Boron monoxide is a one-dimensional polymer

Abstract

It was recently reported that boron monoxide (BO) is formed through the cross-linking of B₄O₂ structural building units. Multiple theoretical phases agree with this description. Using pycnometry, multidimensional ¹⁷O NMR spectroscopy, and plane-wave DFT calculations we determined the likely polymorph to be a one-dimensional polymer initially proposed in 1955.

---

Among the binary oxides, boron monoxide (BO) is perhaps unique in having evaded attempts to determine its structure for nearly a century. The material was initially reported in 1940,^1,2 however, the first synthesis for the preparation of a single-phase BO material was described in 1955 by Wartik and Apple.³ Their synthesis employed the condensation of tetrahydroxydiboron (B₂(OH)₄, Fig. 1) at high temperature and while it was soon discovered that the boron–boron bonds were preserved in the resulting material,^4,5 the long-range structure was never determined. Several potential model structures were reported,^6–8 most exhaustively by Claeyssens et al.,⁹ however, none were supported by experimental evidence.

Fig. 1 The synthesis of BO involves the condensation of B₂(OH)₄ to form structural building units composed of a six-membered B₄O₂ rings that then interlink into a polymeric material. The structural building units can organize into either one-dimensional (A), two-dimensional (B) and (C), or three-dimensional (D) and (E) structures. Three-dimensional structures (D) and (E) feature twisting of the B₂O₄ moieties, which is inconsistent with the observation of symmetry-amplified J splittings. Darker shades indicate moieties in the foreground.

Recently, some of us applied multidimensional ¹¹B NMR spectroscopy to revisit this structural conundrum.¹⁰ Key observations included (1) the detection of a single crystallographically unique boron site; (2) observation of symmetry-amplified ¹¹B–¹¹B J couplings;^11–13 (3) observation of the collinearity of closest B–B bonds from ¹¹B–¹¹B–¹¹B triple-quantum correlations; and (4) observation of diffraction signals that suggested the formation of a layered structure. The work strongly suggested that the main structural building unit was a B₄O₂ ring with local D_2h symmetry.¹⁰ This agrees with the isolation of a B₄O₂(OH)₄ intermediate by Carmalt et al.¹⁴ Owing to the observed diffraction patterns we proposed that the material was two-dimensional, which could have important practical implications;^15–21 however, a great deal of uncertainty on the structure remained. B₄O₂ rings may cross-link in a number of patterns, as evident from the structures compiled by Claeyssens and listed in Fig. 1.⁹ We thus sought to apply ¹⁷O NMR spectroscopy to further narrow the possible structures for BO. While most materials are indistinguishable from ¹¹B NMR (with the exception of structure D and E, which do not have local D_2h symmetry), structure A features a single unique ¹⁷O site while the others feature intraring and interring oxygens that may be distinguishable. All five models further differ in their B–B–O–B dihedral angles that could be probed using ¹⁷O{¹¹B} heteronuclear correlation.

To this aim, we enriched B₂(OH)₄ with ¹⁷O to prepare an ¹⁷O-enriched BO material. Enrichment of tetrahydroxydiboron by direct ¹⁷O exchange with water proved ineffective due to its poor solubility and slow decomposition to form boric acid, a non-isolable impurity. We therefore instead produced tetramethoxydiboron as previously described,²² which that then hydrolyzed with an excess of ¹⁷O-enriched water (39.9%) to yield pure B₂(¹⁷OH)₄. From liquid-state ¹⁷O NMR, we estimate the enrichment level to be ∼3.9% (Fig. S5). Synthesis of B¹⁷O was then carried out in the usual fashion by the self-condensation of B₂(¹⁷OH)₄ at 200 °C, see SI for further details.

We acquired ¹¹B and ¹⁷O multiple-quantum (MQ) magic-angle spinning (MAS) NMR spectra of the resulting material (Fig. 2).^23–25 As expected,¹⁰ we observed a single ¹¹B resonance with an isotropic (δ_iso) chemical shift of 35.1 ppm, a quadrupolar coupling constant (C_Q) of 3.5 MHz, and a quadrupolar asymmetry parameter (η) of 0.5. The ¹⁷O MQMAS spectrum was equally simple, being described with a single site (δ_iso = 178 ppm; C_Q = 3.8 MHz; η = 0.7, see Table 1). These parameters were obtained by fitting the MAS lineshape and the MQMAS shift simultaneously using dmfit.²⁶

Fig. 2 ¹¹B (a) and ¹⁷O (b) 2D MQMAS NMR spectra. There is one well-defined site present for both ¹¹B and ¹⁷O, suggesting a structure containing only a single crystallographically unique site for each element. Sum projections (blue) are overlaid with a fit to a single site with the listed chemical shift and EFG tensor parameters (red).

Table 1 Experimentally-determined and computed BO NMR parameters and densities

	ρ/g cm^−3	Site	δ iso/ppm	|C Q|/MHz	η
a Calculation performed on the Immm polymorph.^9
Expt.	2.08 ± 0.13	B	35 ± 1	3.5 ± 0.1	0.5 ± 0.2
		O	178 ± 2	3.8 ± 0.1	0.7 ± 0.1
A	2.05	B	33.8	3.90	0.32
		O	203.9	4.73	0.52
B	1.53	B	28.2	3.86	0.44
		Ointra	223.1	5.16	0.35
		Ointer	149.8	6.87	0.51
C	1.44	B	25.5	3.65	0.50
		B	31.0	3.95	0.38
		Ointra	217.0	4.84	0.50
		Ointra	219.4	4.48	0.66
		Ointer	161.3	5.03	0.81
D	0.84	B	31.7	3.97	0.38
		B	25.5	5.66	0.95
		Ointra	215.9	4.48	0.70
		Ointra	205.4	4.31	0.40
		Ointer	163.2	4.79	0.30
		Ointer	162.2	4.84	0.36
E	1.35	B	30.9	4.28	0.64
		B	27.7	6.18	0.94
		Ointra	213.6	4.60	0.63
		Ointra	218.8	4.67	0.26
		Ointer	153.6	4.81	0.95

---

Immediately, these results strongly suggest the formation of structure A (see Fig. 1), which uniquely features only one type of chemically distinct oxygen site. We then performed plane-wave density functional theory (DFT) calculations to predict the ¹¹B and ¹⁷O chemical shifts and electric field gradient (EFG) tensor^27,28 so that they may be compared with experiment. We found that the ¹⁷O chemical shifts of sites within ring structures were highly sensitive to the method used, and as such a monomer correction²⁹ was applied using the resolution of identity second-order Møller–Plesset perturbation theory method (RI-MP2) for all calculated ¹⁷O chemical shifts. The results are listed in Table 1. There is a clear distinction between the intraring (δ_iso = 200–225 ppm) and interring oxygen species (δ_iso = 150–165 ppm) suggesting that structure B–F are inconsistent with the ¹⁷O NMR observations.

The differences between the various model are perhaps most clearly depicted by comparing experimental and computed ¹⁷O{¹¹B} dipolar heteronuclear multiple-quantum correlation (D-HMQC)³⁰ spectra (Fig. 3) that depend on the ¹¹B and ¹⁷O EFG tensors (including their relative orientations) in addition to the chemical shifts. We acquired such a spectrum using rotational-echo double-resonance (REDOR) recoupling,³¹ applied at the ¹¹B frequency, and ¹⁷O detection.^32,33 Owing to the strong ¹¹B–¹¹B homonuclear dipolar interactions and the low ¹⁷O concentration, ¹¹B-detection was not feasible. DFT-predicted ¹⁷O{¹¹B} correlation spectra were calculated using SIMPSON^34,35 with the parameters listed in Table 1, Euler angles calculated using MagresView ver. 1.6.2 (Table S2),³⁶ and assuming a 100% efficient magnetization transfer. Clearly only structure A yields a correlation spectrum in close agreement with that measured experimentally, with the others having clearly defined intraring and interring correlations. Three-dimensional structures D and E feature ring distortions that disagree with prior J coupling measurements.^10,11 These distortions further dramatically increase the magnitude of the ¹¹B quadrupolar interactions beyond that observed experimentally. Comparisons between the 1D Hahn echo NMR spectra and the five models are also shown in Fig. S2.

Fig. 3 Experimental (a) and calculated (b)–(g) ¹⁷O{¹¹B} D-HMQC 2D correlation spectra. Simulations were performed using the periodic DFT-calculated chemical shifts and EFG tensor parameters for the structures indicated on the Figure. For the simulation shown in (c), the Euler angles from the DFT calculations were used together with the experimentally-determined EFG tensor parameters and chemical shifts.

To further confirm that structure A is the correct polymorph for the BO materials produced by the condensation of B₂(OH)₄, we performed pycnometry measurements to determine the density of the material. As described by Claeyssens,⁹ and as listed in Table 1, the polymorphs differ greatly in their densities, with structure A being the densest at a predicted 2.05 g cm⁻³. We measured a density of 2.08 ± 0.13, which is within error of the predicted value for A, and far denser than structures B–E.

It thus seems that structure A, a one-dimensional polymer, is indeed the correct model for BO, however, this does not explain the peculiar X-ray reflections that pointed to a two-dimensional material. Claeyssens predicted four different packing arrangements for the polymers in A,⁹ which lead to very different diffraction patterns (Fig. 4), none of which agree with the experimental result. If we randomize the stacking arrangement in a 1 × 1 × 10 P1 supercell,¹⁰ however, we do reproduce a similar powder pattern as that predicted for structure B. There are still differences with the experiment, however, and a broader study of the stacking faults may be required to fully understand the diffraction pattern.

Fig. 4 Comparison of the experimentally-measured PXRD pattern to those predicted using different polymorphs of model A, as indicated on the Figure, including turbostratic models of A and B. None of the predicted patterns agree with the experimental measurement, suggesting that some degree of random stacking is common.

To conclude, density measurements and ¹⁷O{¹¹B} solid-state NMR experiments were used to narrow the potential structural candidates for boron monoxide (BO), which was earlier shown to be made of interconnected B₄O₂ rings. We discovered that the material was, in fact, not a two-dimensional nanomaterial but instead was composed of one-dimensional polymers composed of fused B₄O₂ rings. Interestingly, this exact model was initially suggested by Wartick and Apple in 1955 when the material was first prepared.³ We were nevertheless unable to narrow the possibilities for a space group, with the material seeming to lack long-range order. Learning about the long-range order of the material will require methods that are sensitive to such length scales, such as scanning tunneling microscopy.³⁷

Prof. Aaron J. Rossini is thanked for many fruitful discussions. Qasim Q. Kashif and Alex Kopeny are thanked for help with the DFT analysis. This work was supported by the US. Department of Energy, Office of Science, Basic Energy Sciences, Materials Science and Engineering Division. The Ames National Laboratory is operated for the U.S. DOE by Iowa State University under contract no. DE-AC02-07CH11358.

Conflicts of interest

There are no conflicts to declare.

Data availability

Raw data for this study are available at https://doi.org/10.5281/zenodo.16884299.

Supplementary information (SI): synthesis details, solution-phase NMR data, differential scanning calorimetry, ¹¹B J-resolved NMR data, thermogravimetric analysis, infrared spectroscopy, and further calculation details. See DOI: https://doi.org/10.1039/d5cc04723d.

References

1. V. E. Zintl, W. Morawietz and E. Gastinger, Z. Anorg. Allg. Chem., 1940, 245, 8–11.
2. F. A. Kanda, A. J. King, V. A. Russell and W. Katz, J. Am. Chem. Soc., 1956, 78, 1509–1510.
3. T. Wartik and E. F. Apple, J. Am. Chem. Soc., 1955, 77, 6400–6401.
4. A. L. McCloskey, J. L. Boone and R. J. Brotherton, J. Am. Chem. Soc., 1961, 83, 1766–1767.
5. A. K. Holliday and A. G. Massey, Chem. Rev., 1962, 62, 303–318.
6. D.-Z. Li, H. Bai, Q. Chen, H. Lu, H.-J. Zhai and S.-D. Li, J. Chem. Phys., 2013, 138, 244304.
7. Z. Zhang, L. Pu, Q.-S. Li and R. B. King, Inorg. Chem., 2015, 54, 2910–2915.
8. Y. Liu, C. Liu, Z. Zhang and R. B. King, Chem. Commun., 2017, 53, 3239–3241.
9. F. Claeyssens, N. L. Allan, N. C. Norman and C. A. Russell, Phys. Rev. B: Condens. Matter Mater. Phys., 2010, 82, 094119.
10. F. A. Perras, H. Thomas, P. Heintz, R. Behera, J. Yu, G. Viswanathan, D. Jing, S. A. Southern, K. Kovnir, L. Stanley and W. Huang, J. Am. Chem. Soc., 2023, 145, 14660–14669.
11. F. A. Perras and D. L. Bryce, J. Am. Chem. Soc., 2013, 135, 12596–12599.
12. F. A. Perras and D. L. Bryce, J. Magn. Reson., 2014, 242, 23–32.
13. F. A. Perras and D. L. Bryce, Chem. Sci., 2014, 5, 2428–2437.
14. C. J. Carmalt, W. Clegg, A. H. Cowley, F. J. Lawlor, T. B. Marder, N. C. Norman, C. R. Rice, O. J. Sandoval and A. J. Scott, Polyhedron, 1997, 16, 2325–2328.
15. H. Yamauchi, A. Asano and S. Hayashi, Heterocycles, 2022, 104, 979–986.
16. R. Rahimi and M. Solimannejad, J. Mol. Liq., 2022, 354, 118855.
17. B. Mortazavi, F. Shojaei, F. Ding and X. Zhuang, FlatChem, 2023, 42, 100575.
18. W. Othman, W. Alfalasi, T. Hussain and N. Tit, J. Energy Storage, 2024, 98A, 113014.
19. M. M. Kadhim, N. Sadoon, H. A. Gheni, S. K. Hachim, A. Majdi, S. A. H. Abdullaha and A. M. Rheima, Comput. Theor. Chem., 2023, 1219, 113941.
20. M. Sotudeh, Z. Rastipour, F. Shojaei, A. Mohajeri and H. S. Kang, ACS Appl. Electron. Mater., 2025, 7, 2696–2708.
21. R. Mardanian, A. Kokab and S. B. Touski, Phys. Scr., 2025, 100, 065409.
22. R. Fornwald, A. Yadav, J. Montero Bastidas, M. SmithIII and R. MaleczkaJr, J. Org. Chem., 2024, 89, 6048–6052.
23. L. Frydman and J. S. Harwood, J. Am. Chem. Soc., 1995, 117, 5367–5368.
24. J.-P. Amoureux, C. Fernandez and S. Steuernagel, J. Magn. Reson., Ser. A, 1996, 123, 116–118.
25. Z. Gan and H.-T. Kwak, J. Magn. Reson., 2004, 168, 346–351.
26. D. Massiot, F. Fayon, M. Capron, I. King, S. Le Calvé, B. Alonso, J.-O. Durand, B. Bujoli, Z. Gan and G. Hoatson, Magn. Reson. Chem., 2002, 40, 70–76.
27. C. J. Pickard and F. Mauri, Phys. Rev. B: Condens. Matter Mater. Phys., 2001, 63, 245101.
28. M. Profeta, F. Mauri and C. J. Pickard, J. Am. Chem. Soc., 2003, 125, 541–548.
29. J. Hartman and J. K. Harper, Solid State Nucl. Magn. Reson., 2022, 122, 101832.
30. Z. Gan, J. Magn. Reson., 2007, 184, 39–43.
31. T. Gullion and J. Schaefer, J. Magn. Reson., 1989, 81, 196–200.
32. R. W. Dorn, L. O. Mark, I. Hung, M. C. Cendejas, Y. Xu, P. L. Gor’kov, W. Mao, F. Ibrahim, Z. Gan, I. Hermans and A. J. Rossini, J. Am. Chem. Soc., 2022, 144, 18766–18771.
33. R. W. Dorn, A. L. Paterson, I. Hung, P. L. Gor’kov, A. J. Thompson, A. D. Sadow, Z. Gan and A. J. Rossini, J. Phys. Chem. C, 2022, 126, 11652–11666.
34. M. Bak, J. T. Rasmussen and N. C. Nielsen, J. Magn. Reson., 2000, 147, 296–330.
35. Z. Tošner, R. Andersen, B. Stevensson, M. Edén, N. C. Nielsen and T. Vosegaard, J. Magn. Reson., 2014, 246, 79–93.
36. S. Sturniolo, T. F. G. Green, R. M. Hanson, M. Zilka, K. Refson, P. Hodgkinson, S. P. Brown and J. R. Yates, Solid State Nucl. Magn. Reson., 2016, 78, 64–70.
37. D. Cui, J. M. Macleod, M. Ebrahimi and F. Rosei, CrystEngComm, 2017, 19, 4927–4932.

---