Asymmetric synthesis of cyclic hydroxy ketones derived from enol ethers via sharpless asymmetric dihydroxylation. A study in the correlation of the enol ether chain length and enantioselectivity

Article abstract of DOI:10.1021/jo034854o

The Sharpless asymmetric dihydroxylation reaction of enol ethers derived from their corresponding cyclic ketones, gave α-hydroxyketones with high enantioselectivity. The enantiomeric excess was found to be proportional to the length of the unbranched enol ether chain with a maximum ee for the pentyl enol ether. An efficient synthesis of α-hydroxy chromanone in >90% ee was demonstrated using this method.

Full text of DOI:10.1021/jo034854o

Asym m etr ic Syn th esis of Cyclic Hyd r oxy Keton es Der ived fr om
En ol Eth er s via Sh a r p less Asym m etr ic Dih yd r oxyla tion . A Stu d y
in th e Cor r ela tion of th e En ol Eth er Ch a in Len gth a n d
En a n tioselectivity
Benjamin F. Marcune,* Sandor Karady, Paul J . Reider, Ross A. Miller, Mirlinda Biba,
Lisa DiMichele, and Robert A. Reamer
Department of Process Research, Merck and Co., P.O. Box 2000, Rahway, New J ersey 07765
ben_marcune@merck.com
Received J une 18, 2003
The Sharpless asymmetric dihydroxylation reaction of enol ethers derived from their corresponding
cyclic ketones, gave R-hydroxyketones with high enantioselectivity. The enantiomeric excess was
found to be proportional to the length of the unbranched enol ether chain with a maximum ee for
the pentyl enol ether. An efficient synthesis of R-hydroxy chromanone in >90% ee was demonstrated
using this method.
In tr od u ction
Resu lts a n d Discu ssion
Our approach to hydroxychromanone (3) and its con-
version to cis-aminochromanol (5) are outlined in Scheme
1. The methyl enol ether 2a was prepared as described
in the literature2 by heating the ketone 1 with trimethy-
lorthoformate, excess methanol, and catalytic p-toluene-
sulfonic acid followed by slow distillation.
Our initial asymmetric dihydroxylation reactions re-
quired double the suggested quantity2 of the commer-
cially available AD â-mixture along with additive meth-
anesulfonamide, in a 50/50 mixture of acetonitrile/water
at 0 °C 18 h for completion. (R)-R-Hydroxy chromanone
(3) was produced in 66% LC assay yield and 82% ee. The
AD R-mixture was used on the TBDMS enol ether of
chromanone, yielding the (S)-R-hydroxy ketone in 81%
LC assay yield and 58% ee. In addition to the modest
induction, this method was impractical because for each
gram of 2a , 13 g of reagent and 65 mL of solvent were
required.
Better yields and a more practical process were ob-
tained when we directly prepared the Sharpless AD
â-mixture with a few modifications.6 As reported in the
literature, the use of sodium persulfate as a co-oxidant
requires catalytic amounts of potassium ferricyanide.7 By
applying this method, we were able to reduce the amount
of potassium ferricyanide from 6 to 0.2 equiv by using
only 1.5 equiv of sodium persulfate (for each gram of 2a ,
now only 6 g of reagents and 24 mL of solvent were
required). When applied to the methyl enol ether of
chromanone, the assay yield increased to 90% with the
ee at 83% (R).
R-Hydroxy ketones are important synthons for the
chiral synthesis of natural products, fine chemicals, and
medicines.1 Previously, Sharpless showed that they can
be synthesized from methyl and tert-butyldimethylsilyl
(TBDMS) enol ethers of open-chain hydrocarbons2 and
also methyl and TBDMS enol ethers of cyclic ketones3
using the osmium-catalyzed asymmetric dihydroxylation
(AD) reaction.4
We had been looking for a new efficient chiral prepara-
tion of R-hydroxy ketone 3 as an intermediate to cis-
aminochromanol (5)5 (Scheme 1), an important class of
amino alcohol for the development of biologically active
compounds. When we investigated the Sharpless AD
reaction as a possible method for producing 3, we
discovered a correlation between the enol ether chain
length and enantioselectivity. In this paper, we present
a study of the Sharpless AD reaction of enol ethers
derived from cyclic ketones that examines this correla-
tion. The results of this study have produced a general
method for the asymmetric synthesis of cyclic hydroxy
ketones and an efficient new preparative method for the
synthesis of 3.
(1) (a) Hanessian, S. Total Synthesis of Natural Products: The
Chiron Approach; Pergamon Press: New York, 1983; Chapter 2. (b)
Oppolzer, W. Angew. Chem., Int. Ed. Engl. 1984, 23, 876. (c) Fujita,
M.; Hiyama, T. J . Am. Chem. Soc. 1984, 106, 4629 (c) Girijavallabhan,
V. M.; Ganguly, A. K.; Pinto, P. A.; Sarre, O. Z. Bio. Med. Chem. Lett.
1991, 1, 349.
(2) Hashiyama, T.; Morikawa, K.; Sharpless, B. K. J . Org. Chem.
1992, 57, 5067.
(3) Morikawa, K.; Park, J .; Anderson, P. G.; Hashiyama, T.; Sharp-
less, B. K. J . Am. Chem. Soc. 1993, 115, 8463.
(4) For review, see: Kolb, H. C.; VanNieuwenhze, M. S.; Sharpless,
B. K. Chem. Rev. 1994, 94, 2483.
(5) For literature on work previously performed on the synthesis of
cis-aminochromanol, see: (a) Davies, I. W.; Taylor, M.; Marcoux, J .
F.; Matty, L.; Wu, J .; Hughes, D.; Reider, P. J . Tetrahedron Lett. 2000,
41, 8021. (b) Hansen, K. B.; Rabbat, P.; Springfield, S. A.; Devine, P.
N.; Grabowski, E. J . J .; Reider, P. J . Tetrahedron Lett. 2001, 42, 8743.
(6) For 5 mmol of enol ether, the AD-mixture contained K3Fe(CN)6
(329 mg, 1 mmol), Na2S2O8 (1.79 g, 7.5 mmol), K2CO3 (4.15 g, 30 mmol),
(DHQD)2PHAL for the â-mixture or (DHQ)2PHAL for the R-mixture
(77.9 mg, 0.1 mmol), K2OsO4‚2H2O (18.4 mg, 0.05 mmol), and CH3-
SO2NH2 (500 mg, 5.25 mmol).
(7) Byun, H.-S., Kumar, E. R.; Bittman, R. J . Org. Chem. 1994, 59,
2630.
10.1021/jo034854o CCC: $25.00 © 2003 American Chemical Society
Published on Web 09/16/2003
8088
J . Org. Chem. 2003, 68, 8088-8091

Asymmetric Synthesis of Cyclic Hydroxy Ketones
SCHEME 1
TABLE 1. AD of En ol Eth er s w ith Mod ified r- a n d â-Mixtu r es
yield %8
ee %
entry
compd
R
R-mixture
â-mixture
R-mixture
66 (S)
â-mixture
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
2a
2b
2c
2d
2e
2f
2g
2h
2i
methyla
ethyla
69
90
75
87
74
99
49
70
59
23
75
94
89
79
82
85
95
70
87
99
25
89
47
31
78
50
83 (R)
89 (R)
92 (R)
92 (R)
94 (R)
95 (R)
20 (R)
86 (R)
76 (R)
84 (R)
64 (R)
93 (R)
94 (R)
97 (R)
36 (R)
94 (R)
38 (R)
83 (R)
93 (R)
92 (R)
36 (R)
79 (S)
92 (S)
87 (S)
90 (S)
propyla
butyla
pentyl
decyl
73
77
80 (S)
92 (S)
isopropyl
isobutyl
isopentyl
methoxy ethyl
benzyl
91
12 (S)
2j
2k
6a
6b
6c
6d
6e
6f
8a
8b
8c
8d
10a
10b
12a
12b
methyla
propyl
88
92 (S)
pentyl
53
68
96 (S)
26 (S)
isopropyl
isobutyl
phenyl
methyla
propyl
80
74 (S)
pentyl
53
68
48
49
49
47
94 (S)
26 (S)
76 (R)
90 (R)
83 (R)
92 (R)
isopropyl
methyla
pentyl
methyla
pentyl
a
Enol ethers that were made according to ref 2.
SCHEME 2
isolated in 87% yield. Yields of pure enol ethers of chain
length pentyl and higher were lower (e.g., pentyl ) 50%
yield) because of mixed fractions and large pot residues
due to polymerization.
The enol ethers 2a -k (Table 1, entries 1-11) were
hydroxylated using the modified â-mixture conditions.
Surprisingly, a steady increase in enantiomeric excess
from 83 to 94% was observed with enol ethers from
methyl to pentyl (entries 1-5). Any further increase in
chain length gave no significant improvement in enan-
tioselectivity, as demonstrated by the decyl enol ether,
with only a 1% increase in ee (entry 6).
On the basis of the literature, we did not expect to see
much variation of enantioselectivity with the ether chain
length. However, we decided to examine this by preparing
enol ethers 2a -k (Table 1). The ethyl, propyl, and butyl
enol ethers were prepared as stated above via the
trialkylorthoformate method. When this method was
applied to the higher boiling alcohols, the reactions were
very sluggish and incomplete. Attempts to make the
isopropyl enol ether of chromanone (2g) gave no reaction.
This problem was resolved by implementing a simple
exchange reaction (Scheme 2). The methyl enol ether 2a
was heated with an excess of the higher alcohol and
catalytic p-toluenesulfonic acid while distilling off metha-
nol. Fractional distillation gave the desired higher enol
ethers in moderate to good yields. Enol ether 2g was
When the modified AD â-mixture was applied to 2g
(entry 7), the ee decreased dramatically to 20%, indicat-
ing that branching of the alkyl group is not advantageous
to enantioselectivity. When the branching occurred after
the C1 position, the ee recovered to a level comparable
to the unbranched cases. This is demonstrated by the
isobutyl ether (86% ee) (entry 8) and the isopentyl ether
(76% ee) (entry 9). However, the enantioselectivity of the
branched enols did not exceed that of the unbranched
enols.
J . Org. Chem, Vol. 68, No. 21, 2003 8089

Marcune et al.
SCHEME 3. Absolu te Con figu r a tion Deter m in a tion via MP A Ester Der iva tiza tion of 9
TABLE 2. Ch em ica l Sh ifts of r- a n d γ-P r oton s of MP A
Ester s
δ (R,R)- or
(S,S)-MPA ester
δ (R,S)- or
(S,R)-MPA ester
∆δ (R,S)- or
(S,R)-
compd
3 R 1H
3 R 1H
3 γ 1H
7 R 1H
7 R 1H
7 γ 1H
9 R 1H
9 R 1H
9 γ 1H
4.55
4.45
7.85
3.07
2.30
8.00
2.22
2.09
7.65
4.39
4.27
7.90
3.01
2.20
8.04
2.08
1.98
7.44
0.16
0.18
-0.05
0.06
0.10
-0.04
0.14
0.11
-0.09
synthesis. There are no simple efficient methods for the
asymmetric synthesis of either enantiomer. Therefore,
it was rewarding to observe that our method is applicable
to the synthesis of R-hydroxycyclohexanone (11) and
R-hydroxycycloheptanone (13). The R-hydroxy ketones
were produced in good ee with the R- and â-mixtures.
The pentyl enol ethers 10b and 12b gave an enantiomeric
excess g 90%.
F IGURE 1. Enol Ethers to R-Hydroxy Ketones With Modified
Absolu te Con figu r a tion . Absolute configurations for
3,5b 7,9 and 1110 have been reported in the literature. The
absolute configuration of 9 was assumed on the basis of
the following supportive data.
AD Conditions.
Replacement of one of the carbon atoms in the enol
chain with a heteroatom reduced the selectivity, as
demonstrated by comparing the methoxyethyl ether (84%
ee) (entry 10) with the butyl ether (92% ee) (entry 4).
Finally, when the modified AD â-mixture was applied to
the benzyl ether, the enantiomeric excess was a modest
64% (entry 11), indicating again that branching can be
deleterious to the enantioselectivity. The same trends
were observed with the R-mixture but with an even
greater range of ee when going from the methyl (66% ee)
(entry 1) to the pentyl (92% ee) (entry 5). The isopropyl
ether gave nearly racemic hydroxyketone with an ee of
just 12% (entry 7).
The SFC retention time profile of compounds 3 and 7
showed that the (R)-enantiomers eluted later than the
(S)-enantiomers. We therefore extrapolated that com-
pound 9, being structurally similar to 3 and 7, would have
an analogous SFC profile, with the (R)-enantiomer being
the later elutant.
Derivatization experiments were carried out on 9 on
the basis of the work of Riguera.11 He showed that the
absolute configuration of secondary alcohols can be
1
deduced by H NMR by esterifying both alcohol enanti-
omers with either (R)- or (S)-methoxyphenylacetic acid
(MPA) (Scheme 3). Comparisons of the adjacent protons
of the MPA ester bearing a carbon determine the
configurations as shown by his mnemonic. The results
of these experiments were consistent with the same
experiments as carried out on hydroxy ketones 3 and 7
with known configurations (Table 2).
These results show that optimal enantioselectivity was
obtained with the pentyl enol ether, giving a yield of 99%
and an ee of 94%. The decrease in quantity of reagents
by 80%, along with the high yield and high ee, made the
Sharpless AD reaction a useful, preparative method for
the production of either enantiomer of R-hydroxychro-
manone.
As summarized in Table 1, the method is applicable
for the enantioselective hydroxylation of enol ethers
derived from other cyclic ketones (Figure 1) such as
tetralone and benzosuberone with the same effect of
chain length and branching of the alkyl group as 2. The
pentyl enol ethers 6c and 8c gave excellent enantiose-
lectivity with ees of 97 and 92%, respectively. Interest-
ingly, the phenyl ether 6f (entry 17) with branching
occurring at C1 gave an enantiomeric excess of the same
magnitude as the isopropyl ether 6d (entry 15).
Racemic R-hydroxycyclohexanone is a commercially
available reagent used as a building block in organic
The absolute configuration of 13, made with the
â-mixture, was designated (S)- on the basis of the
assumption of analogous chiral GC retention times and
optical rotations with the known (S)-R-hydroxycyclohex-
anone, also made with the â-mixture.
(8) Standard conditions were used for all AD reactions; therefore,
the yields were not optimized.
(9) Naemura, K.; Wakebe, T.; Hirose, K.; Tobe, Y. Tetrahedron:
Asymmetry 1997, 8, 2585.
(10) Lee, L. G.; Whitesides, G. M. J . Org. Chem. 1986, 51, 25.
(11) Seco, J . M.; Quinoa, E.; Riguera, R. Tetrahedron: Asymmetry
2001, 12, 2915.
8090 J . Org. Chem., Vol. 68, No. 21, 2003

Asymmetric Synthesis of Cyclic Hydroxy Ketones
Gen er a l P r oced u r e for r-Hyd r oxy Keton es 3, 7, 9, 11,
a n d 13. P r ep a r a tion of (R)-r-Hyd r oxych r om a n on e (3)
fr om 1a . A slurry of K3Fe(CN)6 (1.65 g, 5.0 mmol), Na2S2O8
(8.95 g, 37.5 mmol), K2CO3 (20.8 g, 150 mmol), K2OsO4‚2H2O
(92 mg, 0.25 mmol), CH3SO2NH2 (2.5 g, 26.3 mmol), and
(DHQD)2PHAL (390 mg, 0.50 mmol) in CH3CN (50 mL) and
water (50 mL) was aged at 0 °C for 10 min. To the slurry was
added the methyl enol ether 2a (4.05 g, 25 mmol), and the
mixture was aged overnight at 0 °C. Methyl tert-butyl ether
(MTBE) (50 mL) and 20% brine solution (50 mL) were added,
and the layers were separated. The aqueous layer was back-
extracted with MTBE (50 mL) and the combined organic layers
were washed with 20% brine solution. The organic layer was
dried with Na2SO4 and stripped of solvents to give 3 as a yellow
Con clu sion
We have shown that the Sharpless AD reaction, with
a few modifications, gives R-hydroxyketones from enol
ethers of cyclic ketones, with high enantioselectivity. The
degree of the enantiomeric excess is proportional to the
length of the unbranched enol ether chain, giving opti-
mum enantioselectivity with the pentyl moiety. Branch-
ing of the enol ether chain results in decreased selectivity,
especially with branching at the C1 carbon. These results
led to an efficient scalable synthesis of R-hydroxy chro-
manone (99% yield, 94% ee), a precursor of cis-ami-
nochromanol, an important amino alcohol for the syn-
thesis of vital biologically active compounds. The method
has also been shown to be effective for the production of
other enantiomerically enriched R-hydroxy ketones in-
cluding (R)- and (S)-R-hydroxycyclohexanone, an impor-
tant building block in organic synthesis.
oil (3.65 g, 22.5 mmol). [R]20 +71.25 (CHCl3); SFC retention
D
times, 7.6 min for the (R)-enantiomer and 6.7 min for the (S)-
enantiomer. Isolation of 3 as a white crystalline solid was
achieved by crystallization from ethyl acetate and hexanes.
(6R)-5-Oxo-6,7,8,9-tetr a h yd r o-5H-ben zo[a ][7]a n n u len -
6-yl (2S)-(Meth yloxy)(p h en yl)eth a n oa te (14a ): 1H NMR
(CDCl3, 400 MHz) δ 1.80 (m, 1H), 1.98 (m, 1H), 2.04 (m, 1H),
2.08 (m, 1H), 2.95 (m, 2H), 3.46 (s, 3H), 4.40 (s, 1H), 5.42 (m,-
1H), 7.19 (d, J ) 7.6, 1H), 7.30 (t, J ) 7.6, 1H), 7.35 (om, 3H),
7.41 (td, J ) 7.6, 1.5, 1H), 7.45 (m,2H), 7.74 (dd, J ) 7.6, 1.5,
1H); 13C NMR (CDCl3, 100 MHz) δ 23.3, 28.9, 34.0, 57.5, 78.3,
82.2, 126.7, 127.3, 128.6, 128.7, 129.2, 129.9, 132.1, 136.1,
136.5, 141.4, 169.9, 199.1.
Exp er im en ta l Section
Gen er a l P r oced u r e for En ol Eth er s 2e-k , 6b-f, 8b-
d , 10b, a n d 12b. P r ep a r a tion of 4-Isop r op oxy-2H-
ch r om en e (2g). Methyl enol (2a ) (10.0 g, 61.7 mmol), 2-pro-
panol (10 mL), and p-toluenesulfonic acid monohydrate (20 mg,
1.105 mmol) were combined and heated at 80 °C for 2 h. The
temperature was increased to 110 °C allowing the low-boiling
components to distill. A concentric tube column distillation
head was used to collect 10.2 g of a clear colorless oil containing
78 mol % 2g at 112 °C and 4 Torr: 1H NMR (CDCl3, 400 MHz)
δ 1.38 (d, J ) 6.1, 6H), 4.4 (m, 1H), 4.74 (t, J ) 3.9, 1H), 4.88
(d, J ) 3.9, 2H), 6.85 (dd, J ) 8.1, 1.0, 1H), 6.95 (td, J ) 7.5,
1.1 1H), 7.19 (td, J ) 7.7, 1.7, 1H), 7.49 (dd, J ) 7.7, 1.7 1H);
13C NMR (CDCl3, 100 MHz) δ 21.8, 65.8, 69.1, 90.4, 115.6,
120.9, 121.3, 122.5, 129.6, 148.1, 155.7; GCMS m/z 190, 147,
121, 91, 65.
Su p p or tin g In for m a tion Ava ila ble: General experimen-
tal procedure, product data for compounds 2b -f, 2h -k , 7,
6a -f, 8a -d , 9, 10a ,b, 11, 12a ,b, 13, 14b, 15a ,b, and 16a ,b,
1
and copies of H and 13C NMR spectra for compounds 2-16.
This material is available free of charge via the Internet at
http://pubs.acs.org.
J O034854O
J . Org. Chem, Vol. 68, No. 21, 2003 8091