¹¹B NMR Together with Infrared Spectroscopy Provides Insight into Structural Elucidation of Quadrivalent Diazaborines & Cyclic Boronate Esters: Intriguing & Little-Explored

Abstract

Imidazo-fused diazaborines, which serve as intermediary structures somewhat alongside benzene and borazine, had been of particular interest to Dewar and Snyder more than 60 years ago. To this end, Dewar utilised his ‘$\pi$-complex theory’so as to represent ‘borazaros’as a ‘quadrivalent’ species; however, sadly, modern representations have deviated and leapt into ‘trivalent’ counterparts. Bonding in boron species has never been straightforward, to such an extent that the orthodox ‘ethane’ like diborane, i.e., H₃B–BH₃, which conformed to the paradigmatic rules of molecular structure, in particular, hybridisation and electronegativity, was later evolved to a more realistic ‘3-centre 2-electron’ bonding so as to give the lie to the purported diborane structures of X-ray diffractors. Herein ¹¹B NMR together with IR spectroscopy sheds light on the nature of bonding in borazaros, and ‘caged’ cyclic oxazaborons so as to reinforce, and reinvigorate the old literature, which could be of interest to both the synthetic, and medicinal chemist alike.

1. Introduction

Unlike ¹H and ¹³C with a ‘nuclear spin’ I = ½, both naturally occurring isotopes of boron, i.e., ¹⁰B (I = 3) and ¹¹B (I = 5/2), are NMR active; however, the latter is more versatile than the former for NMR investigations by virtue of the fact that ¹⁰B has a natural abundance of only 19.6%, whereas ¹¹B exhibits a natural abundance of 80.4% [1]; as such, ¹⁰B comes with a much lower receptivity, i.e., receiver gain, and a quadrupole moment almost thrice as much as that for ¹¹B, which broadens the ¹⁰B NMR signals. Moreover, the Larmor frequency of ¹⁰B is almost three times lower than that for ¹¹B, resulting in a much lower signal resolution, and coupling constants, i.e., J (in Hz) in ¹⁰B NMR spectra [2]. Despite these drawbacks, ¹⁰B NMR can be particularly useful in niche mechanistic studies such as rearrangements [3]. NMR spectroscopy, much to contemporary efforts, has proved exceptionally advantageous in elucidating structures of boron compounds, and their behaviour in solutions [4,5]. The extrapolation of NMR activities of other common NMR active nuclei, such as ¹H, ¹³C, ¹⁵N, ³¹P, etc., to ¹¹B NMR suggests that the following factors should determine chemical shifts ($\delta$) in ¹¹B NMR spectra:

• Electron density;
• Coordination number;
• Hybridisation;
• Ring current.

However, the above is found not to be true for many ¹¹B NMR spectra. For example, the signals from B¹ and B² in the ¹¹B NMR spectrum of n-B₉H₁₅ are ≈64.7 ppm apart, such that B² resonates at the highest field and B¹ at the lowest field in the aforementioned ¹¹B NMR spectrum [6], in spite of both B¹ and B² in n-B₉H₁₅ exhibiting identical connectivity, hybridisation, and even immediate bonding environment (Figure 1).

Similar, but different to the ¹¹B NMR spectrum of n-B₉H₁₅ Fan et al. [7] have very recently discovered that the ¹¹B{¹H} NMR signals from the boron atoms in B₃H₇ coordinated 4-methylene dihydropyridine and ethyl 4-isopropyl 3-carboxylate dihydropyridine are reversed compared to that of their parent complexes: the top boron atom connected to the pyridinyl nitrogen is downfielded, and the two base boron atoms resonate at a much higher field in the aforementioned ¹¹B{¹H} NMR spectra, in marked contrast to the general trend observed for most B₃H₇ adducts [7].

On the other hand, in stark contrast to other NMR active nuclei, electron density does not appear to preponderate over other factors in determining the position of chemical shifts in ¹¹B NMR spectra, as the boron atoms of higher electron density in B₇${\mathrm{H}}_7^{2-}$ are less shielded than their counterparts of lower electron density in the same cluster [8]. Intriguingly, Ksenofontov et al. [9] have attempted to devise an accurate machine learning method for predicting the ¹¹B NMR chemical shifts of BODIPYs; however, it has been borne out that the peripherally halogenated BODIPYs are incongruent to an extent that nearly 41% of the aforementioned compounds in their dataset were outliers, and that this should be so is a corollary of, inter alia, caveats of the model for not factoring in the effects of spin–orbit interactions. To this end, there are no hard and fast rules as yet to fully predict or explain some of the anomalies that have been observed for the ¹¹B NMR spectra, and each and every single boron compound can give rise to distinctive and unexpected ¹¹B NMR spectra, which have been observed for the novel boronate esters herein and form the basis of the discussions below.

2. Results and Discussion

Recent protein-directed dynamic combinatorial chemistry studies of L-cysteine derivatives with 2-formylphenylboronic acid (2-FPBA) on the surface, i.e., in crystallo of oxacillinases, e.g., OXA-10 (unpublished), brought home an intriguing quadrivalent cyclic boronate ester, i.e., 12 (Figure 2), given that the same observed ‘tetracyclic caged’ structure for 12 has been reported in solution elsewhere [10], it is conceivable that based on the observed spectroscopic and X-ray crystallographic data, boronate esters 11 and 13–18 (Figure 3) should also follow suit.

Inasmuch as both the diazaborines and boronate esters herein presumably almost certainly feature ‘quadrivalent’ boron species, a comparison of chemical shifts (vide infra) in ¹¹B NMR spectra of the thus obtained diazaborines and boronate esters (Figure 3) not only provides structural insight into these intriguing scaffolds, but also paves the way for a better understanding of bonding in such boron species, and salient factors in their ¹¹B NMR resonance frequencies.

Reference has already been made to the unpredictability and irregularity of trends in ¹¹B NMR resonance frequencies (vide supra); however, typically, though it need not of necessity, the vacant 2p_z orbital of boron plays a crucial role in the position of chemical shifts in ¹¹B NMR spectra. Strictly speaking, trivalent boron species resonate at much lower fields than their quadrivalent counterparts. For instance, trimethyl borane as a true trivalent boron species resonates at the lowest field, whereas trimethyl borate as a somewhat quadrivalent boron species with considerable π-bonding character resonates at much higher fields in ¹¹B NMR. Equally, quite recently Li et al. [11] have also observed that increasing the number of SP³ hybridised boron atoms in aqueous solutions of borax shifts the corresponding chemical shifts to a higher field in their observed ¹¹B NMR spectra. The belief that the determinant factor in the position of chemical shifts in ¹¹B NMR spectra is the π-donating ability of substituents on boron, implies that diethylamino substituents on boron should bring about upfield shifts for the ¹¹B signals in their NMR spectra, relative to that from –OMe by virtue of the fact that –NEt₂ is mesmerically more π-donating than –OMe. However, this effect appears to be reversed in aliphatic boron species, as such B(NEt₂)₃ resonates at a much lower field in ¹¹B NMR than that observed for B(OMe)₃ [12].

The chemical shift ${\delta }_R^C$ (in ppm) between a compound C and reference R is given by the following equation:

$${\delta }_R^C={10}^6(H_C-H_R)/H_R\approx {10}^6({\wp }_C-{\wp }_R)$$

(1)

where H_C and H_R are the resonance frequencies of the applied magnetic field for compounds C and R, respectively, and ${\wp }_C$ and ${\wp }_R$ are the resulting shielding tensors. Ramsey [13,14], devised an expression for calculating the NMR shielding tensor ℘; as such, for most solution NMR applications, the shielding tensor ℘ can be replaced by a scalar quantity $\sigma$ to give the following equation:

$$\sigma ={\displaystyle \frac{e^2}{2m{c}^2}}\langle 0\mid \sum _j(x_j^2+y_j^2)/r_j^3\mid 0\rangle$$

(2)

Neglecting the deferential overlaps of the above expression can lead to the following simplified expression for the chemical shift ${\delta }_i$ of the nucleus of atom i [15]:

$${\delta }_i=A{q}_i+\sum _{(j\ne i)}B{q}_j/r_{ij}^3$$

(3)

where q_i is the $\pi$-electron density of atom i, r_ij is the internuclear distance between atom i and j, and A and B are constants. To this end, another salient factor other than the $\pi$-donating capabilities of the substituents attached to boron in ¹¹B NMR would be the ‘electronic cloud’ of the substituents attached to boron, i.e., the availability of lone pairs above and below the plane of the $\pi$-bond [15]. As exemplified by the last equation above (Equation (3), atoms j with lone pairs orthogonal to the plane of the covalent bond would have a significant impact on the chemical shift by virtue of a large q_j and small r_ij (

Table 1. ¹H, ¹³C, ¹¹B and IR Data for 2-FPBA, Diazaborines & Boronate Esters.

Entry	11B ∞	(CHO or $\mathit{\gamma }$-CH or $\mathit{\delta }$-CH) ¶	(CHO or $\mathit{\gamma }$-C or $\mathit{\delta }$-C) ♣	IR: C=O ♠
2-FPBA	29.9	10.05	197.0
1	29.8	8.08	142.7	1673s
2	29.8		146.9	1687s
3	27.7	7.96	140.4
4	16.2 •	8.35 •	175.5 •	1682s
5	28.3			1683s
6	28.6	8.20	140.0	1698s
7	28.4		142.4	1699s
8	28.8	8.15	145.0	1700s
9	29.9	8.13	144.0
10	30.0		147.6
11 ★	11.4	6.16	65.6	1730s
13 ★	11.3	6.25	71.2	1713s
15 ★	11.1		66.8	1715s
17 ★	11.5		76.2	1723s

^★; Boronate esters 11, 13, 15 and 17 are enantiomers of 12, 14, 16 and 18, respectively, with identical spectroscopic data. ^∞; 193 MHz, acetonitrile-$d_3$. ^¶; 600 MHz, acetonitrile-$d_3$. ^♣; 151 MHz, acetonitrile-$d_3$. ^•; benzodiazaborine 4 is much less soluble in acetonitrile-$d_3$; its NMR spectra were recorded in methanol-$d_4$ instead. ^♠; ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR), s stands for sharp.

).

Analysis of the ¹H, ¹³C and ¹¹B NMR data above (Table 1), on looking more closely, reveals that whilst the boron in 2-FPBA has a much greater trivalent character, less shielded and non-aromatic relative to that in quadrivalent diazaborines which should be more shielded, the boron atoms in the diazaborines togehter with that in 2-FPBA all resonate at about the same field in their ¹¹B NMR; on the other hand, the proton of the formyl group in 2-FPBA, resonates at a much lower field than that from $\gamma$-CH of the benzofused-imadazo ring in ¹H NMR by virtue of greater shielding that the $\gamma$-CH moieties of benzofused-imadazo rings experience, which is in marked contrast to that observed for ¹¹B NMR. The same is also true for ¹³C NMR data, the $\gamma$-CH and $\gamma$-CMe of the imidazo rings are significantly upshifted compared to the non-ring carbon of the formyl group in 2-FPBA (Table 1). Intriguingly, the solvation effect of methanol-$d_4$ on benzodiazaborine 4 (vide infra) relative to that of acetonitrile-$d_3$ for its closely related benzodiazaborine 5 has clearly been echoed by both the upshifted and downshifted signals from boron and $\gamma$-CH of benzodiazaborine 4 in its ¹¹B and ¹³C NMR spectra, respectively. The ¹¹B chemical shifts from the boronate esters, in comparison with those from the diazaborines, exhibit a whopping three-fold upfield shift in their spectra, in spite of the fact that both species are quadrivalent, and potentially negatively charged.

Whilst the aromaticity of diazaborines can play a crucial role in the observed data, as illustrated above (Equation (3)), the electronic cloud of heteroatoms with available lone pairs adjacent to the boron centre in question can account for the aforementioned large upfield shift of almost 20 ppm in their ¹¹B NMR spectra (Figure 4).

In view of the above, it is conceivable that the boronate esters are tetracyclic and ‘ring-closed’, as if they were ‘ring-open’ and tricyclic, the thus obtianed C_3h boron species would be less shielded, and would absorb at about the expected field in ¹¹B NMR, just like that of 2-FPBA with a similar chemical environment and shielding effects (Figure 5).

The effects of the electronic cloud of the adjacent substituents on boron (Figure 4) is further evident from the upshifted signal of the boron in benzodiazaborine 4 in its ¹¹B NMR spectrum in methanol-$d_4$, i.e., ${\delta }_B$ 16.2 (Figure 6), relative to that obtained in acetonitrile-$d_3$ for benzodiazaborine 5, i.e., ${\delta }_B$ 28.3 (Table 1).

Further support for the assertion that the aforementioned boronate esters (Figure 3) are ‘ring-closed’ and tetracyclic is provided by the upfield shift of their $\delta$-CH and $\delta$-CMe signals in ¹³C NMR, relative to that observed for the C=O of their precursors i.e., 2-FPBA and 2-acetylphenylboronic acid (2-APBA), which suggests that the D_3h carbon of the formyl or acyl moieties, is now a ${\mathrm{T}}_d$ centre, bonded to four substituents, and fully saturated. This arrangement brings the carboxyl moiety in the ‘ring-open’ structure in close proximity to the boronic acid so as to further cyclise, close the ring and form a ‘tetracyclic caged’ structure (Figure 7). This is in agreement with the observed IR frequencies i.e., ≈1730–1710 $\mathrm{c}{\mathrm{m}}^{-1}$ expected for esters (Table 1), as the parent carboxyl moieties should resonate at lower frequencies i.e., ≈1600–1550 $\mathrm{c}{\mathrm{m}}^{-1}$.

3. Materials and Methods

3.1. General Materials and Instruments

All reagents and solvents in the experiments were of reagent grade quality. They were obtained from commercial sources and used without further purification unless otherwise stated. Diethyl ether, herein, is referred to as ether. Furthermore, 2-FBPA, 2-APBA (purum 95.0+%) and sodium phosphate monobasic (puriss. anhydrous, >99.0%) were from Sigma-Aldrich^® Chemical Co. (Fluka, Germany). Semicarbazide hydrochloride (puriss. 99.0%) was from Alfa Aesar^® (Guilford, England). The water used was from a Millipede Milli-Q EDM Water Purification A10 unit (Merck, Germany) or was HPLC-grade. High performance liquid chromatography–mass spectrometry (LC-MS) analyses were performed using an Agilent UPLC SQD (USA) instrument equipped with a Chromolith^® (ACE, England), C₁₈, 2.1 × 50 mm column. The eluent was 0.1% formic acid ($v/v$) in water (solvent A)/acetonitrile (solvent B) over a 5–95% acetonitrile gradient, with monitoring at 214 nm over 10 min. Preparative reversed phase high performance liquid chromatography (RP-HPLC) was performed using a Shimadzu SPD-20A (Kyoto, Japan) equipped with a Waters Sunfire (USA), C₁₈, 250 × 21.2 mm column. The eluent was 0.1% TFA ($v/v$) in H₂O (solvent A) and 0.1% TFA ($v/v$) in MeCN (solvent B) over various MeCN gradients and flow rates as indicated, with monitoring at 214 nm. Electrospray ionisation mass spectrometry was performed using an Agilent single quadrupole unit equipped with CTC-PAL. High-resolution electrospray mass spectra were obtained on a Waters LCT Premier XE spectrometer (USA). The melting points were determined using a Gallenkamp MFG 595 010 melting point apparatus (Synoptics Ltd., Cambridge, England) and a Reichert Austria hot stage microscope (Vienna, Austria). ¹H, ¹³C and ¹¹B NMR spectra were recorded on a Bruker AVIII HD 400 unit or a Bruker AVIIIHD 600 (Nasdaq, USA) equipped with prodigy N₂ broadband CryoProbe unit, as well as a Bruker AVIII 700 unit (Nasdaq, USA) equipped with TCI H/C/N helium CryoProbe operating (Nasdaq, USA) at 700, 600, 400 MHz, 193, 151, 101 MHz, 225, 193 and 128 MHz, respectively, using CDCl₃, D₂O, methanol-$d_4$ and acetonitrile-$d_3$ as solvents. Chemical shifts ($\delta$) are given in parts per million (ppm) relative to the internal tetramethylsilane (TMS). The coupling constants are given in Hz. The abbreviations s, d, t, td, dd, dq and m correspond to singlet, doublet, triplet, triplet of doublets, doublet of doublets, doublet of quartets, and multiplet, respectively. The IR spectra were recorded on a Bruker Tensor 27 Fourier transform infrared spectrometer (Nasdaq, USA); the abbreviations br, w, m, s and vs correspond to broad, weak, medium, sharp and very sharp, respectively (provisional assignments are given); ${\left[\alpha \right]}_D$ values are given in ${10}^{-1}$ deg cm² g⁻¹ at the specified temperature. The reaction products were identified by their NMR, IR spectra and/or otherwise by LC-MS analyses.

3.2. General Procedure for Synthesis of Boronic Acid and Ester Heterocycles

The benzodiazaborines, diazaborines and boronate esters were generally synthesised according to the literature procedure [16] with the following modification: the boronic acid (50 mM) and the nucleophile (50 mM) in 0.10 M PBS buffer (pH ≈ 7.4) were nutated for 2.0 min at ambient temperature. If a solid residue was precipitated, the slurry was centrifuged for 20 min at 4.0 °C and washed with water twice, followed by lyophilisation to furnish the desired semi-pure product. In cases where the product did not precipitate, it was purified using preparative RP-HPLC–0.1% TFA (v/v) in H₂O (solvent A)/ 0.1% TFA (v/v) MeCN (solvent B) over a 5–50% MeCN gradient and monitored at 124 nm over 40 min at 40 °C. Samples of the semi-pure products from the aforementioned sedimentation process were also purified using the above preparative RP-HPLC for analytical characterisation purposes.

3.3. 1-Hydroxybenzo-[d][1,2,3]-diazaborinine-2(1H)-carboxamide (1)

Benzodiazaborine 1 was synthesised from 2-FPBA and semicarbazide according to the general procedure [16]. The precipitated powder was collected and purified using RP-HPLC (retention time: 12.2–12.5 min) (5.0–50% TFA (0.1% $v/v$) in MeCN/ TFA (0.1% $v/v$) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to furnish the desired compound 1 as a white fluffy powder (76 mg; 80%); mp 197–200 °C (dec.) (lit. [17] 295–299 °C dec.); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3418w (N–H asymmetric), 3140br (N–H symmetric), 2925w (C–H), 1673s (C=O), 1297m (C–N), 1219m (C–B); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 8.17–8.12 (1 H, m, CH-Ar), 8.08 (1 H, s, $\gamma$-CH), 7.79 (2 H, m, 2 × CH-Ar), 7.70 (1 H, td, J 7.0 and 1.5, CH-Ar); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 129.0–135.8 (4 × CH-Ar and 2 × C-Ar), 142.7 ($\gamma$-CH), 164.1 (C=O); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 29.8; LC-MS (retention time: 5.6–5.9 min) $m/z$ (ESI) 190 (M⁺ + H, 100%); HRMS (ESI) calcd. for C₈H₈BN₃O₂ [M + H]⁺$m/z$ = 190.0782, found: 190.0784.

3.4. 1-Hydroxy-4-methylbenzo[d][1,2,3]-diazaborinine-2(1H)-carboxamide (2)

Benzodiazaborine 2 was synthesised from 2-APBA and semicarbazide according to the general procedure [16]. The precipitated powder was collected and purified using RP-HPLC (retention time: 12.7–13.0 min) (5.0–50% TFA (0.1% $v/v$) in MeCN/ TFA (0.1% $v/v$) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to furnish the desired compound 2 as a white fluffy powder (71 mg; 70%); mp 203 °C(dec.); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3389w (N–H asymmetric), 3205br (N–H symmetric), 1687s (C=O), 1311s (C–N), 1155m (C–B); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 8.15–8.13 (1 H, d J 12.0, CH-Ar), 7.90–7.89 (1 H, d, J 6.0, CH-Ar), 7.80–7.77 (1 H, td, J 12.0 and 6.0, CH-Ar), 7.68–7.66 (1 H, td, J 12.0 and 6.0, CH-Ar), 2.56 (3 H, s, C$H_3$); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 20.2 (CH₃), 126.7–135.8 (4 × CH-Ar and 2 × C-Ar), 146.9 ($\gamma$-CH), 164.0 (C=O); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 29.8; LC-MS (retention time: 6.0–6.4 min) $m/z$ (ESI) 204 (M⁺ + H, 100%); HRMS (ESI) calcd. for C₉H₁₀BN₃O₂ [M + H]⁺$m/z$ = 204.0939, found: 204.0941.

3.5. Benzo[d][1,2,3]-diazaborinin-1(2H)-ol (3)

Benzodiazaborine 3 was synthesised from 2-FPBA and tert-butyl carbazate according to the general procedure [16]. The clear reaction mixture was lyophilised, and the residue was purified using RP-HPLC (retention time: 9.7–10.1 min) (5.0–50% TFA (0.1% $v/v$) in MeCN/ TFA (0.1% $v/v$) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to furnish the desired compound 3 as a white powder (38 mg; 31%); mp 232–239 °C (dec.); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3323m (N–H), 1558–1440s (C=C), 1380s (B–O), 1342s (B–N); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 8.10–8.08 (1 H, d, J 8.0, CH-Ar), 7.96 (1 H, s, $\gamma$-CH), 7.71–7.70 (2 H, m, 2 × CH-Ar), 7.62–7.58 (1 H, m, CH-Ar); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 127.8–136.4 (4 × CH-Ar and 2 × C-Ar), 140.4 ($\gamma$-CH); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 27.7; LC-MS (retention time: 4.5 min) $m/z$ (ESI) 147 (M⁺ + H, 100%); HRMS (ESI) calcd. for C₇H₇BN₂O [M + H]⁺$m/z$ = 147.0724, found: 147.0721.

3.6. 8H-Benzo[4,5][1,2,3]diazaborinino[3,2-b]benzo[4,5][1,2,3]diazaborinino[2,3-e][1,3,5,2,6]oxadiazadiborinin-8-one (4)

Benzodiazaborine 4 was synthesised from 2-FPBA and carbazide according to the general procedure [16]. The clear reaction mixture was lyophilised, and the residue was purified using RP-HPLC (retention time: 10.1–10.5 min) (5.0–50% TFA (0.1% $v/v$) in MeCN/ TFA (0.1% $v/v$) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to yield the desired compound 4 as a white powder (11 mg; 7.0%); mp 221–225 °C (dec.); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 2928w (C–H), 1682s (C=O), 1209m (C–B); ${\delta }_H$ (600 MHz, methanol-$d_4$) 8.35 (2 H, s, $\gamma$-CH), 8.23 (2 H, d, J 7.5, 2 × CH-Ar), 7.83–7.75 (4 H, m, 4 × CH-Ar), 7.70 (2 H, t, J 7.5, 2 × CH-Ar); ${\delta }_C$ (151 MHz; methanol-$d_4$) 130.4–133.8 (8 × CH-Ar and 2 × C-Ar), 146.8 ($\gamma$-CH), 175.5 (C=O); ${\delta }_B$ (193 MHz, methanol-$d_4$) 16.2; LC-MS (retention time: 4.06 min) $m/z$ (ESI) 301 (M⁺ + H, 100%); HRMS (ESI) calcd. for C₁₅H₁₂B₂N₄O₃ [M + H]⁺$m/z$ = 301.1063, found: 301.1060.

3.7. 5,11-Dimethyl-8H-benzo[4,5][1,2,3]diazaborinino[3,2-b]benzo[4,5][1,2,3]diazaborinino[2,3-e][1,3,5,2,6]oxadiazadiborinin-8-one (5)

Benzodiazaborine 5 was synthesised from 2-APBA and carbazide according to the general procedure [16]. The clear reaction mixture was lyophilised, and the residue was purified using RP-HPLC (retention time: 12.2–12.5 min) (5.0–50% TFA (0.1% $v/v$) in MeCN/ TFA (0.1% $v/v$) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to furnish the desired compound 5 as a white fluffy powder (3.0 mg; 2.0%); mp 244–258 °C (dec.); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3399br (O–H), 2923w (${\mathrm{C}}_{{sp}^3}$–H), 1683s (C=O), 1210s (C–B); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 8.56–8.54 (2 H, d, J 8.0, 2 × CH-Ar), 8.09–8.07 (2 H, d, J 8.0, 2 × CH-Ar), 7.97–7.94 (2 H, dd, J 12.0 and 8.0, 2 × CH-Ar), 7.88–7.84 (2 H, dd, J 12.0 and 8.0, 2 × CH-Ar), 2.77 (6 H, s, 2 × C$H_3$); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 28.3; LC-MS (retention time: 7.16 min) $m/z$ (ESI) 329 (M⁺ + H, 100%) 346 (M⁺ + H₃O, 20); HRMS (ESI) calcd. for C₁₇H₁₄B₂N₄O₂ [M + H]⁺$m/z$ = 329.1376, found: 329.1373.

3.8. 1-Hydroxythieno[3,2-d][1,2,3]diazaborinine-2(1H)-carboxamide (6)

Diazaborine 6 was synthesised from 2-FTBA and semicarbazide according to the general procedure [16]. The clear reaction mixture was lyophilised, and the residue was purified using RP-HPLC (retention time: 17.1–18.2 min) (5.0–50% TFA (0.1% $v/v$) in MeCN/ TFA (0.1% $v/v$) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to furnish the desired compound 6 as a pale-yellow powder (60 mg; 61%); mp 273 °C (dec.); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3500–3200br (H-bonding O–H), 1698s (C=O), 1425vs (B–O), 1331s (B–N), 1287m (C–B); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 8.20 (1 H, s, $\gamma$-CH), 7.80–7.79 (1 H, d, J 5.0, CH-Ar), 7.58–7.57 (1 H, d, J 5.0, CH-Ar); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 127.3–135.3 (2 × CH-Ar and 2 × C-Ar), 140.0 ($\gamma$-CH), 164.0 (C=O); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 28.6; LC-MS (retention time: 5.8 min) $m/z$ (ESI) 196 (M⁺ + H, 100%), 373 (2 × M⁺ + H, 6.0); HRMS (ESI) calcd. for C₆H₆BN₃O₂S [M + H]⁺$m/z$ = 196.0347, found: 196.0346.

3.9. 1-Hydroxy-4-methylthieno[3,2-d][1,2,3]diazaborinine-2(1H)-carboxamide (7)

Diazaborine 7 was synthesised from 2-ATBA and semicarbazide according to the general procedure [16]. The clear reaction mixture was lyophilised, and the residue was purified using RP-HPLC (retention time: 18.3–19.0 min) (5.0–50% TFA (0.1% $v/v$) in MeCN/ TFA (0.1% $v/v$) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to furnish the desired compound 7 as a white fluffy powder (75 mg; 72%); mp 278 °C (dec.); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3448w (N–H asymmetric), 3261br (N–H symmetric), 1699s (C=O), 1421s (B–O), 1350s (B–N), 1286vs (C–B), 663vs (B–O); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 7.76–7.75 (1 H, d, J 6.0, CH-Ar), 7.57–7.56 (1 H, d, J 6.0, CH-Ar), 2.49 (3 H, s, C$H_3$); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 20.9 (CH₃), 130.3–131.3 (2 × CH-Ar and 2 × C-Ar), 142.4 ($\gamma$-C and C=O); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 28.4; LC-MS (retention time: 6.3 min) $m/z$ (ESI) 210 (M⁺ + H, 100%); HRMS (ESI) calcd. for C₇H₈BN₃O₂S [M + H]⁺$m/z$ = 210.0503, found: 210.0502.

3.10. 1-Hydroxythieno[2,3-d][1,2,3]diazaborinine-2(1H)-carboxamide (8)

Moreover, 3-Formyl-2-thienylboronic acid (0.5 mmol) in acetonitrile (0.10 M) was added to a solution of semicarbazide hydrochloride in acetonitrile (0.1 M), and the resulting mixture was stirred at reflux (90 °C) overnight. Upon completion (LC-MS controlled), the solvent was removed in vacuo, and the residue was washed with plenty of water, filtered and dried in a desiccator to furnish the desired compound 8 as pale brown needle-like crystals (42 mg; 43%); mp 314 °C (dec.); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3445w (N–H asymmetric), 3190br (N–H symmetric), 1700m (C=O), 1449m (B–O), 1396m (B–N), 1309s (C–B), 672m (B–O); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 8.15 (1 H, s, $\gamma$-CH), 7.98 (1 H, d, J 5.0, CH-Ar), 7.52–7.52 (1 H, d, J 5.0, CH-Ar); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 126.7–136.4 (2 × CH-Ar and 2 × C-Ar), 145.0 ($\gamma$-C), 163.8 (C=O); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 28.8; LC-MS (retention time: 5.9 min) $m/z$ (ESI) 196 (M⁺ + H, 100%), 373 (2 × M⁺ + H, 6.0); HRMS (ESI) calcd. for C₆H₆BN₃O₂S [M + H]⁺$m/z$ = 196.0347, found: 196.0347.

3.11. 1-Hydroxybenzo[d][1,2,3]-diazaborinine-2(1H)-carbothioamide (9)

Benzodiazaborine 9 was synthesised from 2-FPBA and thiosemicarbazide according to the general procedure [16]. The needle-like crystals were collected, washed with cold water and purified using RP-HPLC (retention time: 10.2–10.5 min) (5.0–50% TFA (0.1% $v/v$) in MeCN/ TFA (0.1% $v/v$) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to furnish the desired compound 9 as a white fluffy powder (27 mg; 26%); mp 187 °C (dec.); ${\nu }_{max}$/$/\mathrm{c}\mathrm{m}$ (ATR) 3398m (N–H asymmetric), 3263 (N–H symmetric), 3040w (${\mathrm{C}}_{{sp}^3}$–H), 1586s (C=S), 1297m (C–N), 1255m (C–B); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 8.15–8.12 (1 H, d, J 2.0, $\gamma$-CH), 7.79–7.72 (1 H, m, CH-Ar), 7.60–7.56 (1 H, m, CH-Ar), 7.40–7.36 (2 H, m, 2 × CH-Ar); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 128.6–136.6 (4 × CH-Ar and 2 × C-Ar), 142.4 and 145.9 ($\gamma$-CH), 178.8 (C=S); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 29.9; LC-MS (retention time: 5.01 min) $m/z$ (ESI) 206 (M⁺ + H, 100%), 224 (M⁺ + H₃O, 35), 393 (2 × M⁺, 15); HRMS (ESI) calcd. for C₈H₈BN₃OS [M + H]⁺$m/z$ = 206.0554, found: 206.0553.

3.12. 1-Hydroxy-4-methylbenzo[d][1,2,3]diazaborinine-2(1H)-carbothioamide (10)

Benzodiazaborine 10 was synthesised from 2-APBA and thiosemicarbazide according to the general procedure [16]. The precipitated powder was washed with cold water and purified using RP-HPLC (retention time: 16.3–16.6 min) (5.0–50% TFA (0.1% $v/v$) in MeCN/ TFA (0.1% $v/v$) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to furnish the desired compound 10 as a white powder (61 mg; 56%); mp 249 °C (dec.); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3389m (N–H asymmetric), 3251 (N–H symmetric), 1586m (C=S), 1408s (${\mathrm{C}}_{{sp}^2}$=${\mathrm{C}}_{{sp}^2}$), 1303m (C–N); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 8.17–8.15 (1 H, m, CH-Ar), 7.90–7.87 (1 H, m, CH-Ar), 7.80–7.76 (1 H, m, CH-Ar), 7.70–7.66 (1 H, m, CH-Ar), 2.57 (3 H, s, C$H_3$); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 20.2 (CH₃) 126.8–134.9 (4 × CH-Ar and 2 × C-Ar), 147.6 ($\gamma$-C), 186.7 (C=S); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 30.0; LC-MS (retention time: 7.37 min) $m/z$ (ESI) 220 (M⁺ + H, 100%), 420 (2 × M⁺ + H, 10); HRMS (ESI) calcd. for C₉H₁₀BN₃OS [M + H]⁺$m/z$ = 220.0710, found: 220.0710.

3.13. (2aS)-4a-Hydroxy-2,2a,4a,8b-tetrahydro-3H-4-oxa-1-thia-2a¹-aza-4a${\lambda }^4$-borapentaleno[1,6-ab]inden-3-one (11)

Heterocycle 11 was synthesised from 2-FPBA and D-cysteine according to the general procedure [16]. The clear reaction mixture was lyophilised, and the residue was purified using RP-HPLC (retention time: 20.1–21.4 min) (5.0–50% TFA (0.1% $v/v$) in MeCN/ TFA (0.1% $v/v$) in H₂O over 40 min; 3.0 mLmin⁻¹, 40 °C) to yield the desired compound 11 as a white fluffy powder (83 mg; 71%); mp 310–313 °C (dec.); ${\left[\alpha \right]}_D^{25}$ -19.6 (c 0.025 in acetonitrile); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3060w (${\mathrm{C}}_{{sp}^3}$–H), 1730s (C=O), 1320m (C–N), 1204m (C–B); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 7.44–7.42 (1 H, m, CH-Ar), 7.31–7.28 (2 H, m, 2 × CH-Ar), 7.19–7.17 (1 H, m, CH-Ar), 6.16 (1 H, s, $\delta$-CH), 4.75–4.72 (1 H, d, ³$J_{H\alpha ,H\beta }$ 6.0, $\alpha$-CH), 3.59–3.55 (1 H, ABq, ¹$J_{H\beta ,H\beta }$ 12.0 and ³$J_{H\alpha ,H\beta }$ 6.0, $\beta$-CH), 3.46–3.42 (1 H, ABq, ¹$J_{H\beta ,H\beta }$ 12.0 and ³$J_{H\alpha ,H\beta }$ 6.0, $\beta$-CH); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 37.9 ($\beta$-CH₂), 65.6 ($\delta$-CH), 73.4 ($\alpha$-CH), 123.1–143.2 (4 × CH-Ar and 2 × C-Ar), 174.1 (C=O); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 11.4; LC-MS (retention time: 4.58 min) $m/z$ (ESI) 236 (M⁺ + H, 100%), 453 (2 × M⁺ + H, 100); HRMS (ESI) calcd. for C₁₀H₁₀BNO₃S [M + H]⁺$m/z$ = 236.0547, found: 236.0545.

3.14. (2aR)-4a-Hydroxy-2,2a,4a,8b-tetrahydro-3H-4-oxa-1-thia-2a¹-aza-4a${\lambda }^4$-borapentaleno[1,6-ab]inden-3-one (12)

Heterocycle 12 was synthesised from 2-FPBA and L-cysteine according to the general procedure [16]. The clear reaction mixture was lyophilised, and the residue was purified using RP-HPLC (retention time: 20.1–21.4 min) (5.0–50% TFA (0.1% $v/v$) in MeCN/ TFA (0.1% $v/v$) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to yield the desired compound 12 as a white fluffy powder (78 mg; 67%); mp 310–313 °C (dec.); ${\left[\alpha \right]}_D^{25}$ +5.8 (c 0.025 in acetonitrile); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3060w (${\mathrm{C}}_{{sp}^3}$–H), 1730s (C=O), 1320m (C–N), 1204m (C–B); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 7.44–7.42 (1 H, m, CH-Ar), 7.31–7.28 (2 H, m, 2 × CH-Ar), 7.19–7.17 (1 H, m, CH-Ar), 6.16 (1 H, s, $\delta$-CH), 4.75–4.72 (1 H, d, ³$J_{H\alpha ,H\beta }$ 6.0, $\alpha$-CH), 3.59–3.55 (1 H, ABq, ¹$J_{H\beta ,H\beta }$ 12.0 and ³$J_{H\alpha ,H\beta }$ 6.0, $\beta$-CH), 3.46–3.42 (1 H, ABq, ¹$J_{H\beta ,H\beta }$ 12.0 and ³$J_{H\alpha ,H\beta }$ 6.0, $\beta$-CH); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 37.9 ($\beta$-CH₂), 65.6 ($\delta$-CH), 73.4 ($\alpha$-CH), 123.1–143.2 (4 × CH-Ar and 2 × C-Ar), 174.1 (C=O); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 11.4; LC-MS (retention time: 4.58 min) $m/z$ (ESI) 236 (M⁺ + H, 100%), 453 (2 × M₊ + H, 100); HRMS (ESI) calcd. for C₁₀H₁₀BNO₃S [M + H]⁺$m/z$ = 236.0547, found: 236.0545.

3.15. (2aR)-4a-Hydroxy-2,2-dimethyl-2,2a,4a,8b-tetrahydro-3H-4-oxa-1-thia-2a¹-aza-4a${\lambda }^4$-borapentaleno[1,6-ab]inden-3-one (13)

Heterocycle 13 was synthesised from 2-FPBA and L-penicillamine according to the general procedure [16]. The clear reaction mixture was lyophilised, and the residue was purified using RP-HPLC (retention time: 17.5–18.3 min) (5.0–50% TFA (0.1% v/v) in MeCN/ TFA (0.1% v/v) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to yield the desired compound 13 as a white fluffy powder (59 mg; 45%); mp 250–255 °C (dec.); ${\left[\alpha \right]}_D^{25}$ +21.3 (c 0.025 in acetonitrile); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 2935w (${\mathrm{C}}_{{sp}^3}$–H), 1713s (C=O), 1204s (C–N), 1167vs (C–B); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 7.42–7.14 (4 H, m, 4 × CH-Ar), 6.25 (1 H, s, $\delta$-CH), 4.29 (1 H, s, $\alpha$-CH), 1.61 (3 H, s, C$H_3$), 1.41 (3 H, s, C$H_3$); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 23.4 and 29.9 (2 × CH₃), 58.7 ($\beta$-CH₃), 71.2 ($\delta$-CH), 72.1 ($\alpha$-CH), 128.7–143.7 (4 × CH-Ar and 2 × C-Ar), 171.5 (C=O); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 11.3; LC-MS (retention time: 5.4 min) $m/z$ (ESI) 264 (M⁺ + H, 100%), 509 (2 × M⁺ + H, 100); HRMS (ESI) calcd. for C₁₂H₁₄BNO₃S [M + H]⁺$m/z$ = 264.0860, found: 264.0862.

3.16. (2aS)-4a-Hydroxy-2,2-dimethyl-2,2a,4a,8b-tetrahydro-3H-4-oxa-1-thia-2a¹-aza-4a${\lambda }^4$-borapentaleno[1,6-ab]inden-3-one (14)

Heterocycle 14 was synthesised from 2-FPBA and D-penicillamine according to the general procedure [16]. The clear reaction mixture was lyophilised, and the residue was purified using RP-HPLC (retention time: 17.1–17.9 min) (5.0–50% TFA (0.1% v/v) in MeCN/ TFA (0.1% v/v) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to yield the desired compound 14 as a white fluffy powder (67 mg; 51%); mp 250–255 °C (dec.); ${\left[\alpha \right]}_D^{25}$ –16.4 (c 0.025 in acetonitrile); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 2935w (${\mathrm{C}}_{{sp}^3}$–H), 1713s (C=O), 1204s (C–N), 1167vs (C–B); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 7.42–7.14 (4 H, m, 4 × CH-Ar), 6.25 (1 H, s, $\delta$-CH), 4.29 (1 H, s, $\alpha$-CH), 1.61 (3 H, s, C$H_3$), 1.41 (3 H, s, C$H_3$); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 23.4 and 29.9 (2 × CH₃), 58.7 ($\beta$-CH₃), 71.2 ($\delta$-CH), 72.1 ($\alpha$-CH), 128.7–143.7 (4 × CH-Ar and 2 × C-Ar), 171.5 (C=O); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 11.3; LC-MS (retention time: 5.4 min) $m/z$ (ESI) 264 (M⁺ + H, 100%), 509 (2 × M⁺ + H, 100); HRMS (ESI) calcd. for C₁₂H₁₄BNO₃S [M + H]⁺$m/z$ = 264.0860, found: 264.0862.

3.17. (2aS)-4a-Hydroxy-8b-methyl-2,2a,4a,8b-tetrahydro-3H-4-oxa-1-thia-2a¹-aza-4a${\lambda }^4$-borapentaleno[1,6-ab]inden-3-one (15)

Heterocycle 15 was synthesised from 2-APBA and D-cysteine according to the general procedure [16]. The clear reaction mixture was lyophilised, and the residue was purified using RP-HPLC (retention time: 19.3–20.5 min) (5.0–50% TFA (0.1% v/v) in MeCN/ TFA (0.1% v/v) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to yield the desired compound 15 as a white fluffy powder (77 mg; 62%); mp 200 °C (dec.); ${\left[\alpha \right]}_D^{25}$ –14.4 (c 0.025 in acetonitrile); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3416w (O–H H-bonding), 1715s (C=O), 1260m (C–N), 1227m (C–B); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 7.42–7.40 (1 H, d, J 6.0, CH-Ar), 7.34–7.28 (2 H, m, 2 × CH-Ar), 7.24–7.22 (1 H, d, J 6.0, CH-Ar), 4.79–4.77 (1 H, ABX, ³$J_{H\alpha ,H{\beta }_1}$ 8.0 and ³$J_{H\alpha ,H{\beta }_2}$ 4.0, $\alpha$-CH), 3.65–3.62 (1 H, ABq, ¹$J_{H{\beta }_1,H{\beta }_2}$ 12.0 and ₃$J_{H\alpha ,H{\beta }_1}$ 8.0, ${\beta }_1$-CH), 3.42–3.39 (1 H, ABq, ¹$J_{H{\beta }_1,H{\beta }_2}$ 12.0 and ³$J_{H\alpha ,H{\beta }_2}$ 4.0, ${\beta }_2$-CH), 2.01 (3 H, s, C$H_3$); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 29.1 (CH₃), 35.7 ($\beta$-CH₂) 66.8 ($\delta$-C), 85.6 ($\alpha$-CH), 122.0–147.9 (4 × CH-Ar and 2 × C-Ar), 174.0 (C=O); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 11.1; LC-MS (retention time: 3.59 min) $m/z$ (ESI) 250 (M⁺ + H, 100%), 521 (2 × M⁺ + Na, 20); HRMS (ESI) calcd. for C₁₁H₁₂BNO₃S [M + H]⁺$m/z$ = 250.0704, found: 250.0701.

3.18. (2aR)-4a-Hydroxy-8b-methyl-2,2a,4a,8b-tetrahydro-3H-4-oxa-1-thia-2a¹-aza-4a${\lambda }^4$-borapentaleno[1,6-ab]inden-3-one (16)

Heterocycle 16 was synthesised from 2-APBA and L-cysteine according to the general procedure [16]. The clear reaction mixture was lyophilised, and the residue was purified using RP-HPLC (retention time: 19.8–20.9 min) (5.0–50% TFA (0.1% v/v) in MeCN/ TFA (0.1% v/v) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to yield the desired compound 16 as a white fluffy powder (77 mg; 62%); mp 200 °C (dec.); ${\left[\alpha \right]}_D^{25}$ +2.3 (c 0.025 in acetonitrile); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3416w (O–H H-bonding), 1715s (C=O), 1260m (C–N), 1227m (C–B); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 7.42–7.40 (1 H, d, J 6.0, CH-Ar), 7.34–7.28 (2 H, m, 2 × CH-Ar), 7.24–7.22 (1 H, d, J 6.0, CH-Ar), 4.79–4.77 (1 H, ABX, ³$J_{H\alpha ,H{\beta }_1}$ 8.0 and ³$J_{H\alpha ,H{\beta }_2}$ 4.0, $\alpha$-CH), 3.65–3.62 (1 H, ABq, ¹$J_{H{\beta }_1,H{\beta }_2}$ 12.0 and ³$J_{H\alpha ,H{\beta }_1}$ 8.0, ${\beta }_1$-CH), 3.42–3.39 (1 H, ABq, ¹$J_{H{\beta }_1,H{\beta }_2}$ 12.0 and ³$J_{H\alpha ,H{\beta }_2}$ 4.0, ${\beta }_2$-CH), 2.01 (3 H, s, C$H_3$); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 29.1 (CH₃), 35.7 ($\beta$-CH₂) 66.8 ($\delta$-C), 85.6 ($\alpha$-CH), 122.0–147.9 (4 × CH-Ar and 2 × C-Ar), 174.0 (C=O); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 11.1; LC-MS (retention time: 3.59 min) $m/z$ (ESI) 250 (M⁺ + H, 100%), 521 (2 × M⁺ + Na, 20); HRMS (ESI) calcd. for C₁₁H₁₂BNO₃S [M + H]⁺$m/z$ = 250.0704, found: 250.0701.

3.19. (2aS)-4a-Hydroxy-2,2,8b-trimethyl-2,2a,4a,8b-tetrahydro-3H-4-oxa-1-thia-2a¹-aza-4a${\lambda }^4$-borapentaleno[1,6-ab]inden-3-one (17)

Heterocycle 17 was synthesised from 2-APBA and D-penicillamine according to the general procedure [16]. The clear reaction mixture was lyophilised, and the residue was purified using RP-HPLC (retention time: 22.7–24.1 min) (5.0–50% TFA (0.1% v/v) in MeCN/ TFA (0.1% v/v) in H₂O over 40 min; 3.0 mL min⁻¹, 40 °C) to yield the desired compound 17 as a white powder (54 mg; 39%); mp 201 °C (dec.); ${\left[\alpha \right]}_D^{25}$ –4.5 (c 0.015 in acetonitrile); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3040w (${\mathrm{C}}_{{sp}^3}$–H), 1723s (C=O), 1281m (C–N), 1242m (C–B); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 7.38–7.28 (4 H, m, 4 × CH-Ar), 4.47 (1 H, s, $\alpha$-CH), 1.94 (3 H, s, C$H_3$), 1.55 and 0.83 (6 H, s, 2 × C$H_3$); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 26.5 (2 × CH₃), 29.9 (CH₃), 56.4 ($\beta$-C), 76.2 ($\delta$-C), 83.5 ($\alpha$-CH), 123.0–151.6 (4 × CH-Ar and 2 × C-Ar), 171.4 (C=O); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 11.5; LC-MS (retention time: 4.47 min) $m/z$ (ESI) 278 (M⁺ + H, 100%), 537 (2 × M⁺, 40), 577 (2 × M⁺ + MeCN, 20); HRMS (ESI) calcd. for C₁₃H₁₆BNO₃S [M + H]⁺$m/z$ = 278.1017, found: 278.1014.

3.20. (2aR)-4a-Hydroxy-2,2,8b-trimethyl-2,2a,4a,8b-tetrahydro-3H-4-oxa-1-thia-2a¹-aza-4a${\lambda }^4$-borapentaleno[1,6-ab]inden-3-one (18)

Heterocycle 18 was synthesised from 2-APBA and L-penicillamine according to the general procedure [16]. The clear reaction mixture was lyophilised, and the residue was purified using RP-HPLC (retention time: 22.4–23.9 min) (5.0–50% TFA (0.1% v/v) in MeCN/ TFA (0.1% v/v) in H2_O over 40 min; 3.0 mL min⁻¹, 40 °C) to yield the desired compound 18 as a white powder (37 mg; 27%); mp 201 °C (dec.); ${\left[\alpha \right]}_D^{25}$ +8.3 (c 0.015 in acetonitrile); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3040w (${\mathrm{C}}_{{sp}^3}$–H), 1723s (C=O), 1281m (C–N), 1242m (C–B); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 7.38–7.28 (4 H, m, 4 × CH-Ar), 4.47 (1 H, s, $\alpha$-CH), 1.94 (3 H, s, C$H_3$), 1.55 and 0.83 (6 H, s, 2 × C$H_3$); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 26.5 (2 × CH₃), 29.9 (CH₃), 56.4 ($\beta$-C), 76.2 ($\delta$-C), 83.5 ($\alpha$-CH), 123.0–151.6 (4 × CH-Ar and 2 × C-Ar), 171.4 (C=O); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 11.5; LC-MS (retention time: 4.47 min) $m/z$ (ESI) 278 (M⁺ + H, 100%), 537 (2 × M⁺, 40), 577 (2 × M⁺ + MeCN, 20); HRMS (ESI) calcd. for C₁₃H₁₆BNO₃S [M + H]⁺$m/z$ = 278.1017, found: 278.1014.

3.21. (E)-2-(Thiophen-3-ylmethylene)hydrazine-1-carboxamide (19)

Carboxamide 19 was synthesised from 3-FTBA and semicarbazide according to the general procedure [16]. The precipitated powder was washed with cold water and lyophilised to furnish the desired compound 19 as an off-white powder (53 mg; 54%); mp 220 °C (dec.) (lit. [18] 233–234 °C); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3453w (N–H asymmetric), 3106br (N–H symmetric), 1684s (C=O), 1598s (${\mathrm{C}}_{{sp}^2}$–${\mathrm{C}}_{{sp}^2}$), 1170m (C–N); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 7.90 (1 H, s, $\gamma$-CH), 7.62–7.61 (1 H, m, CH-Ar), 7.50–7.49 (1 H, m, CH-Ar), 7.42–7.41 (1 H, m, CH-Ar); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 125.3–127.6 (3 × CH-Ar and 1 × C-Ar), 138.0 ($\gamma$-CH), 138.7 (C=O); LC-MS (retention time: 5.1 min) $m/z$ (ESI) 170 (M⁺ + H, 100%); HRMS (ESI) calcd. for C₆H₇N₃OS [M + H]⁺$m/z$ = 170.0383, found: 170.0382.

3.22. Propane-1-sulfonohydrazide (20)

To a chilled solution of hydrazine monohydrate (4.9 mL, 0.10 mol) in anhydrous THF (20 mL) was added 1-propanesulfonyl chloride (1.1 mL, 10 mmol) in anhydrous THF (10 mL) at 5.0 °C dropwise. The thus obtained clear reaction mixture was stirred at 10–20 °C for 18 h. Upon completion (LC-MS controlled), the clear reaction mixture was extracted with EtOAc (5.0 × 50 mL), the combined organic extracts were washed with aqueous sodium hypochlorite 5.0% w/w (100 mL), dried over Na₂SO₄, filtered and the solvent was removed in vacuo. The crude product was purified by flash column chromatography to yield the desired compound 20 as a white solid (1.3 g; 94%); R_f 0.40 (10% MeOH in CH₂Cl₂); mp 45 °C (lit. [19] 40–41 °C); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3303m (N–H asymmetric), 3243m (N–H symmetric), 1619s (C=O); ${\delta }_H$ (400 MHz, CDCl₃) 3.12–3.04 (2 H, m, $\alpha$-C$H_2$), 1.91–1.80 (2 H, m, $\beta$-C$H_2$), 1.07 (3 H, t, J 7.5, $\gamma$-C$H_3$); ${\delta }_C$ (101 MHz, CDCl₃) 13.1 ($\gamma$-CH₃), 17.1, ($\beta$-CH₂), 51.2 ($\alpha$-CH₂); LC-MS (retention time: 1.94 min) $m/z$ (ESI) 139 (M⁺ + H, 100%).

3.23. 6-Methylthieno[3,2-d][1,2,3]diazaborinin-1(2H)-ol (21)

A solution of bread thiophene (40 mg, 0.30 mmol) in dichloroethane (5.0 mL) was added to hydrazide 20 dropwise, to which was added BBr₃ (0.10 mL, 1.0 mmol) in dichloroethane (0.30 mL) followed by AlCl₃ (4.0 mg, 30 $\mathsf{\mu }$mol) dissolved in dichloroethane (2.0 mL) under argon. The thus obtained yellow reaction mixture was stirred at reflux under argon for 20 min. The reaction mixture was cooled in an ice bath and poured into ice water (3.0 mL); the organic phase was separated, washed with deionised water (2.0 × 0.50 mL) and extracted with 1.0 M aqueous sodium hydroxide (3.0 × 2.0 mL). The aqueous extracts were collected and conc. HCl added until acidic to Congo red. The acidified aqueous extracts were extracted with CH₂Cl₂ (2.0 × 5.0 mL), dried over Na₂SO₄, and the solvent was removed in vacuo to yield a yellow wax. The residue was purified by flash chromatography to furnish the desired compound 21 as a green wax (30 mg; 60%); R_f 0.40 (10% MeOH in CH₂Cl₂); ${\delta }_H$ (600 MHz, acetonitrile-$d_3$) 9.95 (1 H, s, CH=N), 8.01 (1 H, s, C=CH=C), 2.57 (3 H, d,J 1.0, C$H_3$); ${\delta }_C$ (151 MHz; acetonitrile-$d_3$) 29.6 (CH₃), 127.2–135.2 (1 × CH-Ar and 3 × C-Ar), 186.5 (CH=N); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 26.3; LC-MS (retention time: 4.7 min) m/z (ESI) 167 (M⁺ + H, 100%).

3.24. (E)-N^′-((5-Methylthiophen-2-yl)methylene)propane-1-sulfonohydrazide (22)

To a solution of bread thiophene (0.78 mL, 7.2 mmol) and hydrazide 20 (1.0 g, 7.2 mmol) in EtOH (10 mL) was added glacial AcOH (6 drops), and the resulting brown reaction mixture was stirred at ambient temperature for 1.0 h. Upon completion (LC-MS controlled), the solvent was removed in vacuo, and the residue was purified by flash column chromatography to furnish the desired compound 22 as a yellow solid (1.6 g; 94%); R_f 0.35 (7% MeOH in CH₂Cl₂); mp 108 °C (lit. [20] 109–110 °C); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3186m (N–H), 1148vs (S=O), 945s (S–O); ${\delta }_H$ (400 MHz, CDCl₃) 7.95 (1 H, s, γ-CH), 7.75 (1 H, s, α-NH), 7.06 (1 H, d, J 3.5, CH-Ar), 6.75–6.66 (1 H, m, CH-Ar), 3.33–3.15 (2 H, m, α-C$H_2$), 2.49 (3 H, d, J 1.0, C$H_3$), 1.97–1.79 (2 H, m, β-C$H_2$), 1.07 (3 H, t, J 7.5, γ-C$H_3$); ${\delta }_C$ (101 MHz, CDCl₃) 13.0 (γ-CH₃), 15.7 (β-CH₂), 17.0 (CH₃), 52.7 (α-CH₂), 125.8 and 131.1 (CH-Ar), 135.4 (C-Ar), 143.7 (γ-CH), 144.5 (C-Ar); LC-MS (retention time: 7.13 min) $m/z$ (ESI) 247 (M⁺ + H, 100%); HRMS (ESI) calcd. for C₉H₁₅N₂O₂S₂ [M + H]⁺$m/z$ = 247.0570, found: 247.0571.

3.25. 6-Methyl-2-(propylsulfonyl)thieno[3,2-d][1,2,3]diazaborinin-1(2H)-ol (23)

To a solution of FeCl₃ (60 mg, 0.37 mmol) in dry dichloroethane was added hydrazone 22 (0.90 g, 3.7 mmol) in dry dichloroethane (50 mL) and BBr₃ (1.1 mL, 11 mmol) in dry dichloroethane (3.0 mL) simultaneously under argon with vigorous stirring at 70 °C for 15 min. The orange reaction mixture was cooled in an ice bath and poured into ice water (30 mL). The subsequent reaction mixture was washed with deionised water (30 mL), and the organic phase was extracted with 1.0 M aqueous sodium hydroxide (3.0 × 30 mL). The aqueous extracts were combined, and conc. HCl added until a cloudy mixture was obtained (pH ≈ 2–3). The thus obtained colloidal was extracted with CH₂Cl₂ (4.0 × 100 mL), dried over Na₂SO₄, and the solvent was removed in vacuo. The green crude wax was purified by flash column chromatography to furnish the desired compound 23 as a white powder (0.72 g; 72%); R_f 0.30 (5% MeOH in CH₂Cl₂); mp 92–94 °C (dec.) (lit. [21] 85–86 °C); ${\nu }_{max}$/$\mathrm{c}{\mathrm{m}}^{-1}$ (ATR) 3200–2800br (O–H H-bonding), 1361m and 1136s (S=O), 900vs (S–O); ${\delta }_H$ (400 MHz, acetonitrile-$d_3$) 8.19 (1 H, s, $\gamma$-CH), 7.26 (1 H, t, J 1.0, CH-Ar), 3.48–3.42 (2 H, m, $\alpha$-C$H_2$), 2.59 (3 H, d, J 1.0, C$H_3$), 1.71 (2 H, dq, J 15.0 and 7.5, $\beta$-C$H_2$), 0.95 (3 H, t, J 7.5, $\gamma$-C$H_3$); ${\delta }_C$R (101 MHz, acetonitrile-$d_3$) 12.4 ($\gamma$-CH₃), 15.1 ($\beta$-CH₂), 17.3 (CH₃), 53.8 ($\alpha$-CH₂), 128.2 (CH-Ar), 136.8 ($\gamma$-CH), 148.2 (C-Ar); ${\delta }_B$ (193 MHz, acetonitrile-$d_3$) 27.2; LC-MS (retention time: 7.25 min) $m/z$ (ESI) 273 (M⁺ + H, 100%); HRMS (ESI) calcd. for C₉H₁₄BN₂O₃S₂ [M + H]⁺$m/z$ = 273.0533, found: 273.0534.

4. Conclusions

Gao and coworkers [10] have reported that on reacting 2-FPBA and 2-APBA with L-cysteine in 0.10 M phosphate-buffered saline (PBS) buffer at physiological pH, bicyclic or tricyclic heterocycles form, which led to the survey of an array of borazaros and oxazaborons described herein (Figure 3); however, in this work, such reactions exclusively furnished tetracyclic and heteroaromatic structures. Whilst almost all the diazaborines as well as boronate ester 12 described in this work have, one way or another, been reported in the literature, by all means, this work is the first of its kind to delineate a ‘tetracyclic caged’ structure for novel boronate esters 11 and 13–18, and report full chemical characterisation, i.e., mp, optical rotation, IR, NMR (¹H, ¹³C and ¹¹B cf. Supplementary Materials), LC-MS and HRMS of the aforementioned benzodiazaborines, diazaborines and novel boronate esters for which, so far as this work is concerned, there are no known precedents, and the specific applications and the broad implications of these intriguing heterocycles may, hopefully, be of interest to the synthetic and medicinal chemist alike.