Abstract
A Ni-catalyzed halogenation of enol triflates was developed that enables the synthesis of a broad range of alkenyl iodides, bromides, and chlorides under mild reaction conditions. The reaction utilizes inexpensive, bench stable Ni(OAc)2·4H2O as a pre-catalyst and proceeds at room temperature in the presence of sub-stoichiometric Zn and either cod or DMAP.
Keywords: nickel, halogenation, enol triflates, catalysis, alkenyl halides
Graphical Abstract
Alkenyl halides are versatile functional groups that can be used in a variety of carbon–carbon and carbon–heteroatom bond-forming reactions. For example, alkenyl halides are commonly used as substrates in transition metal-catalyzed cross-coupling reactions1 or are converted via metal-halogen exchange to nucleophiles for 1,2-additions to carbonyl compounds (Scheme 1).2 Furthermore, the alkenyl halide moiety appears in some natural products and bioactive molecules.3 Whereas acyclic alkenyl halides are easily prepared from the corresponding alkynes4 or aldehydes,5 most cyclic alkenyl halides are synthesized from the corresponding ketones. A commonly employed method is the Barton alkenyl halide synthesis (and variations thereof),6,7,8 which proceeds through an intermediate hydrazone. These reactions are notoriously capricious: the formation of the requisite hydrazone can be challenging on sterically encumbered substrates and the halogenation step often produces mixtures of alkenyl halide isomers or dihalide side products.9
Scheme 1.
Synthesis and utility of alkenyl halides.
As a result, enol triflates, which can be prepared directly from cyclic ketones under either kinetic or thermodynamic control, have emerged as attractive “pseudohalides” for transition metal-catalyzed cross-coupling processes. Unfortunately, enol triflates cannot be directly converted to the corresponding alkenyllithium or alkenylmagnesium species commonly employed in 1,2-addition reactions. In cases where the Barton alkenyl halide synthesis is poor yielding, a multistep alternative is frequently employed: 1) conversion of the ketone to enol triflate, 2) conversion of the triflate to the stannane, and 3) conversion of the stannane to the halide.10 Direct, mild methods to convert enol triflates to alkenyl halides, without proceeding through organostannane intermediates, can streamline the preparation of these valuable synthons.
Indeed, Buchwald has reported a Pd-catalyzed reaction to convert alkenyl triflates to alkenyl bromides and chlorides;11,12 however, there are no examples of alkenyl iodide formation, and the reaction requires an expensive ligand, temperatures greater than 100 °C, and additives such as fluoride salts or iBu3Al. These additives limit the functional group compatibility of the transformation, particularly with commonly used groups such as silyl ethers. More recently, Hayashi reported a Ru-catalyzed method to convert enol triflates to iodides, bromides, or chlorides that proceeds at ambient temperature; however, the requisite ruthenium catalyst is not commercially available and limited examples of alkenyl iodide formation are reported.13
During our investigations of Ni-catalyzed asymmetric reductive coupling reactions of alkenyl bromides,14 we observed an off-pathway halide exchange process that generated alkenyl chlorides and iodides. Whereas Ni-catalyzed aryl15 and alkenyl halide exchange processes have been previously reported and extensively investigated,15b,16 development of the corresponding reactions of enol triflates have been limited to a single report describing bromination of dihydropyranyl enol triflates.17,18a Having observed promising reactivity with enol triflates in our investigation of asymmetric reductive coupling reactions,19 we hypothesized that an appropriate Ni catalyst and inexpensive halide salts might enable the direct conversion of enol triflates to alkenyl halides under mild conditions. In this communication, we report the development of a Ni-catalyzed triflate-halide exchange (triflex) reaction, which provides access to alkenyl iodides, bromides, and chlorides in good to excellent yields (Scheme 1).
Our reaction development began with enol triflate 1a, prepared in one step from menthone, with the goal of identifying general conditions that could provide the alkenyl iodide, bromide, or chloride simply by changing the halide salt. Initial optimization efforts using Ni(cod)2 as a catalyst revealed that a mixed DMA/THF solvent system and short reaction times were optimal, and that bidentate ligands (phosphine, amine, and pyridine) inhibited the reaction.20 Informed by these results, we screened a variety of Ni(II) pre-catalysts in combination with Zn0 and cod and ultimately found that Ni(OAc)•4H2O afforded promising yields across all three reactions. Whereas the alkenyl bromides and chlorides could be obtained in good yields in 16 h using 5 mol % Ni(OAc2)•4H2O and 10 mol % cod (Method B), the iodination proceeded in modest yield under these conditions (Table 1, entry 9). In an effort to improve the yield of the iodination, an additive screen was conducted. This study revealed that use of 20 mol% DMAP, in conjunction with higher catalyst loadings, (10 mol % Ni(OAc)4, 10 mol% cod) improved the yield of 2a. Since similar yields of 2b and 2c are formed in the presence and absence of DMAP (Table 1, entry 9), this additive was omitted for the standard bromination and chlorination conditions (Method B). The precise role of DMAP in the iodination remains unclear; our hypothesis is that it coordinates ZnI2 salts that are generated as part of the initial Ni(II) reduction. 21,22,23 This is consistent with the observation that addition of exogenous ZnI2 inhibits the reaction, but that reactivity can be restored by addition of DMAP.20 Although 2a could be obtained in good yield with 5 mol % Ni (Table 1, entry 10), incomplete conversions were obtained for many substrates when the scope of the iodination was investigated. Increasing the Ni loading to 10 mol % generally improved the conversion and allowed shorter reaction times, which was necessary to minimize the formation of reduction product (Method A).24
Table 1.
Optimization of reaction conditions.[a]
 | ||||
|---|---|---|---|---|
| Entry | Deviation from Standard Conditions | Yield 2a (%)[b] | Yield 2b (%)[b][c] | Yield 2c (%)[b][c] |
| 1 | None | 87 | 92 | 90 |
| 2 | Ni(acac)2 | 14 | 85 | 90 |
| 3 | Ni(X)2 | 75 | 82 | 0[d] |
| 4 | Ni(cod)2[e] | 70 | 94 | 94 |
| 5 | DMA | 12 | 0 | 0 |
| 6 | THF | 82 | 69 | 40 |
| 7 | DMF | 0 | 18 | 0 |
| 8 | MeCN | 24 | 33 | 8 |
| 9 | -DMAP (I) / + DMAP (Br/Cl) | 33 | 94 | 91 |
| 10 | ½ × Ni(OAc)2•4H2O loading | 90 | 96 | 94 |
| 11 | 2 × Ni(OAc)2•4H2O loading | 87 | 94 | 93 |
Reactions conducted in 1:3 DMA/THF (0.25M) under inert atmosphere on 0.1 mmol scale.
Determined by 1H NMR spectroscopy versus 1,2,4,5-tetrachloronitrobenzene as an internal standard.
Products 2b and 2c were found to be volatile.
NiCl2 was insoluble under these conditions.
No Zn or additional cod were added.
Having identified satisfactory reaction conditions, the substrate scope of the Ni-catalyzed halogenation reaction was investigated (Table 2). The halide exchange was found to be compatible with a variety of common functional groups, including amines (4), carbamates (5, 13), pyridines (20), alkenes (12), dienes (10, 14), esters (19), ketals (6, 14), and enones (11). Chemoselective halogenation of alkenyl triflates was observed in preference to aryl triflates (15, 20), aryl chlorides (9, 21), and aryl boronates (22); however, competitive halide exchange was observed in the presence of aryl bromides and iodides.20
Table 2.
Scope of alkenyl halides.[a]
 |
|---|
 |
Reactions conducted under inert atmosphere on 0.3 mmol scale. Isolated yields. Yields in parentheses were determined by 1H NMR spectroscopy versus 1,2,4,5-tetrachloronitrobenzene as an internal standard.
5 equiv NaI was used in the absence of cod, 36 h reaction time.
Enol nonaflate was employed instead of enol triflate.
Method C: Ni(cod)2 (10 mol %), MX (1.5 equiv), 25% DMA/THF (0.25M), 23 °C, 16 h.
1 mmol scale, benchtop protocol.
DMAP (10 mol %) was used instead of cod.
Reaction was conducted on 0.2 mmol scale.
Reaction was conducted at 0.125 M due to poor solubility of enol triflate.
Although the Ni-catalyzed halogenation exhibits good functional group tolerance, the iodination, bromination, and chlorination did not perform equally well. The iodination conditions (Method A) proved optimal for roughly half of the substrates. In cases where iodination was sluggish (6, 8, 10, 11, 13-16), longer reaction times and 5.0 equiv NaI improved conversions; however, the improved reactivity was often accompanied by increased amounts of protodetriflation (or protodehalogenation, approximately 5–10% yield), and the alkene could be difficult to chromatographically separate from the alkenyl iodides (8a, 10a, 11a, 13a, 15a). For a subset of these substrates (10, 11, 13), the competing reduction was eliminated by use of Ni(cod)2 in conjunction with the enol nonaflate, which afforded access to the pure alkenyl iodides.18
The bromination and chlorination reactions were generally more efficient and robust. For most substrates, complete conversions were achieved with 5 mol% Ni and without the need for DMAP, although addition of DMAP improved the yields for substrates where incomplete conversion was observed with only cod (14-16). In addition, for 1-arylvinyl triflates (19-22, Table 2) the use of Ni(cod)2 (Method C) provided cleaner reaction profiles for the bromination and iodination reactions.25 To demonstrate the scalability of the reaction, bromide 13b was prepared in 95% yield on 1 mmol scale using a benchtop setup (13b).
At this time, the mechanism of the reaction remains unclear. Preliminary investigations of the iodination using Ni(cod)2 as the catalyst revealed that the reaction of 1a exhibits an induction period at low [Ni] (e.g. 0.5 mol %, 1 mol %, see Scheme 2a).26 Plotting Vmax vs. [Ni] revealed that the reaction has a positive-order dependence on [Ni] that negatively deviates from first order at higher [Ni], suggesting the formation of dimeric (or higher order) off-cycle species at higher [Ni].27 No change in the rate of iodination of 1a is observed when the amount of NaI is increased beyond 1 equiv, however the rate dependence on [1a] was more complex.20 One possibility is that the induction period is required to form an active NiI catalyst. During the reaction, a signal is present in the EPR spectrum that is consistent with a NiI-X species. However, quantification of this signal determined that it was only 2% of the total [Ni], and additional experiments to identify the resting state of the catalyst have been inconclusive.
Scheme 2.
Mechanistic investigation.[a]
[a] Reactions conducted under inert atmosphere on 0.2 mmol scale. [b] Determined by GC analysis versus undecane as an internal standard.
A crossover experiment was designed to evaluate the reversibility of triflate-halide exchange: treatment of a 1:1 mixture of 24-OTf and 25-Br with Ni(cod)2 (10 mol %) in 1:3 DMA/THF at 23 °C resulted in complete recovery of 24-OTf and 25-Br, without detection of crossover products 25-OTf or 24-Br (Scheme 2b). Addition of 0.1 or 1.0 equiv LiBr resulted in conversion of 24-OTf to 24-Br in 10% and 90% yield, respectively;28,29 no 25-OTf was detected at any point in either reaction. Monitoring the reaction by 1H NMR spectroscopy confirms that no oxidative addition of 24-OTf occurs in the absence of halide salt. Subjection of alkenyl iodide 25-I to Ni(cod)2 (10 mol %) and metal triflate salts (e.g. NaOTf) did not result in enol triflate formation (Scheme 2c).30 Taken together, these experiments suggest that oxidative addition of the alkenyl triflate is irreversible, or that halide exchange for triflate in the oxidative addition complex is rapid and irreversible. In either scenario, the fact that the enol triflate is irreversibly consumed enables the reaction to proceed in good yield to the respective alkenyl halides. This is in contrast to Ni-catalyzed halide exchange reactions, which are thermodynamically driven equilibrium processes.15b For example, after 2 h, an 85:15 mixture of 25-Br:25-I is obtained for both the Ni-catalyzed reactions of 25-Br with LiI, or 25-I with LiBr.20
In conclusion, a mild Ni-catalyzed halogenation of alkenyl triflates has been developed. By modifying the halide salt, alkenyl iodides, bromides, or chlorides can be obtained using a simple, inexpensive catalyst system. These reactions proceed at room temperature, afford the alkenyl halides in good to excellent yields, and exhibit good functional group tolerance.
Supplementary Material
Acknowledgements
We thank the following Caltech staff for their help: Dr. Scott Virgil and the Caltech Center for Catalysis and Chemical Synthesis for access to experimental and analytical equipment; Dr. Mona Shahgholi and Naseem Torian for assistance with mass spectrometry measurements; and Dr. Paul Oyala for assistance with EPR experiments. We also thank Jordan C. Beck for assistance in the preparation of alkenyl triflates 1e and 1k. Fellowship support was provided by the National Science Foundation (graduate research fellowship to J. L. H. and K. E. P, Grant No. DGE-1144469). S. E. R. is a Heritage Medical Research Institute Investigator. Financial support from the NIH (R35GM118191–01) is gratefully acknowledged.
Contributor Information
Julie L. Hofstra, The Warren and Katharine Schlinger Laboratory of Chemistry and Chemical Engineering, Division of Chemistry and Chemical Engineering, California Institute of Technology Pasadena, CA 91125 (USA).
Kelsey E. Poremba, The Warren and Katharine Schlinger Laboratory of Chemistry and Chemical Engineering, Division of Chemistry and Chemical Engineering, California Institute of Technology Pasadena, CA 91125 (USA).
Alex M. Shimozono, The Warren and Katharine Schlinger Laboratory of Chemistry and Chemical Engineering, Division of Chemistry and Chemical Engineering, California Institute of Technology Pasadena, CA 91125 (USA).
Sarah E. Reisman, The Warren and Katharine Schlinger Laboratory of Chemistry and Chemical Engineering, Division of Chemistry and Chemical Engineering, California Institute of Technology Pasadena, CA 91125 (USA).
References
- 1. Metal-catalyzed Cross-Coupling Reactions; [Google Scholar]; (a) De Meijer A, Diederich F, Eds.; Wiley-VCH Verlag GmbH & Co. KGaA: 2004; Vols. 1 & 2. [Google Scholar]; (b) Nicolaou KC, Bulger PG, Sarlah D, Angew. Chem., Int. Ed 2005, 44, 4442. [DOI] [PubMed] [Google Scholar]
- 2.M. Schlosser, Ed.; John Wiley & Sons, Inc.: 2013.
- 3.(a) G. W. Gribble, Springer-Verlag: Wien, Austria, 1996. [Google Scholar]; (b) G. W. Gribble, Springer-Verlag: Wien, Austria, 2010. [Google Scholar]; (c) Gribble GW, Mar. Drugs 2015, 13, 4004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.(a) Hart DW, Blackburn TF, Schwartz J, J. Am. Chem. Soc 1975, 97, 679. [Google Scholar]; (b) Brown HC, Subrahmanyam C, Hamaoka T, Ravindran N, Bowman DH, Misumi S, Unni MK, Smayaji V, Bhat NG, J. Org. Chem 1989, 54, 6068. [Google Scholar]; (c) Gao F, Hoveyda AH, J. Am. Chem. Soc 2010, 132, 10961. [DOI] [PMC free article] [PubMed] [Google Scholar]; (d) Uehling MR, Rucker RP, Lalic G, J. Am. Chem. Soc 2014, 136, 8799. [DOI] [PubMed] [Google Scholar]; (e) Derosa J, Cantu AL, Boulous MN, O’Duill ML, Turnbull JL, Liu Z, De La Torre DM, Engle KM, J. Am. Chem. Soc 2017, 139, 5183. [DOI] [PubMed] [Google Scholar]
- 5.(a) Takai K, Nitta K, Utimoto K, J. Am. Chem. Soc 1986, 108, 7408. [DOI] [PubMed] [Google Scholar]; (b) Stork G, Zhao K, Tetrahedron Lett. 1989, 30, 2173. [Google Scholar]
- 6.(a) Barton DHR, O’Brien RE, Sternhell S, J. Chem. Soc 1962, 470 Modified Barton procedure: [Google Scholar]; (b) Furrow ME, Myers AG, J. Am. Chem. Soc 2004, 126, 5436. [DOI] [PubMed] [Google Scholar]; (c) Ojha DP, Prabhu KR, Org. Lett 2015, 17, 18. [DOI] [PubMed] [Google Scholar]
- 7. Alkenyl bromides and chlorides can be prepared from ketones using in situ generated P(V) reagents. However, ketones bearing multiple enolizable positions or strained ring systems afford mixtures of isomers: [Google Scholar]; Spaggiari A, Vaccari D, Davoli P, Torre G, Prati F, J. Org. Chem 2007, 72, 2216. [DOI] [PubMed] [Google Scholar]
- 8. For a discussion of the challenges associated with hydrazone synthesis, see ref 5b, and references found therein. [Google Scholar]; (a) Stemke JE, Chamberlin AR, Bond FT, Tetrahedron Lett. 1976, 34, 2947. [Google Scholar]; (b) Chamberlin AR, Stemke JE, Bond FT, J. Org. Chem 1978, 43, 147 For a review, see: [Google Scholar]; (c) Adlington RM, Barrett AGM, Acc. Chem. Res 1983, 16, 55. [Google Scholar]
- 9. For a discussion of the challenges associated with hydrazone synthesis, see ref 5b, and references found therein.
- 10. For a recent example of this sequence in total synthesis, see: [Google Scholar]; Huters AD, Quasdorf KW, Styduhar ED, Garg NK, J. Am. Chem. Soc 2011, 133, 15797. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.(a) Shen X, Hyde AM, Buchwald SL, J. Am. Chem. Soc 2010, 132, 14076. [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Pan J, Wang X, Zhang Y, Buchwald SL, Org. Lett 2011, 13, 4974. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. The conversion of enol triflates to alkenyl iodides using anhydrous magnesium iodide and triethylamine has been reported; however, forcing conditions were required, the scope was limited to four substrates, and alkenyl chlorides and bromides were not prepared. [Google Scholar]; Martinez AG, Alvarez RM, Fraile AG, Subramanian LR, Hanack M, Synthesis 1986, 3, 222. [Google Scholar]
- 13.(a) Shirakawa E, Imazaki Y, Hayashi T, Chem. Commun 2009, 5088. [DOI] [PubMed] [Google Scholar]; (b) Imazaki E, Shirakawa E, Ueno R, Hayashi T, J. Am. Chem. Soc 2012, 134, 14760. [DOI] [PubMed] [Google Scholar]
- 14.(a) Cherney AH, Reisman SE, J. Am. Chem. Soc 2014, 136, 14365. [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Suzuki N, Hofstra JL, Poremba KE, Reisman SE, Org. Lett 2017, 19, 2150. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.(a) Takagi K, Hayama N, Okamoto T, Chem. Lett 1978, 191. [Google Scholar]; (b) Tsou TT, Kochi JK, J. Org. Chem 1980, 45, 1930. [Google Scholar]
- 16.(a) Takagi K, Hayama N, Inokawa S, Chem. Lett 1978, 1435. [Google Scholar]; (b) Hooijdonk MCJM, Peters THA, Vasilevsky SF, Brandsma L, Syn. Commun 1994, 24, 1261 For reviews, see: [Google Scholar]; (c) Petrone DA, Ye J, Lautens M, Chem. Rev 2016, 116, 8003. [DOI] [PubMed] [Google Scholar]; (d) Evano G, Nitelet A, Thilmany P, Dewez DF, Front. Chem 2018, 6:114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.(a) Milne JE, Jarowicki K, Kocienski PJ, Synlett, 2002, 4, 607. [DOI] [PubMed] [Google Scholar]; (b) The authors note that chlorination and iodination is also feasible, but no yields are reported. Application of these conditions to substrate 1a provided products 2a, 2b, and 2c in <10% yield. See the Supporting Information. [Google Scholar]
- 18. A Ni-catalyzed trifluoromethylthiolation of alkenyl triflates was recently reported: [Google Scholar]; Dürr A, Yin G, Kalvet I, Napoly F, Schoenebeck F, Chem. Sci 2016, 7, 1076 Use of an enol nonaflate in place of the enol triflate improved the yield for one example in the above report. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Hofstra JL, Cherney AH, Ordner CM, Reisman SE, J. Am. Chem. Soc 2018, 140, 139. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.See Supporting Information.
- 21. Zn(II) inhibition of Co and Ni mediated reactions has recently been reported. [Google Scholar]; (a) Zhang X. and McNally A, ACS Catal. 2019, 9, 4862. [DOI] [PMC free article] [PubMed] [Google Scholar]; (b) Huang L, Ackerman LKG, Kang K, Parsons AM, Weix DJ, J. Am. Chem. Soc 2019, 141, 10978. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. A XantphosNi(Br)•ZnBr2 adduct has recently been isolated from reduction of the corresponding Ni(II) complex with Zn, demonstrating that Zn salts can alter the Ni speciation. [Google Scholar]; Dicciani J, Hu CT, Diao T, Angew. Chem. Int., Int. Ed 2019, in press. doi: 10.1002/ange.201906005 [DOI] [Google Scholar]
- 23. DMAP is used as an additive in aryl and alkenyl cyanation reactions using Zn(CN)2. [Google Scholar]; (a) Zhang X, Xia A, Chen H, Liu Y, Org. Lett 2017, 19, 2118. [DOI] [PubMed] [Google Scholar]; (b) Gan Y, Wang G, Xie X, Liu Y, J. Org. Chem 2018, 83, 14036. [DOI] [PubMed] [Google Scholar]
- 24.Reduction was observed in up to equimolar amounts with respect to Ni loading; the amount of reduction observed was also dependent on substrate and reaction time. It is possible that the reduction product arises from the initially formed alkenyl halide and not the enol triflate directly.
- 25. The treatment of 1-arylvinyl triflates with LiCl results in a competitive, non-Ni-mediated elimination to afford the aryl acetylenes. [Google Scholar]; Yang, Wu D, Lu Z, Sun H, Li A, Org. Biomol. Chem 2016, 14, 5591–5594. [DOI] [PubMed] [Google Scholar]
- 26.Addition of exogenous cod inhibits the rate of the reaction, indicating that ligand dissociation could also be the source of the induction period.
- 27.Blackmond DG, J. Am. Chem. Soc 2015, 137, 10852. [DOI] [PubMed] [Google Scholar]
- 28. No evidence of oxidative addition is observed when enol triflate 1a and Ni(cod)2 (5 mol %) were mixed in the presence or absence of NaOTf. However, upon addition of a metal halide salt, conversion of 1a to the alkenyl halide (2a-c) was observed concomitantly with the formation of a Ni(I)-X species. The same Ni(I)-X species were detected upon mixing 2a-c and Ni(cod)2 (5 mol %). Oxidative addition of aryl triflates to (dppf)Ni(cod) was previously shown to result in Ni(I) complexes via comproportionation: [Google Scholar]; (a) Bajo S, Laidlaw G, Kennedy AR, Sproules S, Nelson DJ, Organomet. 2017, 36, 1662–1672. [Google Scholar]; (b) Kalvet I, Guo Q, Tizzard GJ, Schoenebeck F, ACS Catal. 2017, 7, 2126. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Relevant mechanistic studies: [Google Scholar]; Beromi MM, Nova A, Balcells D, Brasacchio AM, Brudvig GW, Guard LM, Hazari N, Vinyard DJ, J. Am. Chem. Soc 2017, 139, 922. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.The reaction was sampled every 5 minutes, but at no time was enol triflate 25-OTf detected.
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.