Triphenylbismuth carbonate-mediated oxidation of hydroxylamines to nitrones and in situ 1,3-dipolar cycloaddition

Abstract

Triphenylbismuth carbonate was shown to act as a mild, selective and efficient reagent for the oxidation of hydroxylamines. The developed reaction conditions were shown to be compatible with the in situ trapping of the produced nitrones by a strained alkyne, via a [3 + 2] cycloaddition reaction.

---

Introduction

The development of novel oxidation processes is a fundamental objective in synthetic chemistry as more and more complex molecules are currently produced. Using organobismuth reagents,¹ a wide range of functional groups can be oxidized under very mild conditions.² Indeed, besides the ability of organobismuth(V) reagents to arylate organic compounds with or without copper catalysis,³ the intrinsic Bi^V–Bi^III two-electron transition also provides a convenient oxidizing source for selective oxidation reactions.⁴ Examples of substrates that can be oxidized using triphenylbismuth dichloride or carbonate include alcohols,⁵ thiols, hydrazobenzenes,⁶ and phenols.⁷ We also contributed to the field and reported the oxidation of urazoles to triazolinediones using triphenylbismuth carbonate.⁸ The mildness of the latter transformation also permitted the in situ trapping of the produced triazolinediones by a [4 + 2] cycloaddition reaction.

As part of our longstanding interest in the chemistry of organobismuth reagents,⁹ we sought to investigate the use of triphenylbismuth carbonate for the oxidation of hydroxylamines to nitrones.¹⁰ The latter transformation usually operates in the presence of strong oxidizing agents such as methyltrioxorhenium,¹¹ sodium hypochlorite,¹² manganese dioxide,¹³ mercuric oxide, iodine, ceric ammonium nitrate, or tert-butyl hydroperoxide.¹⁴ More recently, hypervalent iodine reagents¹⁵ and catalytic systems, based for example on nanogold chemistry,¹⁶ have also been devised to perform the oxidation of hydroxylamines. However, extensive heating was sometime required which could be detrimental to the isolated yield of products.

With these features in mind, we conceived that the oxidation reaction of hydroxylamines 1 could be smoothly triggered by Bi^V reagents such as triphenylbismuth carbonate (Ph₃BiCO₃), to afford the corresponding nitrones 2 (Scheme 1). In fact, despite its position in the heavy elements group, bismuth can be seen as an eco-friendly metal with low toxicity.¹⁷ In addition, the ease of handling and relatively low cost of arylbismuth derivatives, make these compounds highly attractive in fine synthetic chemistry. In addition to the oxidation reaction, one could also expect the produced nitrone to react in situ with an alkyne via a 1,3-dipolar cycloaddition reaction.¹⁸ This sequential reaction would provide a “one pot” access to heterocyclic structures analogous to 3 starting from simple hydroxylamines.

Scheme 1 Bismuth-mediated oxidation of hydroxylamines and further 1,3-dipolar cycloaddition.

Results and discussion

The scope of the oxidation reaction of hydroxylamines to nitrones was first evaluated. Ph₃BiCO₃ was prepared as previously described,¹⁹ and reaction conditions were set by working at room temperature in CH₂Cl₂. The reaction of N,N-dibenzyl hydroxylamine (1a) with 1.2 equivalents of Ph₃BiCO₃ cleanly afforded the corresponding nitrone in nearly quantitative yield (95%) after only 2 h at room temperature (Table 1, Entry 1). Running the oxidation reaction on a larger scale (e.g., 5 mmol) provided also a nearly quantitative yield of 2a (97%). The reaction worked well on N-benzyl-N-phenyl hydroxylamine 1b (Entry 2) but the resulting nitrone 2b was shown to be quite labile, thus impacting the isolated yield of product (76%). Alkyl-substituted hydroxylamine 1c (Entry 3) reacted similarly but the process was slower, leading to the corresponding nitrone 2c in moderate yield after 6 h. The regioselectivity of the oxidation reaction was next investigated using hydroxylamine substrates bearing simultaneously benzyl and alkyl groups. Thus, octyl 1d (Entry 4), cyclopentyl 1e (Entry 5) or 3-pentyl-substituted benzyl hydroxylamine 1f (Entry 6) were reacted with the pentavalent bismuth reagent. While substrates substituted with secondary alkyl groups (e.g. 1e and 1f) led only to the oxidation of the benzylic position, benzyl hydroxylamine 1d substituted with a primary alkyl chain afforded a 4 : 6 mixture of regioisomers 2d and 2d′ in satisfactory yields (Entry 4). The influence of the electronic properties of the benzyl groups borne by the hydroxylamines was also scrutinized as regards again the regioselectivity of the oxidation reaction.

Table 1 Scope of the Ph₃BiCO₃-mediated oxidation of hydroxylamines to nitrones^a

Entry	Substrate	1	Product	2	t (h)	Yield^b (%)
a Conditions: 1 (1 equiv.), Ph3BiCO3 (1.2 equiv.), CH2Cl2, room temperature.b Isolated yields.c NMR yield.
1		1a		2a	2	95
2		1b		2b	2	76
3		1c		2c	6	63
4		1d		2d	4	70
				2d′
5		1e		2e	3	91
6		1f		2f	3	94
7		1g		2g	2	99
				2g′
8		1h		2h	2	98
				2h′
9		1i		2i	2	40
10		1j		2j	6	75^c

---

However, hydroxylamines substituted with either an electron withdrawing 4-fluoro (Entry 7) or electron donating 4-methoxy group (Entry 8) both provided a 1 : 1 mixture of oxidized isomers. Thus, there seems to be no major electronic effect of the aryl-substituents on the regioselectivity of bismuth-triggered oxidation of benzyl hydroxylamines.

It is to be noted that in the case of a cyclic hydroxylamine such as the one of Entry 9 (compound 1i), a different behavior of triphenylbismuth carbonate was observed as no nitrone was formed. Instead, the pentavalent bismuth reagent promoted the N-arylation of hydroxylamine 1i, resulting in the formation of N-oxide 2i in moderate yield (Entry 9). Although this difference in reactivity is not fully understood yet, one can hypothesized that geometrical constraints are at the origin of the observed discrepancy. This transformation failed on a 5-membered ring hydroxylamine (i.e. pyrrolidine-derived hydroxylamine) which decomposed under our reaction conditions, but was successful on cyclic hydroxylamine 1j (Entry 10). The arylated product 2j could however not be isolated by flash chromatography because of its intrinsic instability.

The issue of the regioselectivity of the oxidation of 1d was tentatively addressed by changing the reaction conditions (solvent, temperature, reverse addition of the reagents). Thus, hydroxylamine 1d was reacted under the conditions reported in Table 2, and the ratio of the isomeric nitrones 2d and 2d′ was measured by ¹H-NMR analysis of the crude. However, none of the investigated conditions provided a satisfactory selectivity as the two regioisomers were obtained in almost identical proportions in each case, except that of the reverse addition (Entry 6) for which we observed a slightly better 2d/2d′ ratio (3/7). Noteworthy, CH₂Cl₂ was found to be the best solvent as regards the kinetics of the transformation (entries 1–3); the reaction run at 0 °C needed 12 h (Entry 4) and that at reflux (Entry 5) only 2 h to reach ca. 70% conversion (as opposed to 4 h at room temperature).

Table 2 Optimization attempts of the regioselectivity of the oxidation of 1d^a

Entry	Conditions	2d/2d′	t (h)
a Conditions: 1 (1 equiv.), Ph3BiCO3 (1.2 equiv.), solvent and temperature as specified in the table.b Slow degradation of 2d′ was observed under these reaction conditions.c Addition of the hydroxylamine on Ph3BiCO3 over 30 min.
1	CH2Cl2, room temp.	4/6	4
2^b	THF, room temp.	4/6	12
3	CH3CN, room temp.	4/6	6
4	CH2Cl2, 0 °C	4/6	12
5	CH2Cl2, reflux	4/6	2
6^b^,^c	CH2Cl2, room temp., reverse addition	3/7	4

---

A postulated mechanism for the Ph₃BiCO₃-mediated oxidation reaction is illustrated in Scheme 2 for the oxidation of N,N-dibenzyl hydroxylamine (1a). The first step could involve the formation of an oxygen–bismuth bonded intermediate by nucleophilic attack of the oxygen atom of hydroxylamine 1a on Ph₃BiCO₃. The existence of an analogous O–Bi intermediate has been well demonstrated by Barton and co-workers in the case of phenols.²⁰ Subsequent rearrangement, via collapse of the transient species, leads to the formation of nitrone 2a with concomitant release of triphenylbismuth, carbon dioxide, and water. Attempts to run the above reaction catalytically using 10 mol% of Ph₃BiCO₃ in the presence of NBS/K₂CO₃ (ref. 21) were not satisfactory as NBS itself was found to be able to oxidize 1a to nitrone 2a.¹⁴

Scheme 2 Postulated mechanism for the bismuth-mediated oxidation of hydroxylamines.

As our system efficiently promoted the oxidation of hydroxylamines at room temperature, we wanted to further explore the in situ 1,3-dipolar cycloaddition of the produced nitrones with strained alkynes. If operative, this reaction would provide an interesting alternative to the conventional azide–alkyne “click” reaction, as previously investigated by Pezacki and co-workers.¹⁸ Thus, N,N-dibenzyl hydroxylamine (1a) was reacted with 1.2 equivalents of Ph₃BiCO₃ in CH₂Cl₂ at ambient temperature and in the presence of 1 equivalent of benzannulated cyclooctyne 4. After 2 h, we were pleased to isolate the expected cycloadduct 3a in 70% yield (Table 3, Entry 1). Although the two-step in situ process was slightly slower than the direct nitrone–alkyne cycloaddition reaction (i.e. 30 min),¹⁸ the reaction time in our case was mainly governed by the oxidation rate of N,N-dibenzyl hydroxylamine into the corresponding nitrone. It thus seems that the formation of the 1,3-dipole is the rate-limiting step of the oxidation/cycloaddition reaction. The sequential process could even be performed on a gram scale of 1a, providing efficient access to cycloadduct 3a in about comparable yields (76%). While the cycloaddition reaction worked well with a N,N-dibenzyl substrate, it was not operative with a bis-alkyl nitrone. Indeed, N,N-dioctyl hydroxylamine 1c was efficiently converted to nitrone 2c but failed to undergo [3 + 2] cycloaddition with strained alkyne 4 (Entry 2). Nevertheless, the “one pot” oxidation/1,3 dipolar cycloaddition cascade process was further extended to another hydroxylamine substrate (compound 1e) which reacted with triphenylbismuth carbonate and benzannulated cyclooctyne 4 to provide straightforward access to heterocycle 3e (Entry 3).

Table 3 In situ oxidation/1,3 dipolar cycloaddition of hydroxylamines^a

Entry	Substrate	R^1	R^2	Product	t (h)	Yield^b (%)
a Conditions: 1 (1 equiv.), 4 (1 equiv.), Ph3BiCO3 (1.2 equiv.), CH2Cl2, room temperature.b Isolated yields.
1	1a	Ph	Bn	3a	2	70
2	1c	C7H15	C8H17	—	6	—
3	1e	Ph	Cyclopentyl	3e	3	62

---

Experimental

Synthetic procedures

Benzannulated cyclooctyne 4 and Ph₃BiCO₃ were prepared according to ref. 18 and 19, respectively.

General procedure for the oxidation of hydroxylamines

The synthesis of compound 2a is given as a representative example. To a solution of N,N-dibenzyl hydroxylamine 1a (10.7 mg, 0.05 mmol, 1 equiv.) in CH₂Cl₂ (1 mL) was added Ph₃BiCO₃ (30 mg, 1.2 equiv.) in one portion. The reaction mixture was stirred at room temperature for 2 h. The solvent was evaporated under vacuum and the crude product was purified by flash chromatography (silica gel, pentane/EtOAc, 1 : 0 to 1 : 1) to afford compound 2a¹⁵ (10 mg, 95%) as a white solid. ¹H NMR (CDCl₃): δ 8.23–8.18 (m, 2H), 7.52–7.26 (m, 9H), 5.06 (s, 2H) ppm. ¹³C NMR (CDCl₃): δ 134.2, 133.1, 130.4, 130.3, 129.2, 128.9, 128.5, 128.4, 71.1 ppm. IR (neat) 1580, 1457, 1147, 903, 724 cm⁻¹. MS (ES⁺): 212 [M + H]⁺.

General procedure the in situ oxidation/1,3 dipolar cycloaddition of hydroxylamines

The synthesis of 3a is given as a representative example. At room temperature, to a solution of N,N-dibenzyl hydroxylamine 1a (10.7 mg, 0.05 mmol, 1 equiv.) and benzannulated cyclooctyne 4 (10.2 mg, 1 equiv.) in CH₂Cl₂ (0.6 mL) was added Ph₃BiCO₃ (30 mg, 1.2 equiv.) in one portion. The white slurry was stirred at room temperature for 2 h. The solvent was evaporated under vacuum and the crude product was purified by flash chromatography (silica gel, eluent pentane/EtOAc, 1 : 0 to 9 : 1) to afford compound 3a¹⁸ (14.5 mg, 70%) as a white solid. ¹H NMR (CDCl₃): δ 7.53–7.50 (m, 3H), 7.39–7.35 (m, 2H), 7.33–7.23 (m, 7H), 7.18–7.11 (m, 4H), 7.09 (td, J = 7.5, 1.5 Hz, 1H), 7.03 (td, J = 7.5, 1.5 Hz, 1H), 6.97 (dd, J = 7.5, 1.5 Hz, 1H), 5.24 (s, 1H), 4.65 (d, J = 13.0 Hz, 1H), 4.31 (d, J = 13.0 Hz, 1H), 3.43–3.35 (m, 1H), 3.22–3.15 (m, 1H), 3.12–3.05 (m, 1H), 2.97–2.90 (m, 1H) ppm. ¹³C NMR (CDCl₃): δ 147.8, 141.3, 140.9, 138.9, 136.1, 132.4, 130.9, 129.9, 129.6, 129.3, 128.5, 128.4 (2C), 127.8, 127.6, 127.4, 126.9, 126.8, 125.6, 125.4, 110.3, 63.0, 36.8, 32.9 ppm. IR (neat) 1502, 1382, 872, 735 cm⁻¹. MS (ES⁺): 416 [M + H]⁺.

Conclusions

We have demonstrated that hydroxylamine oxidation can be performed using triphenylbismuth carbonate which served as mild, efficient, selective and convenient source of oxidant. The procedure described herein permitted not only the synthesis of nitrones in high yield but also their in situ trapping by strained alkyne via a [3 + 2] cycloaddition reaction. This process could offer an interesting alternative to the development of chemical transformations suitable for chemical biology.²² As a reminder, bismuth is the least toxic of the heavy metals.¹⁷

Acknowledgements

D. V. N. thanks the CEA and “Les Laboratoires Pierre Fabre” for a CTCI PhD grant. P. P. thanks the Enhanced Eurotalents program for support. The “Service de Chimie Bioorganique et de Marquage” belongs to the Laboratory of Excellence in Research on Medication and Innovative Therapeutics (ANR-10-LABX-0033-LERMIT).

Notes and references

1. H. Suzuki, T. Ikegami and Y. Matano, Synthesis, 1997, 249.
2. (a) D. H. R. Barton, J. P. Kitchin, D. J. Lester, W. B. Motherwell and M. T. Barros Papoula, Tetrahedron, 1981, 37, 73; (b) J. P. Kitchin, in Organic Syntheses by Oxidation with Metal Compounds, ed. W. J. Mijs and C. R. H. I. De Jonge, Plenum Press, New York, 1986, p. 817.
3. J. P. Finet, Chem. Rev., 1989, 89, 1487.
4. D. H. R. Barton, J. P. Kitchin and W. B. Motherwell, J. Chem. Soc., Chem. Commun., 1978, 1099.
5. Y. Matano and H. Nomura, Angew. Chem., Int. Ed., 2002, 41, 3028.
6. D. H. R. Barton, D. J. Lester, W. B. Motherwell and M. T. Barros Papoula, J. Chem. Soc., Chem. Commun., 1979, 705.
7. D. H. R. Barton, N. Yadav-Bhatnagar, J. P. Finet, J. Khamsi, W. B. Motherwell and S. P. Stanforth, Tetrahedron, 1987, 43, 323.
8. C. Ménard, E. Doris and C. Mioskowski, Tetrahedron Lett., 2003, 44, 6591.
9. (a) T. Arnauld, D. H. R. Barton and E. Doris, Tetrahedron Lett., 1997, 38, 365; (b) T. Arnauld, D. H. R. Barton and E. Doris, Tetrahedron, 1997, 53, 4137; (c) T. Arnauld, D. H. R. Barton, J. F. Normant and E. Doris, J. Org. Chem., 1999, 64, 6915.
10. For general reviews on the preparation and use of nitrones, see: (a) J. Hamer and A. Macaluso, Chem. Rev., 1964, 64, 473; (b) J. Revuelta, S. Cicchi, A. Goti and A. Brandi, Synthesis, 2007, 485; (c) J. A. Grigoriev, in Nitrile Oxides, Nitrones, and Nitronates in Organic Synthesis, ed. H. Feuer, John Wiley & Sons, New Jersey, 2nd edn, 2008, p. 129; (d) P. Merino, Sci. Synth., 2004, 27, 511; (e) J. Yang, Synlett, 2012, 23, 2293; (f) L. L. Anderson, Asian J. Org. Chem., 2016, 5, 9 For a recent and elegant synthesis of nitrones, see: S. Katahara, S. Kobayashi, K. Fujita, T. Matsumoto, T. Sato and N. Chida, J. Am. Chem. Soc., 2016, 138, 5246.
11. R. Saladino, V. Neri, F. Cardona and A. Goti, Adv. Synth. Catal., 2004, 346, 639.
12. S. Cicchi, M. Corsi and A. Goti, J. Org. Chem., 1999, 64, 7243.
13. S. Cicchi, M. Marradi, A. Goti and A. Brandi, Tetrahedron Lett., 2001, 42, 6503.
14. P. A. S. Smith and S. E. Gloyer, J. Org. Chem., 1975, 40, 2508.
15. C. Matassini, C. Parmeggiani, F. Cardona and A. Goti, Org. Lett., 2015, 17, 4082.
16. G. D'Adamio, C. Parmeggiani, A. Goti and F. Cardona, Eur. J. Org. Chem., 2015, 6541.
17. R. Mohan, Nat. Chem., 2010, 2, 336.
18. C. S. McKay, J. Moran and J. P. Pezacki, Chem. Commun., 2010, 46, 931.
19. D. H. R. Barton, N. Yadav-Bhatnagar, J. C. Blazejewski, B. Charpiot, J. P. Finet, D. J. Lester, W. B. Motherwell, M. T. Barros Papoula and S. P. Stanforth, J. Chem. Soc., Perkin Trans. 1, 1985, 2657.
20. D. H. R. Barton and J. P. Finet, Pure Appl. Chem., 1987, 59, 937.
21. D. H. R. Barton, J. P. Finet, W. B. Motherwell and C. Pichon, Tetrahedron, 1986, 42, 5627.
22. D. A. MacKenzie, A. R. Sherratt, M. Chigrinova, L. L. W. Cheung and J. P. Pezacki, Curr. Opin. Chem. Biol., 2014, 21, 81.

---

Footnote

† Electronic supplementary information (ESI) available: NMR data and copies of the ¹H and ¹³C spectra. See DOI: 10.1039/c6ra18578a

---