Pressure mediated phase transition and dihydrogen bonding formation in trimethylamine borane†

Rongfeng Guan^ab, Jingyan Liu^a, Aditya Kutty^a, Zhihao Yu^a, Yujin Ji^b, Youyong Li*^bc and Yang Song*^ac
aDepartment of Chemistry, The University of Western Ontario, London, Ontario N6A 5B7, Canada. E-mail: yang.song@uwo.ca
^bInstitute of Functional Nano & Soft Materials (FUNSOM), Jiangsu Key Laboratory for Carbon-Based Functional Materials and Devices, Soochow University, Suzhou 215123, Jiangsu, China. E-mail: yyli@suda.edu.cn
^cSoochow University – Western University Centre for Synchrotron Radiation Research, The University of Western Ontario, London, ON N6A 5B7, Canada

First published on 22nd August 2024

---

---

1. Introduction

Hydrogen is widely acknowledged as a promising clean energy source offering a sustainable alternative to fossil fuels due to its abundant availability, high energy content, and minimal environmental impact during the conversion.¹ Over the past several decades, extensive research efforts have been dedicated to the development of solid-state hydrogen storage materials to promote the practical application of hydrogen energy.^2,3 For some time, ammonia borane has emerged as a standout hydrogen storage material with a high gravimetric hydrogen density of 19.6 wt% and a low molecular weight (30.7 g mol⁻¹).^4,5 Ammonia borane also exhibits a moderate dehydrogenation temperature and fulfills the necessary stability requirements for the safe storage of hydrogen. The performance of ammonia borane as a hydrogen storage material is significantly affected by the presence of unconventional hydrogen bonds, known as dihydrogen bonds (DHBs).^6,7 These bonds are characterized by the presence of both protonic H (H^δ+) and hydridic H (H^δ−) in the form of N–H^δ+⋯H^δ−–B, which not only enhance the structural stability of ammonia borane but also facilitate the efficiency of hydrogen release.⁸ Further research has shown that applying high pressure to ammonia borane can dramatically strengthen these dihydrogen bond interactions between molecules, thus introducing new crystal phases and adjusting both the dehydrogenation performance and electronic properties of ammonia borane.^9–17 In particular, Shi et al.¹⁸ showed that ammonia borane is capable of achieving superconductivity under high pressure, and the reinforcement of dihydrogen bonds in the ammonia borane mixed H₂ system drastically decreases the required pressure for achieving superconductivity.

Although ammonia borane exhibits promising properties as a hydrogen storage material, its broader application is limited by certain aspects, such as the need for further optimization of its dehydrogenation temperature for effective energy applications, slow decomposition kinetics and challenges related to recyclability and efficiency due to the formation of harmful by-products.¹⁹ These issues highlight the urgent need to explore the derivatives^20–28 of ammonia borane that may offer improved performance and solutions to existing challenges. For example, extensive research efforts have been conducted to N-methylamine-borane adducts (Me_nH_3−nN·BH₃, n = 1–3), including investigations of dehydrogenation mechanisms and identification of effective hydrogen release catalysts.^29–33 Furthermore, applying external pressure to N-methylamine-borane adducts can significantly influence their properties by altering crystal structures and dihydrogen bonding behaviors, potentially leading to the formation of new stable or metastable phases upon compression. Notably, Szilágyi et al.'s research³⁴ on methylamine borane (MAB) and dimethylamine borane (DMAB) at pressures up to 3 GPa identified the phase transitions at 0.8–1.2 GPa and 0.7 GPa, respectively, which were associated with the strengthening of dihydrogen bonding interactions.

Similar to MAB and DMAB, trimethylamine borane (TMAB) also has a high hydrogen content, making it a potential hydrogen storage material. In addition, TMAB is widely used as a precursor^35,36 for synthesizing boron carbonitride materials, but it has not been studied under high-pressure conditions. The high-pressure stability and phase transitions of TMAB are crucial for its applications in both hydrogen storage and material synthesis. Consequently, there is an urgent need to explore the properties of TMAB under high pressure to fully understand its potential for use in advanced applications. In this work, we systematically investigate the structural stability, electronic properties and pressure-induced phase transition of TMAB up to 35 GPa using a combination of in situ IR spectroscopy, Raman spectroscopy, synchrotron X-ray diffraction (XRD), evolutionary crystal structure prediction, and density functional theory (DFT) calculations. Our findings reveal a new phase of TMAB and provide key insights into its structural stability, bonding, and electronic properties. These insights are crucial for advancing TMAB's potential in hydrogen storage and the synthesis of advanced materials.

2. Methodology

2.1. Sample preparation

Trimethylamine borane with a nominal purity of 97% was purchased from Alfa-Aesar and used without any additional purification. High-pressure Raman experiments were conducted using a symmetrical diamond anvil cell (DAC) equipped with two 400 μm culet Type-I diamonds, while IR experiments employed a set of 300 μm culet Type-II diamonds. The sample was loaded in an MBraun LAB Master 130 glovebox under a N₂ atmosphere (<10 ppm O₂ and H₂O) using silicone oil as the pressure-transmitting medium for Raman and XRD measurements. Before adding the sample, several ruby chips (Cr³⁺ doped α-Al₂O₃) were carefully placed inside the sample chamber to serve as pressure calibrants. Pressure measurements were assessed by monitoring shifts in the R₁ ruby fluorescence line, achieving an accuracy of ±0.05 GPa in quasi-hydrostatic conditions. For IR measurements, high-quality KBr powder was loaded into the DAC to act both as a pressure medium and to dilute the sample. Ruby fluorescence spectra from various chips within the chamber showed consistent pressure without any significant gradients or non-hydrostatic effects.

2.2. In situ characterizations at high pressures

In situ high-pressure Raman measurements were conducted using a specialized Raman micro-spectroscopy system. A 532.10 nm wavelength single longitudinal mode, diode-pumped solid-state (DPSS) green laser served as the excitation source. The laser beam was focused to less than 5 μm on the sample using a 20× Mitutoyo objective, which also collected the Raman signal in a backscattering configuration. Rayleigh scattered light was filtered out using a pair of notch filters, allowing for spectral collection starting from >100 cm⁻¹. An imaging spectrograph, fitted with a 1200 lines per mm grating, dispersed the scattered light achieving a resolution of 0.1 cm⁻¹. The Raman signals were captured by a highly sensitive, back-illuminated charge-coupled device (CCD) detector from Acton, which was cooled by liquid nitrogen. The entire system's accuracy was maintained within ±1 cm⁻¹ using neon line calibration.

In situ high-pressure IR absorption measurements were conducted using a specialized IR micro-spectroscopy system. The system incorporated a commercial Fourier transform infrared (FTIR) spectrometer, the Vertex 80v model from Bruker Optics Inc., which featured a Globar mid-IR light source and operated under a vacuum of less than 5 mbar to effectively eliminate H₂O and CO₂ absorption. The IR beam, after passing through a KBr window of the spectrometer, was directed into a relay box, focused onto the sample via iris optics and a 15× reflective objective lens with a numerical aperture of 0.4. The beam's diameter was precisely matched to the size of the sample (approximately 150 μm) using various iris apertures. The transmitted IR beam was captured by another similar reflective objective used as a condenser and channeled to a wide-band mercury cadmium telluride (MCT) detector with a ZnSe window, covering a spectral range from 400 to 10000 cm⁻¹. All measurements were performed in transmission (or absorption) mode with a resolution of 4 cm⁻¹ and 512 scans to maximize the signal-to-noise ratio. A reference spectrum, i.e., the absorption from diamond anvils filled with KBr but without the sample, was subtracted from each sample spectrum to calculate the absorbance.

In situ angle-dispersive microdiffraction was performed at the high energy wiggler (WHE) beamline of the Brockhouse X-ray Diffraction and Scattering (BXDS) sector of the Canadian Light Source (CLS). The wavelength of the monochromatic X-ray beam was 0.3497 Å with a beam size of ∼50 μm both horizontally and vertically guided by a pinhole. The diffraction geometry was calibrated using nickel powder standard. The 2D Debye–Scherrer diffraction patterns were collected using a Varex XRD 4343CT area detector. The 2D diffraction images were integrated into 1D powder patterns using the Dioptas program³⁷ for further analysis. The 1D XRD patterns were analyzed with the Rietveld refinement method using GSAS-II software.³⁸

2.3. Computational details

Evolutionary algorithms implemented in package USPEX^39,40 were used for prediction of structures of TMAB. Heredity, permutation, mutating molecular orientations, soft-mutations variable operators, and randomly generated structures with various weights were applied in the evolution simulations. The electronic structure calculations for the resulting candidate structures of TMAB were performed at the DFT level using the CASTEP⁴¹ code (Version 20.11). The lattice parameters and atomic positions of the high-pressure candidate TMAB structures were optimized at the target pressure by minimizing the enthalpy. The Perdew–Burke–Ernzerhof (PBE)⁴² generalized gradient approximation was applied for the electronic exchange–correlation functional. Ultrasoft Vanderbilt pseudopotentials⁴³ were used to model the interaction potential between ionic reality and valence electrons. A plane wave cutoff energy of 800 eV and Monkhorst–Pack⁴⁴ k-meshes with a spacing of 2π × 0.03 Å⁻¹ were selected to ensure energy tolerances converged to better than 1 meV per atom and forces smaller than 0.003 eV Å⁻¹. The convergence criterion for self-consistent field (SCF) calculations was set to 1.0 × 10⁻⁹ per atom. The TMAB structures identified as Phases I and II were verified to be true minima by phonon calculations. Electronic property calculations for the predicted stable phases were performed using the VASP code,^45–47 also utilizing the PBE generalized gradient approximation with the all-electron projector-augmented wave (PAW) method.⁴⁸ Post-processing analysis was carried out using VASPKIT.⁴⁹

3. Results and discussion

3.1. IR and Raman spectra of TMAB at 0–35 GPa

As a starting point, the experimental IR and Raman spectra of TMAB were collected under ambient conditions (Fig. 1). The results are consistent with previous research by Durig et al. (Table S1, ESI†),⁵⁰ including the most characteristic vibrational bands: the N–C stretching region, the B–H deformation region, the C–H deformation region, the B–H stretching region, and the C–H stretching region. To explore the behavior of TMAB under compression, we collected the in situ IR spectra of TMAB from ambient pressure up to 35 GPa. The results (Fig. 2) suggest that TMAB undergoes a phase transition at around 9.4 GPa as indicated by the following spectral features: a new peak suddenly appears at 1236 cm⁻¹, acting as the shoulder peak of the B–H deformation mode; the two components of the C–H deformation mode at 1447 cm⁻¹ develop into a triplet; another new peak appears at 2434 cm⁻¹ right in front of the B–H asymmetric mode. Beyond 9.4 GPa, there is a noticeable broadening and softening of all modes, accompanied by diminished peak intensities. At 34.8 GPa, the appearance of an amorphous phase was indicated by extremely weakened and broadened peaks. No further dramatic phase transitions occur in the second pressure region, suggesting that the phase II observed in our high-pressure study has a much wider stability region than the phase I.

The experimental in situ Raman spectra of TMAB were collected up to 30 GPa with the most characteristic spectral regions (Fig. 3). The results point out that a phase transition occurs near 9 GPa with the following spectral features: all the lattice modes become extremely weakened and broadened at 8.9 GPa; lattice mode 1 almost disappears, and the intensity of the sharp B–N stretching mode is significantly reduced; two peaks in the B–H stretching region remarkably fade away, and one of them merges with the second order Raman mode of diamond at 2469 cm⁻¹; the C–H stretching mode at 3018 cm⁻¹ vanishes, whereas a new shoulder peak of the C–H asymmetric stretching mode emerges at 3062 cm⁻¹. These profile changes are indicative of a phase transition near 9 GPa, which is consistent with the results we obtained from the IR measurements.

The suggested phase transition near 9 GPa in the IR and Raman spectra could be further visualized by plotting the characteristic modes as a function of pressure (Fig. 4). Pressure coefficients for the assigned modes were calculated by the least-square fitting of the experimental data are listed in Tables S2 and S3 (ESI†). All the pressure coefficients are positive, indicating that all of the modes display blue shifts in the entire pressure region due to bond stiffening upon compression. This behavior is different from that observed in MAB and DMAB,³⁴ where the frequency of N–H bonds exhibits obvious redshifts under high pressure, indicating the pressure-induced strengthening of the N–H⋯H–B dihydrogen bonding. In particular, the phase transition pressure of TMAB at 9 GPa is in strong contrast to those of MAB and DMAB, which have been shown to undergo pressure-induced phase transitions at considerably lower pressures of 0.8–1.2 GPa and 0.7 GPa, respectively. These results indicate that TMAB exhibits significantly robust structural stability.

3.2. X-ray diffraction patterns of TMAB at 0–22 GPa

The Rietveld refinement of experimental powder XRD patterns at ambient conditions identified the unit cell parameters of TMAB in the R3m space group as a = b = 9.283 Å, c = 5.964 Å, and V = 445.066 ų, which are consistent with the ref. 51. To examine the pressure-induced structural transformations in TMAB, we collected the room temperature synchrotron XRD patterns of TMAB as a function of pressure up to 22 GPa (Fig. 5). Upon compression, the reflections shift to higher angles 2θ relatively smoothly indicating the pressure-induced contraction of unit cells without major structural change (Fig. S1, ESI†). However, the peak at 2θ = 11.08° splits into two peaks when pressures reached 9.15 GPa, indicating a possibly modified crystal structure. Upon decompression, the recovered XRD pattern resembles the one before compression indicating a reversible change. These observations are consistent with the Raman and IR measurements.

3.3. Structure determination of phase II at high pressure

The vibrational spectra and XRD patterns suggested a possible new high-pressure phase of TMAB. To verify and determine the crystal structure of the new phase, we performed structural search calculations using the USPEX package in a pressure range of 0–20 GPa. The calculations yielded five top-ranked candidate structures with the lowest enthalpies belonging to four space groups labeled as follows: P3₁ (A), P3₁ (B), P3₂, R₃ and P₃ (see Table S4 (ESI†) for structural details). These structures were further optimized at target pressures and the calculated enthalpies of these top-ranked candidate structures relative to the enthalpy of the reference phase I-R3m (black curve) are shown in Fig. 6. Among these structures, the relative enthalpies of R3, P3₁ (B) and P3₂ structures increase with pressure and thus can be ruled out as the possible phase II of TMAB at high pressures. At 0 GPa, the P3 structure exhibits higher enthalpy and is less stable compared to the phase I-R3m. As pressure increases to about 3 GPa, it overtakes the phase I in stability and is hence regarded as a possible phase II. Another possible candidate for phase II is the P3₁ (A) structure. The relative enthalpies of the P3₁ (A) and the phase I-R3m structures are very close across the entire pressure range compared to other candidate structures, as illustrated by the overlapped red and black curves and highlighted in Fig. 6 inset. It becomes more stable than the phase I-R3m at around 8 GPa. The phase I-R3m structure at ambient pressure has lattice parameters: a = b = 9.390 Å, c = 5.967 Å, while the P3₁ (A) structure has a = b = 9.405 Å, c = 5.938 Å. These parameters lead to similar unit cell volumes and shapes, contributing to nearly identical relative enthalpies. Besides, the high symmetry in these structures places atoms in special positions, ensuring evenly distributed atomic interactions and thus minimizing enthalpy differences across the entire pressure range.

In order to confirm phase II of TMAB from these two top-ranked candidate structures, we employed Rietveld's method to refine the experimental XRD pattern at 9.15 GPa for both the P3 and the P3₁ (A) structures, respectively. Despite meeting the necessary pressure-enthalpy conditions, the P3 structure was eliminated as a potential phase II due to the relatively poor XRD refinement results (Fig. S2, ESI†). This poor match can be explained by high kinetic barriers that likely prevent the crystal from achieving its most stable configuration under experimental conditions, leading to the discrepancy in the XRD patterns. In contrast, the simulated XRD pattern for phase II refined at 9.15 GPa using the P3₁ (A) structure as input matches the experimental pattern well (Fig. 7), which strongly supports the P3₁ (A) as phase II. Additionally, the IR and Raman spectral simulations of the P3₁ (A) structure at 9 GPa are in good agreement with the experimental vibrational spectra for the most characteristic peaks associated with functional groups (Fig. S3 and S4, ESI†), further supporting the P3₁ (A) structure as phase II. Therefore, we conclude that TMAB likely undergoes a pressure-induced phase transition at about 9 GPa from the R3m (phase I) phase to a P3₁ phase (phase II).

A detailed comparison of the two structural models of the phase I-R3m and the phase II-P3₁ (A) optimized at ambient pressure is shown in Fig. 8. In the R3m structure, atoms are arranged with higher symmetry, occupying special positions that result in a more ordered configuration. This higher symmetry constrains atomic positions and reduces vibrational degrees of freedom. In contrast, the P3₁ (A) structure has lower symmetry, allowing slight shifts in atomic positions and introducing additional vibrational modes due to the less constrained arrangement. These structural differences contributed to the more favorable figures of merit in the Rietveld refinements comparing these two very crystallographically similar structures, indicating P3₁ (A) is the more likely structure than R3m, especially at high pressure although we did not rule out other possibilities or the coexistence of these two phases. Furthermore, these structural difference leads to distinct vibrational modes, as evidenced by the changes in the IR and Raman spectra observed at around 9 GPa.

To investigate the phase stability of TMAB, we analyzed its compressibility under pressure by fitting the experimental and calculated unit cell volumes per TMAB molecule for phases I (0–9 GPa) and II (9–20 GPa) to the third-order Birch–Murnaghan⁵² equation of state (EoS) given by the following expression:

where P is the pressure, V₀ is the reference volume, V is the deformed volume, B₀ is the bulk modulus, and B₀′ is the derivative of the bulk modulus with respect to pressure. The fitted EOS parameters for phase I are V₀ = 161.7 ų and B₀ = 2.1 GPa (experiment) vs. V₀ = 151.4 ų and B₀ = 4.0 GPa (simulation); for phase II, we obtained V₀ = 127.3 ų and B₀ = 30.0 GPa (experiment) vs. V₀ = 136.2 ų and B₀ = 11.9 GPa (simulation). It can be seen that both the experimental and the simulated results (Fig. 9) show that the densities for the phase I and phase II are almost identical, indicating the possibility of the coexistence of these two phases in a broad pressure range. Furthermore, the non-hydrostatic influence at high pressures due to the use of silicon oil as pressure-transmitting medium makes the accurate identification of the phase boundary and quantitative analysis challenging.

3.4. Structure stability and evolutions of the C–H⋯H–B dihydrogen bond frameworks in TMAB at high pressures

As we mentioned above, the impact of dihydrogen bonds and their pressure-induced changes can significantly affect the structural stability and development of new high-pressure polymorphs in ammonia borane and its derivatives.¹⁷ Based on the empirical rule from the Cambridge structural database, a dihydrogen bond is considered to exist when the H⋯H contact distance between two molecules falls within the range of 1.7 and 2.4 Å.^53,54 Notably, the relatively strong N–H⋯H–B dihydrogen bond is absent in TMAB, and only the relatively weak⁵⁵ C–H⋯H–B dihydrogen bond exists in its high-pressure phases. Therefore, we investigated the evolution of the C–H⋯H–B dihydrogen bond in TMAB under high pressures.

At ambient pressure, all the C–H⋯H–B distances in the R3m TMAB are greater than 2.4 Å, indicating the absence of dihydrogen bond (Fig. 10(c)). The number of dihydrogen bonds increases with compression: At 3 GPa, one type of the dihydrogen bond appears due to the pressure-induced strengthening. Then the number of types of dihydrogen bonds gradually increases to two in phase I at 4 GPa (Fig. 10(d)) and remains stable until 8 GPa where four types of dihydrogen bonds that differ in bond lengths and angles are found to be favorable in phase I (Fig. 10(a)). This indicates the start of the phase transition. Then the number of dihydrogen bond types remains at four in the entire phase II region up to 20 GPa (Fig. 10(e)).

An important difference of TMAB from MAB and DMAB is the absence of the relatively strong N–H⋯H–B dihydrogen bonds in TMAB. The N–H⋯H–B dihydrogen bonding networks in MAB and DMAB are sensitive to external pressure, allowing them to undergo phase transitions at lower pressures. Conversely, TMAB requires nearly ten times that pressure to initiate a phase transition, and there is a significant change in the C–H⋯H–B dihydrogen bonds of TMAB from two types to four at 8 GPa, which induces a phase transition that marks the onset of structural changes at this pressure. Of course, the different experimental conditions for MAB and DMAB than that for TMAB may contribute to different pressure effect on dihydrogen bonds. Therefore, it would be interesting to comparatively investigate another ammonia borane derivative containing simply C–H⋯H–B dihydrogen bonds.

3.5. Band structures and energies of TMAB in a pressure range of 0–35 GPa

Finally, we examined the electronic properties of TMAB to better understand its high-pressure behaviors. The calculated electronic band structure and density of states of phase I and phase II at 9 GPa are shown in Fig. 11. Both phases exhibit large direct band gaps, with the phase I structure having a band gap of 6.65 eV at 9 GPa, and the phase II structure having a smaller gap of 6.54 eV at the same pressure. The tops of valence bands in both phases consist mainly of B and H atoms, indicating their significant roles as an active center. Moreover, Fig. 11(a) illustrates the calculated band gaps for TMAB extending up to 35 GPa, the highest pressure achieved in our experiment, and it can be seen that the band gap of TMAB shows a gradual reducing trend in the entire pressure region. In particular, the significant drop in band gap energy at 9 GPa is associated with the pressure-induced phase transition.

3.6. Discussion

The structural stability and nature of the phase transition of TMAB are of great interest. We noticed that the pressure-volume curve for both phase I and phase II is notably smooth across the entire pressure range (Fig. 9). This is in strong contrast to many other ammonia borane derivatives^27,34 which exhibit significant volume collapses during phase transitions, suggesting a substantial contribution from the P–V term as a driving force for their phase changes. For TMAB, the minimal volume changes observed indicate that the transitions are likely of second order, involving gradual structural shifts rather than abrupt changes. This gradual transition supports the coexistence of two phases at high pressures. Furthermore, the minimal difference in enthalpy values between phase I and phase II of TMAB across the entire pressure range indicates minimal thermodynamic advantage in transitioning between the two phases (Fig. 6). This suggests that both phases are possible equally favorable under high-pressure conditions, which also allows for easy reversibility and potential coexistence of the two phases at high pressures.

Our studies reveal that TMAB maintains structural stability up to 9 GPa, which is significantly higher compared to other N-methylamine-borane adducts: MAB exhibits phase transition at 0.8–1.2 GPa, and DMAB exhibits phase transition at 0.7 GPa.³⁴ This difference can be attributed to the unique configuration and interaction of dihydrogen bonds within TMAB. Unlike MAB and DMAB, which undergo phase transitions at lower pressures due to more sensitive N–H⋯H–B dihydrogen bonding networks, the dihydrogen bonds of TMAB only involve C–H⋯H–B interactions. Based on the previous computational studies, C–H⋯H–B has lower stabilization energy than that of N–H⋯H–B at the same H⋯H contact distance of 2.4 Å.⁵³ Therefore, C–H⋯H–B dihydrogen bonds are generally weaker and less sensitive to external pressure compared to N–H⋯H–B dihydrogen bonds, contributing to the enhanced structural stability of TMAB under pressure. Overall, TMAB's robust structure makes it a promising candidate for high-pressure applications and a reliable precursor for chemical synthesis.³⁶

However, the strengthening and reinforcement of the C–H⋯H–B dihydrogen bonds collectively as pressure increases still contribute significantly to the late phase transition of TMAB. At 8 GPa, in particular, the number of types of dihydrogen bonds doubled from two to four (Fig. 10(c)), thus giving rise to the changes in orientation and rotational dynamics of CH and BH groups, leading to the phase transition. This transition is further driven by a decrease in the electronic band gap by 0.11 eV at 9 GPa. As pressures increase up to 35 GPa, this trend of reducing band gap continues, predicting enhanced electronic conductivity at even higher pressures. This consistent narrowing of the band gap, coupled with the stable number of dihydrogen bond types above 8 GPa, strongly suggests that the two phases can coexist and transition smoothly into more conductive states under extreme pressure.

4. Conclusions

In summary, we report the structural stability and pressure-induced polymorphic transitions of TMAB using in situ vibrational spectroscopy and synchrotron X-ray diffraction, supported by DFT calculations. Our experimental and computational evidence reveals that TMAB undergoes a phase transition from the R3m ambient-pressure structure (phase I) to a P3₁ structure (phase II) at around 9 GPa. The identification of the phase transition is substantiated by the IR and Raman spectra, structural refinement with XRD patterns, and relative enthalpies of phase I and II as predicted by DFT calculations. Phase I and phase II may coexist at pressures above 9 GPa, as evidenced by minimal volume changes during phase transitions, similar enthalpies for both phases in the entire pressure region, and the constant number of dihydrogen bonds types after the phase transition. In contrast to other N-methylamine-borane adducts such as MAB and DMAB, TMAB exhibits an unprecedented structural stability due to the absence of the relatively strong N–H⋯H–B dihydrogen bonds. Additionally, the investigation of the C–H⋯H–B dihydrogen bond network evolution of TMAB upon compression clarifies the nature of the polymorphic transition at higher pressure. The calculated band gap energy shows a drop during the phase transition, indicating a trend towards metallization in TMAB under high pressures. These findings highlight the unprecedented stability of TMAB that allows for further investigation into its dehydrogenation mechanisms across a broad pressure–temperature range, and provide new insights into the behavior of TMAB under extreme conditions, which is crucial for its advanced applications in hydrogen storage and as a reliable precursor.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by a Discovery Grant, a Research Tools and Instruments Grant from the Natural Science and Engineering Research Council of Canada, a Leaders Opportunity Fund from the Canadian Foundation for Innovation. The synchrotron experiment described in this paper was performed at the Canadian Light Source, a national research facility of the University of Saskatchewan, which is supported by the Canada Foundation for Innovation (CFI), the Natural Sciences and Engineering Research Council (NSERC), the Canadian Institutes of Health Research (CIHR), the Government of Saskatchewan, and the University of Saskatchewan. Technical assistance from Dr Graham King is greatly appreciated. The calculations were partially supported by the facilities of the Shared Hierarchical Academic Research Computing Network (SHARCNET: https://www.sharcnet.ca) and Compute/Calcul Canada.

References

1. C. Tarhan and M. A. Çil, J. Energy Storage, 2021, 40, 102676 CrossRef .
2. R. P. Prachi, M. W. Mahesh and C. G. Aneesh, Adv. Energy Power, 2016, 4, 11–22 Search PubMed .
3. R. Kumar, A. Karkamkar, M. Bowden and T. Autrey, Chem. Soc. Rev., 2019, 48, 5350–5380 RSC .
4. B. Peng and J. Chen, Energy Environ. Sci., 2008, 1, 479–483 CAS .
5. U. B. Demirci, Int. J. Hydrogen Energy, 2017, 42, 9978–10013 CrossRef CAS .
6. R. H. Crabtree, Science, 1998, 282, 2000–2001 CrossRef CAS .
7. W. T. Klooster, T. F. Koetzle, P. E. M. Siegbahn, T. B. Richardson and R. H. Crabtree, J. Am. Chem. Soc., 1999, 121, 6337–6343 CrossRef CAS .
8. Y. Yao, X. Yong, J. S. Tse and M. J. Greschner, J. Phys. Chem. C, 2014, 118, 29591–29598 CrossRef CAS .
9. Y. Lin, W. L. Mao and H. Mao, Proc. Natl. Acad. Sci. U. S. A., 2009, 106, 8113–8116 CrossRef CAS PubMed .
10. S. Xie, Y. Song and Z. Liu, Can. J. Chem., 2009, 87, 1235–1247 CrossRef CAS .
11. A. Liu and Y. Song, J. Phys. Chem. C, 2012, 116, 2123–2131 CrossRef CAS .
12. Y. Liang and J. S. Tse, J. Phys. Chem. C, 2012, 116, 2146–2152 CrossRef CAS .
13. Y. Lin, H. Ma, C. W. Matthews, B. Kolb, S. Sinogeikin, T. Thonhauser and W. L. Mao, J. Phys. Chem. C, 2012, 116, 2172–2178 CrossRef CAS .
14. L. T. Shi, C. E. Hu, A. Muñoz, L. X. Ji, Y. Y. Huang, Y. K. Wei, G. F. Ji and Y. Cheng, Int. J. Hydrogen Energy, 2020, 45, 33047–33058 CrossRef CAS .
15. H. Liu, C. Liu, Q. Li, Y. Ma and C. Chen, J. Phys. Chem. Lett., 2021, 12, 2036–2043 CrossRef CAS PubMed .
16. S. Nakano, A. Sano-Furukawa, T. Hattori, S. Machida, K. Komatsu, H. Fujihisa, H. Yamawaki, Y. Gotoh and T. Kikegawa, Inorg. Chem., 2021, 60, 3065–3073 CrossRef CAS PubMed .
17. S. Nakano, F. Hiroshi, Y. Hiroshi and T. Kikegawa, J. Chem. Phys., 2022, 157, 234702 CrossRef CAS PubMed .
18. L. T. Shi, R. Turnbull, A. Liang, X. R. Chen and G. F. Ji, J. Mater. Chem. C, 2022, 10, 3081–3088 RSC .
19. U. B. Demirci, Energies, 2020, 13, 3071 CrossRef CAS .
20. A. Staubitz, A. P. M. Robertson and I. Manners, Chem. Rev., 2010, 110, 4079–4124 CrossRef CAS PubMed .
21. Y. Song, Phys. Chem. Chem. Phys., 2013, 15, 14524–14547 RSC .
22. L. Li, Q. Gu, Z. Tang, X. Chen, Y. Tan, Q. Li and X. Yu, J. Mater. Chem. A, 2013, 1, 12263–12269 RSC .
23. R. Moury and U. B. Demirci, Energies, 2015, 8, 3118–3141 CrossRef CAS .
24. A. Torabi, C. Murli, Y. Song and V. N. Staroverov, Sci. Rep., 2015, 5, 13929 CrossRef PubMed .
25. C. A. Castilla-Martinez, R. Moury and U. B. Demirci, Int. J. Hydrogen Energy, 2020, 45, 30731–30755 CrossRef CAS .
26. X. Liu, Y. Wu, S. Wang, Z. Li, X. Guo, J. Ye and L. Jiang, J. Mater. Sci., 2020, 55, 2645–2660 CrossRef CAS .
27. R. Guan, P. Wang, Y. Song and V. N. Staroverov, J. Phys. Chem. C, 2021, 125, 18614–18622 CrossRef CAS .
28. R. Guan, P. Wang, Y. Ji, Y. Li and Y. Song, Phys. Chem. Chem. Phys., 2023, 25, 21860–21867 RSC .
29. S. Caliskan, M. Zahmakiran, F. Durap and S. Özkar, Dalton Trans., 2012, 41, 4976–4984 RSC .
30. L. J. Sewell, M. A. Huertos, M. E. Dickinson, A. S. Weller and G. C. Lloyd-Jones, Inorg. Chem., 2013, 52, 4509–4516 CrossRef CAS PubMed .
31. N. Cao, J. Su, W. Luo and G. Cheng, Int. J. Hydrogen Energy, 2014, 39, 426–435 CrossRef CAS .
32. D. Kundu, G. Pugazhenthi and T. Banerjee, Energy Fuels, 2020, 34, 13167–13178 CrossRef CAS .
33. Y. Karatas, H. Acidereli, M. Gulcan and F. Sen, Sci. Rep., 2020, 10, 7149 CrossRef CAS PubMed .
34. P. A. Szilagyi, S. Hunter, C. A. Morrison, C. C. Tang and C. R. Pulham, J. Alloys Compd., 2017, 722, 953–961 CrossRef CAS .
35. T. Kida, K. Shigezumi, M. A. Mannan, M. Akiyama, Y. Baba and M. Nagano, Vacuum, 2009, 83, 1143–1146 CrossRef CAS .
36. R. Y. Tay, H. Li, S. H. Tsang, M. Zhu, M. Loeblein, L. Jing, F. N. Leong and E. H. T. Teo, Chem. Mater., 2016, 28, 2180–2190 CrossRef CAS .
37. C. Prescher and V. B. Prakapenka, High Press. Res., 2015, 35, 223–230 CrossRef CAS .
38. B. H. Toby and R. B. Von Dreele, J. Appl. Crystallogr., 2013, 46, 544–549 CrossRef CAS .
39. A. R. Oganov and C. W. Glass, J. Chem. Phys., 2006, 124, 244704 CrossRef PubMed .
40. A. R. Oganov, A. O. Lyakhov and M. Valle, Acc. Chem. Res., 2011, 44, 227–237 CrossRef CAS PubMed .
41. S. J. Clark, M. D. Segall, C. J. Pickard, P. J. Hasnip, M. I. Probert, K. Refson and M. C. Payne, Z. Kristallogr. – Cryst. Mater., 2005, 220, 567–570 CrossRef CAS .
42. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865–3868 CrossRef CAS PubMed .
43. D. Vanderbilt, Phys. Rev. B, 1990, 41, 7892–7895 CrossRef PubMed .
44. H. J. Monkhorst and J. D. Pack, Phys. Rev. B, 1976, 13, 5188–5192 CrossRef .
45. G. Kresse and J. Hafner, Phys. Rev. B, 1993, 47, 558–561 CrossRef CAS PubMed .
46. G. Kresse and J. Furthmüller, Phys. Rev. B, 1996, 54, 11169–11185 CrossRef CAS PubMed .
47. G. Kresse and J. Furthmüller, Comput. Mater. Sci., 1996, 6, 15–50 CrossRef CAS .
48. P. E. Blöchl, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 50, 17953–17979 CrossRef PubMed .
49. V. Wang, N. Xu, J. C. Liu, G. Tang and W. T. Geng, Comput. Phys. Commun., 2021, 267, 108033 CrossRef CAS .
50. J. D. Odom, J. A. Barnes, B. A. Hudgens and J. R. Durig, J. Phys. Chem., 1974, 78, 1503–1509 CrossRef CAS .
51. S. Aldridge, A. J. Downs, C. Y. Tang, S. Parsons, M. C. Clarke, R. D. L. Johnstone, H. E. Robertson, D. W. H. Rankin and D. A. Wann, J. Am. Chem. Soc., 2009, 131, 2231–2243 CrossRef CAS PubMed .
52. F. Birch, Phys. Rev., 1947, 71, 809–824 CrossRef CAS .
53. P. C. Singh and G. N. Patwari, Chem. Phys. Lett., 2006, 419, 265–268 CrossRef CAS .
54. C. P. Singh and G. N. Patwari, Chem. Phys. Lett., 2006, 419, 5–9 CrossRef .
55. G. Qi, K. Wang, G. Xiao and B. Zou, Sci. China: Chem., 2018, 61, 276–280 CrossRef CAS .

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4tc02285h

This journal is © The Royal Society of Chemistry 2024