Desymmetrizing Asymmetric Epoxidations of Bis(cis-configured) Divinyl­carbinols: Unusual syn-Selectivity Combined with ee-Enhancement through Kinetic Resolution

Abstract

Bis(cis-1-alkenyl)methanols 2a and b were monoepoxidized by a 1:1:1 mixture of R _tert OOH, an optically active tartaric acid derivative, and a Zr(IV) alkoxide. This provided epoxy alcohols syn-1a,b with perfect diastereoselectivity. These compounds are the minor products of analogous Ti(IV)-mediated epoxidations. The ee of epoxy alcohol syn-1a was increased up to 99% when less enantioenriched material was overoxidized under the reaction conditions (→ up to 66% bisepoxy alcohol syn,syn-5).

Key words

diastereoselectivity - enantioselectivity - epoxy alcohols - kinetic resolution - 1,4-pentadien-3-ols

References and Notes

• 1a Using stoichiometric or near-stoichiometric amounts of titanium tartrate: Katsuki T. Sharpless KB. J. Am. Chem. Soc.  1980,  102:  5976 
• Using molecular sieves and <10 mol% of titanium tartrate:
• 1b Hanson RM. Sharpless KB. J. Org. Chem.  1986,  51:  1922 
• 1c Gao Y. Hanson RM. Klunder JM. Ko SY. Masamune H. Sharpless KB. J. Am. Chem. Soc.  1987,  109:  5765 
• Sharpless oxidations:
• 2a Pfenniger A. Synthesis  1986,  89 
• 2b Johnson RA. Sharpless KB. In Catalytic Asymmetric Synthesis   Ojima I. VCH; New York: 1993.  p.103 
• 2c Katsuki T. Martín VS. Org. React.  1996,  48:  1 
• 2d Johnson RA. Sharpless KB. In Catalytic Asymmetric Synthesis   2nd ed.:  Ojima I. Wiley-VCH; New York: 2000.  p.231-286  
• Tansformations of Sharpless epoxides:
• 2e Hanson RM. Chem. Rev.  1991,  91:  47 
• 2f Pena PCA. Roberts SM. Curr. Org. Chem.  2003,  7:  555 
• 3a Hatakeyama S. Sakurai K. Takano S. J. Chem. Soc., Chem. Commun.  1985,  1759 
• 3b Häfele B. Schröter D. Jäger V. Angew. Chem., Int. Ed. Engl.  1986,  25:  87 ; Angew. Chem. 1986, 98, 89
• 3c Schreiber SL. Schreiber TS. Smith DB. J. Am. Chem. Soc.  1987,  109:  1525 
• 3d Smith DB. Wang Z. Schreiber SL. Tetrahedron  1990,  46:  4793 
• 3e Nakatsuka M. Ragan JA. Sammakia T. Smith DB. Uehling DE. Schreiber SL. J. Am. Chem. Soc.  1990,  112:  5583 
• 3f Jäger V. Hümmer W. Stahl U. Gracza T. Synthesis  1991,  769 
• 4 According to ref. 3c, overoxidation of the monoepoxide increases its ee because the minor epoxide enantiomer is more reactive than the major enantiomer. A more general analysis of the stereochemistry of consecutive asymmetric functionalizations of compounds with two identical prochiral sites led to a complementary conclusion: unless a fast and a slow reacting enantiomer emerge from the first functionalization, the ee does not change during the second functionalization: Rautenstrauch R. Bull. Soc. Chim. Fr.  1994,  131:  515 
• 5a Hatakeyama S. Sakurai K. Numata H. Ochi N. Takano S. J. Am. Chem. Soc.  1988,  110:  5201 
• 5b Schreiber SL. Schreiber TS. Smith DB. J. Am. Chem. Soc.  1987,  109:  1525 
• 6a Herunsalee A. Isobe M. Pikul S. Goto T. Synlett  1991,  199 
• 6b Hatakeyama S. Satoh K. Takano S. Tetrahedron Lett.  1993,  34:  7425 
• 6c Esumi T. Fukuyama H. Oribe R. Kawazoe K. Iwabuchi Y. Irie H. Hatakeyama S. Tetrahedron Lett.  1997,  38:  4823 
• 6d Nishioka T. Iwabuchi Y. Irie H. Hatakeyama S. Tetrahedron Lett.  1998,  39:  5597 
• 6e Masaki H. Maeyama J. Kamada K. Esumi T. Iwabuchi Y. Hatakeyama S. J. Am. Chem. Soc.  2000,  122:  5216 
• 6f Bayer A. Svendsen JS. Eur. J. Org. Chem.  2001,  1769 
• 7 Berkenbusch T. Brückner R. Synlett  2003,  1813 
• 8 Spivey AC. Woodhead SJ. Weston M. Andrews BI. Angew. Chem. Int. Ed.  2001,  40:  769 
• Method:
• 19a Hatakeyama S. Satoh K. Takano S. Tetrahedron Lett.  1993,  34:  7425 
• 19b Esumi T. Fukuyama H. Oribe R. Kawazoe K. Iwabuchi Y. Irie H. Hatakeyama S. Tetrahedron Lett.  1997,  38:  4823 
• 19c Nishioka T. Iwabuchi Y. Irie H. Hatakeyama S. Tetrahedron Lett.  1998,  39:  5597 
• 20 Still WC. Kahn M. Mitra A. J. Org. Chem.  1978,  43:  2923 
• Epoxy alcohol openings at C³:
• 25a Using DIBAL-H: Finan JM. Kishi Y. Tetrahedron Lett.  1982,  23:  2719 
• 25b Using LiBH₄/Ti(Oi-Pr)₄: Dai L.-X. Lou B. Zhang Y. Guo G. Tetrahedron Lett.  1986,  27:  4343 

9

The diastereomeric compositions of monoepoxidation product 1a were determined by comparing the integrals over the following ¹H NMR signals (300 MHz, CDCl₃): δ = 2.99 [dd, J _3,4 = 7.5 Hz, J _3,2 = 4.3 Hz, 3-H (anti-1a [7] )] vs. δ = 3.07 [dd, J _3,4 = 7.4 Hz, J _3,2 = 4.5 Hz, 3-H (syn-1a)]. ‘ds >98:2’ means that the mentioned ¹H NMR signal of the minor diastereomer could not be detected.

10

All new compounds gave satisfactory ¹H NMR and ¹³C NMR spectra and provided correct combustion analyses (C and H ± 0.4%).

11

The ee of monoepoxidation product syn-1a was determined by HPLC. Column: Chiralpak AD; eluent: n-heptane-
i-PrOH 90:10; flow rate: 1.0 mL/min; UV detector: 238 nm; t _R = 48.2 min for syn-1a; t _R = 53.4 min for ent-syn-1a.

12

Synthesis of cis -(2 R ,3 S ,4 R )-2,3-Epoxy-1,7-bis[(4-methoxybenzyl)oxy]-5-hepten-4-ol ( syn -1a).
At -20 °C, l-(+)-DiPT (51 µL, 57 mg, 0.24 mmol, 1.1 equiv) and t-BuOOH (4.43 M in CH₂Cl₂, 100 µL, 0.44 mmol, 2.0 equiv) were added to a suspension of Zr(Oi-Pr)₄·i-PrOH (84 mg, 0.22 mmol, 1.0 equiv) and 4 Å MS (140 mg, powdered) in CH₂Cl₂ (4.5 mL). The mixture was stirred for 1 h at
-20 °C before a solution of divinylcarbinol 2a (85 mg, 0.22 mmol) in CH₂Cl₂ (2.0 ml) was added slowly. After stirring at -20 °C for 4 h, the reaction was quenched at -20 °C with a solution (1.0 mL) prepared from NaOH (30 g), NaCl (5 g), and H₂O (90 mL). The cooling bath was removed and the mixture stirred for 2 h at r.t. t-BuOMe (10 mL) was followed by Na₂SO₄ for drying. After filtration through a pad of Celite^® excess t-BuOOH was removed by azeotropic distillation with toluene (2 × 5 mL). The residue was purified by flash chromatography [20] (cyclohexane-EtOAc, 5:2) to afford syn-1a (56 mg, 63%) in the early fractions and a 68:32-mixture of syn,syn-5/syn-1a [33 mg, composed of 10 mg syn-1a (11%) and 23 mg syn,syn-5 (25%)] thereafter, both samples being colorless liquids. The total yield of syn-1a was 74% and its ee value 82.0% (by HPLC [11] ). Sample of 98.3% ee from the time resolved experiment (Figure [1] ; 16 h, 44% yield): [α]_D ²⁰ -8.7 (c 0.8, CHCl₃), [α]₃₆₅ ²⁰ -33.1 (c 0.8, CHCl₃). ¹H NMR (499.9 MHz, CDCl₃/TMS): δ = 2.67 (br s, OH), 3.07 (dd, J _3,4 = 7.4 Hz, J _3,2 = 4.5 Hz, 3-H), 3.25 (ddd, J _{2,1-H ( A)} = 6.5 Hz, J _2,3 = J _{2,1-H ( B)} = 4.2 Hz, 2-H), AB signal (δ_A = 3.50, δ_B = 3.65, J _AB = 11.4 Hz, in addition split by J _A,2 = 6.6 Hz, J _B,2 = 3.7 Hz, 1-H₂), 3.79 (s, 2 × OCH₃), AB signal (δ_A = 4.02, δ_B = 4.06, J _AB = 12.8 Hz, in addition split by J _A,6 = 5.9 Hz, ⁴ J _A,5 = 1.2 Hz, J _B,6 = 6.2 Hz, ⁴ J _B,5 = 1.5 Hz, 7-H₂), 4.27 (br dd, J _4,5 = J _4,3 = 7.8 Hz, 4-H), 4.39-4.51 (m, 2 × Ar-CH ₂), AB signal (δ_A = 5.66, δ_B = 5.78, J _AB = 11.3 Hz, in addition split by J _A,4 = 8.2 Hz, J _{B,7-H ( A)} = J _{B,7-H ( B)} = 5.9 Hz, A: 5-H, B: 6-H), AA′BB′ signal centered at δ = 6.87 and δ = 7.24 (2 × C₆H₄). ¹³C NMR (125.7 MHz, CDCl₃/CDCl₃): δ = 55.33 (2 × OCH₃), 56.00 (C-2), 58.93 (C-3), 65.83 (C-7), 66.79 (C-4), 67.79 (C-1), 72.41 and 73.02 (2 × benzylic CH₂), 113.92 and 113.96 (2 × C _meta ), 129.52 and 129.56 (2 × C _ortho ), 129.78 (2 × C _ipso ), 130.24 (C-5), 130.38 (C-6), 159.43 and 159.44 (2 × C _para ). Anal. Calcd (%) for C₂₃H₂₈O₆ (400.5): C, 68.98; H, 7.05. Found: C, 69.16; H, 7.16.

13

The diastereomeric composition of monoepoxidation product 1b was determined by comparing the integrals over the following ¹H NMR signals (300 MHz, CDCl₃): δ = 3.02 [dd, J _3,4 = 7.5 Hz, J _3,2 = 4.1 Hz, 3-H (anti-1b [7] )] vs. δ = 3.07 [dd, J _3,4 = 7.4 Hz, J _3,2 = 4.5 Hz, 3-H (syn-1b)]. ‘ds >98:2’ means that the mentioned ¹H NMR signal of the minor diastereomer could not be detected.

14

The ee of monoepoxidation product syn-1b was determined by HPLC. Column: Chiralpak OD-H; eluent: n-heptane-
i-PrOH 200:1; flow rate: 1.0 mL/min; UV detector: 210 nm; t _R = 24.5 min for syn-1b; t _R = 27.8 min for ent-syn-1b.

15

Analytical Data for cis , cis -2,3-5,6-Bisepoxy-1,7-bis[4-methoxybenzyl)oxy]-4-heptanol ( syn , syn -5). Colorless solid (mp 90-91 °C). ¹H NMR (499.9 MHz, CDCl₃/TMS): δ = 2.53 (br d, J _OH,4 = 5.4 Hz, OH), 3.13 (dd, J _3,4 or J _5,4 = 6.0 Hz, respectively, J _3,2 or J _5,6 = 4.5 Hz, respectively, 3-H, 5-H), 3.27 (ddd, J _{2,1-H ( A)} or J _{6,7-H ( A)} = 5.5 Hz, respectively, J _2,3 = J _{2,1-H ( B)} or J _6,5 = J _{6,7-H ( B)} = 4.6 Hz, respectively, 2-H, 6-H), 3.63-3.72 (m, 1-H₂, 4-H, 7-H₂), 3.79 (s, 2 × OCH₃), AB signal (δ_A = 4.42, δ_B = 4.50, J _AB = 11.4 Hz, 2 × Ar-CH ₂), AA′BB′ signal centered at δ = 6.87 und 7.24 (2 × C₆H₄). ¹³C NMR (125.7 MHz, CDCl₃/CDCl₃): δ = 55.33 (C-2, C-6), 55.60 (2 × OCH₃), 56.98 (C-3, C-5), 66.74 (C-4), 67.54 (C-1, C-7), 73.12 (2 × benzylic CH₂), 113.96 (2 × C _meta ), 129.57 (2 × C _ortho ), 129.64 (2 × C _ipso ), 159.48 (2 × C _para ). Anal. Calcd (%) for C₂₃H₂₈O₇ (416.5): C, 66.33; H, 6.78. Found: C, 66.14; H, 6.83.

16

This study was effected starting with 1.56 mmol 2a in 48 mL CH₂Cl₂. After the indicated times, 8 mL aliquots were removed from the reaction mixture and worked up extractively as described in ref. 12. After assessing the yields of the constituents of the crude product [17] the latter was flash-chromatographed [20] for obtaining pure samples of syn-1a for ee determination. [11]

17

The overall weights of the crude products and their ¹H NMR spectra allowed determining the yields of the constituents syn-1a, 2a, and 5 of these mixtures.

18

This experiment was performed before we learnt to prepare epoxy alcohol syn-1a as enantioselectively as documented in Table [1] and Figure [1] .

21

Analytical Data for (2 S ,4 R )-1-[4-Methoxybenzyl)oxy]-6-heptene-2,4-diol [(2 S ,4 R )-7].
Colorless liquid. [α]_D ²⁰ = -7.7 (c 0.92, CHCl₃) when prepared from anti- 6 the ee of which was 96.8% (by HPLC). ¹H NMR (499.9 MHz, CDCl₃/TMS): AB signal (δ_A = 1.55, δ_B = 1.63, J _AB = 14.4 Hz, in addition split by J _A,₄ = 8.9 Hz*, J _A,2 = 3.5 Hz*, J _B,2 = 8.5 Hz**, J _B,4 = 2.8 Hz**, 3-H₂), δ = 2.20-2.31 (m, 5-H₂), 2.54 and 2.79 (2 × br s, 2 × OH), AB signal (δ_A = 3.38, δ_B = 3.47, J _AB = 9.5 Hz, in addition split by J _A,2 = 7.7 Hz, J _B,2 = 3.7 Hz, 1-H₂), 3.80 (s, OCH₃), 3.92-3.98 (m, 4-H), 4.12 (m_C, 2-H), 4.49 (s, Ar-CH ₂), 5.09-5.14 (m, 7-H₂), 5.82 (dddd, J _{6,7-H ( Z )} = 16.3 Hz, J _{6,7-H ( E )} = 11.0 Hz, J _{6,5-H ( A)} = J _{6,5-H ( B)} = 7.2 Hz, 6-H), AA′BB′ signal centered at δ = 6.88 and 7.25 (C₆H₄); *^,** assignments interchangeable. ¹³C NMR (125.7 MHz, CDCl₃/CDCl₃): δ = 38.78 (C-3), 42.20 (C-5), 55.35 (OCH₃), 67.93 and 67.99 (C-2 and C-4), 73.09 (benzylic CH₂), 74.19 (C-1), 113.95 (C _meta ), 118.05 (C-7), 129.48 (C _ortho ), 130.05 (C _ips ), 134.77 (C-6), 159.42 (C _para ). Anal. Calcd (%) for C₁₅H₂₂O₄ (266.3): C, 67.64; H, 8.33. Found: C, 67.41; H, 8.52.

22

The ee of monoepoxidation products anti-6/ent-anti-6 was determined by HPLC. Column: Chiralpak AD; eluent: n-heptane-i-PrOH 85:15; flow rate: 1.0 mL/min; UV detector: 227 nm; t _R = 58.1 min for anti-6; t _R = 52.7 min for ent-anti-6.

23

The diastereomeric composition of monoepoxidation products 6/ent-6 was determined by comparing the integrals over the following ¹H NMR signals (300 MHz, CDCl₃): δ = 2.99 [dd, J _3,4 = 4.6 Hz, J _3,2 = 2.3 Hz, 3-H (syn-6)] vs. 3.02 [dd, J _3,4 = J _3,2 = 2.7 Hz, 3-H (anti-6)] and δ = 3.19 [ddd, J _{2,1-H ( A)} = 5.4 Hz, J _{2,1-H ( B)} = J _2,3 = 2.7 Hz, 2-H (syn-6)] vs. 3.25 [ddd, J _{2,1-H ( A)} = 5.3 Hz, J _{2,1-H ( B)} = J _2,3 = 2.7 Hz, 2-H (anti-6)].

24

The ee of monoepoxidation products syn-6/ent-syn-6 were determined by HPLC. Column: Chiralpak AD-H; eluent: n-heptane-i-PrOH 80:20; flow rate: 1.0 mL/min; UV detector: 227 nm; t _R = 28.6 min for syn-6; t _R = 26.4 min for ent-syn-6.

26

This can be inferred from the coupling constants causing the hyperfine structure of the 4-H resonance in the 500 MHz ¹H NMR spectrum in CDCl₃: 4-H of the major product 9 is a ddd at δ = 4.68 with 3 vicinal H-C-C-H couplings
(J _{4,3-H ( A)} = J _{4,3-H ( B)} = J _4,5 = ca. 6.4 Hz); in contrast, 4-H of the minor product 10 is a ddd at δ = 4.19 with 2 vicinal
H-C-C-H couplings (J _4,5 = 8.5 Hz, J _4,3 = 6.5 Hz) and an additional long-range coupling ⁴ J _4,6 = 1.0 Hz.