Key words
tetrahydroxydiboron - transfer hydrogenation - palladium catalysis - alkene reduction - deuterium - catalytic hydrogenation
Transition-metal-catalyzed hydrogenation is one of the most important reactions in synthetic chemistry and is widely used and studied in both industry and academia.[1] Catalytic hydrogenation is often accomplished by direct application of H2 gas, which is formally ‘byproductless’, but safety and environmental concerns arise due to H2 production, transportation, storage, and handling. To avoid these and other inconveniences associated with the direct application of H2, transfer hydrogenation (TH) allows in situ generation of stoichiometric H2, with the expense of stoichiometric byproduct generation from the transfer reagent.[2] Longstanding transfer reagents include formic acid,[3] primary alcohols,[4] ammonia borane,[5] and silanes.[6]
Since 2016, we and others have developed methods for transition-metal-catalyzed transfer hydrogenation and hydrogenolysis reactions of a variety of organic functional groups using tetrahydroxydiboron-mediated processes with water or alcohols serving as H atom codonors.[7] Among the common diborane reagents,[8] B2(OH)4 is the most atom-economical and is being used industrially for the transition-metal-catalyzed synthesis of aryl boronic acids.[9] More recently there have been examples of transfer hydrogenation or hydrogenolysis using B2(OH)4 without a polar protic additive. For example, in 2020, Lakshman and coworkers reported the Pd/C-catalyzed reduction of aryl halides, aldehydes, alkenes, and alkynes using B2(OH)4 and 4-methylmorpholine (Scheme [1]A).[10]
Scheme 1 Relevant examples of catalytic transfer reductions mediated by tetrahydroxydiboron
Interestingly, 4-methylmorpholine served as a formally aprotic codonor of H atoms. More recently, our lab described a Pd/C-catalyzed transfer deoxygenation of benzylic ketones using B2(OH)4 as the sole H-atom source in THF.[11] Herein, we report a Pd-catalyzed B2(OH)4-mediated alkene transfer hydrogenation method using Pd(OAc)2 as a precatalyst and no H-atom codonor (Scheme [1]C).
The optimized conditions[12] are similar to those we recently published for the ketone deoxygenation.[11] In either case, common aprotic diboron reagents B2pin2 and B2cat2 afford no substrate conversion (see the Supporting Information for details). We evaluated a variety of polar solvents and found that amongst ethers, cyclic monoethers (Scheme [2], first row) are suitable for the reduction of trans-stilbene, whereas 1,4-dioxane and acyclic ethers (row 2) are less so. This may be due to the attenuated polarity of 1,4-dioxane and cyclic monoethers compared to the cyclic monoethers, which, beyond their ability to stabilize the putative metaboric acid byproduct,[11]
[13] may limit their ability to dissolve B2(OH)4. Acetonitrile and triethylamine (row 3) also exhibit some efficacy, while 1,2-dichloroethane, toluene, and DMSO yield no product.[12]
Scheme 2 Solvent effects on hydrogenation yield. Yields of 2a determined by 1H NMR analysis of the crude reaction mixture compared to 1,3,5-trimethoxybenzene as an internal standard.
We then investigated the scope of the transfer hydrogenation of various alkenes as substrates using the optimized conditions and THF as solvent (Scheme [3]A). Di- and trisubstituted stilbenes 1a–e are efficiently reduced, regardless of alkene geometry (cf., 1a and 1d). In contrast, tetraphenylethylene (1f) reacts incompletely even after heating for a full day. Interestingly the water-mediated variant, performed at ambient temperature in dichloromethane, rapidly reduces tetraphenylethylene at room temperature.[7a] A variety of styrenes (1g–o) were evaluated and all afford yields greater than 90%. Furthermore, ethyl cinnamate (1p) undergoes efficient reduction of its α,β-unsaturation, as does chalcone (1q), whereas benzylideneacetone (1r), featuring an enolizable ketone, affords a low C=C reduction yield of 32% due to observed competing carbonyl reduction. Excitingly, dutasteride (1s), a prescription active pharmaceutical ingredient for the treatment of benign prostatic hyperplasia, undergoes selective reduction of its α,β-unsaturation position, albeit at a slow rate. We also evaluated the reduction of a few isolated alkenes using compounds 1t–x. Oleic acid (1t), a Boc-protected dihydropyrrole (1u), and three terminal alkenes (1v, 1w, and 1x) were all efficiently hydrogenated. Diphenylacetylene (3) was also subjected to similar reduction conditions although with twice the amount of additive and catalyst (Scheme [3]B). The major product is cis-stilbene (1d, 43% yield), with trans-stilbene (1a) observed in 16% yield, and just 5% of bibenzyl (2a) produced.
Scheme 3 Substrate scope of the transfer hydrogenation. These reactions were conducted on 0.5 mmol scale. Unless otherwise noted, full conversion was observed and reported yields are of isolated products (PMP = p-methoxyphenyl). a Reaction time was 24 hours. b Due to product volatility, 1H NMR yield is reported (compared to 1,3,5-trimethoxybenzene as an internal standard). c 1H NMR yield compared to 1,3,5-trimethoxybenzene as an internal standard. d Conducted on 0.2 mmol scale.
As shown in Scheme [4] below, we performed additional experiments on trans-stilbene (1a) to assess the fidelity of deuterium isotope incorporation, evaluated a deuterium kinetic isotope effect, and determined the influence of mercury on the catalysis. Excitingly, Scheme [4]A shows that trans-stilbene incorporates deuterium from B2(OD)4 virtually quantitatively. This is exciting because deuterium-enriched compounds are valuable medicinally[14]
[15] and as probes of organic reaction mechanisms.[16] This also validates the hypothesis that the diboron reagent is the sole source of hydrogen atoms. A competition kinetic isotope effect (KIE) study[17,18] was performed using equimolar amounts of B2(OH)4 and B2(OD)4 (Scheme [4]B), resulting in a KIE of 2.3 as determined from the ratio of products 1a/1a-d
2
. Although the transfer of the H atom to form the putative palladium hydride is not likely the rate-determining step, this result informs about the formation of the putative palladium hydride, and especially interesting compared to the competition KIE previously reported for the B2cat2-mediated reduction of diphenylacetylene using equimolar H2O and D2O in dichloromethane, which resulted in a competition KIE of 5.6.[7a]
Scheme 4 Applications and mechanistic studies on trans-stilbene. These reactions were conducted on 0.2 mmol scale. Isolated yields are reported and conversion matched the yield unless otherwise noted. a Mercury was added after eight minutes to allow for catalyst induction. b Yield and % conv. determined by 1H NMR analysis of the crude reaction mixture using 1,3,5-trimethoxybenzene as an internal standard.
Lastly, a mercury drop test was performed (Scheme [4]C). Near-complete conversion is achieved when mercury is added following an eight-minute induction period.[19] In contrast, the yield and conversion decrease slightly (92% each) if mercury is added immediately, suggesting some inhibition of catalyst induction from Pd(OAc)2 (not shown). Considering the putative colloidal ligandless nature of this reaction, the outcome of these mercury drop experiments – perhaps limited in their utility – is difficult to interpret.
In conclusion, we have developed a method for the Pd-catalyzed transfer hydrogenation of a variety of unsaturated C–C bonds mediated by B2(OH)4 using Pd(OAc)2 as a convenient precatalyst,[20]
[21] and quantitative alkene deuteration has been demonstrated using B2(OD)4.[22]
