A DIBAL-Mediated Reductive Transformation of trans Dimethyl Tartrate Acetonide into ε-Hydroxy α,β-Unsaturated Ester and its Derivatives

Abstract

Step-wise, selective DIBAL reduction of the acetonide diester derived from tartaric acid, followed by the Horner-Emmons reaction effectively provided desymmetrized hydroxy mono-olefination products in one-pot operation.

D-(−)- and L-(+)-tartaric acids are well-known chiral molecules possessing two useful stereogenic centers. Both, the natural (L) and unnatural (D) isomers, are relatively inexpensive and widely available even on a bulk scale. Therefore, these molecules have continually been utilized as an important chiral starting material in various natural product synthesis.

Among a number of the derivatives, the acetonide protected diester 2, easily prepared from the acid 1 by simple one- or two-step reactions (Scheme 1),ⁱ is often employed to lead to further symmetrical or unsymmetrical acetonides. In particular, a series of acetonide esters 4 including but not limited to 4a–e (Figure 1) are frequently seen as chiral building blocks for the construction of a variety of synthetic intermediates in total synthesis.^2–8 However, there is no general, efficient path to access these compounds. Indeed, many of the literature routes require a long reaction sequence that includes redundant protection-deprotection and/or oxidation-reduction processes and often results in low overall yields (e.g. 4a: 5 steps from dimethyl L-tartrate,ⁱⁱ4b: 7 steps from cis-2-butene-1,4-diol,ⁱⁱⁱ4c: 6 steps from L-tartaric acid,^iv4d: 7 steps from L-tartaric acid,^v4e: 7 steps from L-arabitol,^vietc.^vii,viii). However, since all these acetonides 4 should technically be accessible from the common acetonide derivative 3 by means of a standard one-step chemical transformation currently available, compound 3 seems to be a highly versatile and potentially useful molecule (Scheme 1). Nonetheless, the synthesis of 3 has been rarely studied and, to the best of our knowledge, no practical route to 3 is yet available,^ix even though efficient approaches for cis-acetonide isomers of 3 have been well-established.^x Thus, we envisioned that exploring a facile, single-pot transformation of acetonide diester 2 into 3 would be highly desirable, and furthermore, enable rapid access to 4 as well as to other useful analogues in just a few steps from tartaric acid 1 or, more conveniently, from a commercially available acetonide diester 2.

Scheme 1.

Short Synthetic Approach to 3 and 4

Figure 1.

A Series of Acetonide Esters 4a–e

A recent literature^xi describes eight-step approach to 3 (3a) via a chiral diol 5 (Scheme 2). Although the starting D-mannitol is a very cheap material, the overall yield is disappointingly low (17%) [the first four steps: 39%^xii and the second four steps: 44%^xi]. Naturally, the diol intermediate 5 can be prepared from D-tartaric acid by two steps,^xiii but the overall reaction efficiency (a total of six steps to 3a) is still not satisfactory for this class of simple molecules.^xiv

Scheme 2.

Synthesis of 3a

Meanwhile, Seebach et al. employed a short-step approach to 3 (3b) starting from an acetonide diester 2 (2a) (Scheme 3).^xv Treatment of 2a with 3 equivalents of DIBAL reagent at −78°C followed by a Soxhlet extraction and further distillation afforded the corresponding lactol/hydroxyaldehyde 6 in 45% yield. Subsequently, Wittig olefination of 6 provided 3b in 80%. While the overall yield was fair (36%)^xvi and the obtained 3b was a mixture of E/Z isomers with a partial epimerization at C4, this protocol still implies a potential feasibility for one-pot reductive transformation of diester 2 into 3 as described in Scheme 1.

Scheme 3.

Synthesis of 3b

^aPh3P=CHCO₂Me.

To begin, the Seebach's protocol under various “single-pot” conditions, that is, without isolation of 6, was initially examined. Even though the desired product 3b was somewhat observed, the E/Z selectivity was, as expected, marginal (E:Z = ~6:4). In order to avoid this stereochemical drawback, the Wittig reagent was exchanged for a Horner-Wadsworth-Emmons (HWE reagent), which is compatible to the use of DIBAL.^xvii The use of [(EtO)₂P(O)CHCOOEt]⁻Na⁺ exclusively gave the desired E-isomer (E:Z = > 25:1);^xviii however the yield of 3c was disappointingly low (13%) (route A in Scheme 4), although, based on the original Seebach's protocol, the entire DIBAL addition temperature was initially maintained at −78°C to prepare the lactol 6 via an organo-aluminum intermediate 7. Due to the poor reaction efficiency of route A, an alternative approach through a presumed hydroxyester 8 was then explored by adding the first two equivalents of DIBAL at r.t. (route B in Scheme 4).^xix Surprisingly, the applied route B conditions greatly improved the overall product yield of 3c to 46%.

Scheme 4.

One-Pot Transformation of 2a into 3c

^a[(EtO)₂P(O)CHCOOEt]⁻Na⁺

As summarized in Table 1, both route A and B conditions were further examined. Although the reaction yield was variable under different conditions, the route B (entries 4–10) still proved to be superior to route A (entries 1–3). The route B was highly solvent/temperature-dependent (entries 4–7), but much less subject to the ester substituent group (entries 6, 9, and 10). A faster DIBAL addition still gave the same results (entry 8). [Note: Not surprisingly, in the crude reaction mixture, two side-products, (1) over-reduced diol acetonide 5 and (2) C₂-symmetrical α,β-unsaturated diester,^xx were always observed (the composition ratio of these depends on the reaction conditions). These were easily separated from product 3c by column chromatography.]

Table 1.

Screening of Route A and B conditions^a

^aReaction conditions: 2 (1.0 mmol), DIBAL (1.0 M in toluene), DIBAL addition rate (1.0 mmol/h), solvent (4.0 mL), HWE reagent (1.5 mmol).

^bHWE reagent was added at r.t.

^cUsed DIBAL (1.0 M in hexane).

^dDIBAL addition rate (2.0 mmol/h).

The dependence of the reaction yield of 3c on variations in the amount of DIBAL at two different temperatures (the first step at room temperature and the second step at −78°C) was also investigated (Figure 2). The results indicate that the route B conditions, adding two equivalents of DIBAL at r.t. (step 1) and another one equivalent at −78°C (step 2), are strictly required to maximize the yield of 3c.

Figure 2.

Dependence of the yield of 3c on the amounts of DIBAL at the first and second steps.

All the entries and reactions discussed/shown so far were operated on a 1 mmol scale. The scalability was therefore examined. First, the reaction was run on a 10 mmol scale and successfully obtained 3c in 44% yield. However, since the first-step DIBAL reduction was a slightly exothermic process that was not preferred for a large-scale operation, the addition temperature was lowered to 0°C to be safe (Scheme 5). Interestingly, this modified condition somewhat improved the yield of 3c (46% → 49% on a 1 mmol scale). More excitedly, even 40 times larger-scale reaction still afforded 3c in 51% yield, which is essentially the same as the one obtained at 1 mmol reaction scale. Thus, the reaction proved to be scalable.

Scheme 5.

Testing scalability (1 mmol vs. 40 mmol)

To explore the scope and limitations of this route B approach, other HWE-type reagents were accordingly employed (Scheme 6). All the cases worked well with similar efficiency and afforded corresponding olefins (3d, 3e, and 3f) in 46–52 % yield, even though the α-cyano phosphonate reagent was not very E stereoselective (E:Z=63:37). Again, the first-step DIBAL addition temperatures (0°C vs. r.t.) affected the product yields. Substrate generality besides diester 2 was also explored. Unfortunately, when dimethyl succinate and dimethyl glutarate were tested, the route B conditions did not give desired olefins in an effective manner (only 15% and 26% yield, respectively). Thus, the presence of the conformational constraint imparted by the acetonide group seems to be important.

Scheme 6.

Different phosphonate reagents

^aYields in parenthesis were obtained when the first-step DIBAL addition was operated at room temperature.

Finally, one-step transformation of compound 3 into 4 was investigated (Scheme 7). Benzylation of 3c by employing Dudley's reagent (2-benzyloxy-1-methylpyridinium triflate)^xxi successfully afforded 4a in 75% yield.^xxii A silyl protecting group was also cleanly introduced on the hydroxyl group of 3c in 97 %. Pd/C-catalyzed hydrogenation quantitatively converted 3c into the reduced ester 4d. One-pot transformation of 3c into 4e was also attempted; however, the yield of 4e was consistently low (<20%). Therefore, following DMP oxidation of 3c, the aldehyde was isolated as a crude product and then used for the next Wittig olefination. This stepwise approach improved the yield to 38%, which is comparable to the reported ones.^xxiii These results strongly support that the now more readily available 3 can play an important role as a versatile intermediate for organic synthesis. In addition, the number of reaction steps to access 4 through our newly established approach are less than half those of literature protocols. Thus, even though the reaction yield from 2 to 3 is fair (up to 52%), these new routes to 4 are still highly competitive and useful from the aspect of overall synthetic efficiency.

Scheme 7.

Transformation of 3 into 4

^a2-benzyloxy-1-methylpyridinium triflate,^b tert-butylchlorodiphenylsilane,^c Dess-Martin periodinane

In summary, applying a unique, stepwise DIBAL addition sequence to trans dimethyl tartrate acetonide 2a most effectively gave the corresponding lactol intermediate 6 that was subsequently exposed to various Horner-Wadsworth-Emmons reagents to provide γ-hydroxy mono-olefination products. This simple protocol now enables a rapid access to a number of useful chiral acetonide derivatives such as 3 and 4 starting from readily available tartaric acid.

EXPERIMENTAL SECTION

Materials and methods

All experiments were performed in flame-dried glassware fitted with rubber septa under argon atmosphere. Toluene was dried by passing through activated alumina. Tetrahydrofuran (THF) was distilled over sodium/benzophenone ketyl. Unless otherwise noted, all other reagents were obtained from commercial sources and used as received. ¹H Nuclear Magnetic Resonance (NMR) spectra were recorded at 300 or 500 MHz. Data are presented as follows: chemical shift (in ppm on the δ scale relative to δH 7.26 for the residual protons in CDCl₃), multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, m = multiplet, br = broad), coupling constant (J/Hz), integration. Coupling constants were taken directly from the spectra and are uncorrected. ¹³C NMR spectra were recorded at 75 or 125 MHz, and all chemical shift values are reported in ppm on the δ scale, with an internal reference of δC 77.0 for CDCl₃. Infrared spectral data are reported in units of cm⁻¹. Analytical TLC was performed on silica gel plates using UV light and/or potassium permanganate stain followed by heating. Flash column chromatography was performed on silica gel 60A (32–63D).

General procedure for the synthesis of 3 (3c) on a 1.0 mmol scale

Into a solution of 2a (183 μL, 1.0 mmol) in dry toluene (4.0 mL) under argon atmosphere was added DIBAL reagent (2.0 mL, 1.0 M solution in toluene, 2.0 mmol) slowly over a period of 2 hours at 0°C by using a syringe pump. After stirring for 1 hour at 0°C, the resulting solution was placed in a dry-ice/acetone cooling bath (−78°C) and another one equivalent of DIBAL reagent (1.0 mL, 1.0 M solution in toluene, 1.0 mmol) was slowly added over a period of 1 hour. Following the addition of Horner-Emmons reagent that was separately prepared from triethyl phosphonoacetate (300μL, 1.5 mmol) and NaH (60 mg, 60 % in mineral oil, 1.5 mmol) in dry toluene (2 mL), the reaction mixture was stirred overnight (−78°C to r.t.) and quenched with saturated Rochelle's salt solution (2 mL). The resulting mixture was vigorously stirred for 2h and diluted with H₂O (2 mL). After the phase separation, the aqueous layer was extracted with Et₂O (2 × 15 mL). The combined organics were then dried over MgSO₄ and concentrated under reduced pressure. The crude product was purified by SiO₂ column chromatography (Hex/EtOAc = 3/1) to afford 3c (113 mg, 49%) as a colorless oil.

(E)-Ethyl 3-[(4S,5S)-5-hydroxymethyl-2,2-dimethyl-1,3-dioxolan-4-yl]prop-2-enoate (3c)

[α]²⁰_D −8.67 (c 1.43, CHCl₃); ¹H NMR (300 MHz, CDCl₃) δ 6.91 (dd, J = 5.8, 15.6 Hz, 1H), 6.14 (d, J = 15.6 Hz, 1H), 4.53 (m, 1H), 4.21 (q, J = 6.0 Hz, 2H), 3.87 (m, J = 7.8 Hz, 2H), 3.68 (m, J = 6.9 Hz, 1H), 2.70 (brs, 1H), 1.46 (s, 3H), 1.45 (s, 3H), 1.30 (t, J = 6.0 Hz, 3H); ¹³C NMR (75 MHz, CDCl₃) δ 165.9, 143.8, 122.8, 110.0, 80.7, 76.0, 60.7, 60.6, 26.9, 26.7, 14.1; IR spectra (neat): 1727, 3480; HRMS (EI) Calcd for C₁₀H₁₅O₅ 215.0919 [M-CH₃]⁺; found 215.0924.

(E)-tert-Butyl 3-[(4S,5S)-5-hydroxymethyl-2,2-dimethyl-1,3-dioxolan-4-yl]prop-2-enoate (3d)

The title compound was prepared according to the general procedure. Column chromatography (Hex/EtOAc = 3/1) yielded 3d (125mg, 48%) as a colorless oil. [α]²⁰_D −6.69 (c 1.27, CHCl₃); ¹H NMR (300 MHz, CDCl₃) δ 6.76 (dd, J = 6.2, 15.8 Hz, 1H), 6.07 (dd, J = 1.2, 15.8 Hz, 1H), 4.46 (m, 1H), 3.84 (m, 2H), 3.64 (m, 1H), 2.34 (brs, 1H), 1.46 (s, 9H), 1.43 (s, 3H), 1.42 (s, 3H); ¹³C NMR (75 MHz, CDCl₃) δ 165.1, 142.4, 124.8, 109.9, 80.74, 80.73, 76.0, 60.6, 27.9, 26.8, 26.7; IR spectra (neat): 1710, 3469; HRMS (EI) Calcd for C₁₂H₁₉O₅ 243.1232 [M-CH₃]⁺; found 243.1231.

(E)-4-[(4S,5S)-5-hydroxymethyl-2,2-dimethyl-1,3-dioxolan-4-yl]but-3-en-2-one (3e)

The title compound was prepared according to the general procedure (note: the HWE reagent was prepared in dry THF (2 mL) instead of dry toluene (2 mL) due to the poor solubility). Column chromatography (Hex/EtOAc = 7/3 to 6/4) yielded 3e (105mg, 52%) as a colorless oil. [α]²⁰_D −9.23 (c 1.29, CHCl₃); ¹H NMR (300 MHz, CDCl₃) δ 6.73 (dd, J = 6.2, 15.9 Hz, 1H), 6.36 (d, J = 15.9 Hz, 1H), 4.52 (m, 1H), 3.86 (m, 2H), 3.69 (brs, 1H), 2.78 (brs, 1H), 2.28 (s, 3H), 1.45 (s, 3H), 1.43 (s, 3H); ¹³C NMR (75 MHz, CDCl₃) δ 198.1, 142.5, 131.1, 110.0, 80.8, 76.2, 60.7, 27.4, 26.8, 26.6; IR spectra (neat): 1679, 3546; HRMS (EI) Calcd for C₉H₁₃O₄ 185.0814 [M-CH₃]⁺; found 185.0813.

3-[(4S,5S)-5-hydroxymethyl-2,2-dimethyl-1,3-dioxolan-4-yl]prop-2-ennitrile (3f)

The title compound was prepared according to the general procedure (note: the HWE reagent was prepared in dry THF (2 mL) instead of dry toluene (2 mL) due to the poor solubility). Column chromatography (Hex/EtOAc = 7/3 to 6/4) yielded 3f [53mg (E-isomer) + 31mg (Z-isomer) = 84mg, 48%] as a colorless oil. (E-isomer): [α]²⁰_D −5.31 (c 1.28, CHCl₃); ¹H NMR (300 MHz, CDCl₃) δ 6.73 (dd, J = 4.8, 16.2 Hz, 1H), 5.75 (dd, J = 1.5, 16.2 Hz, 1H), 4.51 (m, 1H), 3.84 (m, 2H), 3.68 (m, 1H), 2.39 (brs, 1H), 1.44 (s, 3H), 1.41 (s, 3H); ¹³C NMR (75 MHz, CDCl₃) δ 150.6, 116.7, 110.4, 100.7, 80.3, 76.2, 60.7, 26.8, 26.5; IR spectra (neat): 2230, 3475; HRMS (EI) Calcd for C₈H₁₀NO₃ 168.0661 [M-CH₃]⁺; found 168.0662. (Z-isomer): [α]²⁰_D 7.50 (c 0.96, CHCl₃); ¹H NMR (300 MHz, CDCl₃) δ 6.50 (dd, J = 8.3, 11.1 Hz, 1H), 5.53 (d, J = 11.1 Hz, 1H), 4.81 (m, 1H), 3.89 (m, 2H), 3.76 (m, 1H), 2.16 (brs, 1H), 1.47 (s, 3H), 1.46 (s, 3H); ¹³C NMR (75 MHz, CDCl₃) δ 150.5, 114.8, 111.0, 102.3, 80.8, 75.6, 61.1, 26.8, 26.6; IR spectra (neat): 2227, 3485; HRMS (EI) Calcd for C₈H₁₀NO₃ 168.0661 [M-CH₃]⁺; found 168.0659.

Experimental procedure for the synthesis of 3c on a 40.0 mmol scale

Into a solution of 2a (7.33 mL, 40.0 mmol) in dry toluene (160 mL) under argon atmosphere was added DIBAL reagent (80.0 mL, 1.0 M solution in toluene, 80.0 mmol) slowly over a period of 2 hours at 0°C by using a syringe pump. After stirring for 1 hour at the same temperature, the resulting solution was placed in a dry-ice/acetone cooling bath (−78°C) and another one equivalent of DIBAL reagent (40.0 mL, 1.0 M solution in toluene, 40.0 mmol) was slowly added over a period of 1 hour. Following the addition of Horner-Emmons reagent (over 20 min) that was separately prepared from triethyl phosphonoacetate (12.0 mL, 60.0 mmol) and NaH (2.40 g, 60 % in mineral oil, 60.0 mmol) in dry toluene (80 mL), the reaction mixture was stirred for ~12 hours (−78°C to r.t.). The mixture was then quenched with saturated Rochelle's salt solution (80 mL) at 0°C and vigorously stirred for 2 hours at r.t. After diluting with H₂O (80 mL), the biphasic mixture was separated. The separated aqueous layer was extracted with Et₂O (2 × 100 mL). The combined organics were dried over MgSO₄ and concentrated under reduced pressure. The crude product was purified by SiO₂ column chromatography (Hex/EtOAc = 3/1) to afford 3c (4.68 g, 51%) as a colorless oil. {Note: Two side-products, diol 5 (~5% yield by NMR) and C₂-symmetrical α,β-unsaturated diester (7% yield, 780mg), in the crude mixture were easily separated due to the large differences in polarity [Rf (Hex/EtOAc=3/1): 0.43 for the diester, 0.14 for 3c, and ~0 for diol 5].}

One-step transformation of 3 into 4

Benzylation of 3c into 4a

Into a solution of 3c (103 μL, 0.5 mmol) in benzotrifluoride (2 mL) was added 2-benzyloxy-1-methylpyridinium triflate^xxi (349 mg, 1 mmol) and MgO (40 mg, 1 mmol). The reaction mixture was stirred at 83 °C. After stirring for 66 h, the resulting reaction mixture was then filtered through Celite with EtOAc and concentrated under reduced pressure. The crude product was purified by SiO₂ column chromatography (Hex/EtOAc = 10/1) to give the compound 4a as a colorless oil (120 mg, 75 %). This product spectroscopically matched that of the known compound.^xxiv; [α]_D²⁰ = −23.2 (c 1.00, CHCl₃), [lit.^xxiv [α]_D²⁰ = −23.4 (c 1.0, CHCl₃)]; ¹H NMR (300 MHz, CDCl₃) δ 7.38−7.27 (m, 5 H), 6.90 (dd, J = 5.4, 15.8 Hz, 1H), 6.09 (dd, J = 1.5, 15.8 Hz, 1H), 4.60 (s, 2H), 4.43 (ddd, J = 1.5, 5.4, 8.4 Hz, 1H), 4.20 (q, J = 7.2 Hz, 2H), 3.96 (dt, J = 4.8 8.4 Hz, 1H), 3.63 (d, J = 4.8 Hz, 2H),1.46 (s, 3H), 1.44 (s, 3H), 1.29 (t, J = 7.2 Hz, 3H) ; ¹³C NMR (75 MHz, CDCl₃) δ 166.0, 144.0, 137.7, 128.4, 127.8, 127.7, 122.6, 110.2, 79.6, 77.4, 73.6, 69.3, 60.6, 26.9, 26.7, 14.2.

Silyl protection of 3c into 4c

Into a solution of 3c (103 μL, 0.5 mmol) in DMF (5 mL) was added tert-butylchlorodiphenylsilane (133μL, 0.51 mmol) and imidazole (68 mg, 1 mmol). The reaction mixture was stirred at 0 °C for 10 h, then stirring overnight at room temperature. After the phase separation, the aqueous layer was extracted with Et₂O (2 × 15 mL). The combined organics were then washed with brine, dried over MgSO₄, and concentrated under reduced pressure. The crude product was purified by SiO₂ column chromatography (Hex/EtOAc = 20/1) to afford 4c (228 mg, 97 %) as a colorless oil. This product spectroscopically matched that of the known compound.^iv; [α]_D²⁰ = −1.88 (c 3.41, CHCl₃), [lit.^iv [α]_D²⁰ = −1.8 (c 3.4, CHCl₃)]; ¹H NMR (300 MHz, CDCl₃) δ 7.73−7.71 (m, 4 H), 7.43−7.41 (m, 6H), 6.97 (dd, J = 5.7, 15.8 Hz, 1H), 6.14 (d, J = 15.8 Hz, 1H), 4.62 (t, J = 5.7 Hz, 1H), 4.23 (q, J = 7.2 Hz, 2H), 3.88−3.85 (m, 3H), 1.47 (s, 3H), 1.45 (s, 3H), 1.31 (t, J = 7.2 Hz, 3H), 1.10 (s, 9H) ; ¹³C NMR (75 MHz, CDCl₃) δ 166.0, 144.4, 135.60, 135.57, 132.9, 129.9, 129.8, 127.8, 127.7, 122.1, 109.9, 80.7, 77.5, 63.1, 60.5, 26.9, 26.8, 19.2, 14.2.

Hydrogenation of 3c into 4d

Into a solution of 3c (103 μL, 0.5 mmol) in MeOH (5 mL) was added 5% Palladium on charcoal anhydrous (12 mg, 10 wt%). The mixture was stirred under a hydrogen atomosphere at room temperature for 4 h. The reaction mixture was then filtered through Celite with EtOAc and concentrated under reduced pressure. The crude product was purified by SiO₂ column chromatography (Hex/EtOAc = 3/1) to give the compound 4d as a colorless oil (117 mg, quantitative). This product spectroscopically matched that of the known compound.^xxv; [α]_D²⁰ = −24.7 (c 2.60, CHCl₃), [lit.^xxvi [α]_D²³ = −24.1 (c 2.6, CHCl₃)]; ¹H NMR (300 MHz, CDCl₃) δ 4.10 (q, J = 7.2 Hz, 2H), 3.86 (dt = 7.5, 3.6 Hz, 1H), 3.74 (m, 2H), 3.62 (m, 1H), 2.53−2.36 (m, 3H), 1.92 (m, 1H), 1.81 (m, 1H), 1.36 (s, 3H), 1.35 (s, 3H), 1.22 (t, J = 7.2 Hz, 3H); ¹³C NMR (75 MHz, CDCl₃) δ 173.2, 108.8, 81.0, 76.1, 61.8, 60.5, 30.6, 27.9, 27.2, 26.9, 14.1.

DMP oxidation/Wittig olefination of 3c into 4e

Dess-Martin periodinane (467 mg, 1.1 mmol) was added to a solution of 3c (230 mg, 1.0 mmol) in CH₂Cl₂ (8 mL). After stirring 2.5 h at r.t., the reaction mixture was quenched with sat. NaHCO₃ (5 mL) and sat. Na₂S₂O₃ (5 mL) and vigorously stirred for 10 min. After the phase separation, the aqueous layer was extracted with CH₂Cl₂ (2 × 10 mL). The combined organics were then dried over Mg₂SO₄ and concentrated in vacuo. The crude aldehyde was immediately used for the next reaction. Into a suspension of methyltriphenylphosphonium bromide (356 mg, 1.0 mmol) in dry THF (10 mL) at −78°C (acetone/dry-ice) was added n-BuLi (360 μL, 2.5 M solution in hexane, 0.9 mmol) dropwise. After stirring for 30 min, a solution of the aldehyde in dry THF (5 mL) was added slowly over 10 min. After stirring for 1 h, the cooling bath was removed and the resulting mixture was stirred for 1 h at r.t. After quenching with half sat. NH₄Cl (15 mL), the separated aqueous layer was extracted with Et₂O (2 × 15 mL). The combined organics were washed with brine, dried over MgSO₄, and concentrated in vacuo. The crude oil was purified with silica gel column chromatography (Hex/EtOAc=95/5) to afford 85 mg colorless oil (38%). This product spectroscopically matched that of the known compound.^vi; [α]_D²⁰ = +9.48 (c 0.52, CHCl₃), [lit. [α]_D²⁰ = +14.8 (c 1.0, CHCl₃)]; ¹H NMR (300 MHz, CDCl₃) δ 6.86 (dd, J = 5.4, 15.6 Hz, 1H), 6.12 (dd, J = 1.2, 15.6 Hz, 1H), 5.83 (ddd, J =7.2, 10.2, 17.1 Hz, 1H), 5.40 (d, J = 17.1 Hz, 1H), 5.31 (d, J = 10.2 Hz, 1H), 4.28–4.10 (m, 4H), 1.47 (s, 3H), 1.45 (s, 3H), 1.30 (t, J = 7.2 Hz, 3H) ; ¹³C NMR (75 MHz, CDCl₃) δ 166.2, 143.0, 133.8, 123.1, 120.1, 110.2, 82.3, 80.1, 60.8, 27.2, 27.0, 14.4.