Enantio- and diastereoselective boron conjugate addition to α-alkyl α,β-unsaturated esters

Abstract

We developed a copper-catalyzed enantio- and diastereoselective boron conjugate addition to α-alkyl α,β-unsaturated esters under base-free conditions. The approach showed excellent enantioselectivities (87–99% ee) and moderate to good conversions (51–99%), albeit with moderate diastereoselectivities (1 : 1–17 : 1 dr). The synthetic utility of this protocol was demonstrated.

---

Organic boron is one of the most important synthetic blocks in modern organic syntheses.¹ Complex compounds could be assembled and synthesized by modular and general strategies, attributed to the potential of the boron functional group in various transformations.² Introducing the boron group to readily available starting materials is thus of great significance.³ The copper-catalyzed asymmetric boron conjugate addition to α,β-unsaturated compounds has been studied intensively for the last two decades,⁴ providing chiral β-substituted boron products that are versatile building blocks for chemical synthesis and drug development. A typical reaction process^4b is that under the promotion of alkali additives, the copper salt complex reacts with a diboron reagent to form a copper boron intermediate, which adds to an unsaturated substrate via conjugate addition, resulting a β-boron substituted enol copper intermediate. Alcohol protonates the enol copper intermediate to release the final product and alkoxy copper. It is worth noting that the addition of alkali was crucial to the catalytic activity. Due to this issue, α-substituted α,β-unsaturated compounds were not suitable substrates as the epimerization of the product under alkaline conditions would lead to a mixture of diastereoisomers, which limits the practical applicability of these methods.⁵

We previously reported the highly enantio- and diastereoselective boron conjugate addition to α-functionalized α,β-unsaturated compounds, using a copper catalyst (CuCl/(S,R)-ppfa/AgNTf₂) under base-free conditions.⁶ For α-amino unsaturated esters and α-alkyl unsaturated ketones, excellent enantioselectivities and moderate to good diastereoselectivities were obtained, accompanied by good conversions. However, α-alkyl α,β-unsaturated esters gave only 10% conversion under similar conditions due to their comparatively low reactivities (Fig. 1A). Thus, development of a better catalytic method for α-alkyl α,β-unsaturated esters is desirable. Herein, we report an optimized method to achieve the highly enantioselective boron conjugate addition to α-alkyl α,β-unsaturated esters, albeit with moderate diastereoselectivity (Fig. 1B).

Fig. 1 (A) Boron conjugate addition to α-amino α,β-unsaturated esters and α-alkyl α,β-unsaturated ketones (previous work). (B) Boron conjugate addition to α-alkyl α,β-unsaturated esters (this work).

Compared with α,β-unsaturated ketones, the antibonding orbital (π*) energy of α,β-unsaturated esters is higher, while the electrons of the olefin bond are more active. In the copper-catalyzed boron conjugate addition reaction, the reactivity of α,β-unsaturated esters is lower. This shows that the π* orbital of the substrate plays a decisive role in the reaction rate: the substrate acts as the electron acceptor and the other key species (boryl copper compound) acts as the electron donor. In order to increase the reactivity to α,β-unsaturated esters, it is logical to increase the energy of electron donors and decrease the energy of the π* orbital of the substrate. Feringa et al. studied the relationship between the oxidative potential of copper(I) catalysts and the reactivity towards nonactivated substrates in the copper-catalyzed 1,4-addition of Grignard reagents, and they found that a more electron rich catalyst (easier to oxidize) would be beneficial for nonactivated substrates.⁷ They concluded that the difference in reactivity might be related to the energy match of the substrate and catalyst. This could explain that the amino phosphine ligand, (S,R)-ppfa, was proven to be superior to the biphosphine ligands in the copper-catalyzed boron conjugate addition to α-functionalized α,β-unsaturated compounds, as the amino ligand is purely electron donating, while the phosphine ligand could form π-backbonding.⁶ Moreover, Feringa et al. found that the presence of Mg²⁺ and Br⁻ was important to the reactivity and selectivity.⁷ Similarly, we found that additives played critical roles in the reactivity and selectivity, and acidic alcohol additives [CF₃CH₂OH, (CF₃)₂CHOH] were proven to be beneficial for the reactivity.⁶

Inspired by the above analyses, a systematic optimization of the copper precursor, ligand, additive, etc., was performed on the basis of our previous report. We initiated our study with 2,2,2-trifluoroethyl (E)-2-methyl-3-phenylacrylate (1a) as the template substrate which has a lower π* energy than the normal alkyl esters. The effect of the anion exchange reagent as an additive was studied first, with CuCl as the catalyst precursor and tert-butyl substituted amino phosphine (L4) as the ligand (Table 1, entries 1–4). The bulky noncoordinating counterions, such as NTf₂⁻ and BArF⁻, could initiate the reaction with moderate conversion. The relatively low diastereoselectivity (1.6 : 1 dr) obtained by using AgNTf₂ might be attributed to its strong Lewis acidity. For comparison, the weakly coordinating counterions, such as BF₄⁻ and PF₆⁻, gave nearly no conversion.⁸ While NaBArF proved to be superior to other anion exchange reagents, we then studied the effect of alcohol additives (Table 1, entries 5 and 6). The more acidic alcohol, (CF₃)₂CHOH, gave a higher conversion, albeit with lower diastereoselectivity (52% conv., 6 : 1 dr). In the meantime, when ethanol was applied to the system, nearly no conversion was observed. This indicated that the proton on the alcohol was involved in the transition state of the decisive step, probably similar to Mg²⁺ in Feringa's system.⁷ Then we turned to examine copper precursors, including univalent and divalent copper salts with different counterions (Table 1, entries 7–11). Copper salts with different valence states, Cu(OTf)₂ and CuOTf, gave close results with good conversion and diastereoselectivity and excellent enantioselectivity (entries 7 and 8). This might be attributed to the oxidizability of Cu(OTf)₂ and the reducibility of diboron compounds, which led to the in situ formation of the Cu(I) catalyst. The copper carboxylates were less reactive (entries 9 and 10), probably due to the stronger coordination ability of carboxylates. The enolate type copper precursor, Cu(acac)₂, was also reactive and gave similar results to CuOTf (entry 11). With Cu(OTf)₂ as the optimal precursor, we tested anion exchange reagents as additives again (Table 1, entries 12–16). The coordinating anions, such as F⁻ (entries 12 and 13) and ^tBuCO₂⁻ (entry 15), were harmful to the reactivity. When a common activator, NaO^tBu (entry 14), was used in copper-catalyzed boron conjugate addition, only 20% conversion was obtained as well as non-diastereoselectivity (1.2 : 1 dr). The lower conversion with NaO^tBu disclosed that the activation mode with bulky noncoordinating counterions might be different from that of the base additives: while the NaO^tBu additive proved to produce the intermediate CuO^tBu which led to CuBpin,⁹ the NaBArF additive might provide the CuBpin(HFIP) intermediate instead which involves the function of acidic HFIP. The precursor itself, Cu(OTf)₂, could also initiate the reaction, albeit with lower conversion than that with NaBArF (entry 16).

Table 1 Optimization of reaction conditions^a

Entry	R	Cu precursor	L*	Additive	Alcohol	Conv.	dr (syn/anti)	ee (%)
a Conditions: 1a (0.122 mmol), B2pin2 (2 equiv.), Cu precursor (0.012 mmol, 10 mol%), L* (0.018 mmol, 15 mol%), alcohol (1.22 mmol), additive (0.024 mmol), 4 Å MS (30 mg), toluene (1 mL), rt, 48 h. Conversions and diastereoselectivities were determined by ^1H NMR analyses of the crude mixtures. The enantiomeric excess was determined by transforming 2 into the corresponding alcohol using NaBO3·4H2O as the oxidant^4e and then analyzed by chiral HPLC. b A E/Z = 3 : 1 mixture of the substrate was used.
1	OCH2CF3 (1a)	CuCl	L4	AgNTf2	CF3CH2OH	46%	1.6 : 1	ND
2	OCH2CF3	CuCl	L4	NaBArF	CF3CH2OH	42%	10 : 1	ND
3	OCH2CF3	CuCl	L4	AgBF4	CF3CH2OH	<5%	—	—
4	OCH2CF3	CuCl	L4	AgPF6	CF3CH2OH	Trace	—	—
5	OCH2CF3	CuCl	L4	NaBArF	(CF3)2CHOH	52%	6 : 1	ND
6	OCH2CF3	CuCl	L4	NaBArF	CH3CH2OH	Trace	—	—
7	OCH 2 CF 3	Cu(OTf) 2	L4	NaBArF	(CF 3 ) 2 CHOH	84%	7 : 1	97
8	OCH2CF3	(CuOTf)·1/2 toluene	L4	NaBArF	(CF3)2CHOH	80%	5 : 1	98
9	OCH2CF3	CuTc	L4	NaBArF	(CF3)2CHOH	45%	6 : 1	98
10	OCH2CF3	CuOAc	L4	NaBArF	(CF3)2CHOH	47%	3 : 1	91
11	OCH2CF3	Cu(acac)2	L4	NaBArF	(CF3)2CHOH	79%	5 : 1	>99
12	OCH2CF3	Cu(OTf)2	L4	CsF	(CF3)2CHOH	22%	3 : 1	ND
13	OCH2CF3	Cu(OTf)2	L4	KF	(CF3)2CHOH	25%	3 : 1	ND
14	OCH2CF3	Cu(OTf)2	L4	NaO^tBu	(CF3)2CHOH	20%	1.2 : 1	ND
15	OCH2CF3	Cu(OTf)2	L4	CsOPiv	(CF3)2CHOH	Trace	—	—
16	OCH2CF3	Cu(OTf)2	L4	—	(CF3)2CHOH	44%	3 : 1	ND
17	OCH2CF3	Cu(OTf)2	L1	NaBArF	(CF3)2CHOH	67%	4 : 1	90
18	OCH2CF3	Cu(OTf)2	L2	NaBArF	(CF3)2CHOH	36%	5 : 1	89
19	OCH2CF3	Cu(OTf)2	L3	NaBArF	(CF3)2CHOH	63%	6 : 1	90
20	OCH2CF3	Cu(OTf)2	L5	NaBArF	(CF3)2CHOH	62%	10 : 1	97
21^b	OCH2CF3	Cu(OTf)2	L4	NaBArF	(CF3)2CHOH	89%	5 : 1	ND
22	OEt (1b)	Cu(OTf)2	L4	NaBArF	(CF3)2CHOH	42%	10 : 1	91
23	OPh (1c)	Cu(OTf)2	L4	NaBArF	(CF3)2CHOH	>99%	1.5 : 1	94/94
24	NHC6H4(p-OMe) (1d)	Cu(OTf)2	L4	NaBArF	(CF3)2CHOH	4%	—	—

---

We then studied the substitution effect on chiral ligands (Table 1, entries 17–20). When increasing the volume of groups on phosphine atoms, the enantioselectivity and diastereoselectivity could be regularly enhanced. However, there was no obvious law on the effect on conversion. This may be because the copper catalyst was not stable, and the conversion was determined by both the lifetime of the catalyst and the reaction rate. A less bulky catalyst might favour the reaction rate; however, it might be less stable at the same time. A piece of evidence about the relationship between the reaction rate and the catalyst's lifetime was that the conversion could be enhanced at lower temperature (94% conv. at 0 °C, see the ESI† for details), and this was probably due to the higher lifetime of the catalyst at lower temperature, albeit with a slower reaction rate. The effect of the configuration of the substrate on the diastereoselectivity was investigated, and a slightly lower diastereoselectivity was obtained with the E/Z = 3 : 1 mixture (Table 1, entry 21). With the optimal conditions (Table 1, entry 7), we finally compared the types of substrates (Table 1, entries 22–24). The ethyl ester (1b) showed lower reactivity and enantioselectivity, but higher diastereoselectivity. The phenolic ester (1c) gave full conversion, albeit with low diastereoselectivity (1.5 : 1 dr). The amide type substrate was also tested (1d) and only 4% conversion was found. These results were consistent with Feringa's conclusion that the reactivity could be related to the energy match of the substrate and catalyst.⁷

With the optimized conditions in hand, we sought to explore the scope of substrates with α-methyl α,β-unsaturated trifluoroethyl esters (Table 2). As the polarity of 2 was very small, we converted it into the corresponding amide (3a, 3e–v) for separation and identification. Good conversions (84–99%) and enantioselectivities (87–99% ee) were obtained for all the α-methyl substrates, while the α-ethyl substrate (1v) gave only moderate conversion (51%). Changing the substitution group and its position on the phenyl ring did not affect the reactivity significantly (3e–t). However, poor diastereoselectivities were observed in some cases (3i and 3l) without obvious regularity. For the β-methyl substrate, no diastereoselectivity was found (3u, 1 : 1 dr). It disclosed that β-substitution affected the face selectivity in the protonation process that determined the diastereoselectivity. The absolute configuration was determined by an X-ray study of the crystal of 3e, and the kinetically selective syn-product was dominant in the reaction.

Table 2 Substrate scope^a,b

a Conditions: 1 (0.122 mmol), B₂pin₂ (0.244 mmol, 2 equiv.), Cu(OTf)₂ (0.012mmol, 10 mol%), ent-L4 (0.018 mmol, 15 mol%), HFIP (0.13 mL, 1.22 mmol), NaBArF (0.024 mmol), 4 Å MS (30 mg), toluene (1 mL), rt, 48 h. Isolated yields of 3 are given. Diastereoselectivities were determined by ¹H NMR analysis of 3. The enantiomeric excess was determined by chiral HPLC. b Conversions were determined by ¹H NMR analyses of the crude mixtures after the boron conjugate addition step (those of 1).

---

To demonstrate the utility of our method in organic synthesis, the template substrate 1a was transformed into 2a on a 500 mg scale under the optimized conditions (Fig. 2). The trifluoroethyl ester product could be converted directly into the benzyl amide (3a) under the promotion of p-toluenesulfonic acid, or be hydrolyzed into the corresponding acid and then transformed into the amide (5) with the promotion of the condensation agent. The trifluoroborate 4 could be obtained with KHF₂ with an enhanced diastereomeric ratio (8 : 1 dr from 5 : 1 dr) from 3a. A typical Zweifel alkenylation reaction¹⁰ was performed with 5, and the enantioenriched alkene product 6 was obtained after the well-known stereospecific reaction.

Fig. 2 Transformation applications.

As the catalytic mode in this report was probably different from the previous copper-catalyzed boron conjugate addition promoted by a base additive, we proposed a synergistic catalytic mode, involving the copper catalyst and the promotion with an acidic alcohol additive, on the basis of our findings and Feringa's report⁷ (Fig. 3). The active copper species, complex A, is produced after reduction and anion exchange. It interacts with the substrate on two sites: the proton activates the carbonyl groups; in the meantime, the copper catalyst contributes its d electron pair to the π* orbital of the substrate. These interactions are probably synergistic and the transition state in this step is described as a π-complex. The σ-complex (Cu(III) species) is then formed. The product enolate is quickly released as the σ-complex is highly unstable. The diastereomeric ratio is determined via the controlled protonation, in which the volume of the substitution at the β-position affects the efficiency of face selection.⁶ Complex A is regenerated after the reaction between alkoxy copper and diboron.

Fig. 3 Proposed catalytic mode.

Conclusions

In conclusion, a copper-catalyzed enantio- and diastereoselective boron conjugate addition to α-alkyl α,β-unsaturated esters under base-free conditions was developed. A systematic optimization of the copper precursor, ligand, additive, etc., was performed. Unlike the common base-promoted copper-catalyzed boron conjugate addition, the bulky noncoordinating counterions and acidic alcohols proved to be crucial to initiate the reaction. Excellent enantioselectivities and good conversions were obtained, albeit with moderate diastereoselectivities. The synthetic utility of this protocol was also demonstrated. A synergistic catalytic mode, including the copper catalyst and the promotion with an acidic alcohol additive, was proposed on the basis of our results and previous reports.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

We thank NSFC (21702164) for financial support. The project was also supported by the Guangdong Provincial Key Laboratory of Catalysis (Grant No. 2020B121201002).

References

1. For selected examples of chiral organic boron compounds, see: (a) T. Awano, T. Ohmura and M. Suginome, J. Am. Chem. Soc., 2011, 133, 20738–20741; (b) M. Tortosa, Angew. Chem., Int. Ed., 2011, 50, 3950–3953; (c) T. Ohmura, T. Awano and M. Suginome, J. Am. Chem. Soc., 2010, 132, 13191–13193; (d) F. Meng, K. P. McGrath and A. H. Hoveyda, Nature, 2014, 513, 367–374; (e) A. Bonet, M. Odachowski, D. Leonori, S. Essafi and V. K. Aggarwal, Nat. Chem., 2014, 6, 584–589.
2. For selected examples of syntheses with chiral organic boron compounds, see: (a) S. L. Poe and J. P. Morken, Angew. Chem., Int. Ed., 2011, 50, 4189–4192; (b) R. P. Sonawane, V. Jheengut, C. Rabalakos, R. Larouche-Gauthier, H. K. Scott and V. K. Aggarwal, Angew. Chem., Int. Ed., 2011, 50, 3760–3763; (c) S. Nave, R. P. Sonawane, T. G. Elford and V. K. Aggarwal, J. Am. Chem. Soc., 2010, 132, 17096–17098; (d) J. Chang, H. Lee and D. G. Hall, J. Am. Chem. Soc., 2010, 132, 5544–5545; (e) Y. Yamamoto and N. Asao, Chem. Rev., 1993, 93, 2207–2293.
3. For selected examples of the syntheses of chiral organic boron compounds, see: (a) J. K. Park, H. H. Lackey, B. A. Ondrusek and D. T. McQuade, J. Am. Chem. Soc., 2011, 133, 2410–2413; (b) D. Imao, B. W. Glasspoole, V. S. Laberge and C. M. Crudden, J. Am. Chem. Soc., 2009, 131, 5024–5025; (c) E. Hupe, P. Marek and I. Knochel, Org. Lett., 2002, 4, 2861–2863.
4. For selected reviews, see: (a) A. Schiffner, K. Müther and M. Oestreich, Angew. Chem., Int. Ed., 2010, 49, 1194–1196; (b) J. Cid, H. Gulyas, J. J. Carbo and E. Fernandez, Chem. Soc. Rev., 2012, 41, 3558–3570. For selected examples, see: (c) J.-E. Lee and J. Yun, Angew. Chem., Int. Ed., 2008, 47, 145–147; (d) I.-H. Chen, L. Yin, W. Itano, M. Kanai and M. Shibasaki, J. Am. Chem. Soc., 2009, 131, 11664–11665; (e) K.-s. Lee, A. R. Zhugralin and A. H. Hoveyda, J. Am. Chem. Soc., 2009, 131, 7253–7255; (f) G. Palau-Lluch and E. Fernandez, Adv. Synth. Catal., 2013, 355, 1464–1470; (g) Y. Luo, I. D. Roy, A. G. E. Madec and H. W. Lam, Angew. Chem., Int. Ed., 2014, 53, 4186–4190; (h) S. Kobayashi, P. Xu, T. Endo, M. Ueno and T. Kitanosono, Angew. Chem., Int. Ed., 2012, 51, 12763–12766; (i) G. Stavber and Z. Časar, Appl. Organomet. Chem., 2013, 27, 159–165.
5. For acyclic substrates, non-diastereoselectivity could be achieved, see: (a) V. Lillo, A. Prieto, A. Bonet, M. M. Diaz-Requejo, J. Ramirez, P. J. Perez and E. Fernandez, Organometallics, 2009, 28, 659–662; (b) Z.-T. He, Y.-S. Zhao, P. Tian, C.-C. Wang, H.-Q. Dong and G.-Q. Lin, Org. Lett., 2014, 16, 1426–1429. For cyclic substrates, good diastereoselectivity could be achieved, see: (c) K. Kubota, K. Hayama, H. Iwamoto and H. Ito, Angew. Chem., Int. Ed., 2015, 54, 8809–8813; (d) L. Chen, J.-J. Shen, Q. Gao and S. Xu, Chem. Sci., 2018, 9, 5855–5859; (e) H. A. Clement, M. Boghi, R. M. McDonald, L. Bernier, J. W. Coe, W. Farrell, C. J. Helal, M. R. Reese, N. W. Sach, J. C. Lee and D. G. Hall, Angew. Chem., Int. Ed., 2019, 58, 18405–18409; (f) K. Hayama, R. Kojima, K. Kubota and H. Ito, Org. Lett., 2020, 22, 739–744.
6. J.-B. Xie, S. Lin, S. Qiao and G. Li, Org. Lett., 2016, 18, 3926–3929.
7. S. R. Harutyunyan, F. López, W. R. Browne, A. Correa, D. Peña, R. Badorrey, A. Meetsma, A. J. Minnaard and B. L. Feringa, J. Am. Chem. Soc., 2006, 128, 9103–9118.
8. M. Brookhart, B. Grant and A. F. Volpe, Organometallics, 1992, 11, 3920–3922.
9. D. S. Laitar, E. Y. Tsui and J. P. Sadighi, J. Am. Chem. Soc., 2006, 128, 11036–11037.
10. (a) G. Zweifel, H. Arzoumanian and C. C. Whitney, J. Am. Chem. Soc., 1967, 89, 3652–3653; (b) Y. Wang, A. Noble, E. L. Myers and V. K. Aggarwal, Angew. Chem., Int. Ed., 2016, 55, 4270–4274.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 2171509. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d2ob01928k

---