Novel, cyclic and bicyclic 1,3-diols as catalysts for the diethylzinc addition to aldehydes

Article abstract of DOI:10.1039/b110082n

A number of optically pure 1,3-diols have been synthesized and used as catalysts in the asymmetric addition of diethylzinc to aromatic aldehydes. Enantiomeric excesses of up to 92% of (R)-1-phenylpropan-1-ol were obtained with anisylbicyclo[2.2.2]octanedi

Full text of DOI:10.1039/b110082n

View Article Online / Journal Homepage / Table of Contents for this issue
Novel, cyclic and bicyclic 1,3-diols as catalysts for the diethylzinc
addition to aldehydes
Ian Sarvary, Yiqian Wan and Torbjörn Frejd*
1
Department of Organic and Bioorganic Chemistry, Center for Chemistry and Chemical
Engineering, Lund University, P.O. Box 124, S-221 00 Lund, Sweden.
E-mail: Torbjorn.Frejd@orgk1.lu.se
Received (in Cambridge, UK) 5th November 2001, Accepted 25th January 2002
First published as an Advance Article on the web 11th February 2002
A number of optically pure 1,3-diols have been synthesized and used as catalysts in the asymmetric addition of
diethylzinc to aromatic aldehydes. Enantiomeric excesses of up to 92% of (R)-1-phenylpropan-1-ol were obtained
with anisylbicyclo[2.2.2]octanediol (14) as a catalyst. Using 2-picolylbicyclo[2.2.2]ocatanediol (16) as the catalyst
resulted in a reversal of the stereoselectivity, yielding (S)-1-phenylpropan-1-ol in 83% ee. A pronounced positive
non-linear effect was observed when varying the enantiomeric purity of catalyst 14.
Introduction
The asymmetric addition of diethylzinc to aromatic aldehydes
is a convenient test reaction for novel catalysts and several
highly efficient catalysts based on amino alcohols and diols
have been found for this reaction. Both classes of compounds
have been used with and without extra metals, such as Ti().
The amino alcohols are most frequently used as catalysts with-
out addition of extra metals, while in most diol cases Ti() has
been used to form titanate catalysts to obtain high yields and
ees. Without the use of Ti()-coordination the diols themselves
often show lower catalytic activity and selectivity.1,2
Besides the addition of alkylzinc to aldehydes, bicyclic amino
alcohols have been reported to be efficient catalysts for the
diethylzinc addition to imines.3 Recent developments with the
reagent have enabled highly selective addition of diphenylzinc
to aldehydes.4
One of the better diol catalysts is the diphenylglycol 1 (Fig. 1).
In comparison, TADDOL (2) produced only moderate ees.
Also, BINOL (3) alone or derivatives thereof, such as 4, were
poor catalysts in the diethylzinc addition to benzaldehyde.5–7
However, coordinating ether moieties attached to the BINOL
framework gave better catalysts as exemplified by 5 and den-
drimers based on that subunit.8 Other systems, including
polymeric materials, showed even better enhancement, see
below. The use of coordinating alkoxy groups has also been
used in bicyclic diol catalysts such as 6.9 In contrast, the use
of coordinating 2-pyridyl groups, as in 7 and 8, reduced the
selectivity substantially in comparison with 1.10,11
Compound 9, despite its coordinating diaryl ether link, gave
only low ees, although it has only been investigated in auto-
catalytic reactions (with the corresponding dialdehyde as the
aldehyde component) and is therefore a somewhat different
case.12 Some sugar derivatives have also been investigated,
such as modified β-cyclodextrines and catalysts derived from
-mannitol (10). However, these were poor catalysts in the
diethylzinc addition to aldehydes.13,14
Recent developments with BINOL derivatives by Pu and co-
workers15 have shown that extra, bulky and coordinating ether
groups, increase catalytic efficiency. Some of these BINOL
derivatives, both polymers and monomers (12a and 12b, Fig. 2),
show a dramatic increase in reactivity and selectivity. Thus,
12a gave 98% ee and 12b 99% ee in the diethylzinc addition to
benzaldehydes and worked excellently in other reactions as well.
In addition, similar BINOL derivatives have been anchored to
Fig. 1 Diols used as catalysts (without the use of Ti(OPri)4) for the
asymmetric addition of Et2Zn to benzaldehyde. * denotes only used in
autocatalytic reactions where the corresponding dialdehyde was used as
substrate.
DOI: 10.1039/b110082n
J. Chem. Soc., Perkin Trans. 1, 2002, 645–651
This journal is © The Royal Society of Chemistry 2002
645

View Article Online
Results and discussion
In the BODOLs the coordinating hydroxy groups are posi-
tioned at a very defined distance from each other and also in
relation to the rest of the molecule, due to the rigid bicyclic
framework. By removing the ethylene bridge the importance of
this fixation should be revealed. We therefore included the
cyclohexane-1,3-diol derivatives 19 and 20 in this investigation.
We were also interested in where to place an attachment
group in the BODOLs if these ligands/catalysts were to be
covalently anchored to a solid phase. Two obvious positions
were at the bridgeheads, which led to 17 and 18 as test sub-
stances. Moreover, these were reasonable comparison objects
to 19 and 20, with respect to the possible influence of the gem-
dimethyl grouping. Thus, compounds 14–20 were tested as
catalysts in the diethylzinc addition to a number of aldehydes.
First, the choice of solvents had to be made. In either toluene
or mixtures of diethyl ether–hexane (2 : 3) we obtained good
results. Not surprisingly, the coordinating ability of THF
interfered with the complex(es) formed between the diols and
diethylzinc. This resulted in loss of catalytic activity (even at
THF levels below 10% in hexane) since only small amounts of
phenylpropan-l-ol (<10%) formed.
The best compromise was to perform the reactions at 0 ЊC.
At this temperature the yield and ee were 89% and 92% respec-
tively. At Ϫ20 ЊC the yield was much lower (42%) and the ee was
only 2% better then at 0 ЊC. Moreover, the reaction time
was very long. At room temperature the yield was good (93%)
but now the ee was lowered to 72%. We therefore chose to run
the remaining catalytic reactions at 0 ЊC for 40 h.
Fig. 2 A chiral BINOL polymer (12a) and a BINOL derivative (12b)
used as a catalyst, Pu and co-workers.15
polystyrene residues and used with good results in the diethyl-
zinc addition.16
Interestingly, the extra coordinating sites do not have to be
incorporated into the diols to increase their catalytic ability.
Large increases of both activity and selectivity, for BINOL, and
derivatives thereof, were achieved by Mikami and co-workers17
in the diethylzinc addition by adding chiral diimines such as 13
(Fig. 3) as activators. For example, a combination of 4 and 13
As shown, the use of 14 (Fig. 4) as a catalyst in diethyl ether–
Fig. 3 The most efficient combination of BINOL derivative (4) and
diimine activator (13), used by Mikami and co-workers.17
(10 mol% of each) resulted in 99% ee in the diethylzinc addition
to benzaldehyde.
From the cases reported in the literature it seems that
among the diols only 1,2- and 1,4-diols have been found to
act as catalysts. To our knowledge, only one 1,3-diol (11)
has been examined as a catalyst, although without exhibiting
any catalytic effect.6 We asked ourselves whether the 1,3-
diols in general were useless as catalysts in these types of
reactions. This question was all the more of interest to us, as
the 1,3-diol 14 (anisyl-BODOL†‡) was an efficient ligand in
the titanium catalysed asymmetric reduction of ketones with
catecholborane.18,19
Fig. 4 The 1,3-diols used in this investigation.
hexane gave a fair yield and good ee of (R)-1-phenylpropan-1-
ol (Table 1, entry 1A). Changing the solvent to toluene lowered
both the ee and the yield somewhat (entry 1B). The necessity of
an extra coordinating site in the catalyst was demonstrated
in compound 15, in which such coordination was lacking and
which gave a racemic product in low yields independent of
the solvent (entries 1C and 1D). We found slightly more 1-
phenylpropan-1-ol when using 15 as a catalyst, compared with
the background reaction (7%, in hexane–Et2O after 42 h).
To investigate autocatalysis we used the following experiment.
In this report we demonstrate that the BODOLs themselves
may act as catalysts in the diethylzinc addition to aldehydes. It
is also shown that 1,3-diols lacking the rigidity of the bicyclic
backbone were less efficient.
† We suggest that the diols based upon the bicyclo[2.2.2]octane frame-
work be named BODOLs (bicyclo[2.2.2]octanediols).
‡ The IUPAC name for anisyl is methoxyphenyl.
646
J. Chem. Soc., Perkin Trans. 1, 2002, 645–651

View Article Online
Table 1 Addition of diethylzinc to benzaldehyde using catalysts 14–20
Entry
Catalyst
Solventa
Eeb (%)
Yieldc (%)
Config.d
1A
1B
Hexane–Et2O
Toluene
92
80
89
84
R
R
1C
1D
Hexane–Et2O
Toluene
–
–
17
12
–
–
1E
1F
Hexane–Et2O
Toluene
83
82
80
65
S
S
1G
1H
Hexane–Et2O
Toluene
28
46
27
39
S
S
1I
1J
Hexane–Et2O
Toluene
89
56
85
76
R
R
1K
1L
Hexane–Et2O
Toluene
40
61
34
26
S
S
1M
1N
Hexane–Et2O
Toluene
16
36
25
37
S
S
a Hexane–Et2O in the ratio 3 : 2. b Determined by HPLC, Chiralcel OD–H. c Isolated yields. d Established through optical rotation.
By using optically active (R)-1-phenylethan-1-ol as a catalyst,
instead of the BODOLs, only racemic 1-phenylpropan-1-ol
was obtained in 12% yield (GC). As no optical activity could
be detected, this indicated that the background reaction was
slightly enhanced by the addition of alcohols, but that no
autocatalysis occurred.
When using the 2-picolyl-BODOL§ 16 the enantioselectivity
was reversed but a fair yield and good ee of (S)-1-phenyl-
propan-1-ol was obtained (entry 1E). Changing the solvent to
toluene in this case gave a lower yield, but the ee was almost the
same (entry 1F). Similar reversals of enantioselectivity have
been observed when using fenchone-based catalysts.20
When applying a catalyst having a methyl group at the
bridgehead position between the hydroxy groups as in 17 both
yield and ee were lower than with 14 (entry 1G). This may be
explained by the bulk of the methyl group interfering with the
structure of the catalyst. It should also be noted that the
bridgehead methyl in 17 caused a reversal of the enantio-
selectivity in comparison with 14. A change of solvent to
toluene made a considerable improvement in the ee of 17 (from
28 to 46% ee) and had a positive effect on the yield as well
(entry 1H). On the other hand, when the methyl group was
positioned at the back side bridgehead (position 4), as in 18,
the differences in yield and ee compared to those achieved with
14 were negligible (entry 1I). In this case both the yield and
ee of (R)-1-phenylpropan-1-ol decreased substantially when the
reaction was performed in toluene (entry 1J).
The less rigid diols 19 and 20 worked as catalysts but
gave inferior results. Interestingly, 19, with the hydroxy groups
syn-oriented, was better than 20 (entries 1K–1N). More
interesting was the observation that 19 worked better than
17 as a catalyst. Thus, in forming a zinc complex, 17 is
apparently forced to place its bridgehead methyl group
unfavourably, while 19, due to its greater flexibility may com-
pensate for this to some degree, by adjusting the cyclohexane
ring conformation.
It should also be mentioned that the catalysts could be
isolated after the reaction in good yields (>80%). The recycled
catalysts could be used without any significant changes in either
yield or ee.
§ The IUPAC name for picolyl is methylpyridyl.
J. Chem. Soc., Perkin Trans. 1, 2002, 645–651
647

View Article Online
reagents to this aldehyde may result in low ees.24–26 It is believed
that the presence of the coordinating nitrogen close to the
aldehyde group, allows the formation of a bidentate complex
of pyridine-2-carbaldehyde, thereby accelerating the non-
catalysed reaction.26 In the reaction with ferrocene carb-
aldehyde no product could be isolated when performing the
reaction under the standard conditions. Increasing the tem-
perature to 22 ЊC for 24 h produced racemic 1-ferrocenyl-
propan-1-ol in 72% yield (entry 2F). As expected, having an
electron donating methoxy-group in the 4-position gave a lower
ee and yield in comparison with benzaldehyde (entry 2G).
Table 2 Addition of diethylzinc to aromatic aldehydes using anisyl-
BODOL
Entry
R
Eea (%)
92
Yieldb (%)
86
Config.d
Investigation of non-linear effects
2A
R
Not long after the first observation of a NLE (Non Linear
Effect) was reported in asymmetric catalysis,27 a strong asym-
metric amplification in the catalytic asymmetric diethylzinc
addition to aldehydes was found.28,29 Therefore, the catalytic
asymmetric organozinc additions to carbonyl compounds,
with amino alcohols as catalysts, are closely associated with
NLEs. However, to our knowledge, only two such investigations
have been performed with diols as catalysts in the addition of
diethylzinc to aldehydes. These investigations were performed
with the catalysts 12b and 6. A positive NLE was found when
using the latter, but no obvious NLE was noticed with 12b.15
As a NLE requires that either the catalyst itself or a precatalyst
(in equilibrium with the true catalyst) must incorporate at least
two ligands, it seems likely that the bulk of 12b may prevent
formation of dimers or larger complexes.
We have previously observed dimer formation between 14
and Ti(OPri)4 in toluene-d8, benzene-d6 and THF-d8 using
NMR spectroscopy,19 a positive NLE was indeed found in the
asymmetric reduction of ketones with catecholborane, cata-
lysed by a mixture of 14 and Ti(OPri)4. Naturally, we wanted to
see if this also was the case with 14 in the asymmetric addition
of diethylzinc to benzaldehyde. Unfortunately, structural
information of the complex(es) formed between 14 and Et2Zn
in toluene-d8 could not be acquired by NMR spectroscopy,
due to a multitude of low intensity peaks in addition to peak
broadening. Allowing the solutions to mature or addition of
more Et2Zn did not improve the spectra. Also, efforts to obtain
X-ray quality crystals were unsuccessful. The solid material
obtained from mixtures of 14 and Et2Zn produced only
an amorphous material not suitable for X-ray crystallographic
analysis.
2B
2C
92
88
88
80
R
R
2D
2E
79c
82
54
R
R
<5e
2F
–
72f
74
R
R
2G
65
a Determined by HPLC, Chiralcel OD–H. b Isolated yields. c Deter-
mined by GC, Supelco β-DEX. d Established through optical rotation.
e Determined by optical rotation. f Isolated after 24 h at 22 ЊC.
Indications of the presence of dimeric (or higher) species
came from the pronounced positive NLE that was seen when
performing the reaction with a varying enantiomeric purity of
14 (Fig. 5). When performing the reactions with 5 mol% of the
catalyst having an ee below 30% the ee determinations of (R)-1-
phenylpropan-1-ol became too uncertain, therefore, 10 mol%
of 14 was used throughout the NLE investigation.
The catalytic addition to some other aromatic aldehydes
As anisyl-BODOL appeared to be the best catalyst it was used
in the addition of diethylzinc to some aromatic aldehydes other
than benzaldehyde. The results, using diethyl ether–hexane
as the solvent under the standard conditions (0 ЊC, 40 h) are
shown in Table 2. No significant changes in the ee were
observed when the electron withdrawing Cl-substituents were
located in either the 4- or 3-positions of the benzaldehyde
(entries 2A and 2B). With the sterically more challenging
naphthalene-1-carbaldehyde a somewhat lower ee was obtained
(entry 2C) and thiophene-2-carbaldehyde gave a still lower ee
(entry 2D). In all four cases the yields were quite high.
In the thiophene case some 2-thienylmethanol could also be
isolated (10%). It is not unusual to observe a reduction of
the aldehyde to the corresponding primary alcohol when the
diethylzinc addition becomes slower.21 This was also noticed for
a few of the other catalysts (15, 17, 19 and 20).
The use of pyridine-2-carbaldehyde gave a very rapid
addition, which was complete within a few minutes (entry 2E).
However, the ee in this reaction was very small and was not
analysed in further detail. Soai’s group has reported auto-
catalysis with several nitrogen-containing hetero-aromatic
aldehydes, but does not mention pyridine-2-carbaldehyde.22,23
Other authors have also observed that the addition of zinc
Both the MLn model of Girard and Kagan30 and the exten-
sion of Noyori and co-workers 31 suggest the formation of equi-
librium mixtures of heterochiral and homochiral complexes.
In the Girard–Kagan model the equilibrium between these
complexes, in combination with their different reaction rates as
catalysts, led to the observed NLE. For the diethylzinc addition
catalysed by amino alcohols, the model proposed by Noyori
and co-workers, is similar to the MLn-model, but a mechanistic
extension is introduced, in which the larger complexes are in
equilibrium with monomeric species, which then are the active
catalysts. Thus, in this model the catalytically incompetent
larger complexes (both hetero- and homochiral) can be viewed
as storage forms of the monomeric catalyst. The equilibria
between these large and monomeric complexes dictate the
proportions of the active, monomeric catalysts. More detailed
discussions concerning the kinetics and mathematical models
involved in autocatalysis and NLE have been published by
Blackmond et al.32–36 In our case, no further interpretation of
648
J. Chem. Soc., Perkin Trans. 1, 2002, 645–651

View Article Online
would require a planar arrangement of the aromatic ring and
the cationic site at C1–C2ϩ–C3. This may be hindered, due
to steric interference, between the 1-methyl group and the
aromatic substituent.
An unambiguous structure determination of the diastereo-
mers 19 and 20 could not be made by NMR spectroscopy.
However, the structures of 20 and 23b were ascertained by X-
ray crystallography, which thereby cleared the structure of 19.
The crystallographic data will be published elsewhere.40
Experimental
All reactions were carried out in oven-dried glassware under an
argon atmosphere using anhydrous solvents (distilled from
CaH2 and dried over molecular sieves). Compounds 14, 15 and
16 were synthesised according to our previous procedures.19
Also compounds 21a,b and 22 were synthesised according
to the literature.37,39 Benzaldehyde was distilled prior to use
and the other aldehydes and diethylzinc were used as delivered
(Aldrich). TLC was carried out on silica gel (60 F254, Merck)
and spots were visualised with UV light and then with a solu-
tion of p-methoxybenzaldehyde (10 ml), concentrated sulfuric
acid (50 ml) and ethanol (95%, 950 ml). Flash column chroma-
tography was performed on Matrex (25–70 µm) silica gel. Melt-
ing points were taken on a Sanyo Gallenkamp melting point
apparatus (MPD.350.BM3.5) and are uncorrected. Optical
rotations were measured with a Perkin-Elmer 241 polarimeter
at 22 ЊC and are given in 10Ϫ1 deg cm2 gϪ1. Elemental analyses
of C, H, and N were performed by H. Kolbe Mikroanalytisches
Laboratorium. HPLC analyses were performed on a Daicel
OD–H column (250 × 4.6 id, 5 µm) and GC analyses were
performed on a β-DEX column (Supelco, 30 m × 0.25 mm id,
25 µm film thickness). All NMR data were recorded on a
Bruker DRX 400 spectrometer, if not otherwise mentioned,
and the chemical shifts were measured using the solvents as
internal references {CHCl3 (1H, 7.27 ppm), CDCl3 (13C, 77.23
ppm), C6H6 (1H, 7.16 ppm), C6D6 (13C, 128.39 ppm)}.
Fig. 5 The positive NLE, using 10% of 14 as a catalyst.
the NLE has been performed other than the obvious notion of
dimeric or larger complexes.
As the diols 14–20 were synthesised by methods essentially
described earlier, only a few comments concerning their syn-
thesis will be made here. The alcohols 21a and 21b were syn-
thesized according to literature procedures.37 After protecting
them as TBDMS-ethers, the addition of the anisyllithium–
CeCl3 reagent was completely stereoselective.18,19,38
Compound 22 was also synthesized according to literature
procedures.39 In contrast to 21a and 21b, the addition of
the anisyllithium–CeCl3 reagent was not stereoselective when
applied on compound 22. Instead, a 3 : 1 diastereomeric
mixture of 23a and 23b was isolated (Scheme 1). However,
(1R,2R,4S,6S)-2-(2-Methoxyphenyl)-1-methylbicyclo[2.2.2]-
octane-2,6-diol (17)
Keto-alcohol 21a (0.45 g, 2.9 mmol) was added to dry DMF
(2 ml) followed by imidazole (0.49 g, 7.3 mmol) and TBDMSCl
(0.55 g, 3.7 mmol). The mixture was kept overnight at rt,
and then diethyl ether (60 ml) and water (40 ml) were added.
The organic phase was separated, washed with HCl (25 ml,
0.5 M), aqueous saturated NaHCO3 (25 ml) and brine (25 ml).
The organic solution was then dried (Na2SO4), filtered and
concentrated at reduced pressure. Chromatography (SiO2,
heptane–EtOAc: 90 : 10) of the residue gave (1R,4S,6S)-6-
tert-butyldimethylsilyloxy-1-methylbicyclo[2.2.2]octane-2-one
(0.78 g, quantitatively) as a clear oil, which was used in the next
step.
A suspension of CeCl3 (2.0 g, 5.4 mmol) in THF (15 ml) was
prepared as previously described.19 The resulting mixture was
stirred for 5 min at 0 ЊC and then for 5 h at rt. This should result
in a “milky suspension” without any large fragments. The
suspension was then cooled to Ϫ78 ЊC, followed by addition
of anisyllithium (prepared by addition of n-BuLi (3.7 ml,
5.9 mmol) to anisole (0.87 ml, 8.0 mmol) in 6 ml THF at rt) and
the resulting yellow mixture was stirred for 1 h at Ϫ78 ЊC. A
solution of (1R,4S,6S)-6-tert-butyldimethylsilyloxy-1-methyl-
bicyclo[2.2.2]octane-2-one (0.49 g, 1.9 mmol in THF 5 ml) was
then added and the mixture was allowed to reach rt overnight.
Thereafter aqueous saturated NH4Cl (25 ml) was added. The
phases were separated and the water phase was extracted with
ether (2 × 20 ml). The combined organic phase was dried
(Na2SO4) and filtered through a pad of SiO2, in order to remove
inorganic cerium salts. The solvent was removed under vacuum,
and the residue was then diluted with cold, dry THF (30 ml).
Scheme 1 a) TBDMSCl, imidazole, DMF. b) Anisyllithium, CeCl3,
THF. c) Bu4NF, THF.
these compounds could easily be separated by column
chromatography.
Removal of the TBDMS groups with Bu4NF then gave the
desired compounds 14–20 in high yields. It was noticed that
the monocyclic diols 19 and 20 were not as prone to acid
degradation as the bicyclic compounds 14–16,19 possibly due to
the absence of steric strain. Also 17 was remarkably stable
towards acid. If we assume that this degradation proceeds via
carbocation formation at the benzylic position, the cation
stabilisation (by delocalisation of charge into the aromatic ring)
J. Chem. Soc., Perkin Trans. 1, 2002, 645–651
649

View Article Online
Bu4NF (0.75 g, 2.4 mmol) was added to this solution and the
reaction mixture was stirred overnight. TLC analysis confirmed
that the deprotection was complete. The reaction mixture was
concentrated under reduced pressure to yield an orange oil,
which was diluted with 50 ml of EtOAc and the organic solu-
tion was washed with brine (3 × 20 ml) and dried (Na2SO4).
Filtration and removal of the solvent under reduced pressure
gave an oil, which was purified by flash chromatography (SiO2,
heptane–EtOAc: 60 : 40, Rf = 0.29) to yield the title compound
as crystals (0.37 g, 49% from 21a). Mp 169 ЊC, [α]D ϩ61.9
(c 0.512, MeOH). δH(C6D6) 0.8 (1H, m), 1.03 (3H, s), 1.2–1.4
(3H, m), 1.77 (1H, m), 1.94 (1H, m), 2.20 (1H, m), 2.31 (1H, m),
2.47 (1H, m), 2.98 (3H, s), 3.68 (1H, m), 4.55 (1H, br d, J 11
Hz), 6.3 (1H, br s), 6.40 (1H, dd, J 1.1 and 8.2 Hz), 6.82 (1H,
m), 6.98 (1H, m), 7.12 (1H, br d). δC(C6D6) 19.8, 25.5, 27.2,
29.6, 40.7, 41.2, 45.5, 55.5, 76.4, 83.5, 112.7, 121.4, 128.3, 130.1,
132.8, 158.0 (Found: C, 73.18; H, 8.48. C16H22O3 requires C,
73.25; H, 8.45%).
(1S,3S)-1-(2-Methoxyphenyl)-2,2-dimethylcyclohexane-1,3-diol
(19)
Compound 23b (0.41 g, 1.1 mmol) was dissolved in THF
(30 ml). Bu4NF (0.43 g, 1.4 mmol) was added and the reaction
mixture was stirred overnight. TLC analysis confirmed that
the deprotection was complete. The reaction mixture was con-
centrated under reduced pressure to yield an orange oil, which
was diluted with 50 ml of EtOAc and the organic solution was
washed with brine (3 × 20 ml) and dried (Na2SO4). Filtration
and removal of the solvent under reduced pressure gave an oil,
which was purified by flash chromatography (SiO2, 10% Et2O in
CH2Cl2) to yield the title compound as crystals (0.28 g, 97%).
Mp 77–78 ЊC, [α]D Ϫ19.8 (c 1.26 in CHCl3). δH(CDCl3) 0.8 (3H,
s), 1.06 (3H, s), 1.58 (1H, m), 1.76 (2H, m), 1.91 (1H, m), 2.20
(1H, m), 2.50 (1H, m), 3.52 (1H, m), 3.91 (3H, s), 6.5 (1H, br s),
6.94–7.05 (2H, m), 7.17 (1H, br d, 7.1 Hz), 7.29–7.24 (1H, m).
δC(CDCl3) 15.9, 22.8, 24.8, 29.3, 34.4, 42.2, 56.0, 78.7, 83.0,
112.3, 121.2, 128.6, 129.4, 131.2, 157.7 (Found: C, 71.83;
H, 8.92. C15H22O3 requires C, 71.97; H, 8.86%).
(1R,2R,4S, 6S)-2-(2-Methoxyphenyl)-4-methylbicyclo[2.2.2]-
octane-2,6-diol (18)
(1R,3S)-1-(2-Methoxyphenyl)-2,2-dimethylcyclohexane-1,3-diol
(20)
Following the method mentioned above for the synthesis of 17,
the keto-alcohol 21b, gave after flash chromatography (SiO2,
pentane–acetone: 80 : 20, Rf = 0.22) the title compound (27%
from 21b) as crystals. Mp 85–87 ЊC, [α]D ϩ46.7 (c 0.6, MeOH).
δH(C6D6) 0.90 (3H, s), 1.08–0.90 (2H, m), 1.29–1.08 (2H,
m), 1.89 (1H, m), 1.96 (1H, m), 2.05–2.20 (2H, m), 2.66
(1H, m), 3.13 (3H, s), 4.18 (1H, m), 4.47 (1H, s), 4.63 (1H, d,
J 11.2 Hz), 6.53 (1H, m), 6.87 (1H, m), 7.09 (1H, m), 7.22
(1H, dd, J 1.8 and 7.8 Hz). δC(C6D6) 22.6, 28.1, 30.6, 31.0, 40.3,
46.5, 50.1, 55.0, 71.7, 79.0, 112.2, 121.2, 126.9, 128.3, 135.0,
158.1 (Found: C, 73.18; H, 8.48. C16H22O3 requires C, 73.25;
H, 8.45%).
Following the method for the synthesis of 19, using 23a, with
the exception that the reaction mixture had to be refluxed to
remove the protecting group, gave the title compound (97%) as
crystals. Mp 119 ЊC, [α]D ϩ58.1 (c 1.26 in CHCl3). δH(CDCl3)
0.75 (3H, s), 1.08 (3H, s), 1.97–1.53 (4H, m), 2.20 (2H, m), 2.50
(1H, br s), 3.51 (1H, dd, J 14.0 and 7.02 Hz), 3.91 (3H, s), 4.89
(1H, br s), 6.97 (2H, m), 7.26 (2H, m). δC(CDCl3) 17.0, 19.7,
21.5, 31.0, 34.4, 45.4, 56.2, 74.1, 82.0, 112.6, 121.1, 128.6, 130.2,
132.0, 158.2 (Found: C, 72.05; H, 8.78. C15H22O3 requires C,
71.97; H, 8.86%).
Addition of Et2Zn to aldehydes (general procedure)
(1R,3S)- and (1S,3S)-1-(2-Methoxyphenyl)-3-tert-butyldi-
Under an argon atmosphere the catalyst (50 µmol) was dis-
solved in diethyl ether (2.0 ml) and cooled to 0 ЊC. Thereafter
diethylzinc (3.0 ml, 1.0 M in hexanes, 3.0 mmol) was added and
the mixture was allowed to stir at 0 ЊC for 15 min whereafter the
aldehyde (1 mmol, neat) was added. After stirring the mixture
at 0 ЊC for 40 h the reaction was quenched by the addition of
saturated NH4Cl (7 ml) at 0 ЊC. Stirring at this temperature was
continued for a few minutes before water (25 ml) was added.
The organic phase was separated and the aqueous phase was
extracted with EtOAc (2 × 25 ml). The combined organic
phases were dried (Na2SO4), filtered and concentrated under
reduced pressure. Chromatography of the residue then gave
the desired product. The chemical structures were established
methylsiloxy-2,2-dimethylcyclohexanol (23a) and (23b)
A solution of (S)-3-tert-butyldimethylsilyloxy-2,2-dimethyl-
cyclohexanone (22) (1.6 g, 6.0 mmol in THF 5 ml) was added to
an anisyllithium–CeCl3 reagent prepared as mentioned {from
CeCl3 (3.6 g, 9.8 mmol), THF (25 ml) anisole (1 ml, 9 mmol)
and Bu–Li (6.6 ml 11 mmol in hexane)}. The mixture was
allowed to reach rt overnight, thereafter aqueous saturated
NH4Cl (20 ml) was added and the phases were separated and
the water phase was extracted with diethyl ether (3 × 20 ml).
The combined organic phase was dried (Na2SO4), filtered
and concentrated under reduced pressure. Chromatography
(SiO2, CH2Cl2) of the residue gave the title compounds 23a
(Rf = 0.67, 1.45 g, 66.5%) and 23b (Rf = 0.54, 0.46 g, 21%) as
crystals.
1
by H NMR spectroscopy and the enantiomeric excesses were
determined by enantioselective HPLC-, GC-chromatography
or by specific rotation (see below).
23a. Mp: 81 ЊC, [α]D ϩ42 (c 0.95 in CHCl3). δH(CDCl3) 0.08
(6H, s), 0.75 (3H, s), 0.88 (3H, s), 0.91 (9H, s), 1.60 (4H, m),
1.95 (1H, m), 2.40 (1H, m), 3.89 (3H, s), 4.05 (1H, dd, J 11.1
and 4.7 Hz), 5.85 (1H, br s), 6.96 (2H, m), 7.25 (2H, m).
δC(Bruker ARX 300, CDCl3) Ϫ4.6, Ϫ3.7, 17.2, 18.3, 19.4, 21.9,
26.1, 31.2, 34.2, 45.6, 55.9, 74.2, 82.0, 112.3, 120.8, 128.2, 129.9,
132.0, 157.9 [Found (HRMS, CI, CH4): Mϩ, 363.2348.
C21H35O3Si requires M, 363.2355].
Determination of ee by HPLC
HPLC: Chiralcel OD–H (Daicel) 250 × 4.6 mm, rt, flow rate:
0.5 ml minϪ1, detection at 245 nm. Anisyl-BODOL (14).
Solvent, hexane–PriOH: 80 : 20, 13 min (1S,2S,4R,6R), 17 min
(1R,2R,4S,6S).
1-Phenylpropan-1-ol, [α]D ϩ43 (c in 3.2, CHCl3), {lit.41 (S)
[α]D Ϫ48.5 (c 4 in CHCl3)}. Solvent, hexane–PriOH: 97.5 : 2.5,
(R) 25.4 min, (S) 27.1 min.
23b. Mp 88–89 ЊC, [α]D ϩ51.9 (c 1.08 in CHCl3). δH(CDCl3)
0.13 (6H, s), 0.89 (3H, s), 0.91 (3H, s), 0.98 (9H, s), 1.60
(4H, m), 2.05 (2H, m), 3.30 (1H, m), 3.77 (3H, s), 5.7 (1H, br s),
6.88 (2H, dd, J 8.16 and 1.11 Hz), 6.96 (1H, m), 7.21 (1H,
m), 7.94 (1H, dd, J 7.95 and 1.84 Hz). δC(Bruker ARX 300,
CDCl3) Ϫ4.9, Ϫ4.4, 16.2, 18.2, 22.9, 24.9, 26.0, 29.4, 33.1, 42.2,
55.5, 79.6, 80.5, 112.2, 120.3, 128.0, 131.9, 133.0, 157.3 [Found
(HRMS, CI, CH4): Mϩ, 365.2359. C21H35O3Si requires M,
363.2355].
1-(4-Chlorophenyl)propan-1-ol, [α]D ϩ27 (c 3.8 in C6H6)
{lit.41 (S) [α]D Ϫ28.2 (c 5.01 in C6H6)}. Solvent, hexane–PriOH:
95 : 5, (S) 14.8 min, (R) 15.4 min.
1-(3-Chlorophenyl)propan-1-ol, [α]D ϩ28.0 (c 2.23 in C6H6)
{lit.42 (R) [α]D ϩ24.2, toluene}. Solvent, hexane–PriOH: 95 : 5
(S) 14.7 min, (R) 15.4 min.
1-Naphthylpropan-1-ol, [α]D ϩ45.8 (c 2.51 in CHCl3) {lit.43
(R) [α]D ϩ55.6, CHCl3}. Solvent, hexane–PriOH: 90 : 10, (S)
15.8 min, (R) 26.8 min.
650
J. Chem. Soc., Perkin Trans. 1, 2002, 645–651

View Article Online
12 K. Soai, T. Hayase, C. Shimada and K. Isobe, Tetrahedron:
1-(4-Methoxyphenyl)propan-1-ol, [α]D ϩ17.6 (c 2.8 in C6H6),
{lit.16 [α]D ϩ31.8 (c 3.5 in C6H6)}. Solvent, hexane–PriOH:
95 : 5, (R) 20.4 min, (S) 23.7 min.
Asymmetry, 1994, 5, 789.
13 L. Bonetto, R. Fornasier and U. Tonellato, Gazz. Chim. Ital., 1995,
125, 63.
14 J.-S. You, M.-Y. Shao and H.-M. Gau, Organometallics, 2000, 19,
3368.
Determination of ee by GC
15 W.-S. Huang, Q.-S. Hu and L. Pu, J. Org. Chem., 1999, 64, 7940.
16 B. H. Lipshutz and Y.-J. Shin, Tetrahedron Lett., 2000, 41, 9515.
17 K. Ding, A. Ishii and K. Mikami, Angew. Chem., Int. Ed., 1999,
38, 497.
18 F. Almqvist, L. Torstensson, A. Gudmunsson and T. Frejd, Angew.
Chem., Int. Ed., 1997, 36, 376.
19 I. Sarvary, F. Almqvist and T. Frejd, Chem. Eur. J., 2001, 7, 2158.
20 B. Goldfuss, M. Steigelmann, S. I. Khan and K. N. Houk, J. Org.
Chem., 2000, 65, 77.
Column: β-DEX (Supelco) 30 m × 0.25 mm id, 25 µ film
thickness.
1-Thiophenepropan-1-ol, [α]D ϩ19 (c 2.1 in CHCl3) {lit.44 (R)
[α]D ϩ25.9 (c 2.1 in CHCl3)} GC: isothermal 110 ЊC, 30 min
(R) 15.4 min (S) 16.1 min.
Determination of ee by optical rotation
21 E. J. Corey and F. Hannon, Tetrahedron Lett., 1987, 28, 5233.
22 K. Soai, T. Shibata, H. Morioka and K. Choji, Nature, 1995,
378, 767.
23 K. Soai, T. Shibata and I. Sato, Acc. Chem. Res., 2000, 33, 382.
24 C. Bolm and K. Muñiz, Chem. Commun., 1999, 1295.
25 W.-M. Dai, H.-J. Zhu and X.-J. Hao, Tetrahedron: Asymmetry, 2000,
11, 2315.
26 M. Ishizaki and O. Hoshino, Chem. Lett., 1994, 1337.
27 C. Puchot, O. Samuel, E. Dunach, C. Zhao, C. Agami and
H. B. Kagan, J. Am. Chem. Soc., 1986, 108, 2353.
28 N. Oguni, Y. Matsuda and T. Kaneko, J. Am. Chem. Soc., 1988, 110,
7877.
29 M. Kitamura, S. Okada, S. Suga and R. Noyori, J. Am. Chem. Soc.,
1989, 111, 4028.
Since baseline separation could not be achieved for the com-
pounds below and the ees were low the enantiomeric purity
was determined by optical rotation with a Perkin-Elmer 241
polarimeter at 22 ЊC.
1-(Pyridin-2-yl)propan-1-ol, [α]D ϩ5.2 (c 2.1 in CHCl3)
{lit.25 (R) [α]D ϩ5.7, (c 2.25 in CHCl3, (5.4% ee))}.
1-Ferrocenylpropan-1-ol, [α]D Ϫ0.1 (c 1.7 in C6H6) {lit.45(S)
[α]D ϩ56, (c 1.4 in C6H6)}.
Acknowledgements
We thank The Swedish Research Council and the Royal Physio-
graphic Society in Lund for economic support.
30 C. Girard and H. B. Kagan, Angew. Chem., Int. Ed., 1998, 37, 2922.
31 M. Kitamura, S. Suga, M. Niwa and R. Noyori, J. Am. Chem. Soc.,
1998, 120, 9800.
32 D. G. Blackmond, J. Am. Chem. Soc., 1997, 119, 12934.
33 D. G. Blackmond, J. Am. Chem. Soc., 1998, 120, 13349.
34 D. G. Blackmond, J. Am. Chem. Soc., 2001, 123, 545.
35 D. G. Blackmond, C. R. McMillan, S. Ramdeehul, A. Schorm and
J. B. Brown, J. Am. Chem. Soc., 2001, 123, 10103.
36 T. Rosner, P. J. Sears, W. A. Nugent and D. G. Blackmond, Org.
Lett., 2000, 2511.
37 K. Mori and E. Nagano, Biocatalysis, 1990, 3, 23.
38 T. Imamoto, Y. Sigiura and N. Takiyama, Tetrahedron Lett., 1984,
25, 4233.
References
1 L. Pu and H.-B. Yu, Chem. Rev., 2001, 757.
2 R. Noyori and M. Kitamura, Angew. Chem., Int. Ed. Engl., 1991, 30,
49.
3 P. Pinho and P. G. Andersson, Tetrahedron, 2001, 57, 1615.
4 C. Bolm, N. Hermanns, J. P. Hildebrand and K. Muñiz, Angew.
Chem., Int. Ed., 2000, 39, 3465.
5 C. Rossini, L. Franzini, D. Pini and P. Salvadori, Tetrahedron:
Asymmetry, 1990, 1, 587.
6 K. R. K. Prasad and N. N. Joshi, Tetrahedron: Asymmetry, 1996, 7,
1957.
39 K. Mori and H. Wanatabe, Tetrahedron, 1986, 42, 273.
40 I. Sarvary, Y. Wan, C. Svensson and T. Frejd, manuscript in
preparation, 2002.
7 J. Long and K. Ding, Angew. Chem., Int. Ed., 2001, 40, 544.
8 Q.-H. Fan, G.-H. Liu, X.-M. Chen, G.-J. Deng and A. S. C. Chan,
Tetrahedron: Asymmetry, 2001, 12, 1559.
9 Y.-Y. Cherng, F.-M. Fang and T.-J. Lu, J. Org. Chem., 1999, 64,
3207.
41 M. Watanabe, S. Araki and Y. Butsugan, J. Org. Chem., 1991, 56,
2218.
42 P. A. Chaloner and P. Renuka, Tetrahedron Lett., 1987, 28, 3013.
43 J. Kang, J. W. Lee and J. I. Kim, J. Chem. Soc., Chem. Commun.,
1994, 2009.
10 M. A. Cabras, G. Chelucci, G. Giacomelli and F. Soccolini, Gazz.
Chim. Ital., 1994, 124, 23.
11 H. Kotsuki, Y. Nakagawa, N. Moriya, H. Tateishi, M. Ochi,
T. Suzuki and K. Isobe, Tetrahedron: Asymmetry, 1995, 6, 1165.
44 S.-K. Kang, J.-H. Jeon, T. Yamaguchi, J.-S. Kim and B.-S. Ko,
Tetrahedron: Asymmetry, 1995, 6, 2139.
45 N. Arroyo, U. Halsinger, K. Mereiter and M. Widhalm, Tetrahedron:
Asymmetry, 2000, 11, 4207.
J. Chem. Soc., Perkin Trans. 1, 2002, 645–651
651