Electrochemical Synthesis of Dimeric λ³-Bromane: Platform for Hypervalent Bromine(III) Compounds

Abstract

A straightforward and scalable approach to a previously unreported class of cyclic hypervalent Br(III) species capitalizes on the anodic oxidation of aryl bromide to dimeric benzbromoxole that serves as a versatile platform to access a range of structurally diverse Br(III) congeners such as acetoxy-, alkoxy-, and ethynyl-λ³-bromanes as well as diaryl-λ³-bromanes. The synthetic utility of dimeric λ³-bromane is exemplified by photoinduced Minisci-type heteroarylation reactions and benzylic oxidation.

Hypervalent halide reagents are an indispensable part of the organic synthesis tool kit. Among hypervalent halides, bromine(III) reagents possess superior reactivity to that of iodine(III) counterparts owing to stronger electrophilicity and better nucleofugality of the bromanyl moiety.¹ The unmatched reactivity of λ³-bromanes has allowed for the development of a series of unprecedented synthetic transformations, e.g., an unusual Hofmann rearrangement of sulfonamides,² a rare Bayer–Villiger-type oxidation of open-chain aliphatic aldehydes,³ the oxidative coupling of alkynes and primary alcohols to conjugated enones,⁴ and the regioselective C–H functionalization of nonactivated alkanes.⁵ Recently, the application of diaryl-λ³-bromanes in catalysis⁶ and cycloaddition reactions⁷ was also reported. Notwithstanding these application examples, the remarkable synthetic potential of λ³-bromanes appears to be unexploited due to challenges associated with their preparation. Thus, high oxidation potentials of aryl bromides (e.g., 2.0 V vs Ag/0.01 M AgNO₃ for PhBr in CH₃CN)⁸ and poor stability of the resulting bromine(III) species complicate the direct two-electron oxidation by chemical oxidants, a method that is routinely employed for the synthesis of iodine(III) counterparts from iodoarenes.⁹

In 1984, Frohn demonstrated that the challenging direct oxidation of Br(I) to Br(III) species could be circumvented by a ligand exchange reaction between a suitable Br(III) source such as BrF₃ and arylsilane to afford difluoro-λ³-bromane 1 (eq 1, Figure 1).¹⁰ Since then, the unstable and moisture-sensitive Frohn’s reagent 1 has served as a key starting material for the synthesis of various Br(III) species,¹¹ including bench-stable imino-λ³-bromane 2.¹² Recently, the latter was employed by Miyamoto as the starting material in the preparation of a new series of λ³-bromanes such as 3 and 4 (eq 1, Figure 1).¹³ Notwithstanding these notable achievements, all of the above-mentioned syntheses rely on the use of BrF₃ as the ultimate starting material. Given the high toxicity and extreme reactivity of liquid BrF₃ that require specialized equipment and experimental techniques for safe handling, the synthesis of λ³-bromanes is possible only in laboratories that are capable of handling the BrF₃ reagent.

Figure 1.

Electrochemical approach to benzbromoxoles 8–13.

In a quest to solve the challenge of λ³-bromane synthesis, we recently disclosed an inherently safe, inexpensive, and straightforward approach to chelation-stabilized hypervalent bromine(III) reagents 6 (Martin’s bromanes) by anodic two-electron oxidation of parent aryl bromides 5 under constant current conditions in an undivided cell (eq 2, Figure 1).¹⁴ The suitability of hypervalent Br(III) reagents 6 for the oxidative amination of anilines and the homocoupling of electron-rich arenes was also demonstrated. However, the remarkably stable doubly chelated structure of Martin’s bromanes 6 limits their application in reactions involving ligand exchange at the hypervalent Br(III) center, e.g., functional group transfer reactions.¹⁵ Herein, we report on the further development of an anodic oxidation methodology for the synthesis of benzbromoxoles 8 and 9. Notably, λ³-bromane 9 is a stable crystalline material that serves as a platform for the synthesis of benzbromoxoles with various ligands at the exocyclic position such as acetoxy-λ³-bromane 10, phenyl-λ³-bromane 11, alkynyl-λ³-bromane 12, and tert-butoxy-λ³-bromane 13 (eq 3, Figure 1). Electrochemical properties of all synthesized hypervalent Br(III) species 8–13 are also disclosed together with representative synthetic application examples, as shown below.

The published conditions for the electrochemical generation of bromine(III) species 6(14) (glassy carbon as the working electrode and platinum foil as the counter electrode in a TBA-BF₄/HFIP electrolyte; current density j = 10 mA cm^–2, passed charge equivalents 2F) were used as the starting point for the anodic oxidation of aryl bromide 7 (Table S1, SI). NMR spectra of an electrolyzed solution of 7 in HFIP (with an added equal volume of CHCl₃-d) revealed the presence of starting aryl bromide 7 (77% conversion) together with a new product (70% NMR yield) that could be isolated from the electrolysis mixture by aqueous workup and separated from the unreacted 7 by reversed-phase chromatography.¹⁶ A structure of hypervalent bromine(III) reagent 8 was assigned on the basis of NMR and MS data (Figure 2). λ³-Bromane 8 is a crystalline material that could be stored at −18 °C (freezer) for 8 weeks. It is stable in organic solvents such as CHCl₃-d (τ_1/2 = 96 h) and MeCN-d₃ (τ_1/2 = 168 h) at room temperature. The stability of 8 is notable, given that the electrochemically generated HFIP-containing hypervalent iodine(III) species were insufficiently stable to be isolated in a pure form.¹⁷ Optimization of the anodic oxidation conditions (increase in current density to 20 mA cm^–2 and passed charge equivalents to 2.5F) allowed us to isolate λ³-bromane 8 in 57% yield. (For an overview of the optimization experiments, see Table S1 in the SI.)

Figure 2.

Synthesis of λ³-bromanes 8–13. Stability was measured in a 0.02 M solution at room temperature. Ellipsoids are shown at 50% probability, with hydrogen atoms omitted for clarity. Selected bond distances (Å) and angles (deg) for λ³-bromane 9: Br1–O2, 2.034(2); Br1–O1, 1.9240(16); Br1–C7A, 1.931(3); O1–Br1–O2, 173.63(8); Br1–O1–Br1′, 114.32(14); O2–Br1–C7A, 83.06(10); C7A–Br1–O1, 90.73(9); and Br1–O2–C3–C3A, 1.4(3). Selected bond distances (Å) and angles (deg) for diaryl-λ³-bromane 11: Br1–O2, 2.365(2); Br1–C6, 1.933(2); Br1–C5, 1.959(2); C6–Br1–O2, 176.7(1); C6–Br1–C5, 98.76(10); O2–Br1–C5, 78.1(1); and Br1–O2–C3–C4, −8.8(2). For a discussion of crystalline state structures of dimeric λ³-bromane 9 and diaryl-λ³-bromane 11, see the Supporting Information.

With the optimized electrolysis conditions in hand (Figure 2), the postelectrolysis conversion of relatively unstable 8 into an easier-to-handle λ³-bromane was addressed by the exchange of the 1,1,1,3,3,3-hexafluoroisopropoxy ligand for acetate. Gratifyingly, the addition of AcOH to the electrolysis mixture followed by concentration and extractive workup afforded crystalline acetate 10 (Figure 2) that was stable at −18 °C for more than 2 months. The corresponding trifluoroacetate-containing λ³-bromane IM1 was also observed by NMR when 8 was reacted with TFA in CHCl₃-d. However, attempted isolation of trifluoroacetyl-λ³-bromane IM1 unexpectedly delivered crystalline dimer 9, whose structure was assigned by NMR and MS methods and confirmed by X-ray analysis (Figure 2). Presumably, λ³-bromane IM1 undergoes ligand exchange with water to form transient IM2 that readily condenses into dimer 9.¹⁸ Importantly, bench-stable dimer 9 can be stored at 4 °C for 4 months without any decomposition. The superior stability of crystalline 9 compared to that of λ³-bromanes 8 and 10 renders dimer 9 the most suitable product of the anodic oxidation reaction. Accordingly, additional optimization of electrolysis conditions (replacement of TBA–BF₄ with water-miscible TEA–BF₄ to ease the extractive removal of supporting electrolyte) combined with the acidic aqueous workup allowed for the scaling up of the electrolysis to furnish 2.42 g (61% yield from 7) of dimeric λ³-bromane 9 in a single run (Figure 2).

Easy-to-handle, stable, and crystalline dimeric λ³-bromane 9 served as a key starting material for the synthesis of a series of bromine(III) congeners 10–13 (Figure 2). Thus, the treatment of 9 with acetic anhydride in the presence of catalytic TMS-OTf (10 mol %) afforded acetoxy-λ³-bromane 10. The reaction of 9 with alkynylsilane and stoichiometric BF₃·OEt₂ furnished β-(triisopropylsilyl)ethynyl-λ³-bromane 12.¹⁹ Notably, both 10 and 12 turned out to be sufficiently stable (Figure 2) to be purified by reversed-phase column chromatography.²⁰ Crystalline diaryl-λ³-bromane 11 was also readily obtained from 9 and phenyl trimethylsilane in the presence of stoichiometric BF₃·OEt₂, and its structure was proven by X-ray analysis.²¹ Dimer 9 can be converted back to alkoxy-λ³-bromane 8 simply by dissolution in HFIP. Finally, activation with TMS-OTf was required for the reaction of dimeric λ³-bromane 9 with tert-BuOH to afford tert-butoxy-λ³-bromane 13 (Figure 2).

Electrochemical properties of the 7/8 redox couple and reduction potentials of λ³-bromanes 9–13 were studied with cyclic voltammetry (CV) in MeCN or HFIP and 0.1 M NBu₄BF₄ as an electrolyte, a glassy carbon working electrode, and a Ag/0.01 M AgNO₃ reference (E₀ = −87 mV vs Fc/Fc⁺ couple).²² Bromide 7 exhibits a single irreversible anodic feature with a half-peak potential (E_P/2 = 2.62 V; see Figure 3B and the SI),²³ which is higher than that of aryl bromide 5 (R=CF₃; E_P/2 = 2.54 V), a precursor of Martin’s bromane.¹⁴ As anticipated, the corresponding iodide 7-I is oxidized at a considerably lower potential (E_P/2 = 2.01 V; see Figure 3B).²⁴ A CV study of the cathodic reduction (Figure 3A and the SI) showed that each of the λ³-bromane 8–13 exhibits a single irreversible feature, indicating that electron transfer is followed by a rapid chemical step. Of note, the reduction potentials of λ³-bromanes 8–13 span a relatively wide range from −0.36 to −1.38 V with dimer 9 possessing the highest half-peak potential of −0.36 V (Figure 3A), which is comparable to that of Martin’s bromane 6 (R=CF₃; E_P/2 = −0.37 V).^14b λ³-Bromanes 8 and 10 are not as easy to reduce (E_P/2 = −0.55 and −0.58 V, respectively), while λ³-bromanes 11–13 are reduced at even lower potentials (E_P/2 = −1.38, −1.30, and −1.34 V, respectively) than are required for λ³-iodane 9-I (E_P/2 = −0.91 V; Figure 3A). Interestingly, the reduction potentials are highly sensitive to the electronic properties of ligands at the Br(III) center as evidenced by the remarkable 0.79 V difference between half-peak potentials for structurally related HFIP- and tert-butoxy-substituted benzbromoxoles 8 and 13 (Figure 3A). Overall, the electrochemical reduction of benzbromoxoles is relatively difficult as it proceeds at negative potentials; therefore, λ³-bromanes 8–13 could be considered to be weaker SET oxidants than hypervalent iodine(III) reagents such as widely used PIFA and Koser’s reagent PhI(OH)OTs (E_P/2 = −0.33 and +0.14, respectively; see the SI, p S23). In the meantime, λ³-bromane 9 (E_P/2 = −0.36 V) is a much stronger SET oxidant than the analogous λ³-iodane 9-I (E_P/2 = −0.91 V), so considerable SET oxidizing ability is anticipated for hitherto unreported bromine(III) analogs of Koser’s reagent and PIFA.²⁵

Figure 3.

(A) Background and iR drop-corrected linear sweep voltammograms (LSV) of λ³-bromanes 8–13 and λ³-iodane 9-I (c = 1 mM) were recorded at 100 mV·s^–1 in MeCN. (B) Oxidation half-peak potentials E_P/2 are for bromide 7 and iodide 7-I in HFIP (c = 5 mM).

Despite the comparable reduction potentials of λ³-bromanes 8–10 and Martin’s reagent 6 (Figure 3A), benzbromoxoles 8–10 are considerably less stable than the doubly chelated λ³-bromane 6. For example, the thermolysis of dimer 9 in benzene at 80 °C for 2 h generated a 4:1 mixture of reduction product 7 and ketone 14, respectively, together with biphenyl (Figure 4A). Possibly, λ³-bromane 9 undergoes homolytic cleavage of the exocyclic Br–O bond^13,26 to generate a pair of radicals IM3 and IM4 that abstract a hydrogen atom from benzene, leading to Br(III) intermediate IM2 and bromide 7, respectively (Figure 4A). Subsequent homolysis of the Br–O bond in IM2 leads to 7 via radical IM4. Alternatively, IM2 may also dimerize into 9. A competing decomposition pathway for oxygen-centered radical IM4 apparently involves β-scission to form ketone 14 and a CF₃ radical, which upon reaction with benzene produced trifluorotoluene (observed in the reaction mixture by ¹⁹F NMR and GC–MS assays). An alternative mechanistic scenario would involve the SET reduction of λ³-bromane 9 by benzene. We regard this as a less likely scenario because dimer 9 is a weaker oxidant than PIFA (E_P/2 = −0.33; see the SI), which requires activation by a Lewis acid to effect the SET oxidation of electron-rich arenes.²⁷ Furthermore, no signs of decomposition were observed for doubly chelated λ³-bromane 6 (R=CF₃) in benzene at 80 °C even after 18 h. Given the virtually similar reduction potentials for the relatively unstable dimer 9 and stable Martin’s bromane 6 (R=CF₃), SET oxidation could be ruled out in favor of homolytic cleavage of the exocyclic Br–O bond as the underlying mechanism for thermal decomposition. Notably, clean conversion of dimeric λ³-bromane 9 to bromide 7 was observed upon light irradiation at 365 nm under ambient temperature within 1 h.

Figure 4.

Reactivity of λ³-bromane 9. ^aThe yield of 15 was determined by ¹H NMR using mesitylene as an internal standard and was calculated based on λ³-bromane 9. ^bReaction conditions: heterocycle (0.5 mmol), 9 (0.5 mmol), TFA (2.5 mmol), and THF (3 mL); ^c0.6 mmol of 9; ^d0.8 mmol of 9; ^e0.6 mmol of 9, 1:1 cyclohexane/EtOAc (3.0 mL); ^fheterocycle (0.5 mmol), 9 (0.6 mmol), TFA (0.5 mmol), 1:1 formamide/EtOAc (3.0 mL); ^g7 h; and ^h2.5 mmol of TFA, 3 h.

The propensity for the homolytic Br–O bond cleavage renders dimeric λ³-bromane 9 a convenient source of electrophilic radicals for application in hydrogen atom transfer (HAT) reactions²⁸ such as benzylic oxidation²⁹ (Figure 4B) and photoinduced Minisci-type heteroarylations (Figure 4C).³⁰ Indeed, indane was readily oxidized to 1-indanone 15 at 40 °C under oxygen in the presence of dimer 9. Likely, the oxidation is initiated by the abstraction of a benzylic hydrogen atom by radical intermediate IM3 or IM4 to produce the corresponding benzyl radical^26c that undergoes autoxidation with dioxygen. λ³-Bromane 9 is also suitable for late-stage oxidative functionalization of complex natural products and pharmaceuticals, as demonstrated by the benzylic oxidation of natural diterpene leelamine (dehydroabietylamine) and antidepressant citalopram to the corresponding ketones 16 and 17 (Figure 4B). Notably, doubly chelated Martin’s bromane 6 and λ³-iodane 9-I were completely inefficient as initiators of alkane aerobic oxidation reactions. Dimeric λ³-bromane 9 also effected the cross-dehydrogenative Minisci-type alkylation of heterocycles with THF and amidation with formamide upon light irradiation at 365 nm (Figure 4C).³¹ Both alkylation products 18–21 and amides 22–25 were readily formed in 38–98% yields within 1 h at room temperature.³² It should be noted that alkylation product 18 was isolated in only 8% yield (at 10% conversion of the starting 4-methyl quinoline) when dimeric λ³-iodane 9-I was employed. Possibly, the higher reactivity of λ³-bromane 9 as compared to that of λ³-iodane 9-I can be attributed to a lower BDE for the Br–O bond. Importantly, recovery of aryl bromide 7 (98% in the formation of ketone 15 and 94% in the synthesis of quinoline 18) followed by electrochemical oxidation allowed for multiple reuses of λ³-bromane 9 reagent in benzylic oxidation and photoinduced Minisci-type heteroarylations. Further studies to expand the application scope of λ³-bromanes 8–13 are ongoing in our laboratory.

In summary, we have disclosed an efficient, reliable, and inexpensive methodology toward a previously unreported class of cyclic hypervalent bromine(III) species. Our approach capitalizes on the anodic oxidation of chelating ortho-substituent-containing aryl bromide to a dimeric benzbromoxole derivative under constant current conditions in an undivided cell. Easy-to-handle, stable, and crystalline dimeric benzbromoxole serves as a versatile platform to access a wide range of structurally diverse bromine(III) congeners such as acetoxy-, alkoxy-, and ethynyl-λ³-bromanes as well as diaryl-λ³-bromanes. All synthesized bromine(III) species are sufficiently stable to be handled under ambient conditions and purified using standard techniques such as column chromatography. The electrochemical oxidation can be easily scaled-up to afford gram quantities of the dimeric benzbromoxole derivative in a single batch. Cyclic voltammetry studies indicate that oxygen- and carbon-ligand-containing benzbromoxoles are relatively difficult to reduce with potentials spanning a range from −0.36 to −1.38 V. In the meantime, benzbromoxoles are prone to homolysis of the exocyclic Br–O bond under elevated temperatures as well as upon photoexcitation at 365 nm. Hence, the cyclic λ³-bromanes are suitable sources of electrophilic oxygen-centered radicals in, for example, hydrogen atom transfer (HAT) reactions such as Minisci-type amidation and cross-dehydrogenative alkylation as well as benzylic oxidation. Importantly, the recovery of aryl bromide postreaction followed by electrochemical oxidation allowed for multiple reuse of the λ³-bromane reagent. Overall, we believe that the disclosed methodology is a breakthrough in the preparation of λ³-bromanes, a hitherto difficult-to-access class of hypervalent halide species of unmatched reactivity and remarkable application potential in organic synthesis and pharmaceutical chemistry.

Data Availability Statement

The data underlying this study are available in the published article and its Supporting Information.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.orglett.3c00405.

• Experimental procedures, analytical and spectroscopic data for new compounds, copies of NMR spectra, and X-ray crystallographic data for λ³-bromanes 9 and 11 (PDF)

The authors declare no competing financial interest.