A General Organocatalyzed Michael-Michael Cascade Reaction Generates Functionalized Cyclohexenes

Abstract

While β-dicarbonyl compounds are regularly employed as Michael donors, intermediates arising from the Michael addition of unsaturated β-ketoesters to α,β-unsaturated aldehydes are susceptible to multiple subsequent reaction pathways. We designed cyclic unsaturated β-ketoester substrates that enabled the development of the first diphenyl prolinol silyl ether-catalyzed Michael-Michael cascade reaction initiated by a β-dicarbonyl Michael donor to form cyclohexene products. The reaction conditions we developed for this Michael-Michael cascade reaction were also amenable to a variety of linear unsaturated β-ketoester substrates, including some of the same linear unsaturated β-ketoester substrates that were previously ineffective in Michael-Michael cascade reactions. These studies thus revealed that a change in simple reaction conditions, such as solvent and additives, enables the same substrate to undergo different cascade reactions, thereby accessing different molecular scaffolds. These studies also culminated in the development of a general organocatalyzed Michael-Michael cascade reaction that generates highly functionalized cyclohexenes with up to four stereocenters, in up to 97% yield, 32:1 dr, and 99% ee, in a single step from a variety of unsaturated β-ketoesters.

Introduction

Organocatalytic cascade reactions are an efficient, green chemical method for rapidly building molecular complexity from simple starting materials.¹ Particularly useful for this purpose are diphenyl prolinol silyl ether catalysts (i.e., 1a, Scheme 1). These catalysts have been used to generate carbocyclic products via cascade reactions in which multiple C-C single bonds are formed by a combination of iminium and/or enamine catalyzed reactions.^2–12 While the use of β-dicarbonyl compounds as Michael donors in organocatalytic conjugate additions is widespread,^13–20 they had previously initiated only one 1a-catalyzed Michael-Michael cascade reaction (Reaction 1).²¹

Scheme 1.

1a-Catalyzed Cascade Reactions.

Michael reactions initiated by related β-dicarbonyl compounds resulted in other cascade reactions. Unsaturated β-dicarbonyl compounds without substitution at the 2-position (5) underwent a Michael-Morita-Baylis-Hillman cascade reaction instead (Reaction 2).²² These unhindered substrates faciliate conjugate addition of the amine catalyst to the 2-position (vida infra) and contain a removable proton at the 1-position, both prerequisites for the Morita-Baylis-Hillman reaction. Alternatively, β-dicarbonyl compounds with aryl substituents at the 2-position (7) underwent a Michael-acetalization cascade reaction in the presence of catalytic p-NO₂C₆H₄CO₂H in CH₂Cl₂ (Reaction 3).²³ These substrates are thermodynamically predisposed to undergo the acetalization pathway, which generates products (8) in which the extended conjugation is preserved.

We recently reported a Michael-Michael cascade reaction, catalyzed by 1a, that formed highly functionalized, fused carbocycles (10).²⁴ This was the first example of the formation of cyclohexene products via a 1a-catalyzed Michael-Michael cascade reaction initiated by a β-dicarbonyl Michael donor. These findings prompted further questions, primarily: are these reaction conditions limited to cyclic unsaturated β-dicarbonyl compounds of type 9? We set out to answer this question and report the full substrate scope of this transformation herein.

Results and Discussion

A. Reactions with Cyclic Unsaturated β-Dicarbonyl Substrates

Cyclic unsaturated β-dicarbonyl compounds, 9, were designed to preclude the Morita-Baylis-Hillman pathway (through substitution at the 1-position) and to disfavor the acetalization pathway (by being void of aryl substituents at the 2-position). When studies began, substrate 9a generated only 13% of the Michael-Michael product after ten days (entry 1, Table 1). Use of benzoic acid, an additive known to accelerate the turnover of catalyst 1a, nearly doubled the yield of the Michael-Michael product (entry 2). In 1,2-dichloroethane (DCE) and other solvents (toluene, Et₂O, THF, MeCN), the initial Michael addition went to completion overnight, whereas the subsequent Michael addition required prolonged reaction times. Additionally, in all solvents, the ratio of 10a:10b was approximately 1:1. Ethanol as solvent further doubled the yield of the Michael-Michael product, presumably by activating the second Michael addition through hydrogen bonding interactions with the ketoester moiety (entry 3). Use of a stronger hydrogen bonding solvent, trifluoroethanol, ultimately proved optimal after consideration of the results for this solvent in the context of the catalytic cycle (Scheme 2).

Table 1.

Summary of Key Optimizations.^a

entry	additive	solvent	time (h)	conversionb (%)	ratiob (10a:10b)	drb (10:11)	% eec (10a)	% eec (10b)
1	--	DCE	240	13	1:1.1	4:1	nd	nd
2	PhCO2H	DCE	168	25	1.5:1	5:1	99	99
3	PhCO2H	EtOH	168	61	1:1.1	4:1	99	99
4	PhCO2H	CF3CH2OH	2	12	1:29	8:1	nd	nd
5	PhCO2H	CF3CH2OH	41	17	1:5	10:1	99	99
6	--	CF3CH2OH	17	85	1:1.1	9:1	99	99
7d	--	CF3CH2OH	17	87	1:1.1	10:1	99	99

^aReaction conditions: 3a (1 equiv.), 9a (1 equiv.), 1a (20 mol %), additive (20 mol %), solvent (0.3 M), rt.

^bDetermined by ¹H NMR.

^cDetermined by chiral phase HPLC.

^d10 mol% 1a used.

Scheme 2.

Catalytic Cycle

Although trifluoroethanol as solvent led to a rapid Michael-Michael cascade reaction (entry 4 vs. entry 1), longer reaction times led only to a nominal improvement in conversion (entry 5 vs. entry 4). This was partly because the initial Michael addition had still not gone to completion, which was indicative of either a preference for the “catalyst release 1” pathway (12→3a) over the productive Michael addition pathway (12→14), or of a reversible Michael addition (14→3a). Resubjecting single Michael adduct 14 to reaction conditions resulted in rapid conversion to 10 only. This revealed that the formation of 14 was effectively irreversible and confirmed that the “catalyst release 1” pathway was predominating under these conditions. It also revealed that the second Michael addition (pathway A) was rapid in trifluoroethanol. This was substantiated by the formation of cascade product in this solvent after just two hours (entry 4). The plateau in conversion can be explained by the predominance of the “catalyst release 2” pathway over the productive pathway (pathway A) using benzoic acid in trifluoroethanol. Additionally, under these conditions, 3a is present in sufficient quantities (as explained above) to compete with intermediate 14 for the catalyst, thereby inhibiting 14 from reentering the catalytic cycle and proceeding to product.

Finally, while the conversion to 10a and 10b did not change dramatically over the course of 39 hours in trifluoroethanol, the ratio of 10a to 10b did (entries 4 and 5). This suggested that 10b formed first and was slowly epimerizing to, and equilibrating with, 10a. This epimerization is believed to occur via an enamine intermediate (16) and not via an enol intermediate, since catalyst 1a was required for epimerization, which did not occur in the presence of benzoic acid alone. The dramatically reduced rate of epimerization in these conditions versus in other solvents can be explained by the predominance of “catalyst release 3” over formation of enamine 16 either from 15 prior to catalyst release or from 10b (via 15) after its ejection from the catalytic cycle.

Thus, whereas in all other solvents an exceedingly sluggish second Michael addition was the primary culprit in the low yields, in trifluoroethanol, multiple predominant “catalyst release” (i.e., turnover) pathways were detrimental to this transformation. Running this reaction in the absence of benzoic acid, an additive known to accelerate the turnover of catalyst 1a, resulted in nearly complete conversion to cascade products overnight and restored the 1:1 ratio of 10a:10b (entry 6). Reducing the catalyst loading to 10 mol% provided the optimal reaction conditions (entry 7), although the catalyst loading could be reduced to 1 mol% before the conversion and diastereoselectivity were impacted.

The reaction was run with a variety of substrates, the results of which are summarized in Table 2. In all cases, the ratio of a:b was approximately 1:1. Also, α,β-unsaturated aldehydes with an o-nitro-phenyl or an alkyl R group were unreactive.

Table 2.

Cyclic Unsaturated β-Dicarbonyl Substrates.^a–^e

^aReaction conditions: 3 (1 equiv.), 9 (1 equiv.), 1a (10 mol %), CF₃CH₂OH (0.3M), rt.

^bYield = isolated yield.

^cdr determined by ¹H NMR.

^d ee determined by chiral phase HPLC.

^edr = ratio of MAJOR:MINOR.

B. Reactions with Linear Alkyl-Substituted Unsaturated β-Dicarbonyl Substrates

We next sought to determine whether linear alkyl-substituted unsaturated β-dicarbonyl substrates of type 25a (Table 3) would furnish Michael-Michael adducts under these conditions. Compound 25a lacks an aromatic substituent at the 2-position and therefore, for reasons explained earlier, may not be prone to a Michael-acetalization reaction pathway. It does, however, contain a proton at the 1-position, and may be susceptible to the Michael-Morita-Baylis-Hillman reaction pathway. Pleasingly, using the optimal reaction conditions for β-ketoesters of type 9, β-ketoester 25a generated the desired Michael-Michael cascade product in high conversion and good diastereoselectivity (entry 1). There was also non-insignificant conversion (13%) to the corresponding Michael-Morita-Baylis-Hillman product. However, the ee of both epimers of the major Michael-Michael adduct (26a and 26b) was excellent and, notably, the ee of the minor diastereomer (27) was also high. Additionally, epimers 26a and 26b were present in a 1:8 ratio, which is in contrast to what was observed with products arising from cyclic unsaturated β-dicarbonyl substrates. In light of the high selectivity of Michael-Michael reactions using 25a, efforts were directed towards minimizing the formation of the Michael-Morita-Baylis-Hillman byproduct.

Table 3.

Optimizations to Minimize Byproduct Formation.^a

entry	cat.	additive	conc. (M)	conversionb,c (%)	drb (26:27)	% eed (26a)	% eed,e (26b)
1	1a	--	0.30	87 (13)	4:1f	93	94 (82)
2	1a	--	0.15	79 (21)g	4:1	nd	nd
3	1a h	--	0.30	84 (16)	4:1	nd	nd
4	1b	--	0.30	trace	nd	nd	nd
5	1c	--	0.30	89 (11)	5:1	90	91
6	1d	--	0.30	89 (8)	5:1	84	84
7	1e	--	0.30	91 (7)	7:1	85	88
8	1a	4Å MS	0.30	77 (9)	4:1	nd	nd
9	1a	PhCO2HI	0.30	78 (22)	4:1	nd	nd

^aReaction conditions: 25a (1 equiv.), 3a (1 equiv.), cat. (10 mol %), CF₃CH₂OH, rt, 2 days.,

^bDetermined by ¹H NMR.

^cNumber in parentheses is conversion to products of type 6.

^dDetermined by chiral phase HPLC.

^eNumber in parentheses is ee of 27.

^fRatio of 26a:26b = 1:8.

^gReaction time = 1 day.

^h5 mol % 1a used.

^I10 mol % PhCO₂H used.

Consideration of the competing reaction pathways for the direct product of the initial Michael addition (28, Scheme 3) identified variables for optimization. If 28 undergoes an intramolecular Michael addition, intermediate 29 will be generated and will ultimately furnish the desired Michael-Michael product, 26a. However, if 28 undergoes a catalyst turnover, both the catalyst and aldehyde 30 will be liberated. Subsequent intermolecular conjugate addition of the catalyst to 30 would initiate the Morita-Baylis-Hillman reaction, leading to byproducts of type 6a. Thus, conditions that would impact intermolecular reactions between catalyst and substrate, as well as catalyst turnover were varied.

Scheme 3.

Competing Pathways for the Single Michael Adduct.

Dilute reaction concentration, lower catalyst loading and electronically or sterically less nucleophilic catalysts were examined in an effort to hamper the intermolecular conjugate addition of the catalyst to 30 (entries 2–7). Additionally, additives that affect catalyst turnover were also examined (entries 8 and 9). In all cases, either the amount of 6a formed did not decrease, or it decreased at the expense of decreased conversion to, or ee of, the desired product. The original conditions (i.e., those reported for cyclic unsaturated β-ketoesters; entry 1) were therefore chosen as optimal for linear alkyl-substituted unsaturated β-ketoesters.

Using these conditions, several linear alkyl-substituted unsaturated β-ketoesters were examined (Table 4). In all cases, the Michael-Michael adduct was obtained in high conversion and dr, and the β-epimer of the major diastereomer was formed in greater proportion and in excellent ee (entries 1–4). Notably, this cascade reaction is amenable to β-ketoesters with sterically demanding R substituents, such as i-Pr, which gave the best results (entry 4). β-Ketoesters of type 25 where R = H (i.e., 5, Scheme 1) were not compatible with these conditions.

Table 4.

Linear Alkyl-Substituted Unsaturated β-Dicarbonyl Substrates^a–^e

^aReaction conditions: 3a (1 equiv.), 25 (1 equiv.), 1a (10 mol %), CF₃CH₂OH (0.3M), rt.

^bConversion, dr, and ratio of a:b determined by ¹H NMR.

^dee determined by chiral phase HPLC.

^edr = ratio of MAJOR:MINOR.

C. Reactions with Linear Aryl-Substituted Unsaturated β-Dicarbonyl Substrates

To our delight, β-ketoesters of type 7—the exact same substrates that form Michael-acetalization products—generated the Michael-Michael cascade product in high conversion and selectivity using our reaction conditions (Table 5)! These β-ketoesters furnished the desired product with α,β-unsaturated aldehydes with electron-deficient, -neutral, and –rich aromatic R groups (entries 1–3). Additionally, an α,β-unsaturated aldehyde with an ortho-substituted-phenyl R group was tolerated, without detriment to the yield or selectivity of this reaction (entry 4). This in contrast to what was observed with cyclic unsaturated β-dicarbonyl compounds; as mentioned previously, no reaction occurred with an α,β-unsaturated aldehyde with an o-nitrophenyl R group. α,β-Unsaturated aldehydes with heteroaromatic and non-aromatic R groups also furnished the desired products in high yield and dr and in excellent ee (entries 5 and 6). Even α,β-unsaturated aldehydes with aliphatic R groups were amenable to this reaction (entry 7), whereas no reaction occurred when used in conjunction with cyclic unsaturated β-dicarbonyl compounds. This may be because linear compounds of type 7 have a proton at the 1-position, proximal to the nucleophilic β-ketoester moiety, whereas cyclic compounds of type 9 have alkyl substitution at the 1-position. Compounds of type 7, being less hindered, may be more reactive nucleophiles, and thus capable of reacting with less reactive α,β-unsaturated aldehydes, such as those with ortho-substituted-phenyl and alkyl R groups.

Table 5.

Linear Aryl-Substituted Unsaturated β-Dicarbonyl Substrates^a–e

^aReaction conditions: 3 (1 equiv.), 7 (1 equiv.), 1a (10 mol %), CF₃CH₂OH (0.3M), rt.

^bYield = isolated yield.

^cConversion, dr, and ratio of a:b determined by ¹H NMR.

^dee determined by chiral phase HPLC.

^edr = ratio of MAJOR:MINOR.

β-Ketoesters of type 7 with electron-rich, electron-deficient or ortho-substituted phenyl Ar groups all produced the desired Michel-Michael products in high yield and excellent ee, and in moderate to high dr (entries 8–10). Finally, β-ketoesters of type 7 with a heteroaromatic Ar group also gave the desired product in excellent ee, albeit in reduced yield and dr (entry 11).

As with linear alkyl-substituted unsaturated β-ketoesters, linear aryl-substituted unsaturated β-ketoesters formed the β-epimer of the major diastereomer in greater proportion in all but three cases (entries 5, 8, and 10). The absolute stereochemistry of 42 was established by X-ray crystallography.²⁵ Additionally, as with 10a and 10b, we established that 26a and 26b were epimers by resubjecting pure 26a to 1a in CF₃CH₂OH, which produced a mixture of 26a and 26b. All other stereochemical assignments were made by analogy.

For all β-ketoesters of type 7, no Michael-Morita-Baylis-Hillman product and no Michael-acetalization product was observed. The direct product of the initial Michael addition of β-ketoester 7 to enal 3, activated by catalyst 1a through iminium ion formation, is 45 (Scheme 4). In a non-polar aprotic solvent such as CH₂Cl₂, the ketoenol equilibrium of the β-ketoester moiety may favor the enol form, as in 46, due to stabilization by extended conjugation and by an intramolecular hydrogen bond. Moreover, in the presence of catalytic acid, the enamine moiety in 45 can be protonated to from an iminium species, as in 46. Alternatively, the catalytic acid can facilitate catalyst turnover, liberating the corresponding free aldehyde. Intermediate 46 (or its corresponding free aldehyde) undergoes an intramolecular acetalization to form dihydropyrans (8) in moderate to good yields and in good to excellent enantioselectivities.²³ On the other hand, a polar protic solvent such as CF₃CH₂OH disrupts the intramolecular hydrogen bond in 46, biasing the keto-enol equilibrium of the β-ketoester moiety towards the keto form. Moreover, CF₃CH₂OH presumably activates the Michael acceptor moiety towards the intramolecular conjugate addition through hydrogen bonding. Intermediate 45 undergoes a subsequent Michael addition to form cyclohexenes (34–44) in moderate to excellent yields and in excellent enantioselectivities.

Scheme 4.

Solvent and Additives Determine Product Structure.

This is illustrative of a powerful concept: the same substrates, using the same organocatalyst and identical catalyst loadings, can produce completely different molecular scaffolds in high yields and selectivities by simply changing the solvent and additives used. Since, as mentioned, β-dicarbonyl compounds are widely employed as Michael donors in organocatalytic conjugate addition reactions, it is anticipated that this idea can be exploited in many other organocatalytic cascade reactions.

Conclusion

In conclusion, despite the fact that unsaturated β-dicarbonyl compounds are susceptible to multiple reaction pathways, we have developed general conditions for a 1a-catalyzed Michael-Michael cascade reaction that are compatible with cyclic unsaturated β-dicarbonyl compounds as well as linear unsaturated β-dicarbonyl compounds with both alkyl and aryl substituents. This cascade reaction generates highly functionalized cyclohexene rings with up to four chiral centers in up to 97% yield, 32:1 dr and 99% ee in a single step from achiral starting materials. Moreover, this cascade reaction demonstrates that the same starting materials and catalyst enable efficient access to diverse molecular scaffolds through modification of simple reaction conditions, a principle that should be applicable to many other organocascade reactions.

Experimental Section

General Information

¹H and ¹³C NMR spectra were collected using a 400 MHz spectrometer. The NMR data herein uses the following abbreviations: s=singlet, d=doublet, t=triplet, q=quartet, m=multiplet, dd=doublet of doublets, ddd = doublet of doublets of doublets, td=triplet of doublets, dt=doublet of triplets, qd = quartet of doublets. Enantiomeric excesses were determined using chiral phase HPLC with Chiralpak AD-H (0.46 × 25 cm) and Chiralpak AS-H (0.46 × 25 cm) columns. Optical rotations were determined using a polarimeter. IR spectra were collected using a FT-IR. High-resolution mass spectra were collected using a MS spectrometer. Flash chromatography was carried out with F60, 40–63 m 60 Å silica gel and with EMD silica 60 F₂₅₄ glass TLC plates. Solvents were dried and kept air free in a solvent purification unit, or stored under an argon atmosphere. Solvents were evaporated using a standard rotovapor and a high vacuum. All reactions were carried out in oven-dried glassware and conducted under an argon atmosphere.

Characterization and Supporting Information for 10a–b, 11, and 17–24

Full chracterization of 10a–b, 11, and 17–24 as well as copies of ¹H NMR spectra, ¹³C NMR spectra, HPLC chromatograms, and X-ray crystal data for compound 10a are contained in the Supporting Information for reference 24.

Preparation of catalysts (1a–e), enals (3c–3d), and β-ketoesters (7 and 25)

Catalysts 1a–e²⁶ were prepared from the corresponding diarylprolinols²⁷ using known procedures. Characterization of 1a–e via ¹H NMR was in agreement with that in the literature. (E)-3-(p-tolyl)acrylaldehyde (3c) and (E)-3-(2-fluorophenyl)acrylaldehyde (3d) were prepared using a known procedure.²⁸ Characterization of 3c and 3d via ¹H NMR was in agreement with that in the literature. β-ketoesters of type 25 and type 7 (except 7a, 7b and 7c whose preparation and characterization are described below) were prepared according to a known literature procedure.²⁹ Characterization of β-ketoesters of type 25 and type 7 via ¹H NMR was in agreement with that in the literature.

(E)-3-(4-bromophenyl)acrylaldehyde, 3b

Enal 3b was prepared by adapting a known procedure.³⁰ A solution of 4-bromobenzaldehyde (2.41 g, 13.0 mmol) and triphenylphosphoranylidene acetaldehyde (4.75 g, 15.6 mmol) in toluene (103 ml) was heated at 80 oC for 16 h. To the reaction mixture was added triphenylphosphoranylidene acetaldehyde (250 mg, 0.06 mmol) again and the reaction mixture was further refluxed for 19 h. After the reaction mixture was concentrated in vacuo, the crude residue was chromatographed on silica gel eluting with Et₂O/petroleum ether (7.5:92.5) to give 3b (1.82 g, 66%) as a yellow solid. The ¹H and ¹³C NMR spectra for 3b were in agreement with that in the literature (see Supporting Information for spectra).³¹

(E)-ethyl 5-(2-fluorophenyl)-3-oxopent-4-enoate, 7b

To a round-bottom flask was added THF (22 mL) and diisopropylamine (844 μL 6.6 mmol) and cooled to −78 °C. n-BuLi (2.5M in hexanes, 6.6 mmol) was added slowly and stirred for 15 minutes. Ethyl acetate (586 μL, 6.0 mmol) was then added dropwise and the solution stirred for 50 minutes. (E)-3-(2-fluorophenyl)acrylaldehyde (3d) (6.0 mmol) was added dropwise and stirred for 10 minutes. The reaction was quenched by the addition of sat. aq. NH4Cl (1.7 ml), followed by immediate transfer to a separatory funnel containing diethyl ether (25 ml). The mixture was extracted with diethyl ether (25 mL), washed with brine (2 × 25 ml) and water (2 × 25 mL), and dried over MgSO₄. Removal of solvent yielded the corresponding hydroxyester in >99% yield. No further purification was needed before the oxidation with Jones' Reagent. Jones' Reagent was prepared by the addition of concentrated H₂SO₄ (1.8 mL) to CrO₃ (2.0 g) followed by careful dilution with water to give a total volume of 15 mL. Then, Jones' Reagent (9.0 mL, 9.0 mmol) was added dropwise to a stirred solution of the β-hydroxyester (6.0 mmol) in acetone (24 mL) at 0 °C. After complete addition of the oxidizing agent, the reaction was stirred for 10 minutes, when the absence of starting material was determined by thin-layer chromatography. Methanol (1.5 mL) was added slowly to quench excess Jones' Reagent. The reaction mixture was poured into a separatory funnel and extracted with diethyl ether (30 mL). The organic extracts were washed with water (3 × 20 mL) and then brine (2 × 20 mL). The organic layer was dried (MgSO₄) and filtered and the solvent removed under reduced pressure. Purification of the crude product mixture was achieved by flash chromatography on silica gel eluting with Et₂O/petroleum ether (2.5:97.5), which gave 7b as a white crystalline solid (338 mg, 24% yield). 7b: white solid. m.p.: 57–58°C; IR (thin film, KBr): 2983, 1741, 1649, 1596, 1486, 1458, 1422, 1235, 1148, 1094, 1039, 969, 799, 755 cm⁻¹; ¹H NMR (400 MHz, CDCl₃) (1:2 ratio of keto : enol) δ 11.99 – 11.94 (d, J = 1.3 Hz, 1H, enol), 7.81 – 6.81 (m, 10H), 6.59 – 6.50 (dd, J = 16.1, 1.5 Hz, 2H), 5.21 – 5.16 (s, 1H, enol), 4.29 – 4.17 (qd, J = 7.1, 4.6 Hz, 4H), 3.73 – 3.68 (s, 2H, keto), 1.36 – 1.24 (m, 6H); ¹³C NMR (101 MHz, CDCl₃) δ 192.1, 172.8, 168.9, 167.3, 162.9, 162.4, 159.9, 137.0, 136.9, 132.4, 132.3, 130.7, 130.6, 129.5, 129.4, 129.2, 129.1, 128.7, 128.6, 127.4, 127.4, 124.6, 124.6, 124.5, 124.4, 124.4, 124.3, 123.5, 123.4, 116.4, 116.2, 116.2, 116.0, 92.5, 61.5, 60.3, 47.6, 14.3, 14.1 ppm; HRMS (ESI) calcd for C₁₃H₁₃F O₃ [M]⁺ 236.0849, found 236.0860.

(E)-ethyl 5-(furan-2-yl)-3-oxopent-4-enoate, 7c²³

β-ketoester 7c was prepared by a modification of the same procedure used for 7b. To a round-bottom flask was added THF (22 mL) and diisopropylamine (844 μL 6.6 mmol). The reaction was cooled to −78 °C. n-BuLi (2.5M in hexanes, 6.6 mmol) was added slowly and stirred for 15 minutes. Ethyl acetate (586 μL, 6.0 mmol) was then added dropwise and the solution was stirred for 50 minutes. (E)-3-(furan-2-yl)acrylaldehyde (732.7 mg, 6.0 mmol) was added dropwise and stirred for 10 minutes. The reaction was quenched by the addition of sat. aq. NH₄Cl (1.7 ml), followed by immediate transfer to a separatory funnel containing diethyl ether (25 ml). The mixture was extracted with diethyl ether (25 mL), washed with brine (2 × 25 ml) and water (2 × 25 mL), then dried over MgSO₄. Removal of solvent yielded the hydroxyester. No further purification was needed before the oxidation with DMP. However, purification could be achieved on silica gel with 15%–25% Et₂O in petroleum ether. To a solution of the corresponding hydroxyester (420.5 mg, 2.0 mmol) in CH₂Cl₂ (55.5 mL) and MeCN (77.1 mL) at r.t. was added NaHCO₃ (330.6 mg, 3.9 mmol). The mixture was cooled to 0 °C and Dess-Martin Periodinane (848.3 mg, 2.0 mmol) was added. The reaction mixture was stirred at 0 °C for 1 hr. TLC (10% EtOAc in pet. ether) indicated ~10% starting material remaining, so the reaction mixture was warmed to room temperature and stirred for an additional 30 minutes. TLC at this point indicated all starting material had been consumed. The reaction was quenched with a 1:1 solution (110 mL) of saturated NaHCO₃ and saturated Na₂S₂O₃ and was stirred vigorously until the organic layer was no longer cloudy. The quenched reaction mixture was then poured into a separatory funnel. The aqueous layer was extracted with CH₂Cl₂ (2 × 40 mL). The organic layers were combined, dried over Na₂SO₄, filtered and concentrated. Purification by column chromatography (silica gel, 5% Et₂O/petroleum ether) provided 7c as a yellow solid (220.7 mg, 53% yield). The ¹H and ¹³C NMR spectra for 7c were in agreement with that in the literature (see Supporting Information for spectra).²³

(E)-ethyl 5-(4-methoxyphenyl)-3-oxopent-4-enoate, 7d³²

β-ketoester 7d was prepared according to the same procedure used for the synthesis of 7c and was collected in 55% yield (273.1 mg). The ¹H NMR spectra for 7d was in agreement with that in the literature (see Supporting Information for spectra).³²

General Procedure for synthesis of carbocycles (26, 31–44)

To an oven-dried flask was added catalyst 1a (13.0 mg, 0.04 mmol), CF₃CH₂OH (1.33 mL), β-ketoester 7a (79.3 mg, 0.4 mmol), and enal 3a (70.8 mg, 0.4 mmol). The reaction was allowed to stir for the indicated time at room temperature. The reaction mixture was concentrated and filtered through a plug of silica, followed by concentration again. The percent conversion of the crude reaction mixture was determined using an internal standard (allyl alcohol). The diastereomeric ratio was determined through comparison of the relative integrations of the aldehyde peaks in this ¹H NMR spectrum. The diastereomeric mixture was then purified via column chromatography (silica gel, 95/5, petroleum ether/Et₂O, unless noted otherwise) and an isolated yield of the diastereomeric mixture was determined. Pure major diastereomer 26a was obtained by one further chromatography (silica gel, 9/1, petroleum ether/Et₂O). Pure minor diastereomer 26b was obtained after two additional chromatographies (silica gel, 9/1, petroleum ether/Et₂O). Carbocycles 31–44 were prepared and purified using the same procedure. In certain cases (34, 35, 36), the major epimer of the major diastereomer was inseparable from the minor epimer of the major diastereomer and the epimers were characterized as a mixture. Racemic samples of the Michael-Michael products 26 and 31–44 were prepared in a similar manner using racemic catalyst 1a.

(1S,5S,6S)-ethyl 5-butyl-6-formyl-3-hydroxy-4'-nitro-1,4,5,6-tetrahydro-[1,1'-biphenyl]-2-carboxylate, 26b

Clear oil. [α]_D²⁶ = +80.4 (c=0.60, CHCl₃, 93% ee); IR (thin film, KBr): 2957, 2928, 2858, 1721, 1653, 1519, 1347, 1277, 1230, 1094, 1044, 854 cm⁻¹; ¹H NMR (400 MHz, CDCl₃) δ 12.56 (s, 1H), 9.83 (s, 1H), 8.17 (d, J = 8.7 Hz, 2H), 7.36 (d, J = 8.7 Hz, 2H), 4.40 (d, J = 2.7 Hz, 1H), 4.02 (qd, J = 7.1, 1.7 Hz, 2H), 2.50–2.68 (m, 2H), 2.32 (dd, J = 18.8, 10.8 Hz, 1H), 1.95 (dd, J = 16.3, 6.2 Hz, 1H), 1.42–1.54 (m, 2H), 1.21–1.22 (m, 4H), 0.96 (t, J = 7.1 Hz, 3H), 0.83 (dd, J = 9.1, 4.6 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 201.5, 173.2, 171.4, 152.1, 146.7, 128.6, 123.7, 97.2, 60.7, 56.6, 38.9, 33.0, 31.9, 30.1, 29.4, 22.5, 13.9, 13.8 ppm; the enantiomeric excess was determined by HPLC with an AD-H column (n-hexane: i-PrOH = 99:1), 0.1 mL/min; major enantiomer t_R = 22.9 min, minor enantiomer t_R = 20.3 min; HRMS (ESI) calcd for C₂₀H₂₅NO₆ [M]⁺ 375.1682, found 375.1687.

(1S,5S,6R)-ethyl 5-butyl-6-formyl-3-hydroxy-4'-nitro-1,4,5,6-tetrahydro-[1,1'-biphenyl]-2-carboxylate, 26a

Clear oil. [α]_D²³ = +26.2 (c=0.59, CHCl₃, 94% ee); IR (thin film, KBr): 2957, 2930, 2860, 1722, 1655, 1621, 1520, 1347, 1278, 1221, 1156, 1109, 854, 705 cm⁻¹; ¹H NMR (400 MHz, CDCl₃) δ 12.47 (s, 1H), 9.12 (d, J = 4.4 Hz, 1H), 8.18 (d, J = 8.7 Hz, 2H), 7.31 (d, J = 8.7 Hz, 2H), 4.33 (d, J = 5.6 Hz, 1H), 4.02 (dd, J = 12.5, 7.1 Hz, 2H), 2.82 (dd, J = 18.6, 5.7 Hz, 1H), 2.50–2.55 (m, 1H), 2.15–2.35 (m, 2H), 1.14–1.40 (m, 6H), 1.00 (t, J = 7.1 Hz, 3H), 0.89 (t, J = 6.9 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 204.3, 172.5, 170.9, 148.0, 147.1, 129.9, 123.5, 99.0, 60.7, 54.9, 41.2, 33.9, 33.7, 28.2, 27.9, 22.7, 13.9, 13.8 ppm; the enantiomeric excess was determined by HPLC with an AS-H column (n-hexane: i-PrOH = 90:10), 0.5 mL/min; major enantiomer t_R = 20.2 min, minor enantiomer t_R = 37.8 min. HRMS (ESI) calcd for C₂₀H₂₅NO₆ [M]⁺ 375.1682, found 375.1690.

(1S,5S,6S)-ethyl 6-formyl-3-hydroxy-5-methyl-4'-nitro-1,4,5,6-tetrahydro-[1,1'-biphenyl]-2-carboxylate, 31

Clear oil. [α]_D²⁵ = +59.8 (c=0.45, CHCl₃, 98% ee); IR (thin film, KBr): 2962, 2925, 1722, 1651, 1519, 1348, 1277, 1225, 1099, 853, 736, 700 cm^–1; ¹H NMR (400 MHz, CDCl₃) δ 12.56 (s, 1H), 9.81 (d, J = 0.6 Hz, 1H), 8.10–8.19 (m, 2H), 7.32–7.40 (m, 2H), 4.40 (d, J = 3.8 Hz, 1H), 4.01 (qd, J = 7.1, 2.9 Hz, 2H), 2.62 (dd, J = 18.5, 5.6 Hz, 1H), 2.53 (t, J = 3.4 Hz, 1H), 2.33 (dd, J = 18.1, 10.0 Hz, 1H), 2.22 (dd, J = 6.8, 3.4 Hz, 1H), 1.13 (d, J = 6.9 Hz, 3H), 0.95 (t, J = 7.1 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 201.2, 172.9, 171.4, 152.3, 146.7, 128.6, 123.6, 97.2, 60.7, 57.9, 38.5, 34.9, 24.9, 17.4, 13.8 ppm; the enantiomeric excess was determined by HPLC with an AD-H column (n-hexane: i-PrOH = 99:1), 0.2 mL/min; major enantiomer t_R = 38.4 min, minor enantiomer t_R = 43.8 min. HRMS (ESI) calcd for C₁₇H₁₉NO₆ [M-H]⁻ 332.1139, found 332.1121.

(1S,5S,6S)-ethyl 5-ethyl-6-formyl-3-hydroxy-4'-nitro-1,4,5,6-tetrahydro-[1,1'-biphenyl]-2-carboxylate, 32

Clear oil. [α]_D²³ = +62.9 (c=1.67, CHCl₃, 95% ee); IR (thin film, KBr): 2965, 2933, 2877, 1721, 1652, 1519, 1348, 1275, 1222, 853, 756, 737, 701 cm⁻¹; ¹H NMR (400 MHz, CDCl₃) δ 12.57 (s, 1H), 9.83 (s, 1H), 8.17 (d, J = 8.8 Hz, 2H), 7.36 (d, J = 8.6 Hz, 2H), 4.41 (d, J = 2.7 Hz, 1H), 4.02 (qd, J = 7.1, 1.7 Hz, 2H), 2.53–2.70 (m, 2H), 2.31 (dd, J = 18.8, 10.9 Hz, 1H), 1.79–1.95 (m, 1H), 1.54 (ddd, J = 29.7, 14.3, 6.9 Hz, 2H), 0.97 (t, J = 7.1 Hz, 3H), 0.88 (t, J = 7.4 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 201.5, 173.2, 171.4, 152.1, 146.7, 128.6, 123.7, 97.2, 60.7, 56.5, 38.9, 32.6, 31.8, 25.2, 13.8, 11.8 ppm; the enantiomeric excess was determined by HPLC with an AD-H column (n-hexane: i-PrOH = 97:3), 0.5 mL/min; major enantiomer t_R = 25.8 min, minor enantiomer t_R = 32.7 min. HRMS (ESI) calcd for C₁₈H₂₁NO₆ [M]⁺ 347.1369, found 347.1360.

(1S,5R,6S)-ethyl 6-formyl-3-hydroxy-5-isopropyl-4'-nitro-1,4,5,6-tetrahydro-[1,1'-biphenyl]-2-carboxylate, 33

Clear oil. [α]_D²⁴ = +71.7 (c=1.71, CHCl₃, >99% ee); IR (thin film, KBr): 2962, 2927, 2872, 1720, 1655, 1519, 1348, 1301, 1281,1264, 1223, 1096, 1033, 854, 832 cm⁻¹; ¹H NMR (400 MHz, CDCl₃) δ 12.55 (s, 1H), 9.85 (s, 1H), 8.18 (d, J = 8.8 Hz, 2H), 7.34 (d, J = 8.5 Hz, 2H), 4.42 (s, 1H), 4.04 (qd, J = 7.1, 0.8 Hz, 2H), 2.76 (s, 1H), 2.66 (dd, J = 19.1, 6.1 Hz, 1H), 2.31 (dd, J = 19.0, 12.1 Hz, 1H), 1.78–1.94 (m, 1H), 1.35–1.48 (m, 1H), 0.99 (t, J = 7.1 Hz, 3H), 0.92 (d, J = 6.6 Hz, 3H), 0.78 (d, J = 6.6 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 201.7, 173.7, 171.3, 151.6, 146.8, 128.4, 123.7, 96.6, 60.8, 54.0, 39.4, 36.9, 31.7, 31.6, 29.7, 21.1, 20.6, 13.9 ppm. the enantiomeric excess was determined by HPLC with an AS-H column (n-hexane: i-PrOH = 99:1), 0.3 mL/min; major enantiomer t_R = 31.6 min, minor enantiomer t_R = 50.4 min. HRMS (ESI) calcd for C₁₉H₂₂NO₆ [M-H]⁻ 360.1447, found 360.1448.

(1'S,2'S,3'S)-ethyl 2'-formyl-5'-hydroxy-4"-nitro-1',2',3',6'-tetrahydro-[1,1':3',1"-terphenyl]-4'-carboxylate, 34:

White amorphous solid. [α]_D²³ = +247.5 (c=0.93, CHCl₃, 99% ee (34), >99% ee (epi-34)); IR (thin film, KBr): 2983, 1722, 1653, 1618, 1596, 1518, 1403, 1347, 1259, 1217, 1065, 853, 829, 753, 700 cm^{−1 1}H NMR (400 MHz, CDCl₃) (10:1 mixture of 18 to its C4 epimer) δ 12.69 (s, 1H, 34), 12.55 (s, 1H, epi-34), 9.68 (s, 1H, 34), 8.88 (d, J = 3.8 Hz, 1H, epi-34), 8.28 – 8.17 (m, 4H, 34+epi-34), 7.50 - 7.04 (m, 14H, 34+epi-34), 4.66 – 4.49 (m, 2H, 34+epi-34), 4.07 (m, 4H, 34+epi-34), 3.55 – 3.26 (m, 2H, 34+epi-34), 3.06–3.19 (m, 1H, epi-34), 3.05 – 2.75 (m, 4H, 34+epi-34), 2.66 (dd, J = 19.0, 11.4 Hz, 1H, epi-34), 1.06 – 0.93 (m, 6H, 34+epi-34); ¹³C NMR (101 MHz, CDCl₃) (10:1 mixture of 18 to its C4 epimer) δ 203.4 (epi-18), 200.8, 172.5, 172.2 (epi-18), 172.1 (epi-18), 171.4, 152.1, 146.9, 140.5 (epi-18), 139.2, 130.2 (epi-18), 129.3 (epi-18), 128.9, 128.6, 127.7 (epi-18), 127.6 (epi-18), 127.4, 127.3, 123.9, 123.6 (epi-18), 97.1, 60.8, 58.5, 54.3 (epi-18), 41.4 (epi-18), 38.8, 37.5 (epi-18), 35.3 (epi-18), 34.3, 31.0, 30.9 (epi-18), 13.8 ppm; the enantiomeric excess was determined by HPLC with an AS-H column (n-hexane: i-PrOH = 95:5), 1.0 mL/min; major diastereomer: major enantiomer t_R = 27.9 min., minor enantiomer t_R = 35.9 min.; minor diastereomer: major enantiomer t_R = 48.7 min., minor enantiomer t_R = 56.1 min. HRMS (ESI) calcd for C₂₂H₂₁NO₆ [M-H]⁻ 394.1291, found: 394.1271. Purification on silica gel was best achieved using 60/40 CH₂Cl₂/petroleum ether.

(1'S,2'S,3'S)-ethyl 2'-formyl-5'-hydroxy-1',2',3',6'-tetrahydro-[1,1':3',1"-terphenyl]-4'-carboxylate, 35:

Clear oil. [α]_D²⁶ = +86.2 (c=2.0, CHCl₃, 99% ee); IR (thin film, KBr): 3060, 3027, 2907, 2829, 2732, 1722, 1652, 1621, 1495, 1452, 1403, 1370, 1351, 1292, 1259, 1216, 1066, 1032, 831, 756, 700 cm⁻¹; ¹H NMR (400 MHz, CDCl₃) (5:1 mixture of 35 to its C4 epimer) δ 12.68 (s, 1H, 35), 12.53 (s, 1H, epi-35), 9.72 (s, 1H, 35), 8.84 (d, J = 4.4 Hz, 1H, epi-35), 7.05–7.44 (m, 20H, 35 + epi-35), 4.39–4.51 (m, 2H, 35 + epi-35), 3.98–4.17 (m, 4H, 35 + epi-35), 3.58 (td, J = 11.9, 6.2 Hz, 1H, epi-35), 3.36–3.48 (m, 1H, 35), 2.73–3.10 (m, 6H, 35 + epi-35), 1.01 (t, J = 7.1 Hz, 6H, 35 + epi-35); ¹³C NMR (101 MHz, CDCl₃) (5:1 ratio of major 35 to its C4 epimer) δ 204.7 (epi-19), 201.9, 171.9, 171.9, 171.4 (epi-19), 144.0, 141.2 (epi-19), 140.2, 139.5 (epi-19), 129.3 (epi-19), 129.1 (epi-19), 128.7, 128.5, 128.4 (epi-19), 127.7 (epi-19), 127.6, 127.4 (epi-19), 127.4, 127.1 (epi-19), 127.0, 126.6, 99.9 (epi-19), 97.9, 60.5, 59.0, 54.6 (epi-19), 41.7 (epi-19), 38.9, 37.5 (epi-19), 35.2 (epi-19), 34.1, 31.1, 13.8 ppm; the enantiomeric excess of a mixture of 35 and epi-35 was determined by HPLC with an AS-H column (n-hexane: i-PrOH = 97:3), 0.5 mL/min; major diastereomer: major enantiomer t_R = 14.8 min., minor enantiomer t_R = 17.8 min.; minor diastereomer: major enantiomer t_R = 59.6 min., minor enantiomer t_R = 28.5 min. HRMS (ESI) calcd for C₂₂H₂₂O₄ [M]⁺ 350.1518, found 350.1505.

(1'S,2'S,3'S)-ethyl 2'-formyl-5'-hydroxy-4"-methyl-1',2',3',6'-tetrahydro-[1,1':3',1"-terphenyl]-4'-carboxylate, 36:

Clear oil. [α]_D²³ = +101.3 (c=2.0, CHCl₃, 99% ee (36) 95% ee (epi-36)); IR (thin film, KBr): 2979, 2922, 2730, 1723, 1652, 1619, 1510, 1497, 1454, 1403, 1365, 1259, 1215, 1096, 1065, 821, 759, 699; ¹H NMR (400 MHz, CDCl₃) (2.5:1 ratio of 36 to its C4 epimer) δ 12.63 (s, 1H, 36), 12.48 (s, 1H, epi-36), 9.69 (d, J = 0.6 Hz, 1H, 36), 8.80 (d, J = 4.4 Hz, 1H, epi-36), 6.96–7.36 (m, 18H, 36 + epi-36), 4.29–4.42 (m, 2H, 36 + epi-36), 3.97–4.13 (m, 4H, 36 + epi-36), 3.27–3.64 (m, 2H, 36 + epi-36), 2.69–3.04 (m, 6H, 36 + epi-36), 2.29–2.38 (m, 6H, 36 + epi-36), 1.02 (td, J = 7.1, 1.1 Hz, 6H, 36 + epi-36); ¹³C NMR (101 MHz, CDCl₃) (2.5:1 ratio of 36 to its C4 epimer) δ 204.9 (epi-20), 202.1, 172.0, 171.8, 171.5 (epi-20), 171.2 (epi-20), 141.3 (epi-20), 140.9, 140.3, 136.6 (epi-20), 136.4 (epi-20), 136.0, 129.2, 129.1 (epi-20), 129.0 (epi-20), 128.7, 127.7 (epi-20), 127.5, 127.4, 127.0 (epi-20), 100.1 (epi-20), 98.1, 60.5, 59.1, 54.6 (epi-20), 41.2 (epi-20), 38.5, 37.5 (epi-20), 35.2 (epi-20), 34.1, 31.1, 21.0, 13.8 ppm; the enantiomeric excess of a mixture of 36 and epi-36 was determined by HPLC with an AS-H column (n-hexane: i-PrOH = 97:3), 0.5 mL/min; major enantiomer of major diastereomer t_R = 14.3 min, minor enantiomer t_R = 17.9 min; major enantiomer of minor diastereomer t_R = 47.3 min., minor enantiomer t_R = 27.3 min. HRMS (ESI) calcd for C₂₃H₂₄O₄ [M]⁺ 364.1675, found 364.1675.

(1'S,2'S,3'R)-ethyl 2"-fluoro-2'-formyl-5'-hydroxy-1',2',3',6'-tetrahydro-[1,1':3',1"-terphenyl]-4'-carboxylate, 37:

Clear oil. [α]_D²³ = +82.5 (c=2.0, CHCl₃, 98% ee); IR (thin film, KBr): 2982, 2929, 2734, 1724, 1620, 1654, 1486, 1454, 1404. 1259, 1216, 1094, 1066, 1034, 699 cm⁻¹; ¹H NMR (400 MHz, CDCl₃) δ 12.68 (s, 1H), 9.71 (s, 1H), 6.99–7.39 (m, 9H), 4.80 (s, 1H), 4.09 (tdd, J = 10.8, 7.1, 3.6 Hz, 2H), 3.26–3.40 (m, 1H), 3.02 (dd, J = 18.3, 12.4 Hz, 2H), 2.79 (dd, J = 18.4, 5.5 Hz, 1H), 1.02 (t, J = 7.1 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 201.6, 172.9, 171.9, 161.8, 159.3, 140.3, 129.5, 129.0, 127.7, 127.4, 124.2, 124.2, 115.9, 115.7, 97.2, 60.9, 56.9, 34.9, 32.3, 32.3, 31.2, 14.1 ppm; the enantiomeric excess was determined by HPLC with an AD-H column (n-hexane: i-PrOH = 97:3), 0.5 mL/min; major enantiomer t_R = 27.1 min, minor enantiomer t_R = 18.8 min. HRMS (ESI) calcd for C₂₂H₂₁FO₄ [M]⁺ 368.1424, found 368.1426.

(1S,2S,3R)-ethyl 2-formyl-3-(furan-2-yl)-5-hydroxy-1,2,3,6-tetrahydro-[1,1'-biphenyl]-4-carboxylate, 38:

Yellow amorphous solid. [α]_D²³ = +123.4 (c=1.7, CHCl₃, 98% ee); IR (thin film, KBr): 2926, 2851, 2731, 1724, 1654, 1620, 1498, 1277, 1239, 1218, 1096, 1060, 828, 761, 741, 701 cm⁻¹; ¹H NMR (400 MHz, CDCl₃) δ 12.49 (s, 1H), 9.07 (d, J = 4.0 Hz, 1H), 7.10–7.40 (m, 6H), 6.31 (d, J = 1.9 Hz, 1H), 6.08 (d, J = 3.0 Hz, 1H), 4.43 (d, J = 4.9 Hz, 1H), 4.15 (q, J = 7.1 Hz, 2H), 3.55 (td, J = 11.8, 6.2 Hz, 1H), 2.93 (dt, J = 12.4, 4.5 Hz, 1H), 2.81 (dd, J = 18.9, 6.2 Hz, 1H), 2.52 (dd, J = 18.9, 11.3 Hz, 1H), 1.15 (t, J = 7.1 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 203.1, 171.9, 171.4, 153.9, 141.9, 141.2, 129.1, 127.7, 127.4, 110.3, 108.5, 97.9, 60.7, 54.4, 37.5, 36.4, 35.1, 14.0 ppm; the enantiomeric excess was determined by HPLC with an AS-H column (n-hexane: i-PrOH = 97:3), 0.5 mL/min; major enantiomer t_R = 32.7 min, minor enantiomer t_R = 36.5 min. HRMS (ESI) calcd for C₂₀H₂₀O₅ [M]⁺ 340.1311, found: 340.1311.

(1S,2S,3R)-diethyl 2-formyl-5-hydroxy-1,2,3,6-tetrahydro-[1,1'-biphenyl]-3,4-dicarboxylate, 39:

Clear oil. [α]_D²³ = +18.5 (c=1.1, CHCl₃, 91% ee); IR (thin film, KBr): 2982, 2936, 2736, 1724, 1660, 1624, 1406, 1371, 1264, 1218, 1183, 1069, 1032, 830, 758, 737, 699 cm⁻¹; ¹H NMR (400 MHz, CDCl₃) δ 12.44 (s, 1H), 9.54 (s, 1H), 7.34 (dt, J = 12.5, 7.8 Hz, 5H), 4.13–4.36 (m, 4H), 4.01 (d, J = 1.5 Hz, 1H), 3.43–3.54 (m, 1H), 3.30 (s, 1H), 2.77 (ddd, J = 23.4, 18.3, 8.6 Hz, 2H), 1.30 (dd, J = 12.9, 7.1 Hz, 6H); ¹³C NMR (101 MHz, CDCl₃) δ 200.2, 173.6, 171.7, 171.5, 139.3, 128.9, 127.4, 95.5, 61.4, 60.8, 53.5, 39.4, 36.4, 30.7, 14.4, 14.1 ppm; the enantiomeric excess was determined by HPLC with an AS-H column (n-hexane: i-PrOH = 97:3), 0.5 mL/min; major enantiomer t_R = 21.2 min, minor enantiomer t_R = 32.3 min; HRMS (ESI) calcd for C₁₉H₂O₆ [M-H]⁻ 345.1338, found 345.1345.

(1S,2S,3R)-ethyl 3-butyl-2-formyl-5-hydroxy-1,2,3,6-tetrahydro-[1,1'-biphenyl]-4-carboxylate, 40:

Clear oil. [α]_D²³ = −44.5 (c=2.23, CHCl₃, 98% ee); IR (thin film, KBr): 2956, 2930, 2871, 1722, 1617, 1403, 1261, 1213, 1064, 831, 698 cm⁻¹; ¹H NMR (400 MHz, CDCl₃) δ 12.45 (s, 1H), 9.60 (s, 1H), 7.32 (ddd, J = 17.9, 12.8, 5.6 Hz, 5H), 4.27 (dddd, J = 25.1, 10.8, 7.1, 3.7 Hz, 2H), 3.36–3.53 (m, 1H), 2.94–3.08 (m, 2H), 2.89 (d, J = 1.7 Hz, 1H), 2.64 (dd, J = 18.5, 5.8 Hz, 1H), 1.72–1.84 (m, 1H), 1.29–1.54 (m, 8H), 0.96 (dd, J = 9.7, 3.9 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 202.8, 172.2, 170.8, 140.9, 128.8, 127.5, 127.0, 100.6, 60.6, 53.2, 34.6, 34.2, 32.8, 31.0, 30.1, 22.4, 14.2, 14.0 ppm; the enantiomeric excess was determined by HPLC with an AS-H column (n-hexane: i-PrOH = 99:1), 0.3 mL/min; major enantiomer t_R = 26.3 min, minor enantiomer t_R = 29.5 min. HRMS (ESI) calcd for C₂₀H₂₆O₄ [M+H]⁺ 331.1904, found 331.1904.

(1S,2S,3R)-ethyl 2-formyl-3-(furan-2-yl)-5-hydroxy-4'-methoxy-1,2,3,6-tetrahydro-[1,1'-biphenyl]-4-carboxylate, 41:

Clear oil. [α]_D²³ = +106.1 (c=0.47, CHCl₃, 97% ee); IR (thin film, KBr): 2927, 2837, 1723, 1653, 1513, 1306, 1278, 1250, 1214, 1179, 1062, 1035, 1012, 831, 737 cm⁻¹; ¹H NMR (400 MHz, CDCl₃) δ 12.47 (s, 1H), 9.04 (d, J = 4.2 Hz, 1H), 7.29–7.37 (m, 1H), 7.08 (d, J = 8.6 Hz, 2H), 6.83 (d, J = 8.6 Hz, 2H), 6.30 (dd, J = 2.9, 1.9 Hz, 1H), 6.07 (d, J = 3.2 Hz, 1H), 4.40 (d, J = 5.0 Hz, 1H), 4.14 (q, J = 7.1 Hz, 2H), 3.76 (s, 3H), 3.50 (td, J = 11.7, 6.1 Hz, 1H), 2.69–2.96 (m, 2H), 2.48 (dd, J = 18.9, 11.3 Hz, 1H), 1.15 (t, J = 7.1 Hz, 3H); 13C NMR (101 MHz, CDCl₃) δ 203.4, 172.0, 171.4, 158.8, 153.9, 141.8, 133.1, 128.7, 114.5, 110.3, 108.4, 97.9, 60.7, 55.2, 54.6, 37.7, 35.6, 35.2, 14.0 ppm; the enantiomeric excess was determined by HPLC with an AS-H column (n-hexane: i-PrOH = 90:10), 0.5 mL/min; major enantiomer t_R = 33.7 min, minor enantiomer t_R = 54.0 min. HRMS (ESI) calcd for C₂₁H₂₂O₆ [M+H]⁺ 371.1489, found 371.1481.

(1'S,2'S,3'S)-ethyl 4,4"-dibromo-2'-formyl-5'-hydroxy-1',2',3',6'-tetrahydro-[1,1':3',1"-terphenyl]-4'-carboxylate, 42:

Colorless crystals. m.p.: 118–120°C; [α]_D²⁴ = +81.6 (c=0.53, CHCl₃, 99% ee); IR (thin film, KBr): 2924, 2853, 1723, 1653, 1488, 1406, 1287, 1258, 1215, 1095, 1072, 1009, 821 cm⁻¹; ¹H NMR (400 MHz, CDCl₃) δ 12.63 (s, 1H), 9.64 (s, 1H), 7.44 (dd, J = 22.2, 8.5 Hz, 4H), 7.12 (d, J = 8.3 Hz, 2H), 6.99 (d, J = 8.3 Hz, 2H), 4.38 (d, J = 1.7 Hz, 1H), 4.06 (tt, J = 7.1, 3.5 Hz, 2H), 3.22–3.30 (m, 1H), 2.81–3.02 (m, 2H), 2.74 (dd, J = 18.4, 5.4 Hz, 1H), 1.02 (t, J = 7.1 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 200.9, 171.9, 171.6, 142.9, 138.9, 131.8, 131.7, 129.3, 129.1, 121.1, 120.6, 97.5, 60.8, 58.7, 38.4, 33.7, 30.9, 13.9 ppm; the enantiomeric excess was determined by HPLC with an AS-H column (n-hexane: i-PrOH = 97:3), 0.5 mL/min; major enantiomer t_R = 22.0 min, minor enantiomer t_R = 31.0 min. HRMS (ESI) calcd for C₂₂H₂₀Br₂O₄ [M-H]⁻ 504.9650, found 504.9665.

(1S,2S,3R)-ethyl 2'-fluoro-2-formyl-3-(furan-2-yl)-5-hydroxy-1,2,3,6-tetrahydro-[1,1'-biphenyl]-4-carboxylate, 43

Clear oil. [α]_D²⁶ = +120.3 (c=0.71, CHCl₃, 93% ee); IR (thin film, KBr): 2983, 2930, 2826, 2731, 1725, 1654, 1622, 1491, 1406, 1307, 1232, 1217, 1097, 1061, 1038, 758 cm⁻¹; ¹H NMR (400 MHz, CDCl₃) δ 12.48 (s, 1H), 9.14 (d, J = 3.6 Hz, 1H), 7.33 (dd, J = 1.8, 0.8 Hz, 1H), 6.95–7.25 (m, 4H), 6.30 (dd, J = 3.2, 1.9 Hz, 1H), 6.09 (d, J = 3.2 Hz, 1H), 4.46 (d, J = 5.0 Hz, 1H), 4.15 (q, J = 7.1 Hz, 2H), 3.84 (td, J = 11.8, 6.3 Hz, 1H), 2.95–3.13 (m, 1H), 2.81 (dd, J = 18.8, 6.3 Hz, 1H), 2.59 (dd, J = 18.8, 11.3 Hz, 1H), 1.16 (t, J = 7.1 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 202.5, 171.7, 171.4, 162.0, 159.6, 153.7, 141.9, 129.1, 129.1, 129.0, 128.9, 128.1, 127.9, 124.8, 124.7, 116.1, 115.9, 110.4, 108.5, 97.9, 60.7, 53.6, 53.5, 35.4, 34.8, 30.3, 14.0 ppm; the enantiomeric excess was determined by HPLC with an AS-H column (n-hexane: i-PrOH = 97:3), 0.5 mL/min; major enantiomer t_R = 31.3 min, minor enantiomer t_R = 18.5 min. HRMS (ESI) calcd for C₂₀H₁₉FO₅ [M]⁺ 358.1217, found 358.1218.

(4S,5S,6R)-ethyl 5-formyl-4,6-di(furan-2-yl)-2-hydroxycyclohex-1-enecarboxylate, 44

Yellow oil. [α]_D²⁶ = +65.3 (c=0.58, CHCl₃, 95% ee); IR (thin film, KBr): 2925, 2853, 1723, 1654, 1407, 1276, 1217, 1094, 1067, 1012, 812, 736 cm⁻¹; ¹H NMR (400 MHz, CDCl₃) δ 12.57 (s, 1H), 9.69 (s, 1H), 7.32–7.38 (m, 2H), 6.31 (ddd, J = 19.6, 3.2, 1.8 Hz, 2H), 6.10–6.16 (m, 1H), 6.00 (dt, J = 3.2, 0.8 Hz, 1H), 4.49 (s, 1H), 4.02–4.25 (m, 2H), 3.39–3.56 (m, 1H), 3.22–3.35 (m, 1H), 2.61–2.84 (m, 2H), 1.16 (t, J = 7.1 Hz, 3H); ¹³C NMR (101 MHz, CDCl₃) δ 201.2, 171.7, 171.0, 156.1, 154.1, 141.8, 141.7, 110.4, 110.2, 107.0, 106.1, 96.9, 60.7, 52.3, 32.6, 30.1, 30.0, 14.0 ppm; the enantiomeric excess was determined by HPLC with an AS-H column (n-hexane: i-PrOH = 95:5), 1.0 mL/min; major enantiomer t_R = 17.9 min, minor enantiomer t_R = 25.2 min. HRMS (ESI) calcd for C₁₈H₁₈O₆ [M+Na]⁺ 353.1000, found 353.0992.