verified predominance of the α-isomer, with only minor traces of alternative species, indicating high isomeric purity of the product. Microscopic analyses further revealed a rough and nanostructured surface morphology, which is favorable for interfacial interactions during catalytic reactions. Collectively, these characterization results demonstrate that the synthesized LiPOM possesses a stable framework, high purity, and a controlled surface architecture. Owing to its robust Dawson-type structure, pronounced electronic transitions, and enlarged effective surface area, the lithium-stabilized polyoxometalate exhibits strong potential as a functional catalyst. In particular, its performance in promoting hydrogen evolution from NaBH₄ methanolysis highlights the promise of lithium-based Dawson-type POMs as efficient catalytic materials for environmentally benign hydrogen generation and related energy conversion applications.

Author contributions

Yasemin Torlak: Writing – original draft, Methodology, Conceptualization. Ebru Halvacı: Writing – original draft, Methodology, Visualization, Conceptualization. Ayşenur Aygün: Writing – review & editing, Visualization, Formal analysis. Fatih Sen: Writing – review & editing, Conceptualization, Supervision, Investigation.

Funding

The author gratefully acknowledges that no specific funding was received for this work.

Data availability

All data generated or analysed during this study are included in this published article.

Declarations

Competing interests

The authors declare no competing interests.