A new synthetic route to (S)-chroman aldehyde, a key chiral precursor of vitamin E

Article abstract of DOI:10.1055/s-2001-14642

A highly stereocontrolled synthetic route to (S)-chroman aldehyde has been developed which starts with trimethyl-p-hydroquinone and (R)-glycerladehyde acetonide and employs as key steps the Michael addition of the aromatic copper species onto the (R)-4-acryloyl-1,3-dioxolane followed by the highly diastereoselective methylation of the resulting ketone adduct.

Full text of DOI:10.1055/s-2001-14642

LETTER
989
A New Synthetic Route to (S)-Chroman Aldehyde, a Key Chiral Precursor of
Vitamin E
Hisashi Mikoshiba, Koichi Mikami, Takeshi Nakai*
Department of Chemical Technology, Tokyo Institute of Technology, Meguro-ku, Tokyo 152-8552, Japan
Fax +81-(0)3-5734-2885; E-mail: takeshi@o.cc.titech.ac.jp
Received 30 January 2001
Dedicated with honor to Prof. Ryoji Noyori
Abstract: A highly stereocontrolled synthetic route to (S)-chroman
HO
BnO
aldehyde has been developed which starts with trimethyl-p-hydro-
quinone and (R)-glycerladehyde acetonide and employs as key
steps the Michael addition of the aromatic copper species onto the
(R)-4-acryloyl-1,3-dioxolane followed by the highly diastereoselec-
tive methylation of the resulting ketone adduct.
O
O
CHO
Vitamin E
(S)-1
OH
O
Key words: chroman aldehyde, diastereoselectivity, glycerketone,
Michael addition, vitamin E
BnO
MeO
MeLi/SnCl4
3
O
(S)-1
O
0Me
(3S)-3
OH
O
OH
2
O
route a
route b
In recent years considerable effort has been devoted to the
development of efficient methods for the asymmetric syn-
thesis of vitamin E ( -tocopherol) and its key intermedi-
ate, (S)-chroman aldehyde (1).1,2 The most crucial
problem in their syntheses is associated with the stereose-
lective creation of the “quaternary” chiral center in the
chroman ring (or its tert-alcoholic precursor). While we
have recently developed a highly diastereoselective proto-
col for the methylation of “glycerketone acetonides” (4-
acyl-2,2-dimethyl-1,3-oxolanes) which affords the tertia-
ry alcohols (terpenetriols) in enantiopure form,3 we be-
came interested in developing a new synthetic route to the
vitamin E precursor (S)-1 by applying our “glycerketone’s
methylation protocol” as a key stereocontrolling step. Our
synthetic plan is depicted in Scheme 1. For the preparation
of the requisite “glycerketone” 4, routes a and b are avail-
able. Apparently, route a is more convergent and efficient
than route b, since bromide 6 required as the organometal-
lic precursor for the former is more easily accessible than
bromide 6’ for the latter.4 We now wish to report the suc-
cessful realization of the synthetic plan to eventually fur-
nish (S)-chroman aldehyde (1) in enantiopure form and
relatively high overall yield.
MeO
+
Br
O
O
O
O
OMe
OMe
5
HO
Br
0Me
6
4
OH
O
MeO
+
8
H
O
O
7
6'
Scheme 1
Michael addition reaction of the Grignard reagent gener-
ated from bromide 6 [Mg (2.0 equiv), (CH2Br)2 (1.0
equiv), THF, reflux] onto vinyl ketone 5 in the presence
of CuI and Me3SiCl in THF/HMPA at 70 °C to give the
desired ketone 4 in 64% isolated yield.7 Interestingly, the
absence of Me3SiCl led to a lower yield (39%). The next
step is the stereochemically crucial methylation of ketone
4 using our recently-reported protocol.3 Thus, ketone 4
was treated with an excess of MeLi/SnCl4 complex in
dichloromethane at -70 °C to give the tert-alcohol 3 with
high diastereomeric ratio (94: 6) in quantitative yield. The
1H NMR spectrum (CDCl3) of 3 showed two peaks due to
3-Me at 1.33 for the major and 1.20 for the minor.8 The
major product was assigned to the desired (3S)-isomer by
analogy with the NMR data reported for the epimeric
pairs of the closely related triol acetonides.3
Our synthesis began with the preparation of the aromatic
bromide 6 required for generating the organometallic spe-
cies. Thus, trimethyl-p-hydroquinone (8) was methylated
with dimethyl sulfate [(n-Bu)4NI, aq. KOH, r.t.] followed
by bromination (Br2, CCl4, 0 °C) to give bromide 62i in al-
most quantitative yield. On the other hand, the requisite
vinyl ketone (4-acryloyl-1,3-dioxolane) 5 was prepared
from the commercially available (R)-aldehyde 7 in >95%
overall yield via the reaction with vinylmagnesium bro-
mide and the Swern oxidation of the resulting alcohol
[DMSO, (COCl)2,TEA, -70 °C].5,6 It is interesting to note
that vinyl ketone 5 is rather stable enough to be isolated
by flash column chromatography. Next, we carried out the
With the highly enantio-enriched (3S)-alcohol 3 in hand,
further effort was directed to its transformation into the
target aldehyde 1 (Scheme 2). Thus, alcohol 3 was treated
with the cerium ammonium salt following the reported
procedure2i to give the quinone form 9.9 Acid treatment of
9 gave the cyclized intermediate as crystals and its recrys-
tallization from ethanol gave the stereochemically pure
compound in 94% yield. Then, the intermediate was hy-
drogenated to afford the chroman derivative 10.10 It
Synlett 2001 SI, 989–990 ISSN 0936-5214 © Thieme Stuttgart · New York

990
H. Mikoshiba et al.
LETTER
should be noted that the original (3S)-chirality in alcohol
3 is completely retained on the chroman ring of 10. After
re-protections of the diol moiety as the acetonide and the
phenolic hydroxy group as the benzyl ether, the resulting
intermediate 1111 was subjected to oxidative cleavage to
furnish (R)-aldehyde 1 in enantiopure form and 83% over-
all yield from alcohol 3; mp 50-52 °C; [ ]D +11.9 (CHCl3,
c = 1.66, 18 °C). The physical data were in agreement
References and Notes
(1) For asymmetric synthesis of vitamin E, see, e.g.: (a) Olson, L.;
Cheung, H.-C.; Morgan, K.; Saucy, G. J. Org. Chem. 1980,
45, 803. (b) Cohen, N.; Lopresti, R. J.; Saucy, G. J. Am. Chem.
Soc. 1979, 101, 6710.
(2) For selected asymmetric synthesis of (R)-chroman aldehyde 1
(or synthetic equivalents), see: (a) Tietze, L. F.; Görlitzer, J.
Synlett 1996, 1041. (b) Mizuguchi, E.; Achiwa, K. Synlett
1995, 1255. (c) Takano, S.; Yoshimitsu, T.; Ogasawara, K.
Synlett 1990, 451 (1990). (d) G. Solladie, G.; Moine, G. J. Am.
Chem. Soc. 1984, 106, 6097. (e) Takabe, K.; Okisaka, K.;
Uchiyama, Y.; Katagiri, T.; Yoda, H. Chem. Lett. 1985, 561.
(f) Fuganti, C.; Grasselli, P. J. Chem. Soc., Chem. Commun.
1982, 205. (g) Sakito, Y.; Suzukamo, G. Tetrahedron Lett.
1982, 23, 4953. (h) Cohen, N.;. Lopresti, R. J.; Neukom, C. J.
Org. Chem. 1981, 46, 2445. (i) Barner, R.; M. Schmid, M.
Helv. Chim. Acta 1979, 62, 2384.
(3) Mikoshiba, H.; Mikami, K.; Nakai, T. Heterocycles 2001, in
press.
(4) In fact, the reported synthetic method for the precursors of
bromide 6’ requires seven steps from 8: Smith, L. I.; Maullen,
C. W. J. Am. Chem. Soc. 1936, 58, 629.
1
with the literature values, and the H NMR spectral data
are also in accord with the reported ones.1b,12 LIS-NMR
2e,g
analysis using Eu(hfc)3 confirmed that aldehyde (S)-1
is enantiomerically pure.
OH
HO
O
b, c
a
O
(3S)-3
O
O
O
OH
OH
9
10
BnO
d, e
f
[ 2 ]
(S)-1
O
(5) Vinyl ketone 5: [ ]D +83.5° (CHCl3, c = 2.06, 19 °C);
1H NMR (CDCl3), 1.42 (s,3H), 1.47 (s, 3H), 3.96-4.37 (m,
2H), 4.63 (t, 1H), 5.86 (dd, 1H, J = 1.8, 11.6 Hz), 6.40 (dd, 1H,
J = 1.8, 17.4 Hz), 6.83 (dd, 1H, J = 11.6, 17.4 Hz).
(6) The attempted oxidation of the allylic alcohol with PCC
failed, while the oxidation with MnO2 gave ketone 5 in ca.
30% yield.
O
O
11
Scheme 2 Reagents: (a) Ce(NH4)2(NO3)6, MeCN (100%); (b) HCl,
aq. MeOH (96%); (c) H2, Pd/C, EtOH (100%); (d) Me2C(OMe)2, PT-
SA, Et2O (94%); (e) BnBr, K2CO3, DMF (99%); (f) H5IO6, aq. THF
(99%).
(7) Ketone 4: [ ]D +29.3° (CHCl3, c = 1.92, 20 °C); 1H NMR
(CDCl3), 1.37 (s, 3H), 1.47 (s, 3H), 2.17 (s, 6H), 2.20 (s,
3H), 2.83 (t, 4H, J = 3.0 Hz), 3.35 (s, 6H), 3.97 (dd, 1H,
J = 6.0, 6.8 Hz), 4.17 (dd, 1H, J = 6.8, 6.8 Hz), 4.43 (dd, 1H,
J = 6.0, 6.8 Hz).
In summary, we have developed a new, short synthetic
route to (R)-chroman aldehyde (1), a key chiral precursor
of vitamin E from commercially available trimethyl-p-hy-
droquinone and (R)-glycerladehyde acetonide. This syn-
thesis highlights the use of the Michael addition to
introduce the aromatic part into the chiral aliphatic frame-
work and the use of our methylation protocol to create the
crucial “quaternary” chirality in high diastereoselectivity.
Thus, this route seems to be attractive for the practical
asymmetric synthesis of vitamin E.
(8) The major isomer of alcohol 3: 1H NMR (CDCl3), 1.33 (s,
3H), 1.37 (s, 3H), 1.43 (s, 3H), 2.17 (s, 6H), 2.23 (s, 3H), 2.53-
3.00 (m 4H), 3.63 (s, 3H), 3.68 (s, 3H), 3.68 (s, 3H), 3.83-4.17
(m, 3H).
(9) Quinone 9: IR (neat), 1640 cm-1 (C=O); [ ]D +13.5° (CHCl3,
c = 0.56, 20 °C); 1H NMR (CDCl3), 1.27 (s, 3H), 1.33 (s,
3H), 1.40 (s, 3H), 1.43-1.80 (m, 2H), 1.97 (s, 6H), 2.03 (s,
3H), 2.17 (br, s, 1H), 2.37-2.83 (m, 2H), 3.77-4.10 (m, 3H).
(10) Diol 10: [ ]D 5.4° (MeOH, c = 0.88, 19 °C); 1H NMR
(DMSO-d6), 1.10 (s, 3H), 1.75 (t, 2H, J = 6.8 Hz), 2.00 (s,
3H), 2.03 (s, 3H), 2.06 (s, 3H), 2.27-2.73 (t, 2H, J = 6.8 Hz),
4.20-4.43 (m, 2H), 4.73-4.90 (m, 1H), 7.37 (s, 1H).
(11) Intermediate 11: mp, 105-107 °C; [ ]D +55.0° (CHCl3,
c = 0.39, 22 °C); 1H NMR (CDCl3), 1.20 (s, 3H), 1.37 (s,
3H), 1.47 (s, 3H), 1.87 (t, 2H, J = 6.8 Hz), 2.07 (s, 3H), 2.13
(s, 3H), 2.20 (s, 3H), 2.63 (t, 2H, J = 6.8 Hz), 3.87-4.33 (m,
3H), 4.07 (s, 2H), 7.27-7.57 (m, 5H).
(12) Aldehyde (S)-1: 1H NMR (CDCl3), 1.40 (s, 3H), 1.77 (t, 2H,
J = 6.8 Hz), 2.10 (s, 3H), 2.17 (s, 3H), 2.23 (s, 3H), 2.57 (t, 2H,
J = 6.8 Hz), 4.70 (s, 2H), 7.24-7.63 (m, 5H), 9.60 (s, 1H).
Article Identifier:
1437-2096,E;2001,0,SI,0989,0990,ftx,en;Y03601ST.pdf
Synlett 2001, SI, 989–990 ISSN 0936-5214 © Thieme Stuttgart · New York