Iridium-Catalyzed Hydrochlorination and Hydrobromination of Alkynes by Shuttle Catalysis.

Described herein are two different methods for the synthesis of vinyl halides by a shuttle catalysis based iridium-catalyzed transfer hydrohalogenation of unactivated alkynes. The use of 4-chlorobutan-2-one or tert-butyl halide as donors of hydrogen halides allows this transformation in the absence of corrosive reagents, such as hydrogen halides or acid chlorides, thus largely improving the functional-group tolerance and safety profile of these reactions compared to the state-of-the-art. This method has granted access to alkenyl halide compounds containing acid-sensitive groups, such as tertiary alcohols, silyl ethers, and acetals. The synthetic value of those methodologies has been demonstrated by gram-scale synthesis where low catalyst loading was achieved.

Introduction

Alkenyl halides have found widespread application in organic synthesis.1 Among them, vinyl chlorides have gained increasing attention in recent years because of their significant synthetic utility and occurrence in natural products, pharmaceuticals, and agrochemicals.2 Moreover, vinyl chlorides have emerged as efficient coupling partners in cross‐coupling reactions, such as the Suzuki–Miyaura coupling reaction3 and the Buchwald–Hartwig amination.4 Thus, the development of methods for the preparation of vinyl chlorides is of great synthetic significance. Vinyl chlorides are generally accessed from either carbonyl compounds5 or alkynes.6 Although a catalytic process using carbonyl compounds as starting materials has been realized,5h the requirement for toxic phosphorus reagents like PCl5 5a, 5b or POCl3 5d or a large excess of CrCl2 5c limits their applications in complex molecule synthesis.

The catalytic synthesis of vinyl chlorides from broadly accessible alkynes is a powerful alternative to the traditional synthetic methods. A first approach, carbochlorination, the addition of a C−Cl bond across an alkyne, mostly involves acid chlorides as reagents.6 The second approach, alkyne hydrochlorination,7 a reaction which requires otherwise harsh reaction conditions in the absence of a catalyst, provides a powerful and complementary strategy to transform alkynes into vinyl chlorides. However, most of the catalytic hydrochlorination methods rely on corrosive and acidic HCl as a reagent,8 leading to a limited functional‐group tolerance (Scheme 1). For example, Dérien and co‐workers have reported a ruthenium‐catalyzed hydrochlorination of alkynes with HCl (Scheme 1 a),9 and Hammond, Xu, and co‐workers have recently employed HCl (as a DMPU/HCl adduct) in a gold‐catalyzed reaction (Scheme 1 b).10 However, the use of HCl as a reagent inherently limits the functional‐group tolerance of these methods. A palladium‐catalyzed directed anti‐hydrochlorination of unactivated alkynes has recently been realized by Engle and co‐workers (Scheme 1 c).11 Through the introduction of a bidentate directing group, they could efficiently control the regioselectivity of the transformation. This method is, however, limited to the synthesis of vinyl chloride products that have a suitably positioned amino group for the introduction of the directing group. The use of an acid chloride as an HCl precursor also limits the functional‐group tolerance of this reaction. Thus, the discovery of a functional group tolerant catalytic hydrochlorination still remains a challenge in organic synthesis. Herein, we report an iridium‐catalyzed transfer hydrochlorination and hydrobromination of alkynes using 4‐chlorobutan‐2‐one and tert‐butyl chloride/bromide as a formal HX donor, a feature that allows an unusually broad functional‐group tolerance in the synthesis of vinyl halide species (Scheme 1 d). Furthermore, it also represents a rare example of hydrofunctionalization of unactivated internal alkynes by a simple iridium catalyst.

Scheme 1

Context of the work.

Results and Discussion

Evaluation of the Halide Donor

Recently, our group reported several reactions that addressed safety problems associated with the use of hazardous and toxic reagents, such as HCN‐free hydrocyanation,12 cyanide‐free cyanation of aryl chloride,13 and CO/HCl‐free hydrochlorocarbonylation.14 These reactions proceed by a shuttle catalysis15a, 15b paradigm wherein a chemical moiety is transferred between two stable organic molecules. Accordingly, we sought a suitable catalytic system that could enable an HCl molecule to be formally transferred between a simple alkyl chloride and an alkyne without any direct use of corrosive HCl. We initially explored reagents that are similar in structure to those previously employed in other examples of shuttle catalysis by reaction with a simple aliphatic terminal alkyne (1; see Table 1).15 We started our investigations with iridium catalysts because of their expected high reactivity toward the oxidative addition of C−X bonds.16 Unfortunately, both isobutyl chloride (entry 1) and butyl chloride (entry 2) did not show any reactivity, probably because of the inertness of the C−Cl bond. Interestingly, the use of tert‐butyl chloride (2 c) as the HCl donor gave trace amounts of a vinyl chloride product (3) under these reaction conditions (entry 3), albeit together with some decomposition product. We reasoned that the installation of a polar group at a neighbouring position on the reagent backbone could possibly facilitate the initial activation step through either coordination of a metal and/or electronic activation of the C−Cl bond. Through the evaluation of several reagents and catalysts (see the Supporting Information for the optimization), we found 4‐chlorobutan‐2‐one (2 d), which contains an acyl group on the β‐carbon center relative to the chlorine atom, as a suitable donor of HCl for the transfer hydrochlorination of pent‐4‐yn‐1‐ylbenzene (1). The ideal catalytic system for this transformation emerged as a combination of [IrCl(cod)]2 as a metal precursor and CPhos as a ligand at 80 °C in toluene, giving rise to 3 as a single regioisomer in 85 % yield (entry 4). To the best of our knowledge, this is the first example of an iridium‐catalyzed hydrochlorination reaction. Moreover, the alkene by‐product, methyl vinyl ketone, was obtained in 87 % yield, a result which strongly suggests that the reaction proceeds through shuttle catalysis. A similar reagent 1‐chloropentan‐3‐one (2 e) also showed high reactivity under the same reaction conditions, leading to 3 in 75 % yield (entry 5). Among those reagents bearing various carbonyl groups, only the one containing a carboxylic acid (2 g) showed limited reactivity in this hydrochlorination reaction, delivering 3 in 15 % yield (entry 7). As expected, the reagent lacking β‐hydrogen atoms did not react under the reaction conditions (entry 8). A reagent bearing the acyl group on the γ carbon to the chlorine atom (2 h) was also tested in the reaction. However, only trace amounts of product were detected, further demonstrating the beneficial effect of the acyl group on the reactivity of the chloride reagent (entry 9).

Table 1

Exploring the reagent for the transfer hydrochlorination of alkynes.[a]

[a] 1 (0.1 mmol), 2 (0.2 mmol), [IrCl(cod)]2 (2.5 mol %), CPhos (7.5 mol %), and toluene (0.25 mL) at 80 °C for 5 h. [b] Yields determined by NMR spectroscopy using dibromomethane as an internal standard. [c] 2 c (1.0 mmol), toluene (0.5 mL) at 110 °C, 12 h. cod=1,5‐cyclooctadiene, CPhos= 2′‐(dicyclohexylphosphanyl)‐N ,N ,N ,N ‐tetramethyl‐[1,1′‐biphenyl]‐2,6‐diamine.

Hydrochlorination using 4‐Chlorobutan‐2‐One

With 4‐chlorobutan‐2‐one (2 d) as the optimized reagent, we then moved to explore the substrate scope of this iridium‐catalyzed transfer hydrochlorination reaction. As illustrated in Table 2, a wide variety of alkyne derivatives could successfully undergo hydrochlorination reaction using our system to give the corresponding vinyl chlorides in good yield. Initially, we focused on the evaluation of aliphatic terminal alkynes. A broad range of functional groups, including nitriles (8), chlorides (9), esters (10), amines (12), nitro (16), aryl chloride (17), aryl fluorides (18), iodide (19), aldehyde (22), and ketone (29) were tolerated. Likewise, α,β‐unsaturated esters, which serve as an important building block in organic synthesis, were tolerated under the reaction conditions (15). Interestingly, a substrate bearing two terminal alkyne moieties could undergo successful hydrochlorination reaction giving the product containing two vinyl chloride moieties (7). To our delight, acid‐sensitive functional groups such as tertiary alcohols (13), silyl ethers (14), and acetals (23), which are usually not tolerated by other methods employing hydrochloric acid as a reagent, remained untouched during the reaction, showcasing the excellent functional‐group tolerance of our methodology. Alkynes containing heterocycles such as pyridine and thiophene reacted smoothly, leading to the corresponding vinyl chlorides in 83 and 78 % yield, respectively (20, 21). It is worth noting that alkynes derived from natural products such as estrone and cholic acid could also be efficiently converted into the corresponding vinyl chlorides in high yields (29, 30), further demonstrating the diversity and practicality of this novel protocol. An aromatic terminal alkyne was subsequently evaluated and improved yield of product were obtained using Ruphos as the ligand (24). Next, we were pleased to find that slightly modifying our protocol, both a symmetric diarylethyne (25) and an asymmetric internal alkyne (26, 28) could be converted into the corresponding vinyl chlorides in good selectivities. In the case of symmetric aliphatic internal alkynes, excellent yield of product was obtained, but in a low Z/E ratio of 64:36 (27). In contrast, an ester‐substituted alkyne led to product formation with high yield, and regio‐ and stereoselectivity (26).

Table 2

Scope of iridium‐catalyzed transfer hydrochlorination of alkynes with 4‐chlorobutan‐2‐one.[a]

[a] Alkyne (0.2 mol), 2 d (0.4 mmol, 2.0 equiv), [IrCl(cod)]2 (2.5 mol %), CPhos (7.5 mol %), and toluene (0.5 mL) at 80 °C for 5 h. Yields of isolated products given. [b] 2 d (4.0 equiv). [c] CPhos (15 mol %). [d] [IrCl(cod)]2 (5 mol %), tBu‐Xantphos (20 mol %), 12 h. [e] Ruphos (7.5 mol %), 12 h. [f] 110 °C, 12 h.

Hydrochlorination using tert‐Butyl Chloride

Although the method using 4‐chlorobutan‐2‐one showed broad applicability and enabled the hydrochlorination without the use of corrosive reagent, its high price and the generation of highly reactive methyl vinyl ketone as a by‐product can possibly limit its synthetic applicability. Thus, the discovery of more sustainable donors of HCl still remains highly significant. As shown in Table 1, the use of tert‐butyl chloride as a donor of HCl can produce the vinyl chloride product in a trace amount under the reaction conditions with more equivalents of the reagent, higher temperature, and longer reaction time (entry 3). Encouraged by this result, we explored the reactivity of tert‐butyl chloride in the hydrochlorination of alkynes because of its higher availability and lower price than 4‐chlorobutan‐2‐one. Furthermore, the use of tert‐butyl chloride as a donor of HCl produces a nontoxic and benign by‐product, isobutene. After screening of reaction conditions (see details in the Supporting Information), we found the optimal system for hydrochlorination reaction of internal alkynes using tert‐butyl chloride under ligand‐free conditions with [IrCl(cod)]2 as a catalyst at 110 °C. A range of internal alkynes, either symmetric or asymmetric, aliphatic or aromatic, activated or unactivated, were successfully converted into the corresponding vinyl chlorides in good to excellent yields even without using any external ligands (Table 3, 25, 26, 27, 31–37). Regarding the symmetric internal alkynes, diarylethynes bearing either electron‐donating (32, 33) or electron‐withdrawing groups (25, 34, 35, 36) were well tolerated, reacting with tert‐butyl chloride to give vinyl chlorides in excellent yields with good E/Z selectivity. The symmetric aliphatic internal alkyne also reacted to afford the corresponding vinyl chlorides with high efficiency and moderate Z/E ratio (27). Besides, some asymmetric internal alkynes also showed good reactivity in the transformation, being converted into the vinyl chlorides in good yields and selectivities (26, 28, 37). Terminal alkynes, 5‐phenyl‐1‐pentyne, could also undergo hydrochlorination to give 3 in 42 % yield, albeit tris(pentafluorophenyl)phosphine is required to successfully promote this transformation.

Table 3

Scope of iridium‐catalyzed transfer hydrochlorination of alkynes with tert‐butyl chloride.[a]

[a] Alkyne (0.2 mol), 2 c (2.0 mmol, 10.0 equiv.), [IrCl(cod)]2 (2.5 mol %), and toluene (1.0 mL) at 110 °C for 12 h. Yields of isolated products are given. [b] Tris(pentafluorophenyl)phosphine (7.5 mol %). [c] 5 h.

Hydrobromination using tert‐Butyl Bromide

Catalytic hydrobromination of alkynes represents an efficient approach to synthesize vinyl bromides. However, current reaction conditions for this process mainly rely on the use of either corrosive and gaseous HBr or in situ generated HBr,17 which is not ideal for laboratory‐scale synthesis. Recently, an alternative method using a transfer hydrofunctionalization strategy was reported by the groups of Lautens18 and Oestreich.19 They achieved the hydrobromination of 1,6‐enynes and alkynes through the transfer of HBr from Et3N⋅HBr and 1‐(2‐bromoethyl)‐1,4‐dihydro‐1,1′‐biphenyl, respectively. In this context, we were delighted to see that our protocol for the transfer hydrochlorination using tert‐butyl chloride could also be applied to the transfer hydrobromination using tert‐butyl bromide as the source of HBr. We then decided to evaluate the scope and the potential of this transformation. As shown in Table 4, an aliphatic terminal alkyne could successfully undergo the hydrobromination reaction to give the branched vinyl bromide 48 in moderate yield. Internal alkynes, including symmetric diarylethynes and dialkylethynes, and several asymmetric substrates, were also converted into the corresponding vinyl bromides (38–47) in good to excellent yields.

Table 4

Scope of iridium‐catalyzed transfer hydrobromination of alkynes with tert‐butyl bromide.[a]

[a] Alkyne (0.2 mol), 2 j (2.0 mmol, 10.0 equiv), [IrCl(cod)]2 (2.5 mol %), and toluene (1.0 mL) at 110 °C for 12 h. Yields of isolated products are given. [b] Tris(pentafluorophenyl)phosphine (7.5 mol %).

Application on Preparative Scale

To demonstrate the robustness and applicability of those methodologies, we performed several scale‐up experiments (Scheme 2). To our delight, both methods can be applied to a gram‐scale synthesis with high efficiency. Furthermore, in the specific case of the methodology based on tert‐butyl halide, we reduced the catalyst loading to 0.2 mol % with a turnover number (TON) of at least 480. This observation demonstrates the simplicity, efficiency, and the synthetic value of this protocol.

Scheme 2

Gram‐scale experiments.

Mechanistic Study

The lack of previously reported examples of iridium‐catalyzed hydrohalogenation reactions raises questions regarding the mechanism of our processes. Thus, we became interested in investigating the mechanism of our novel methodologies performing both stoichiometric and catalytic experiments. We started evaluating the reactivity of the iridium catalyst, the phosphine ligand and the first reagent, 4‐chlorobutan‐2‐one (2 d). After forming the Ir‐phosphine in situ, which was confirmed by NMR spectroscopy, we added a stoichiometric amount of 4‐chlorobutan‐2‐one and monitored the reaction over time and at different temperatures (see the Supporting Information). Upon heating at 80 °C, trace amounts of methyl vinyl ketone were observed together with formation of a new phosphorous compound, which became the major product after 24 hours. Unfortunately, despite our efforts, this product could not be isolated. Interestingly, after 3 hours, trace amounts of an Ir‐H species were observed (see the Supporting Information) which, however, subsequently disappeared upon further heating. Next, we investigated whether a different behavior was taking place with the other class of reagents, tert‐butyl chloride and bromide. Reacting these species at 80 °C with [Ir(cod)Cl]2 led, this time, to a significant formation of isobutene (see the Supporting Information). In all the cases, no significant amounts of an Ir−H species were observed, raising questions about the possible intermediacy and stability of such species under our reaction conditions.

As we could not gather further information from the stoichiometric experiments, we wondered whether the reagent could just serve as a donor of HCl which then could be oxidatively added to the iridium and generate the active intermediate. Replacing our reagents, 4‐chlorobutan‐2‐one or tert‐butyl chloride, with an HCl solution in ether, led to product formation with similar conversion and yield (Scheme 3 a). Besides being a new synthetic method on its own, this result suggests that both reactions could proceed through a similar Cl‐Ir‐H intermediate. Interestingly, and in contrast to the normal reaction conditions, the reaction with HCl could even be run at room temperature in the absence of the phosphine ligand. These results suggest that the most challenging (and slower) part of the catalytic cycle in the shuttle process, the one which requires higher temperature and an electron‐rich phosphine ligand or only higher temperature, is the activation of either 4‐chlorobutan‐2‐one or tert‐butyl halide by the iridium catalyst (Scheme 3 b). To test the thermal stability of those halide reagents and any possible in situ formation of HCl, we performed the reaction in the absence of the catalyst and it showed no significant conversion of either chlorobutan‐2‐one or the tert‐butyl halide, as well as no formation of the desired vinyl halide product. This result, combined with the high tolerance of the reaction to acid‐sensitive groups, makes the in situ generation of significant amounts of free HCl highly unlikely (Scheme 3 c).

Scheme 3

Control experiments.

To gain more information regarding the actual alkyne addition step, a deuterium‐labelled alkyne substrate was synthesized and subjected to the reaction conditions (Scheme 4 a). Interestingly, a nearly equimolar mixture of the two possible isomers was obtained with both methodologies. This result was later confirmed by another labelling experiment wherein DCl was employed with a non‐labelled alkyne substrate, which led to the formation of a similar ratio of both isomers. However, as expected, the other isomer was formed as the major product (Scheme 4 b). While an initial Ir−H migratory insertion cannot be excluded at this stage, the results obtained, combined with the known propensity of RhIII−Cl and IrIII−Cl species to undergo rapid M−Cl insertion into unsaturated substrates,6t support a mechanism where an initial chloroiridation step occurs through competing syn and anti‐addition pathways. In further support of this rationale, we treated the reaction with an exogenous source of Cl anion (Scheme 4 a). As expected, a significant change in the observed ratio in favor of the anti‐addition product was observed.

Scheme 4

Deuterium‐labelling experiments.

Finally, we performed an experiment using a fully deuterated tert‐butyl chloride (98 % D) as the reagent to determine the level of deuterium incorporation in the product (Scheme 5 a). Interestingly, only 85 % incorporation of deuterium in the product was observed, probably the result of a side reaction with the cod ligand on the iridium catalyst. This side reaction might proceed through rapid and reversible Cl‐Ir‐D insertion into the alkene groups of cod, a process that could result in the loss of deuterium. To get evidence for the side reaction, an experiment using an equimolar mixture of diphenylacetylene and cyclooctene as the substrate was performed (Scheme 5 b). As expected, the deuterated cyclooctene was detected by deuterium NMR spectroscopy, a result which suggests the presence of a short‐lived Ir‐D species under the catalytic reaction conditions.

Scheme 5

Deuterium‐labelling experiments.

Conclusion

In conclusion, we have presented an iridium‐catalyzed transfer hydrohalogenation reaction which proceeds through a shuttle catalysis strategy. The method uses chlorobutan‐2‐one as an HCl surrogate, enabling this milder protocol to tolerate the widest range of functional groups reported to date in a hydrochlorination reaction. Alternatively, a protocol using inexpensive tert‐butyl chloride can be used with decreased catalyst loadings. The latter transformation can also be efficiently extended to hydrobromination reactions. In a broader context, this work highlights both the possibility to use the shuttle catalysis concept to elude the use of mineral acids, and the previously untapped potential of iridium complexes to catalyze hydrohalogenation reactions. Despite our extensive efforts, the isolation of relevant catalytic intermediates was hampered by the short life of those iridium species. Thus, further studies will be necessary to unravel the mechanism of these novel catalytic reactions.

Conflict of interest

The authors declare no conflict of interest.

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re‐organized for online delivery, but are not copy‐edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

Supplementary

Click here for additional data file.