Stereoselective alcohol silylation by dehydrogenative Si-O coupling: Scope, limitations, and mechanism of the Cu-H-catalyzed non-enzymatic kinetic resolution with silicon-stereogenic silanes

Article abstract of DOI:10.1002/chem.200801377

Ligand-stabilized copper(I)hydride catalyzes the dehydrogenative Si-O coupling of alcohols and silanes-a process that was found to proceed without racemization at the silicon atom if asymmetrically substituted. The present investigation starts from this pivotal observation since silicon-stereogenic silanes are thereby suitable for the reagent-controlled kinetic resolution of racemic alcohols, in which asymmetry at the silicon atom enables discrimination of enantiomeric alcohols. In this full account, we summarizeour efforts to systematically examine this unusual strategy of diastereoselective alcohol silylation. Ligand (sufficient reactivity with moderately electron-rich monophosphines), silane (reasonable diastereocontrol with cyclic silanes having a distinct substitution pattern) as well as substrate identification (chelating donor as a requirement) areintroductorily described. With these basic data at hand, the substrate scope was defined employing enantiomerically enriched tert-butyl-substituted 1-silatetraline and highly reactive 1-si-laindane. The synthetic part is complemented by the determination of the stereochemical course at the silicon atom in the Si-O coupling step followed by its quantum-chemical analysis thus providing a solid mechanistic picture of this remarkable transformation.

Full text of DOI:10.1002/chem.200801377

DOI: 10.1002/chem.200801377
Stereoselective Alcohol Silylation by Dehydrogenative Si–O Coupling:
Scope, Limitations, and Mechanism of the Cu–H-Catalyzed Non-Enzymatic
Kinetic Resolution with Silicon-Stereogenic Silanes
Sebastian Rendler,[a, b] Oliver Plefka,[b] Betꢀl Karatas,[a] Gertrud Auer,[b]
Roland Frçhlich,[a] Christian Mꢀck-Lichtenfeld,[c] Stefan Grimme,[c] and
Martin Oestreich*[a, b]
Abstract: Ligand-stabilized copper(I)–
hydride catalyzes the dehydrogenative
Si–O coupling of alcohols and silanes—
a process that was found to proceed
without racemization at the silicon
atom if asymmetrically substituted. The
present investigation starts from this
pivotal observation since silicon-stereo-
genic silanes are thereby suitable for
the reagent-controlled kinetic resolu-
tion of racemic alcohols, in which
asymmetry at the silicon atom enables
discrimination of enantiomeric alco-
hols. In this full account, we summarize
our efforts to systematically examine
this unusual strategy of diastereoselec-
tive alcohol silylation. Ligand (suffi-
cient reactivity with moderately elec-
tron-rich monophosphines), silane (rea-
sonable diastereocontrol with cyclic si-
lanes having a distinct substitution pat-
tern) as well as substrate identification
(chelating donor as a requirement) are
introductorily described. With these
basic data at hand, the substrate scope
was defined employing enantiomeri-
cally enriched tert-butyl-substituted 1-
silatetraline and highly reactive 1-si-
laindane. The synthetic part is comple-
mented by the determination of the
stereochemical course at the silicon
atom in the Si–O coupling step fol-
lowed by its quantum-chemical analysis
thus providing a solid mechanistic pic-
ture of this remarkable transformation.
Keywords:
asymmetric catalysis
kinetic resolution · silicon
alcohols
·
·
·
copper
Introduction
The Si–O linkage is an attractive connectivity for synthetic
organic chemists. On the one hand, the strong Si–O bond
ensures its compatibility with the majority of synthetic
methods and, on the other hand, the availability of orthogo-
nal strategies for (chemo)selective bond formation and
cleavage account for the extensive utilization of silyl ethers
to temporarily protect hydroxy groups in complex molecule
synthesis.[1] For these reasons, the development of stereose-
lective alcohol silylation methodology is of considerable in-
terest. Such strategies would substantially increase the value
of silyl ether formation and, thus, complement existing
asymmetric carbonyl hydrosilylation techniques.[2]
[a] Dr. S. Rendler, Dr. B. Karatas, Dr. R. Frçhlich,+
Prof. Dr. M. Oestreich
Organisch-Chemisches Institut
Westfꢀlische Wilhelms-Universitꢀt Mꢁnster
Corrensstrasse 40, 48149 Mꢁnster (Germany)
Fax : (+49)251-83-36501
E-mail: martin.oestreich@uni-muenster.de
[b] Dr. S. Rendler, Dipl.-Chem. O. Plefka, Dr. G. Auer,
Prof. Dr. M. Oestreich
Institut fꢁr Organische Chemie und Biochemie
Albert-Ludwigs-Universitꢀt Freiburg
Albertstrasse 21, 79104 Freiburg im Breisgau (Germany)
[c] Dr. C. Mꢁck-Lichtenfeld, Prof. Dr. S. Grimme
Theoretische Organische Chemie
Following a report by Ishikawa and co-workers,[3] stereo-
selective alcohol silylation has increasingly attracted atten-
tion in recent years.[4] We and, shortly thereafter, Hoveyda
and Snapper et al. devised fundamentally different concepts
for stereoselective Si–O coupling.[5,6] From a synthetic per-
spective, both approaches constitute novel strategies for
non-enzymatic kinetic resolution of racemic mixtures of al-
Organisch-Chemisches Institut
Westfꢀlische Wilhelms-Universitꢀt Mꢁnster
Corrensstrasse 40, 48149 Mꢁnster (Germany)
E-mail: grimmes@uni-muenster.de
[+] X-ray crystal structure analysis.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200801377.
11512
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 11512 – 11528

FULL PAPER
cohols[7] or desymmetrization of meso diols,[8] respectively,
an area, in which asymmetric acylation[9] has been the pre-
vailing methodology thus far. Our strategy for kinetic reso-
lution of racemic alcohols relies on a reagent-controlled,
congruously diastereoselective, dehydrogenative Si–O cou-
pling[10,11] with recyclable silicon-stereogenic silanes
(Scheme 1).[5] In this process, a transition-metal hydride, for
at the silicon atom facilitated promising stereoselectivity in
kinetic resolution of secondary alcohols.
In this full account, we present a comprehensive investiga-
tion of the factors governing the stereoselectivity in the ki-
Figure 1. Silicon-stereogenic silanes for kinetic resolution of secondary al-
cohols.
netic resolution of secondary donor-functionalized alcohols
by Cu–H-catalyzed dehydrogenative Si–O coupling. A more
refined ligand screening, the influence of the tether length
in the donor-functionalized alcohol, and structural features
of the resolving silane are investigated. Subsequently, the
substrate scope with regard to both substituents and the
donor is included; results of the kinetic resolution with
highly reactive 1-silaindane (SiR)-5a are discussed as well. A
quantum-chemical investigation combined with experimen-
tal data on the stereochemical course of the Si–O coupling
step gives mechanistic insight into this copper(I) catalysis.
Scheme 1. Cu–H-catalyzed diastereoselective alcohol silylation: Kinetic
resolution with recyclable silicon-stereogenic silanes. Si represents an
asymmetrically substituted silicon atom.
example, ligand-stabilized copper(I) hydride,[12] facilitates a
diastereodiscriminating reaction of enantiopure chiral silane
(SiR)-A with one out of two possible enantiomers of a
donor-functionalized chiral alcohol rac-B to form silyl ether
(SiS,S)-C. Notably, dihydrogen is formed as the sole by-prod-
uct.[13] Along with diastereoenriched silyl ether (SiS,S)-C, the
slow-reacting enantiomer of the alcohol (R)-B remains in
enantioenriched form. As delineated in the assumed transi-
tion state D, s-bond metathesis of a copper(I) alkoxide with
the silane might account for diastereoselective Si–O cou-
pling.[14] The virtue of s-bond metathesis for enantiospecific
transformations of silicon-stereogenic silanes (retention of
configuration is observed) had already been disclosed in ear-
lier contributions by our group.[15,16] Two-point binding by
coordination of the pending donor in the substrate was
shown to be essential in order to achieve good diastereose-
lectivity. A second synthetic operation then completes the
concept: Enantiospecific reductive cleavage[17] of the Si–O
bond with DIBAL-H liberates the resolving reagent (SiR)-A
Results and Discussion
Parameters influencing diastereocontrol: At the initial
stages of this project, the reaction setup of the Cu–H-cata-
lyzed dehydrogenative Si–O coupling[13] involved standard
substrate rac-6, silane rac-4a, and a catalyst generated in
situ as reported earlier by Buchwald[14] and Leighton[21] [rac-
6!
(SiS*,S*)-7a, Table 1]. As an advantage of the reagent-
ACHTUNGTRENNUNG
controlled nature of this process, all optimization experi-
ments were conducted with readily available racemic silane;
determination of the diastereomeric ratio of the formed silyl
1
ether 7a by H NMR spectroscopy was sufficiently accurate.
Importantly, the diastereomeric excess, which is independent
of the degree of conversion in the racemic case, correlates
directly with the enantiomeric excess of the slow-reacting al-
cohol at exactly 50% conversion in a hypothetical kinetic
resolution with enantiopure silane.[22]
along with alcohol (S)-B [(SiS,S)-C!
(SiR)-A].[5a]
ACHTUNGTRENNUNG
Our initial studies[5a,15a,16c] also underlined the importance
of the substitution pattern at silicon in order to achieve
useful levels of diastereoinduction in the Si–O coupling
event. Acyclic triorganosilanes such as (SiR)-1[18] and (SiR)-
2[19] as well as silatetraline (SiR)-3[20a] invariably showed in-
sufficient reagent control. From these insights, silane (SiR)-
4a[20b] emerged as the reagent of choice (see Figure 1). Em-
bedded into a rigid cyclic framework and equipped with
three substituents of different steric demand, the asymmetry
Catalyst optimization by ligand screening: In previous stud-
ies, monodentate triarylphosphines were identified as the li-
gands of choice in order to form copper(I) complexes that
combine sufficient reactivity with reasonable diastereoselec-
tivity.[5a] The results compiled in Table 1 give a more refined
picture for this ligand class. Compared to parent Ph3P (L1a)
Chem. Eur. J. 2008, 14, 11512 – 11528
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11513

M. Oestreich et al.
Table 1. Ligand screening for Cu–H-catalyzed diastereoselective Si–O coupling.[a]
the donor site (Table 2). For
this, rac-6 (n=1, entry 2) was
compared with pyridyl metha-
nol rac-8 (n=0, entry 1) and
pyridyl propanol rac-9 (n=2,
entry 3). Not surprisingly, rac-
8—likely to involve a five-mem-
bered chelate ring—also under-
went the desired Si–O coupling
to give silyl ether (SiS*,R*)-10a
(n=0) in markedly decreased
diastereoselectivity (d.r. 81:19,
entry 1). A pronounced effect
on both reactivity and diaste-
reoselectivity was seen in the
dehydrogenative coupling of
rac-9 (n=2, d.r. 79:21, entry 3);
it reacted only sluggishly at ele-
vated temperatures and pro-
longed reaction times [rac-9!
Entry
Ligand L
L/CuCl
t
T
Conv
[%][b]
d.r.[c]
[h]
[8C]
1
2
3
4
5
6
7
8
Ph3P (L1a)
ACHTUNGTRENNUNG(4-MeC6H4)3P (L1b)
(4-tBuC6H4)3P (L1c)
(3,5-xylyl)3P (L1d)
(4-F3CC6H4)3P (L1e)
2:1
2:1
2:1
2:1
2:1
2:1
2:1
2:1
1:1
1:1
48
9
8
20
60
60
13
16
2
20
20
20
20
70
70
20
45
60
85
42
38
48
50
38
34
33
50
40
10
90:10
92:8
88:12
92:8
83:17
86:14
91:9
90:10
76:14
55:45
A
ACHTUNGTRENNUNG(F3C)2C6H3]3P (L1 f)
ACHTUNGTRENNUNG
tBuPh2P (L2)
IMes·HCl (L3a)
IPr·HCl (L3b)
9[d]
10[d]
2
[a] Unless otherwise noted, all reactions were conducted using CuCl (5.0 mol%), the indicated ligand
(10 mol% for L1 and L2, 5.0 mol% for L3), and NaOtBu (5.0 mol%) with a substrate concentration of 0.1m
in toluene at the indicated temperature. [b] Monitored by 1H NMR spectroscopy and determined by integra-
tion of baseline-separated resonance signals of 6 at d=5.18 ppm and 7a at d=4.96/5.05 ppm. [c] Determined
AHCTUNGTRENNUNG
1
chelate might account for this.
Yet, a temporary copper(I)–ni-
trogen interaction is neverthe-
less likely since the observed
diastereoselectivity is still sig-
nificantly higher than for a re-
by H NMR spectroscopy of the crude reaction mixture by integration of baseline-separated resonance signals
of (SiS*,S*)-7a at d=4.96 and (SiS*,R*)-7a at d=5.05 ppm. [d] Substoichiometric amounts of NaOtBu
(30 mol%) used. IMes=1,3-bis(2,4,6-trimethylphenyl)imidazol-2-ylidene, IPr=1,3-bis(2,6-diisopropylphenyl)-
imidazol-2-ylidene.
(entry 1), alkyl substitution on the aromatic rings had only
minor effect on diastereoselectivity but considerably im-
proved conversion; slightly more electron-rich phosphines
L1b, L1c and L1d (entries 2–4) all resulted in smooth con-
version at room temperature. In particular, (3,5-xylyl)3P
(L1d) led to complete conversion within 20 h along with the
highest diastereoselectivity observed (d.r. 92:8) (entry 4).
Conversely, electron-withdrawing groups at the arenes re-
sulted in significantly dimished reactivity; only partial con-
version at elevated temperatures was achieved with trifluoro-
methyl-substituted ligands L1e and L1 f (entries 5 and 6). In
contrast, phosphine L1g equipped with an electron-rich aro-
matic ring formed a reactive and selective yet unstable cata-
lyst, that decomposed before conversion of silane rac-4a
was complete (entry 7). We also considered the use of com-
mercially available, bulky diarylalkylphosphine L2 (entry 8).
As opposed to disappointing results with Buchwald-type li-
gands as well as trialkylphosphines,[5a] L2 showed good reac-
tivity and appreciable diastereoselectivity. The poor diaste-
reoselectivities obtained with N-hetereocyclic carbene li-
gands L3a and L3b (entries 9 and 10) contrast with our
recent findings for the related rhodium(I)-catalyzed proto-
col;[5a,c] this might indicate different coordination spheres of
the transition metals (CuI tetrahedral versus RhI square-
planar).
lated alcohol devoid of a pending donor; 1,2-diphenyl etha-
nol gave a 60:40 mixture of diastereomers.[5a]
Silane-substitution pattern: The distinct influence of the
silane design in this kinetic resolution is evident from a
mechanistic perspective: The silicon atom as the sole source
of stereochemical information is responsible for enantiodis-
crimination between the enantiomeric copper(I) alkoxide
complexes (cf. D, Scheme 1). In order to identify silane lead
structures, a broad selection of chiral silanes was surveyed
(Table 3). Attempts to utilize known chiral silanes, for ex-
ample, Sommerꢃs acyclic rac-1,[18] were hampered by the in-
adequate design for the present purpose with the diastereo-
selection in the reaction with rac-6 not exceeding 57:43 [rac-
6!
(SiS*,S*)-12, Table 3, entry 1].[5a] While in this case poor
ACHTUNGTRENNUNG
diastereoselectivity might be attributed to insufficient steric
differentiation of the three substituents at silicon, phenyl
and 1-naphthyl are similar, such arguments do not hold true
for
acyclic
congener
rac-2[19b]
[rac-6!
(SiS*,S*)-13,
ACHTUNGTRENNUNG
entry 2];[16c] low diastereocontrol in the formation of
(SiS*,S*)-13 is in sharp contrast to its successful utilization in
a highly diastereoselective palladium(II)-catalyzed hydrosi-
lylation.[16c,d] Incorporating the stereogenic silicon center
into a rigid cyclic framework provided a solution to this di-
lemma; however, silatetraline rac-3[20a] equipped with a exo-
cyclic phenyl group performed almost as poor as any of the
Chelate ring size: With the catalyst system consisting of
CuCl, L1d and NaOtBu at hand, we next turned our atten-
tion to the tether length between the hydroxy oxygen and
acyclic reagents [rac-6!
(SiS*,S*)-14, entry 3].[5a] We rea-
ACHTUNGTRENNUNG
soned that, again, insufficient steric differentiation of sub-
stituents, phenyl versus phenyl in back-bone, might account
11514
www.chemeurj.org
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 11512 – 11528

Stereoselective Alcohol Silylation
FULL PAPER
Table 2. Influence of chelate ring size on diastereoselectivity.[a]
selection of pyridyl ethanols.
For this, highly enantiomeri-
cally enriched silane 4a (93–
99% ee) was used. This privi-
leged silane (cf. Table 3) was
made available in both enantio-
meric forms by a classic resolu-
tion of its corresponding (ꢀ)-
menthyl ethers (99% ee) or—at
the early stages of this proj-
ect—by reversed kinetic resolu-
tion with an enantiopure pyrid-
yl alcohol (ꢁ96% ee).[20b,25]
Routinely, 0.60 equiv of enan-
tioenriched silane (SiS)-4a or
(SiR)-4a, respectively, were used
along with a catalyst loading of
5.0 mol% at room temperature.
After full consumption of the
resolving agent, conversion of
approximately 55% was deter-
Entry
Alcohol
t
T
[8C]
Conv
[%][b]
Silyl ether
Yield
[%][c]
d.r.[d]
[h]
1
2
3
rac-8 (n=0)
rac-6 (n=1)
rac-9 (n=2)
24
20
62
20
20
50
50[e]
47[f]
45[f]
A
37 (43)
44 (50)
34 (40)
81:19
92:8
79:21
AHCTUNGTRENNUNG
A
[a] All reactions were conducted using CuCl (5.0 mol%), ligand L1d (10 mol%), and NaOtBu (5.0 mol%)
with a substrate concentration of 0.1m in toluene at the indicated temperature. [b] Monitored by 1H NMR
spectroscopy and determined by integration of baseline-separated resonance signals. [c] Isolated yield of ana-
lytically pure material after flash chromatography; numbers in parentheses refer to reisolated alcohol. [d] De-
termined by 1H NMR spectroscopy of the crude reaction mixture by integration of baseline-separated reso-
nance signals of the diastereomers. [e] 0.50 equiv of rac-4a used. [f] 0.55 equiv of rac-4a used.
for this outcome. This finally led to the design of privileged
1
silane rac-4a, in which the exocyclic phenyl group was re-
placed by a bulky tert-butyl group.[20b] Indeed, the combina-
tion of both features, a rigid cyclic framework and the steric
differentiation at the silicon atom, facilitated promising dia-
stereoselectivity of 92:8 in the dehydrogenative Si–O cou-
mined by H NMR spectroscopy; diastereomeric ratios were
determined by H NMR spectroscopic analysis of the crude
1
reaction mixture as well.
This survey commenced with variation of the substituent
at the stereogenic carbon. Tables 4 and 5 summarize the re-
sults for kinetic resolutions arranged by the silane enantio-
mer (SiS)-4a (Table 4) and (SiR)-4a (Table 5). Since the dia-
stereomeric ratios for the obtained silyl ethers are biased by
conversion when using enantioenriched silane (Tables 4 and
5, column 8), the diastereomeric ratios of silyl ethers for the
reaction with racemic silane rac-4a are also listed (column
9) in order to compare the efficiency of kinetic resolution
for different substrates. As expected, phenyl-substituted al-
cohol rac-6 cleanly underwent the kinetic resolution when
exposed to CuCl/L1d/NaOtBu and 0.60 equiv of silane (SiS)-
pling [rac-6!
ACHTUNGTRENNUNG
(SiS*,S*)-7a, entry 4].[5a] An even more refined
survey on this basis clearly demonstrates the scope and limi-
tations of this general structural motif (entries 5–7). With an
isopropyl instead of a tert-butyl group (rac-4a!rac-4b[16b]),
diastereoselectivity decreased yet being by far higher than
those for acyclic silanes [rac-6!
A similar effect was observed with silaindane rac-5a[23]
equipped with a bulky tert-butyl group [rac-6!
(SiS*,S*)-
(SiS*,S*)-7b, entry 5].
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
15a, entry 6]. While diastereoselection was slightly dimin-
ished, a distinct acceleration of the reaction rate was moni-
tored. The higher reactivity of the silane rac-5a in compari-
son to silatetralines rac-4 is likely to be connected with the
increased Lewis acidity of the silicon atom in a strained sila-
cycle.[24] Therefore, rac-5a constitutes an alternative for the
resolution of alcohols otherwise unreactive in this catalysis.
Recently, a kinetic resolution of tertiary propargylic alcohols
with highly enantioenriched (SiR)- or (SiS)-5a[23] was ach-
ieved following exactly this strategy.[5b] It is, however, inter-
esting to note that a further decrease of steric bulk results in
a complete collapse of diastereoselectivity; isopropyl-bear-
ing silaindane rac-5b[16d] yielded silyl ether (SiS*,S*)-15b in a
diastereomeric ratio of 52:48 (entry 7). The high reactivity
of rac-5b also made this silane highly susceptible to hydroly-
sis with traces of water present in the reaction mixture; ad-
dition of powdered molecular sieves effectively suppressed
this side-reaction.
4a (94% ee) [rac-6!
(SiR,R)-7a, Table 4, entry 1]. At 58%
ACHTUNGTRENNUNG
conversion, a promising enantiomeric excess for the slow-re-
acting enantiomer (S)-6 (88% ee) was detected as well as a
reduced diastereomeric ratio for silyl ether (SiR,R)-7a com-
pared to the racemic series due to >50% conversion. Grati-
fyingly, chemical yields of both silyl ether (SiR,R)-7a and al-
cohol (S)-6 were almost quantitative. Straightforward sepa-
ration of alcohol and silyl ether by flash column chromatog-
raphy underscores the preparative simplicity of this method-
ology. As supported by 2- as well as 1-naphthyl substituted
alcohols rac-16 and rac-24, respectively, aryl substitution at
the stereogenic carbon is tolerated (Table 4, entry 2;
Table 5, entry 1). At a conversion of 54%, the remaining al-
cohol (S)-16 showed 75% ee [rac-16!
(SiR,R)-20a, Table 4,
ACHTUNGTRENNUNG
entry 2]. Kinetic resolution of related rac-24 furnished silyl
ether (SiS,S)-29a in good diastereoselectivity (d.r. 92:8 in the
racemic series) along with enantioenriched alcohol in (R)-24
Kinetic resolution of 2-(pyridin-2-yl)ethanols with sterically
hindered silatetraline: After having identified suitable con-
ditions for diastereoselective dehydrogenative Si–O cou-
pling, this novel kinetic resolution process was tested with a
(80% ee at 57% conversion) [rac-24!
(SiS,S)-29a, Table 5,
ACHTUNGTRENNUNG
entry 1]. These findings were substantiated in the kinetic
resolution of other substituents with sp2-hybridization at
carbon, allylic alcohols rac-17 and rac-25 [Table 4, entry 3;
Chem. Eur. J. 2008, 14, 11512 – 11528
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11515

M. Oestreich et al.
Table 3. Influence of silane substitution on diastereoselectivity.[a]
Entry
Silane
(equiv)
t
Conv
[%][b]
Silyl ether
Yield
[%][c]
d.r.[d]
N
[h]
1
rac-1 (0.55)
rac-2 (0.55)
rac-3 (0.55)
3
51
52
52
51 (47)
51 (34)
51 (45)
57:43
59:41
66:34
2
12
3
4
4
rac-4a (0.50)
rac-4b (1.20)
rac-5a (0.55)
rac-5b (1.20)
20
24
1
47
44 (50)
92:8
5
100
94
83:17
86:14
52:48
6
52
49 (47)
7[e]
4
100
85
For footnotes [a]–[d], see Table 2. [e] Reaction performed in the presence of powdered molecular sieves 4 ꢄ.
www.chemeurj.org ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
11516
Chem. Eur. J. 2008, 14, 11512 – 11528

Stereoselective Alcohol Silylation
FULL PAPER
Table 5, entry 2]. In the former case, diminished diastereose-
lection in the racemic series (d.r. 88:12) was also manifested
in the kinetic resolution with (SiS)-4a (94% ee) delivering
ly lowered reactivity when exposed to silane (SiS)-4a: Silyl
ether (SiR,R)-22a was only formed at elevated temperatures
with modest diastereoselectivity (d.r. 81:19 in the racemic
series); thus the optical purity of the remaining alcohol (S)-
18 (47% ee) failed to meet a synthetically useful level
(Table 4, entry 4). For comparison, rac-27 equipped with a
cyclohexyl group emerged as poor as well and required ele-
(S)-17 in moderate enantiopurity (69% ee) [rac-17!
(SiR,R)-
ACHTUNGTRENNUNG
21a, Table 4, entry 3]. Similarly, cinnamyl alcohol rac-25 un-
derwent kinetic resolution with (SiR)-4a to give (R)-25
(74% ee) with essentially the same stereoselectivity [rac-
25!
rac-26 was tested with an increased amount of silane (SiR)-
4a (0.65 equiv) [rac-26!
(SiS,S)-31a, Table 5, entry 3].
A
vated reaction temperatures for turnover [rac-27!
ACHTUNGTRENNUNG
(SiS,S)-
32a, Table 5, entry 4]. A substantial loss in reactivity also
complicated dehydrogenative Si–O coupling of sterically de-
manding tert-butyl group-containing alcohol rac-19 (Table 4,
entry 5). Low conversion of 46% was only amenable in the
presence of 10 mol% of copper catalyst and 1.2 equiv of re-
solving agent (SiS)-4a. As a consequence—irrespective of
the promising diastereoselectivity of isolated silyl ether
(SiR,R)-23a (d.r. 94:6 as against 95:5 in the racemic series)—
the enantiomeric purity of isolated alcohol (S)-19 remained
disappointing. Conversely, a simple methyl group as in rac-
28 was tolerated by stereogenic silane (SiR)-4a although the
optical purity of (S)-28 at 58% conversion was only moder-
AHCTUNGTRENNUNG
Higher conversion (63%) in turn led to the isolation of the
remaining alcohol (R)-26 in good enantioselectivity, al-
though the diastereoselectivity was only moderate (d.r.
85:15 in the racemic series). Concomitant (Z)-selective
alkyne reduction was observed, which might be explained
by a directed hydrogenation in the presence of dihydrogen
and a copper catalyst; this is in accordance with a report by
Stryker and coworkers.[26] For the (Z)-cinnamyl alcohol, an
enantiomeric excess of 57% ee was detected; minor
amounts of the (Z)-cinnamyl silyl ether showed an diaste-
reomeric ratio of 90:10.
ate [rac-28!
(SiS,R)-33a, Table 5, entry 5].
ACHTUNGTRENNUNG
Since an enantioimpure reagent was used in almost all ki-
netic resolutions (see below), the assignment of selectivity
factors according to Kagan is not straightforward.[27] In
In contrast, pyridyl ethanols with branched alkyl substitu-
ents emerged as poor substrates (Table 4, entries 4 and 5;
Table 5, entry 4). Isopropyl substitution in rac-18 significant-
Table 4. Kinetic resolution of pyridyl ethanols with (SiS)-4a.[a]
Entry
Racemic
alcohol
R
Silane
Silyl
ether
Recovered
alcohol
Conv
[%][h]
s[i]
s’[j]
ee
T
[8C]
Yield
[%][c]
d.r.[d,e]
d.r.[d,f]
Yield
[%][c]
ee
[%][b]
[%][b,g]
1
2
rac-6
94
99
20
20
G
57
50
84:16
85:15
92:8
91:9
(S)-6
42
42
88 (ꢀ)
58
54
30
26
13
10
rac-16
(S)-16
75 (ꢀ)
3
4
rac-17
rac-18
94
93
20
60
N
48
47
88:12
80:20
88:12
81:19
(S)-17
(S)-18
50
38
69 (ꢀ)
47 (+)
50
51
17
11
8.2
4.4
5[k]
rac-19
94
110
(SiR,R)-23a
40
94:6
95:5
(S)-19
53
68 (+)
46
58
18
[a] All reactions were conducted using CuCl (5.0 mol%), ligand L1d (10 mol%), and NaOtBu (5.0 mol%) with a substrate concentration of 0.1m in tolu-
ene at the indicated temperature. [b] Determined by HPLC analysis using Daicel Chiralcel or Chiralpak columns providing baseline separation of enan-
tiomers. [c] Isolated yield of analytically pure material after flash chromatography; yield calculation is based on initial amount of racemic alcohol. [d] De-
1
termined by H NMR spectroscopy of the crude reaction mixture by integration of baseline-separated resonance signals of the diastereomers. [e] Diaste-
reomeric ratio in the enantioenriched series is biased by conversion (>50%). [f] Diastereomeric ratio in the racemic series (not biased by conversion).
[g] Sense of optical rotation is given in parentheses. [h] Monitored by 1H NMR spectroscopy and determined by integration of baseline-separated reso-
nance signals. [i] Theoretical selectivity factor based on enantiopure silane calculated from s=ln
N
ACHTUNGTRENNUNG
emic series corresponds to ee of the recovered alcohol in the enantiomeric series at exactly 50% conversion (provided that both follow identical kinet-
ics). [j] Apparent selectivity factor when using an enantioimpure reagent calculated from: s’=ln
C=conversion/100.[7b] [k] 1.2 equiv of (SiS)-4a and a catalyst loading of 10 mol% used.
G
E
N
ACHTUNGTRENNUNG
Chem. Eur. J. 2008, 14, 11512 – 11528
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11517

M. Oestreich et al.
Table 5. Kinetic resolution of pyridyl ethanols with (SiR)-4a.[a]
Entry
Racemic
alcohol
R
Silane
Silyl
ether
Recovered
alcohol
Conv
[%][h]
s[i]
s’[j]
ee
T
[8C]
Yield
[%][c]
d.r.[d,e]
84:16
d.r.[d,f]
92:8
Yield
[%][c]
ee
[%][b]
[%][b,g]
1
rac-24
93
20
(SiS,S)-29a
57
(R)-24
43
80 (+)
57
30
9.6
2
rac-25
rac-26
93
93
20
20
G
56
51
87:13
74:26
88:12
85:15
(R)-25
(R)-26
43
31
74 (ꢀ)
89 (ꢀ)
57
63
17
12
7.6
9.0
3[k,l]
4[k]
5
rac-27
rac-28
94
93
60
20
N
55
57
73:27
76:24
n.d.
n.d.
(R)-27
(S)-28
43
42
42 (+)
73 (+)
56
58
n.d.
n.d.
2.9
6.8
For footnotes[a]–[j], see Table 4. [k] 0.65 equiv of (SiR)-4a used. [l] Reaction accompanied by partial Z-selective alkyne reduction: (SiS,S)-31a contami-
nated with 7% of Z-alkene [Z/E 93:7, d.r. (Z Isomer) 90:10] and (R)-26 contaminated with 21% of Z-alkene (57% ee). n.d.=not determined.
Tables 4 and 5, two different calculations are used to assign
a selectivity factor: s[7b] and, additionally, the apparent selec-
tivity factor s’.[27] The former value s is calculated based on
the diastereomeric ratio observed in the racemic series; the
diastereomeric excess should correspond to the enantiomer-
ic excess at exactly 50% conversion when an enantiopure
silane is used and is thus not biased by conversion and the
somewhat critical determination of the latter by 1H NMR
spectroscopy.[28] However, this assumption requires that both
reaction setups follow identical kinetics. The latter value s’
considers the enantiopurity of the resolving chiral silane as
being 100% and is calculated based on the enantiomeric
purity of the recovered alcohol at a given conversion.[7b]
Thus, s’ may be regarded as a minimal value. For suitable
substrates containing aryl, alkenyl and alkynyl moieties at
the stereogenic carbon, s ranges between 12 and 30; s’ is in
the range of 7.6 to 13. Alkyl substitution in turn leads to dif-
ficult-to-resolve substrates; this is not only reflected in 2.9
< s’ < 6.8 (except s’=18 for tBu) but also by the insuffi-
cient reactivity of these derivatives.
secondary alcohols rac-34–rac-40. The “pyridine-type motif”
offered considerable generality (Table 6, entries 1–3): Secon-
dary alcohol rac-34 equipped with a 2-picolinyl group gave
almost identical results as parent alcohol rac-6 [rac-34!
AHCTUNGTRENNUNG
(SiR,R)-41a, entry 1]; at 60% conversion, remaining alcohol
(S)-34 was isolated in useful enantiomeric purity (89% ee,
s=30, s’=11). Not surprisingly, 2-quinolyl substitution was
also tolerated; kinetic resolution of rac-35 with (SiS)-4a
(99% ee) delivered enantioenriched (S)-35 (82% ee at 57%
conversion, s=26, s’=10) along with silyl ether (SiR,R)-42a
(d.r. 81:19 or 91:9 in the racemic series) (entry 2). In con-
trast, the isoquinoline derivative rac-36 displayed significant-
ly reduced stereoselectivity when subjected to (SiS)-4a (97%
ee): At 52% conversion, diastereoselectivity of silyl ether
(SiR,R)-43a (d.r. 83:17) and enantiomeric excess of the slow-
reacting (S)-36 (54% ee or s’=5.0) was disappointing
(entry 3). Subsequently, secondary alcohols rac-37 and rac-
38 containing an oxazole subunit were subjected to the Si–O
coupling (entries 4 and 5). To our delight, 4,5-dimethyloxa-
zol-2-yl ring exhibited good donor capability for copper(I);
the latter fact was apparent from the observed diastereose-
lectivity for rac-37 in the racemic series (d.r. 89:11). When
using enantiopure silane (SiS)-4a (99% ee), moderate enan-
tiomeric excess for recovered (S)-37 (65% ee, s=19, s’=6.2)
Variation of the pending donor: In order to explore the
scope of the present concept, a survey of potential donors
other than the 2-pyridyl unit was of particular interest. For
this purpose, several heteroaromatic substituents with a
Lewis basic site were screened to explore reactivity and ste-
reoselectivity (Table 6). Initially, phenyl substitution at the
stereogenic carbon was retained as a common motif for all
was detected [rac-37!
(SiR,R)-44a, entry 4]. As opposed to
ACHTUNGTRENNUNG
rac-37, benzoxazole derivative rac-38 emerged as incompati-
ble to the reaction conditions: The low solubility of alcohol
rac-38 in toluene required a solvent switch to benzene (or
11518
www.chemeurj.org
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 11512 – 11528

Stereoselective Alcohol Silylation
FULL PAPER
Table 6. Variation of the donor (Do) in the kinetic resolution (SiS)-4a.[a]
Entry
Racemic
alcohol
Donor
Silane
Silyl
ether
Recovered
alcohol
Conv
[%][h]
s[i]
s’[j]
ee
T
[8C]
Yield
[%][c]
d.r.[d,e]
77:23
d.r.[d,f]
92:8
Yield
[%][c]
ee
[%][b]
[%][b,g]
1
2
rac-34
rac-35
98
99
20
20
(SiR,R)-41a
57
52
(S)-34
(S)-35
32
36
89 (ꢀ)
82 (ꢀ)
60
57
30
26
11
10
(SiR,R)-42a
81:19
83:17
91:9
n.d.
3
rac-36
97
40
A
46
(S)-36
47
54 (ꢀ)
52
n.d.
5.0
4
rac-37
rac-38
99
–
20
20
G
50
8
79:21
–
89:11
83:17
(S)-37
rac-38
41
0
65 (ꢀ)
55
18
19
6.2
–
5[k,l]
–
9.1
6[k,l]
rac-39
rac-40
–
–
20
20
G
26
24
–
–
92:8
rac-39
rac-40
4
–
–
48
27
30
–
–
7
65:35
70
2.4
For footnotes[a]–[j], see Table 4. [k] Low solubility of rac-38 and rac-39 in toluene; instead benzene was used as a solvent. [l] Reaction was accompanied
by decomposition of alcohols rac-38 and rac-39 under the reaction conditions.
toluene/THF). Nevertheless, efficient kinetic resolution was
thwarted by a competitive decomposition under the stan-
dard reaction conditions; retro-1,2-addition predominantly
led to methyl benzoxazole along with benzaldehyde. For this
reason, no alcohol rac-38 was recovered in the reaction with
rac-4a; poor conversion to silyl ether (SiR*,R*)-45a (d.r.
83:17) was achieved (entry 5). The same situation was seen
for related benzothiazole derivative rac-39; despite good di-
astereoselection in the formation of silyl ether (SiR*,R*)-46a
(d.r. 92:8), almost no alcohol was recovered from the reac-
tion mixture (entry 6). A comparison with the thiophene de-
rivative rac-40 suggests that sulfur in rac-39 is less likely to
coordinate to copper(I); low conversion and diasterereose-
crowded substrates its reactivity emerged as insufficient.[5b]
Based on an analysis of the assumed transition state, s-bond
metathesis D (cf. Scheme 1), we reasoned that increased
Lewis acidity at the silicon atom might enhance its reactivity
since Si–O interaction is believed to be operating prior to
the metathesis step. Therefore, we anticipated that embed-
ding silicon into a strained carbocycle (strain release Lewis
acidity[24]) ought to produce higher reactivity of silaindane
derivatives 5 towards less reactive substrates. The validity of
this reasoning was recently demonstrated to facilitate the
Cu–H-catalyzed kinetic resolution of tertiary propargylic al-
cohols.[5b]
A brief survey of the synthetic value of virtually enantio-
pure silane (SiR)-5a[23] as a resolving agent for donor-func-
tionalized secondary alcohols revealed that, compared to si-
latetraline (SiR)-4a, reactivity was significantly increased
(Table 7). Kinetic resolutions were generally completed
within a few hours or even minutes compared to approxi-
mately one day for 4a; lower reaction temperatures were
lectivity corroborate this interpretation [rac-40!
(SiR*,R*)-
ACHTUNGTRENNUNG
47a, entry 7].
Kinetic resolution with sterically hindered silaindane: Privi-
leged tert-butyl-substituted silatetraline 4a was compatible
with a reasonable substrate range. However, for sterically
Chem. Eur. J. 2008, 14, 11512 – 11528
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11519

M. Oestreich et al.
Table 7. Kinetic resolution with silane (SiR)-5a.[a]
Entry Racemic
alcohol
R
Donor
Silyl
ether
Recovered
alcohol
Conv s[i] s’[j]
[%][h]
Yield [%][c] d.r.[d,e] d.r.[d,f]
Yield [%][c] ee [%][b,g]
1[j,k]
2
3
4
5
rac-6
Ph
2-pyridyl
N
77:23 86:14 (R)-6
45
38
41
40
41
83 (+)
69 (+)
75 (+)
74 (+)
65 (+)
53
58
56
58
57
13 17
12 5.9
17 8.4
15 7.1
13 5.5
rac-16
rac-24
rac-35
rac-37
2-naphthyl 2-pyridyl
1-naphthyl 2-pyridyl
Ph
Ph
76:24 85:15 (R)-16
77:23 88:12 (R)-24
75:25 87:13 (R)-35
74:26 86:14 (R)-37
2-quinolyl
4,5-dimethyl-oxazol-2-yl ACHTGNUTRENNUNG
For footnotes[a]–[j], see Table 4. [j] 0.57 equiv of (SiR)-5a used. [k] Reaction performed at 08C.
also possible using 5a yet having only minor effect on ste-
reoselectivity. Substrates rac-6, rac-16 and rac-24 with a 2-
pyridyl donor and different aromatic R groups at the stereo-
genic carbon gave inferior results (Table 7, entries 1–3). At
conversions of 53–58%, enantiomeric excesses of the re-
maining alcohols ranged from 69–83% ee; this is also re-
flected by the moderate diastereoselectivities for silyl ethers
(SiS,S)-15a, (SiS,S)-48a–49a in the racemic series (85:15 ꢁ
d.r. ꢁ 88:12). Isolated yields of both silyl ethers (SiS,S)-15a,
Scheme 2. Enantiospecific Si–O coupling and reductive Si–O cleavage.
(SiS,S)-48a–49a and alcohols (R)-6, (R)-16 and (R)-24 indi-
cated again a good mass balance. Similar observations were
made when two different donor functionalities were
screened (entries 4 and 5). Kinetic resolution of quinolyl
substituted rac-35 and oxazole-bearing rac-37 delivered
enantioenriched alcohols (R)-35 and (R)-37 along with silyl
ethers (SiS,S)-50a and (SiS,S)-51a in good yields and the ex-
pected stereoselectivites. Notably, the same enantiomers of
the racemic alcohols were consumed by silatetraline (SiR)-4a
and silaindane (SiR)-5a as secured by the sense of optical ro-
tation.
tectable erosion of enantiomeric purity. The reductive cleav-
age using DIBAL-H at elevated temperature then liberated
silane (SiR)-4a;[17,29] a comparison of HPLC analytical data
confirmed retention of configuration for the overall se-
quence. This outcome is also in agreement with the expected
double retention pathway for the two-step procedure.
At the beginning, the absolute configuration of the slow-
reacting alcohols and newly formed silyl ethers were as-
signed by comparing the sign of optical rotation of recov-
ered (S)-6 with literature data[30] (cf. Table 4, entry 1).[5a]
Relative and absolute configuration of silyl ether (SiR,R)-7a
was further validated by stereospecific reductive cleavage of
silyl ether (SiR,R)-7a that delivered (R)-6 and (SiS)-4a
(Scheme 3). The latter reaction also demonstrates the feasi-
bility of recovering the resolving silane in preserved enan-
tiomeric purity.
Stereochemical course at the silicon atom: In order to rigor-
ously exclude any racemization processes under the reaction
conditions of the kinetic resolution, initial considerations
had resulted in choosing a s-bond metathesis reaction path-
way instead of an approach based on chlorosilanes.[15b] Al-
though we anticipated an enantiospecific Si–O bond forma-
tion by Cu–H-catalyzed dehydrogenative coupling, experi-
mental proof had not been accomplished yet. A simple two-
step sequence consisting of methanolysis and subsequent re-
ductive cleavage of a highly enantiomerically enriched
silane (SiR)-4a clarified this issue [(SiR)-4a!
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
(SiR)-4a, Scheme 2].
Since methanol is, of course, not donor-functionalized, the
Cu–H-catalyzed methanolysis proceeded only sluggishly at
higher catalyst loadings in the presence of excess of metha-
nol [(SiR)-4a!
ACHTUNGTRENNUNG
(SiS)-52a]. Methyl ether (SiS)-52a was conta-
minated with minor amounts of phosphine L1d. Along with
Scheme 3. Recycling of the resolving reagent by stereospecific reductive
Si–O cleavage.
(SiS)-52a, starting silane (SiR)-4a was isolated without de-
11520
www.chemeurj.org
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 11512 – 11528

Stereoselective Alcohol Silylation
FULL PAPER
Thus, absolute and relative configurations of both silyl
ethers and alcohols are predictable provided that the abso-
lute configuration of the resolving silane is known. Based on
model substrate rac-6, configurational assignments for all
other substrates were made. Crystalline material suitable for
X-ray diffraction was obtained as well; the major diastereo-
mer of silyl ether (SiR*,R*)-43a (Figure 2) that was prepared
from isoquinolyl-substituted alcohol rac-36 and rac-4a and
subsequent diastereoenrichment by flash chromatography
confirmed the predicted relative configuration.
much smaller extent than prevailing catalyst-controlled pro-
cesses. In the former scenario, the chiral resolving reagent
should be literally enantiopure. At the beginning of this
project, we were unable to prepare sufficient quantities of
enantiopure silanes; enantiomeric excesses typically ranged
from 93 to 96%.[5a,20b] Yet, this issue is a superable problem
since Kaganꢃs precise predictions for reagent-controlled ki-
netic resolutions also give satisfying results when an enan-
tioimpure reagent is used and (pseudo-)first order kinetics
are prevalent.[27] Consequently, a mutual kinetic resolu-
tion[7b] should be operative. Thus, not only the enantiomeric
excess of the remaining alcohol but also of the resolving
silane is expected to increase with conversion. To test for
this, a kinetic resolution of standard substrate rac-6 with
silane (SiR)-4a of low enantiopurity (32% ee) was monitored
1
by H NMR spectroscopy; aliquots of the reaction mixture
were individually separated into silane, silyl ether and re-
maining alcohol by flash chromatography and subsequently
analyzed by HPLC using a chiral stationary phase. Not only
that the enantiomeric purity of the alcohol (R)-6 gradually
increased with time until reaching 35% ee at 52% conver-
sion, the enantiomeric excess of unreacted silane (SiR)-4a
also rose to 48% ee at 51% conversion (Figure 3). These ex-
perimental observations are in accordance with Kaganꢃs pre-
diciton for a (pseudo-)first order kinetic resolution.[27]
Figure 2. Molecular structure of (SiR*,R*)-43a.
Reductive degradation of the pyridine donor: There is only
a handful of methods available for the enzymatic[30,31] and
chemical[32] kinetic resolution of pyridine-containing alco-
hols. As our procedure provides an enantioselective access
to several, potentially synthetically useful motifs, we briefly
looked into their subsequent chemistry. Danishefsky and co-
workers had used 2,6-lutidines as 1,5-dicarbonyl equivalents
in total synthesis.[33] In our case, Birch reduction followed by
acidic hydrolysis transformed (S)-34 (89% ee) into diketone
(S)-53 (69% ee) in moderate chemical yield (Scheme 4),
which compares well with the reported systems.[33] Unfortu-
nately, partial racemization thwarted any further applica-
tions.
~
*
:
Figure 3. Mutual kinetic resolution using a scalemic silane: : (R)-6,
(SiR)-4a.
Quantum-chemical calculations: Density functional theory
(DFT) and improved Moller–Plesset perturbation (SCS-
MP2) calculations were performed with TURBOMOLE[34]
on the model reaction between (H3P)2CuOR (R=Me and
tBu) and Me3SiH. We used the PBE[35] functional for geom-
etry optimizations and obtained single point energies with
the hybrid functional B3LYP[36] and the wavefunction
method SCS-MP2.[37] For all calculations, a triple zeta AO
basis with additional polarization functions on the heavy
atoms (TZVPP) was used.[38]
Scheme 4. Danishefsky pyridine route: Access to enantioenriched 1,5-di-
carbonyl compounds.
Mutual kinetic resolution: The present methodology holds
interesting mechanistic peculiarties, which are closely linked
to the reagent-controlled nature of this kinetic resolution
strategy—a line of research that has been addressed to a
With R=Me, we have identified a reaction path
(Figure 4) for the Cu–H-catalyzed Si–O coupling via an
transition structure (TS), in which the alkoxide is trans-
ferred to silicon yielding a four-center-intermediate (INT).
Chem. Eur. J. 2008, 14, 11512 – 11528
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11521

M. Oestreich et al.
Although the intermediate has been characterized as a mini-
mum by vibrational analysis, no according transition struc-
ture for its fragmentation towards the product complex was
found. Note that in the course of the reaction (fragmenta-
tion of INT), the Me3Si moiety must be rotated to achieve
the tetrahedral geometry of the product. Relative energies
for intermediates and TS are given in Table 8.
occurs between a chiral Si–H compound and a chiral cop-
per(I) alkoxide.
To get an estimate of steric effects in the reaction, we
have re-calculated the reaction with a tert-butoxy substitu-
ent. The larger substituent has a pronouned impact on the
energy profile, as can be seen in Table 9. With OR=OtBu,
we were not able to locate a four-center-intermediate as
with OR=OMe. Instead, opti-
mization yielded the reactive or
product complexes. These have
higher relative energies com-
pared to the isolated reactants
or products than the corre-
sponding methoxy complexes.
Table 8. Relative energies [kcalmolꢀ1] of stationary points for reaction in Figure 4.
Method[a]
Reactants
r-Complex
TS
Intermediate
p-Complex
Products
DFT-PBE
DFT-B3LYP
SCS-MP2
4.8
1.5
6.2
0.0
0.0
0.0
4.0
5.6
2.3
3.2
5.3
0.2
ꢀ11.7
ꢀ13.7
ꢀ15.7
ꢀ7.3
ꢀ12.1
ꢀ8.4
[a] TZVPP AO basis, DFT-PBE optimized geometries.
Although the reaction barrier and energies differ by some
kcalmolꢀ1 with the three methods, the qualitative descrip-
tion of the energy profile is the same: The energetic distinc-
tion between INT and TS is only very small. It is therefore
not advisable to consider the reaction a “two-step”-process.
With the SCS-MP2 method, the intermediate is almost iso-
energetic to the r-complex and the barrier lower than 3 kcal
molꢀ1. In summary, the prereactive r-complex and the prod-
uct p-complex stabilize reactants and products, the dehydro-
genative Si–O coupling proceeds through an intermediate/
transition structure with a very low barrier. It is exothermic
and irreversible since the back reaction (Si–O reductive
cleavage) requires a substantially larger activation energy
than the Si–O coupling.
Figure 5. Optimized distances in ꢄ (DFT-PBE/TZVPP, bold) and Mullik-
en overlap populations (DFT-PBE/TZVP, italic) for the intermediate
(INT) of the methoxide reaction. For comparison, optimized single bond
lengths of the reactants and products at this level are Cu–O: 1.85 ꢄ, Cu–
H: 1.54 ꢄ, Si–H: 1.51 ꢄ, and Si–O: 1.68 ꢄ.
The barrier of the reaction was estimated by calculating re-
action paths of the Si–O bond formation and taking the
highest point [rACTHNUTRGNEUNG(Si–O)=1.89 ꢄ] as an approximation for the
TS. Most notable are the decreased exothermicity and
larger reaction barrier for (H3P)2CuOtBu. Although the cal-
culated reaction is only a model using very small metal li-
gands, the barrier has more than doubled with all methods
employed. The “best” method (SCS-MP2) predicts a 2.5-
fold higher energy of the transition structure than in the re-
action of the smaller copper(I) alkoxide. The retardation ob-
served in the experiment using CuCl/NaOtBu with L1d is
therefore nicely reproduced in this calculation.
Figure 4. Energy profile (DFT-PBE/TZVPP) for the reaction of
(H3P)2CuOMe and Me3SiH.
Proposed mechanism: Based on literature precedent,[12–14,39]
our stereochemical analysis, and the quantum-chemical cal-
culations, we postulate a complete catalytic cycle. It com-
prises a four-step propagation as it is depicted in Scheme 5.
The catalysis is initiated by phosphine-stabilized copper(I)
alkoxide[11c,40] rac-55 that is generated from 54 by alkoxide
exchange. A single catalytic turnover generates copper(I)
hydride 57, which is the catalytically active species.[12–14,39] In
step i) dihydrogen is liberated from alcohol rac-6 and cop-
per(I) hydride 57 upon which copper(I) chelate rac-55 is
formed. Reversible and, likely, rate-limiting dissociation of a
The front-side attack at the silicon atom confirms the ex-
perimental observation that the substitution occurs under
retention at the silicon atom. Although the four-center-inter-
mediate (Figure 5) is energetically close to the reactant com-
plex, the bond lenghts and bond orders indicate an almost
synchronous bond transfer that has proceeded about half
the way towards the product. The compact nature of the in-
termediate is also reflected in the rather short Cu–Si dis-
tance (2.81 ꢄ). It is therefore reasonable to expect a high
degree of stereochemical discrimination when the reaction
11522
www.chemeurj.org
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 11512 – 11528

Stereoselective Alcohol Silylation
FULL PAPER
Table 9. Relative energies [kcalmolꢀ1] of stationary points for reaction in Figure 4 with (H3P)2CuOtBu as cop-
per(I) reagent.
electron-rich monophosphines
generate sufficiently reactive
copper(I) complexes, 2) two-
Method[a]
Reactants
r-Complex
TS[b]
p-Complex
Products
DFT-PBE
DFT-B3LYP
SCS-MP2
3.6
0.1
5.4
0.0
0.0
0.0
9.4
12.6
5.9
ꢀ7.4
ꢀ9.9
ꢀ10.8
ꢀ4.1
ꢀ9.1
ꢀ4.4
point binding of the substrate
to the transition metal, that is
formation of a six-membered
[a] TZVPP AO basis, DFT-PBE optimized geometries. [b] Highest point of the reaction coordinate for Si–O
bond formation [r(Si–O)=1.89 ꢄ].
chelate by coordination of both
ACHTUNGTRENNUNG
the alkoxy group and an sp2-hy-
bridized nitrogen donor, is es-
sential, and 3) the use of steri-
cally hindered, silicon-stereogenic silanes with silatetralines
being less reactive but more selective than the correspond-
ing silaindanes. With ideal catalyst–substrate–silane combi-
nations a selectivity factor of s=30 are possible.
The Si–O coupling step itself is a particularly interesting
feature of this Cu–H catalysis. It proceeds with stereoreten-
tion at the silicon atom passing through an assumed four-
centered transition state. Quantum-chemical calculations do
indeed support a s-bond metathesis, yet its precise nature—
transition state or, more likely, transient intermediate—re-
mains slightly uncertain.
Experimental Section
General information: Reagents for starting material syntheses, CuCl as
well as ligands L1a, L1g and L2 were purchased from commercial sour-
ces. NaOtBu was freshly prepared and stored under argon atmosphere.
Silanes rac-1,[18] rac-2,[19] rac-3,[20a] 4a,[20b] rac-4b,[16b] rac-5b,[16d] ligands
L1b–L1 f,[41] as well as L3a and L3b[42] were prepared according to
known procedures. All reactions were performed in flame-dried glass-
ware under a static pressure of argon. Liquids and solutions were trans-
ferred with syringes or double-ended needles. Toluene (Na/benzophe-
none) and CH2Cl2 (CaH2) were dried by continuous distillation from the
indicated drying reagent prior to use. Analytical thin-layer chromatogra-
phy (TLC) was performed on silica gel SIL G-25 glass plates by Macher-
ey-Nagel (Germany); for flash column chromatography, silica gel 60 (40–
63 mm, 230–400 mesh, ASTM) by Merck (Germany), cyclohexane and
tert-butyl methyl ether as solvent were used. 1H and 13C NMR spectra
were recorded in CDCl3 on Bruker AM 400, AV 400, DRX 500, Varian
INOVA 500 and Unity plus 600 spectrometers. HPLC analyses were per-
formed with an Agilent 1200 or a Merck-Hitachi LaChrom 7100 instru-
ment using chiral stationary phases (Daicel Chiralcel OD-H, AD-H, OJ-
R, OJ-RH or Chiralpak IB columns). Mass spectra were recorded with
Finnigan MAT TSQ 7000 (EI, CI) or Bruker MicroTOF (ESI) instru-
ments. IR spectra were recorded on Perkin–Elmer Paragon 1000 or
Varian 3100 FT-IR instruments. Optical rotations were measured in a 1
dm cuvette on a Perkin–Elmer 341 polarimeter. Elemental analyses were
conducted on a Vario EL instrument from Elementaranalysensysteme
GmbH. Experimental details for the synthesis and resolution of silane
5a, as well as for the preparation of racemic secondary alcohols and char-
acterization data for all silyl ethers are included in the Supporting Infor-
mation of this article.
Scheme 5. Proposed catalytic cycle.
phosphine ligand in step ii) generates a 16-electron inter-
mediate that is prone to coordinative saturation by weakly
coordinating silane 4a in step iii). Step ii) could moreover
facilitate a pseudo-first order scenario that is essential for
good stereoselectivity in the kinetic resolution. At this stage,
diastereodiscrimination between enantiomeric copper(I)
alkoxide complexes must occur. Hence, one out of the two
enantiomers of 6 is preferentially funneled out of the reac-
tion mixture via rapid, exergonic s-bond metathesis, transi-
ent intermediate or transition state 56 in step iv).[12–14,39] By
this, the major diastereomer (SiS*,S*)-7a is formed along
with the regeneration of copper(I) hydride 57. Importantly,
the diastereodiscriminating event is step iv) and thus all pre-
ceding steps must be in equilibrium. A relative to the irre-
versible step rapid exchange of both enantiomers of pyridyl
alcohol 6 at copper(I) chelate 55 is essential for this.
(SiR)-1-tert-Butyl-1-silaindane [(SiR)-5a]:[23] Rf =0.77 (cyclohexane);
Conclusion
27.8 min [(SiS)-5a]; 1H NMR (400 MHz, CDCl3): d=1.02 (s, 9H), 1.04
(dddd, J=15.2, 7.9, 6.0, 3.0 Hz, 1H), 1.16 (dddd, J=15.2, 8.8, 7.1, 3.2 Hz,
1H), 3.08–3.14 (m, 2H), 4.45 (dd, J=J=3.1 Hz, 1H), 7.20 (m, 1H), 7.24–
7.28 (m, 1H), 7.32 (m, 1H), 7.61 ppm (m, 1H); 13C NMR (100 MHz,
CDCl3): d=5.3, 17.1, 26.8, 32.5, 125.7, 125.8, 129.7, 133.4, 135.2,
154.5 ppm; IR (film): n˜ =3056 (s), 2923 (s), 2108 (s), 1589 (s), 1463 (s),
In summary, we presented a full account of our investiga-
tions towards a unique kinetic resolution strategy. The im-
portant parameters governing reactivity and diastereoselec-
tivity of this reagent-controlled process are 1) moderately
Chem. Eur. J. 2008, 14, 11512 – 11528
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11523

M. Oestreich et al.
1441 cmꢀ1 (s); HRMS (EI): m/z: calcd for C12H18Si [M+]: 190.1178,
found: 190.1173; elemental analysis calcd (%) for C12H18Si (190.4): C
75.71, H 9.53; found: C 75.72, H 9.52.
(49.1 mg, 0.240 mmol, 0.600 equiv, 94% ee), CuCl (2.0 mg, 0.020 mmol,
0.050 equiv), L1d (13.9 mg, 0.0400 mmol, 0.100 equiv) and NaOtBu
(1.9 mg, 0.020 mmol, 0.050 equiv) in toluene (4.0 mL), silyl ether (SiR,R)-
21a (67 mg, 48%, d.r. 88:12) and alcohol (S)-17 (30 mg, 50%, 69% ee)
were isolated as analytically pure materials. (S)-17: M.p. 468C (CH2Cl2);
Rf =0.41 (CH2Cl2/methanol 95:5); [a]2D0 =ꢀ1.15 (c=0.420, CHCl3),
[a]25078 =ꢀ0.77, [a]52046 =+0.38, [a]42036 =+10.0, [a]23605 =+44.2; HPLC (Daicel
General procedure for Cu–H-catalyzed kinetic resolution of donor-func-
tionalized alcohols (GP 1): A flame-dried Schlenk tube equipped with a
magnetic stirring bar was charged with a mixture of CuCl (0.050 equiv)
and L1d (0.10 equiv) under argon atmosphere. The solids were suspend-
ed in degassed, anhydrous toluene. After the additon of NaOtBu
(0.050 equiv), a pale yellow solution formed within 1–2 min stirring at
Chiralcel OD-H, 208C, n-heptane/iPrOH 98:2, flow rate 0.80 mLminꢀ1
,
l=230 nm): tR =21.8 min [(S)-17], 24.2 min [(R)-17]; 1H NMR (500 MHz,
CDCl3): d=2.93 (dd, J=14.8, 8.6 Hz, 1H), 3.02 (dd, J=14.8, 3.2 Hz,
1H), 4.59 (ddddd, J=8.5, 5.7, 3.2, J=J=1.6 Hz, 1H), 5.11 (ddd, J=10.4,
J=J=1.6 Hz, 1H), 5.23 (brs, 1H), 5.31 (ddd, J=17.0, J=J=1.6 Hz, 1H),
5.94 (ddd, J=17.0, 10.4, 5.7 Hz, 1H), 7.13–7.18 (m, 2H), 7.62 (ddd, J=
J=7.6, 1.9 Hz, 1H), 8.50 ppm (brd, J=4.4 Hz, 1H); 13C NMR (125 MHz,
CDCl3): d=43.3, 72.0, 114.7, 121.7, 123.9, 136.9, 140.3, 148.7, 159.8 ppm;
IR (cuvette/CDCl3): n˜ = 3317 (s), 3088 (s), 3018 (s), 2079 (s), 2915 (m),
1849 (w), 1645 (w), 1597 (s), 1571 (s), 1476 (s), 1438 (s), 1366 (w), 1327
(w), 1284 (w), 1236 (w), 1198 (m), 1150 (m), 1111 (m), 1099 (m), 1078
(s), 1051 (s), 1032 (s), 992 (s), 963 (s), 843 (s), 808 cmꢀ1 (s); LRMS (CI/
NH3): m/z: 150 [M+H+]; elemental analysis calcd (%) for C9H11NO
(149.2): C 72.26, H 7.43, N 9.39; found: C 72.24, H 7.50, N 9.20.
room temperature. Subsequently,
a solution of racemic alcohol
(1.0 equiv) in toluene (0.2m) and silane (0.60 equiv) in toluene (0.24m)
were sequentially added via syringe. The resulting bright yellow solution
(0.1m relating to the alcohol) occasionally showed gas evolution and was
maintained at the indicated temperature until complete consumption of
1
silane as monitored by H NMR spectroscopic analysis. After addition of
tert-butyl methyl ether (8 mL) and a small portion of silica gel, the sol-
vents were evaporated. Purification by flash column chromatography on
silica gel using cyclohexane/tert-butyl methyl ether mixtures afforded the
analytically pure silyl ether along with the unreacted alcohol as oils or
solids.
(S)-1-Phenyl-2-(pyridin-2-yl)ethanol [(S)-6] (Table 4, entry 1): According
to GP 1, starting from rac-6 (79.7 mg, 0.400 mmol, 1.00 equiv), (SiS)-4a
(49.1 mg, 0.240 mmol, 0.600 equiv, 94% ee), CuCl (2.0 mg, 0.020 mmol,
0.050 equiv), L1d (13.9 mg, 0.0400 mmol, 0.100 equiv) and NaOtBu
(1.9 mg, 0.020 mmol, 0.050 equiv) in toluene (4.0 mL), silyl ether (SiR,R)-
7a (92 mg, 57%, d.r. 84:16) and alcohol (S)-6 (33 mg, 42%, 88% ee)
were isolated as analytically pure materials. (S)-6: M.p. 1078C (CH2Cl2/
(S)-3-Methyl-1-(pyridin-2-yl)butan-2-ol [(S)-18] (Table 4, entry 4): Ac-
cording to GP 1, starting from rac-18 (66.1 mg, 0.400 mmol, 1.00 equiv),
(SiS)-4a (49.1 mg, 0.240 mmol, 0.600 equiv, 93% ee), CuCl (2.0 mg,
0.020 mmol, 0.050 equiv), L1d (13.9 mg, 0.0400 mmol, 0.100 equiv) and
NaOtBu (1.9 mg, 0.020 mmol, 0.050 equiv) in toluene (4.0 mL), silyl ether
(SiR,R)-22a (69 mg, 47%, d.r. 80:20) and alcohol (S)-18 (25 mg, 38%,
47% ee) were isolated as colorless oils. (S)-18: Rf =0.11 (cyclohexane/
tert-butyl methyl ether); Rf =0.12 (cyclohexane/tert-butyl methyl ether
tert-butyl methyl ether 1:1); [a]2D0 =+5.32 (c=0.470, CHCl3), [a]25078
=
20
436
1:1); [a]2D0 =ꢀ31.4 (c=1.05, CHCl3), [a]25078 =ꢀ32.7, [a]25406 =ꢀ35.9, [a]
=
+5.96, [a]25046 =+7.87, [a]42306 =+25.5, [a]32605 =+78.5; HPLC (Daicel Chiral-
cel OD-H, 208C, n-heptane/iPrOH 98:2, flow rate 0.80 mLminꢀ1, l=
230 nm): tR =12.7 min [(R)-18], 14.4 min [(S)-18]; 1H NMR (400 MHz,
CDCl3): d=0.98 (d, J=6.7 Hz, 3H), 1.01 (d, J=6.8 Hz, 3H), 1.74 (qqd,
J=J=6.7, 5.9 Hz, 1H), 2.83 (dd, J=14.8, 9.5 Hz, 1H), 2.90 (dd, J=14.8,
2.4 Hz, 1H), 3.77 (ddd, J=9.4, 5.7, 2.6 Hz, 1H), 5.03 (brs, 1H), 7.12–7.17
(m, 2H), 7.61 (ddd, J=J=7.7, 1.9 Hz, 1H), 8.48 ppm (ddd, J=4.8, 1.8,
0.8 Hz, 1H); 13C NMR (100 MHz, CDCl3): d=18.1, 18.7, 33.6, 40.3, 76.0,
121.5, 123.8, 136.8, 148.6, 160.9 ppm; IR (cuvette/CDCl3): n˜ =3337 (s),
3142 (w), 3089 (w), 3076 (w), 3018 (w), 2963 (s), 2933 (s), 2910 (s), 2896
(s), 2874 (s), 1596 (s), 1570 (s), 1475 (s), 1438 (s), 1384 (m), 1367 (m),
1340 (w), 1311 (w), 1279 (m), 1238 (w), 1198 (m), 1170 (w), 1150 (m),
1130 (w), 1097 (m), 1079 (m), 1051 (s), 1039 (s), 1007 (s), 997 (s), 921 (s),
887 (w), 845 (m), 795 cmꢀ1 (m); LRMS (CI/NH3): m/z: 166 [M+H+]; ele-
mental analysis calcd (%) for C10H15NO (165.2): C 72.69, H 9.15, N 8.48;
found: C 72.47, H 9.22, N 8.34.
ꢀ48.9, [a]23065 =ꢀ37.0; HPLC (Daicel Chiralcel OD-H, 208C, n-heptane/
iPrOH 90:10, flow rate 0.80 mLminꢀ1, l=230 nm): tR =12.4 min [(R)-6],
1
18.1 min [(S)-6]; H NMR (400 MHz, CDCl3): d=3.08–3.20 (m, 2H), 5.18
(dd, J=8.3, 3.9 Hz, 1H), 5.69 (brs, 1H), 7.11 (brd, J=8.0 Hz, 1H), 7.18
(ddd, J=7.6, 4.9, 1.1 Hz, 1H), 7.24–7.29 (m, 1H), 7.32–7.37 (m, 2H),
7.41–7.45 (m, 2H), 7.61 (ddd, J=J=7.6, 1.0 Hz, 1H), 8.54 ppm (ddd, J=
4.9, 1.6, 1.0 Hz, 1H); 13C NMR (100 MHz, CDCl3): d=45.8, 73.4, 121.8,
123.9, 125.9, 127.3, 128.4, 136.9, 144.2, 148.7, 158.9 ppm; IR (cuvette/
CDCl3): n˜ =3308 (s), 3089 (s), 3066 (s), 3031 (s), 2906 (m), 1952 (w), 1882
(w), 1760 (w), 1597 (s), 1570 (s), 1495 (m), 1475 (s), 1453 (s), 1438 (s),
1338 (m), 1279 (m), 1280 (m), 1150 (w), 1100 (m), 1078 (w), 1054 (m),
1028 (m), 1019 (m), 1001 (w), 903 (m), 816 (w), 735 cmꢀ1 (w); LRMS
(CI/NH3): m/z: 200 [M+H+]; elemental analysis calcd (%) for C13H13NO
(199.3): C 78.36, H 6.58, N 7.03; found: C 78.13, H 6.63, N 6.95.
(S)-1-(Naphth-2-yl)-2-(pyridin-2-yl)ethanol [(S)-16] (Table 4, entry 2):
According to GP 1, starting from rac-16 (99.7 mg, 0.400 mmol,
1.00 equiv), (SiS)-4a (49.1 mg, 0.240 mmol, 0.600 equiv, 99% ee), CuCl
(2.0 mg, 0.020 mmol, 0.050 equiv), L1d (13.9 mg, 0.0400 mmol,
0.100 equiv) and NaOtBu (1.9 mg, 0.020 mmol, 0.050 equiv) in toluene
(4.0 mL), silyl ether (SiR,R)-20a (90 mg, 50%, d.r. 85:15) and alcohol (S)-
16 (42 mg, 42%, 75% ee) were isolated as analytically pure materials.
(S)-16: M.p. 99–1008C (cyclohexane/tert-butyl methyl ether); Rf =0.28
(cyclohexane/tert-butyl methyl ether 1:2); [a]2D0 =ꢀ21.2 (c=1.01, CHCl3),
[a]25078 =ꢀ23.0, [a]52046 =ꢀ25.4, [a]42036 =ꢀ34.0, [a]23605 =ꢀ27.7; HPLC (Daicel
(S)-3,3-Dimethyl-1-(pyridin-2-yl)butan-2-ol [(S)-19] (Table 4, entry 5):
According to GP 1, starting from rac-19 (71.7 mg, 0.400 mmol,
1.00 equiv), (SiS)-4a (98.1 mg, 0.480 mmol, 1.20 equiv, 94% ee), CuCl
(4.0 mg, 0.040 mmol, 0.10 equiv), L1d (27.7 mg, 0.0800 mmol,
0.200 equiv) and NaOtBu (3.8 mg, 0.040 mmol, 0.10 equiv) in toluene
(4.0 mL), silyl ether (SiR,R)-22a (61 mg, 40%, d.r. 94:6) and alcohol (S)-
19 (38 mg, 53%, 68% ee) were isolated as colorless oils. (S)-19: Rf =0.59
(cyclohexane/tert-butyl methyl ether 1:2); [a]2D0 =+18.1 (c=0.36, CHCl3),
[a]25078 =+18.6, [a]25406 =+23.3, [a]42036 =+61.1, [a]23605 =+164; HPLC (Daicel
Chiralcel AD-H, 208C, n-heptane/iPrOH 80:20, flow rate 1.00 mLminꢀ1
,
Chiralcel AD-H, 208C, n-heptane/iPrOH 98:2, flow rate 0.80 mLminꢀ1
,
l=230 nm): tR =14.0 min [(R)-16], 16.6 min [(S)-16]; 1H NMR (400 MHz,
CDCl3): d=3.21 (dd, J=15.2, 4.7 Hz, 1H), 3.23 (dd, J=15.2, 7.5 Hz,
1H), 5.35 (dd, J=7.4, 4.7 Hz, 1H), 5.88 (brs, 1H), 7.11 (ddd, J=7.8, 2.1,
1.0 Hz, 1H), 7.19 (ddd, J=7.6, 4.9, 1.2 Hz, 1H), 7.46 (m, 2H), 7.55 (dd,
J=8.6, 1.6 Hz, 1H), 7.61 (ddd, J=J=7.7, 1.9 Hz, 1H), 7.81–7.85 (m,
3H), 7.89–7.91 (m, 1H), 8.56 ppm (ddd, J=4.9, 1.8, 0.9 Hz, 1H);
13C NMR (100 MHz, CDCl3): d=45.7, 73.5, 121.8, 123.9, 124.3, 124.5,
125.7, 126.0, 127.7, 128.1, 132.9, 133.5, 137.0, 141.6, 148.7, 159.8 ppm; IR
(cuvette/CDCl3): n˜ =3305 (m), 3060 (w), 1595 (s), 1475 (s), 1437 (s),
1058 cmꢀ1 (s); HRMS (EI): m/z: calcd for C17H15NO [M+]: 249.1154,
found: 249.1160; elemental analysis calcd (%) for C17H15NO (249.3): C
81.90, H 6.06, N 5.62; found: C 81.64, H 6.10, N 5.55.
l=230 nm): tR =15.2 min [(R)-19], 29.9 min [(S)-19]; 1H NMR (400 MHz,
CDCl3): d=1.00 (s, 9H), 2.76 (dd, J=14.7, 10.5 Hz, 1H), 2.92 (dd, J=
14.7, 1.9 Hz, 1H), 3.66 (dd, J=10.5, 1.9 Hz, 1H), 5.03 (brs, 1H), 7.14 (dd,
J=7.6, 5.0 Hz, 1H), 7.16 (brd, J=7.9 Hz, 1H), 7.62 (ddd, J=J=7.6,
1.9 Hz, 1H), 8.48 ppm (ddd, J=4.8, 1.6, 0.9 Hz, 1H); 13C NMR
(100 MHz, CDCl3): d=25.9, 34.8, 38.7, 78.7, 121.5, 123.8, 136.9, 148.6,
161.2 ppm; IR (cuvette/CDCl3): n˜ = 3325 (s), 3142 (w), 3090 (w), 3076
(w), 3063 (w), 3017 (w), 2961 (s), 2910 (s), 2870 (s), 1595 (s), 1570 (m),
1478 (s), 1437 (s), 1389 (w), 1364 (s), 1328 (w), 1312 (w), 1279 (w), 1243
(m), 1213 (w), 1201 (w), 1177 (w), 1150 (w), 1101 (m), 1071 (s), 1049 (m),
1013 (s), 966 (m), 837 (s), 806 cmꢀ1 (s); HRMS (ESI): m/z: calcd for
C11H18NO [M+H+]: 180.1383, found: 180.1396; elemental analysis calcd
(%) for C11H17NO (179.3): C 73.70, H 9.56, N 7.81; found: C 73.42, H
9.74, N 7.67.
(S)-1-(Pyridin-2-yl)but-3-en-2-ol [(S)-17] (Table 4, entry 3): According to
GP 1, starting from rac-17 (59.7 mg, 0.400 mmol, 1.00 equiv), (SiS)-4a
11524
www.chemeurj.org
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 11512 – 11528

Stereoselective Alcohol Silylation
FULL PAPER
(R)-1-(Naphth-1-yl)-2-(pyridin-2-yl)ethanol [(R)-24] (Table 5, entry 1):
According to GP 1, starting from rac-16 (99.7 mg, 0.400 mmol,
1.00 equiv), (SiR)-4a (49.1 mg, 0.240 mmol, 0.600 equiv, 93% ee), CuCl
(2.0 mg, 0.020 mmol, 0.050 equiv), L1d (13.9 mg, 0.0400 mmol,
0.100 equiv) and NaOtBu (1.9 mg, 0.020 mmol, 0.050 equiv) in toluene
(4.0 mL), silyl ether (SiS,S)-29a (102 mg, 57%, d.r. 84:16) and alcohol
(R)-24 (43 mg, 43%, 80% ee) were isolated as analytically pure materi-
als. (R)-24: M.p. 1148C (cyclohexane/tert-butyl methyl ether); Rf =0.17
(cyclohexane/tert-butyl methyl ether 1:1); [a]2D0 =+171 (c=0.320, CHCl3),
[a]52078 =+178, [a]52406 =+203, [a]42306 =+345, [a]23605 =+504; HPLC (Daicel
(w), 1100 (s), 1053 (s), 1032 (s), 1001 (s), 936 (m), 926 (s), 919 (s), 841
(s), 795 cmꢀ1 (s); HRMS (EI): m/z: calcd for C15H13NO [M+]: 223.0993,
found: 223.0997. (Z)-alkene: HPLC (Daicel Chiralcel OD-H, 208C, n-
heptane/iPrOH 90:10, flow rate 0.80 mLminꢀ1, l=230 nm): tR =10.9 min
(S enantiomer of Z-alkene), 13.9 min (R enantiomer of Z-alkene);
1H NMR (500 MHz, CDCl3): d=3.02 (dd, J=14.9, 3.5 Hz, 1H), 3.07 (dd,
J=15.1, 8.2 Hz, 1H), 5.01 (dddd, J=J=8.5, 3.2, 1.2 Hz, 1H), 5.57 (brs,
1H), 5.77 (dd, J=11.3, 9.1 Hz, 1H), 6.57 (brd, J=11.3 Hz, 1H), 7.13
(brd, J=7.9 Hz, 1H), 7.16–7.29 (m, 2H), 7.29–7.37 (m, 4H), 7.62 (ddd,
J=J=7.6, 1.9 Hz, 1H), 8.50 ppm (brd, J=5.0 Hz, 1H); 13C NMR
(125 MHz, CDCl3): d=43.3, 67.4, 121.8, 124.1, 127.2, 128.2, 128.3, 129.0,
131.1, 133.5, 136.8, 148.7, 159.9 ppm; HRMS (EI): m/z: calcd for
C15H15NO [M+]: 225.1154, found: 225.1150.
Chiralcel OD-H, 208C, n-heptane/iPrOH 90:10, flow rate 0.80 mLminꢀ1
,
l=230 nm): tR =18.6 min [(S)-24], 24.0 min [(R)-24]; 1H NMR (400 MHz,
CDCl3): d=3.25 (dd, J=15.3, 9.7 Hz, 1H), 3.35 (dd, J=15.4, 2.7 Hz,
1H), 5.85 (brs, 1H), 5.98 (dd, J=9.0, 2.7 Hz, 1H), 7.11 (brd, J=7.8 Hz,
1H), 7.22 (ddd, J=7.6, 5.0, 1.1 Hz, 1H), 7.47–7.54 (m, 3H), 7.63 (ddd,
J=J=7.6, 1.9 Hz, 1H), 7.77–7.81 (m, 2H), 7.88–7.91 (m, 1H), 8.12–8.16
(m, 1H), 8.60 ppm (ddd, J=5.0, 2.0, 1.0 Hz, 1H); 13C NMR (100 MHz,
CDCl3): d=44.9, 70.1, 121.9, 123.0, 123.3, 123.8, 125.4, 125.7, 126.0, 127.8,
129.0, 130.3, 133.8, 137.0, 139.5, 148.8, 160.0 ppm; IR (cuvette/CDCl3):
n˜ =3314 (s), 3062 (s), 2906 (w), 1599 (s), 1570 (s), 1512 (m), 1475 (s),
1438 (s), 1394 (m), 1359 (w), 1325 (w), 1281 (w), 1189 (s), 1102 (m), 1067
(w), 1003 (w), 891 (s), 883 (s), 805 cmꢀ1 (s); LRMS (EI): m/z: 249 [M+];
elemental analysis calcd (%) for C17H15NO (249.3): C 81.90, H 6.06, N
5.62; found: C 81.63, H 6.12, N 5.50.
(R)-1-Cyclohexyl-2-(pyridin-2-yl)ethanol [(R)-27] (Table 5, entry 4): Ac-
cording to GP 1, starting from rac-27 (82.1 mg, 0.400 mmol, 1.00 equiv),
(SiR)-4a (53.1 mg, 0.260 mmol, 0.650 equiv, 93% ee), CuCl (2.0 mg,
0.020 mmol, 0.050 equiv), L1d (13.9 mg, 0.0400 mmol, 0.100 equiv) and
NaOtBu (1.9 mg, 0.020 mmol, 0.050 equiv) in toluene (4.0 mL), silyl ether
(SiS,S)-32a (90 mg, 55%, d.r. 73:27) and alcohol (R)-27 (35 mg, 43%,
42% ee) were isolated as analytically pure materials. (R)-27: M.p. 358C
(cyclohexane/tert-butyl methyl ether); Rf =0.26 (cyclohexane/tert-butyl
20
546
methyl ether 1:1); [a]D20 =ꢀ5.82 (c=0.550, CHCl3), [a]25708 =ꢀ6.36, [a]
=
ꢀ7.64, [a]24306 =ꢀ18.5, [a]32605 =ꢀ50.4; HPLC (Daicel Chiralcel OD-H,
208C, n-heptane/iPrOH 98:2, flow rate 0.80 mLminꢀ1, l=230 nm): tR =
7.8 min [(R)-27], 8.7 min [(S)-27]; 1H NMR (400 MHz, CDCl3): d=1.05–
1.31 (m, 5H), 1.38–1.47 (m, 1H), 1.63–1.70 (m, 1H), 1.72–1.82 (m, 3H),
1.94 (m, 1H), 2.85 (dd, J=14.9, 9.2 Hz, 1H), 2.92 (dd, J=14.9, 2.6 Hz,
1H), 3.77 (ddd, J=9.2, 5.9, 2.6 Hz, 1H), 5.04 (brs, 1H), 7.11–7.16 (m,
2H), 7.61 (ddd, J=J=7.7, 1.9 Hz, 1H), 8.47 ppm (dd, J=5.6, 1.9 Hz,
1H); 13C NMR (100 MHz, CDCl3): d=26.3, 26.4, 26.7, 28.6, 29.1, 40.5,
43.6, 75.3, 121.5, 123.8, 136.8, 148.6, 160.9 ppm; IR (cuvette/CDCl3): n˜ =
3333 (s), 3142 (w), 3089 (w), 3076 (w), 3018 (w), 2929 (s), 2854 (s), 1597
(s), 1570 (m), 1475 (s), 1450 (s), 1437 (s), 1345 (w), 1312 (w), 1277 (m),
1262 (m), 1228 (m), 1193 (m), 1175 (m), 1150 (m), 1131 (m), 1102 (m),
1051 (m), 1034 (s), 1002 (m), 990 (m), 900 (s), 841 (m), 809 cmꢀ1 (m);
LRMS (CI/NH3): m/z: 206 [M+H+]; elemental analysis calcd (%) for
C13H19NO (205.3): C 76.06, H 9.33, N 6.82; found: C 75.95, H 9.50, N
6.64.
ACHTUNGTRENNUNG(R,E)-4-Phenyl-1-(pyridin-2-yl)but-3-en-2-ol [(R)-25] (Table 5, entry 2):
According to GP 1, starting from rac-25 (90.1 mg, 0.400 mmol,
1.00 equiv), (SiR)-4a (49.1 mg, 0.240 mmol, 0.600 equiv, 93% ee), CuCl
(2.0 mg, 0.020 mmol, 0.050 equiv), L1d (13.9 mg, 0.0400 mmol,
0.100 equiv) and NaOtBu (1.9 mg, 0.020 mmol, 0.050 equiv) in toluene
(4.0 mL), silyl ether (SiS,S)-30a (99 mg, 56%, d.r. 87:13) and alcohol (R)-
25 (38 mg, 43%, 74% ee) were isolated as analytically pure materials.
(R)-25: Rf =0.24 (cyclohexane/tert-butyl methyl ether 1:2); [a]2D0 =ꢀ67.5
20
365
(c=0.440, CHCl3), [a]25708 =ꢀ71.4, [a]25046 =ꢀ82.5, [a]42306 =ꢀ163, [a]
=
ꢀ326; HPLC (Daicel Chiralcel OD-H, 208C, n-heptane/iPrOH 90:10,
flow rate 0.80 mLminꢀ1
, l=230 nm): tR =17.9 min [(R)-25], 20.5 min
[(S)-25]; 1H NMR (500 MHz, CDCl3): d=3.03 (dd, J=15.5, 8.5 Hz, 1H),
3.10 (dd, J=15.2, 3.5 Hz, 1H), 4.78 (dddd, J=8.5, 6.0, 3.5, 1.3 Hz, 1H),
5.48 (brs, 1H), 6.31 (dd, J=15.7, 5.8 Hz, 1H), 6.68 (dd, J=15.7, 1.6 Hz,
1H), 7.16–7.20 (m, 2H), 7.20–7.24 (m, 1H), 7.28–7.32 (m, 2H), 7.36–7.39
(m, 2H), 7.64 (ddd, J=J=7.6, 1.9 Hz, 1H), 8.52 ppm (ddd, J=4.7, 1.9,
1.0 Hz, 1H); 13C NMR (125 MHz, CDCl3): d=43.6, 71.8, 121.8, 123.9,
126.6, 127.5, 128.6, 130.0, 131.7, 136.9, 137.1, 148.7, 159.7 ppm; IR (cuv-
ette/CDCl3): n˜ =3306 (s), 3143 (w), 3084 (m), 3063 (m), 3028 (s), 2908
(m), 1596 (s), 1570 (s), 1494 (s), 1475 (s), 1438 (s), 1326 (w), 1307 (w),
1188 (m), 1150 (m), 1101 (m), 1088 (m), 1069 (s), 1050 (s), 1029 (s), 1001
(s), 968 (s), 921 (s), 904 (m), 892 (m), 888 (m), 809 cmꢀ1 (s); LRMS (EI):
m/z: 225 [M+]; elemental analysis calcd (%) for C15H15NO (225.3): C
79.97, H 6.71, N 6.22; found: C 79.82, H 6.66, N 6.24.
(S)-1-(Pyridin-2-yl)propan-2-ol [(S)-28] (Table 5, entry 5): According to
GP 1, starting from rac-28 (54.9 mg, 0.400 mmol, 1.00 equiv), (SiR)-4a
(49.1 mg, 0.240 mmol, 0.600 equiv, 93% ee), CuCl (2.0 mg, 0.020 mmol,
0.050 equiv), L1d (13.9 mg, 0.0400 mmol, 0.100 equiv) and NaOtBu
(1.9 mg, 0.020 mmol, 0.050 equiv) in toluene (4.0 mL), silyl ether (SiS,R)-
33a (77 mg, 57%, d.r. 76:24) and alcohol (R)-27 (23 mg, 42%, 73% ee)
were isolated as analytically pure materials. (S)-28: Rf =0.25 (CH2Cl2/
20
546
MeOH 97:3); [a]D20 =+5.71 (c=0.350, CHCl3), [a]25708 =+6.29, [a]
=
+5.15, [a]24306 =ꢀ3.43, [a]32605 =ꢀ43.7; HPLC (Daicel Chiralcel AD-H,
208C, n-heptane/iPrOH 98:2, flow rate 0.80 mLminꢀ1, l=230 nm): tR =
34.9 min [(S)-28], 43.3 min [(R)-28]; 1H NMR (500 MHz, CDCl3): d=
1.26 (d, J=6.3 Hz, 3H), 2.84 (dd, J=15.0, 8.4 Hz, 1H), 2.89 (dd, J=15.0,
3.3 Hz, 1H), 4.23 (dqd, J=8.4, 6.3, 3.3 Hz, 1H), 5.08 (brs, 1H), 7.12–7.17
(m, 2H), 7.62 (ddd, J=J=7.7, 1.9 Hz, 1H), 8.49 ppm (ddd, J=4.9, 1.8,
0.9 Hz, 1H); 13C NMR (125 MHz, CDCl3): d=32.1, 45.0, 67.2, 121.6,
123.7, 136.8, 148.7, 160.3 ppm; IR (cuvette/CDCl3): n˜ =3343 (s), 3142 (w),
3087 (w), 3018 (w), 2974 (s), 2932 (m), 2903 (m), 1596 (s), 1570 (s), 1476
(s), 1438 (s), 1374 (w), 1331 (w), 1283 (w), 1198 (m), 1150 (m), 1128 (m),
1100 (w), 1078 (m), 1052 (s), 1027 (m), 1001 (m), 937 (s), 930 (s), 840 (s),
795 cmꢀ1 (s); HRMS (ESI): m/z: calcd for C8H11NONa [M+Na+]:
160.0733, found: 160.0729.
(R)-4-Phenyl-1-(pyridin-2-yl)but-3-in-2-ol [(R)-26] (Table 5, entry 3): Ac-
cording to GP 1, starting from rac-26 (89.3 mg, 0.400 mmol, 1.00 equiv),
(SiR)-4a (53.1 mg, 0.260 mmol, 0.650 equiv, 93% ee), CuCl (2.0 mg,
0.020 mmol, 0.050 equiv), L1d (13.9 mg, 0.0400 mmol, 0.100 equiv) and
NaOtBu (1.9 mg, 0.020 mmol, 0.050 equiv) in toluene (4.0 mL), silyl ether
(SiS,S)-31a (87 mg, 51%, d.r. 74:26) and alcohol (R)-26 (38 mg, 31%,
89% ee) were isolated as yellowish oils. Silyl ether (SiS,S)-31a was conta-
minated with 7% the silyl ether of Z-alkene [Z/E 93:7, d.r. (Z isomer)
90:10]; (R)-26 contained 21% of (Z)-alkene (57% ee) as determined by
1H NMR spectroscopy. (R)-26: Rf =0.29 (cyclohexane/tert-butyl methyl
ether 1:2); [a]D20 =ꢀ3.94 (c=0.330, CHCl3), [a]25708 =ꢀ5.15, [a]52406 =ꢀ6.67,
[a]24036 =ꢀ20.6; HPLC (Daicel Chiralcel OD-H, 208C, n-heptane/iPrOH
90:10, flow rate 0.80 mLminꢀ1
, l=230 nm): tR =15.9 min [(R)-26],
(S)-1-Phenyl-2-(6-methylpyridin-2-yl)ethanol [(S)-34] (Table 6, entry 1):
According to GP 1, starting from rac-34 (85.3 mg, 0.400 mmol,
1.00 equiv), (SiS)-4a (49.1 mg, 0.240 mmol, 0.600 equiv, 98% ee), CuCl
(2.0 mg, 0.020 mmol, 0.050 equiv), L1d (13.9 mg, 0.0400 mmol,
0.100 equiv) and NaOtBu (1.9 mg, 0.020 mmol, 0.050 equiv) in toluene
(4.0 mL), silyl ether (SiR,R)-41a (94 mg, 57%, d.r. 77:23) and alcohol (S)-
34 (27 mg, 32%, 89% ee) were isolated as analytically pure materials.
(S)-34: M.p. 378C (tert-butyl methyl ether); Rf =0.26 (cyclohexane/tert-
butyl methyl ether 1:1); [a]2D0 =ꢀ30.9 (c=0.555, CHCl3), [a]25708 =ꢀ32.2,
24.4 min [(S)-26]; 1H NMR (500 MHz, CDCl3): d=3.24 (dd, J=15.1,
7.2 Hz, 1H), 3.30 (dd, J=15.1, 3.8 Hz, 1H), 5.08 (dd, J=7.2, 3.8 Hz, 1H),
5.57 (brs, 1H), 7.16–7.29 (m, 5H), 7.29–7.37 (m, 2H), 7.67 (ddd, J=J=
7.8, 1.9 Hz, 1H), 8.52 ppm (brd, J=4.7 Hz, 1H); 13C NMR (125 MHz,
CDCl3): d=43.6, 62.3, 84.6, 89.7, 122.0, 122.9, 123.9, 128.3, 128.3, 131.8,
137.0, 148.7, 159.0 ppm; IR (cuvette/CDCl3): n˜ =3305 (s), 3084 (m), 3063
(m), 3020 (s), 2977 (s), 2930 (m), 1597 (s), 1571 (s), 1491 (s), 1476 (s),
1438 (s), 1389 (w), 1367 (w), 1325 (w), 1283 (w), 1235 (w), 1200 (m), 1151
Chem. Eur. J. 2008, 14, 11512 – 11528
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11525

M. Oestreich et al.
[a]25046 =ꢀ35.3, [a]42306 =ꢀ44.3; HPLC (Daicel Chiralpak IB, 208C, n-hep-
tane/iPrOH 85:15, flow rate 0.80 mLminꢀ1, l=230 nm): tR =7.6 min [(R)-
34], 8.9 min [(S)-34]; 1H NMR (400 MHz, CDCl3): d=2.54 (s, 3H), 3.04
(dd, J=15.0, 4.1 Hz, 1H), 3.09 (dd, J=15.0, 8.0 Hz, 1H), 5.12 (dd, J=8.0,
4.1 Hz, 1H), 6.32 (brs, 1H), 6.89 (d, J=7.7 Hz, 1H), 7.02 (d, J=7.7 Hz,
1H), 7.24 (dd, J= J=7.3 Hz, 1H), 7.33 (dd, J=J=7.3 Hz, 2H), 7.42 (d,
J= J=7.3 Hz, 2H), 7.48 ppm (dd, J=J=7.7 Hz, 1H); 13C NMR
(100 MHz, CDCl3): d=24.2, 45.3, 73.2, 120.6, 121.2, 125.7, 127.1, 128.2,
137.1, 144.1, 157.2, 159.0 ppm; IR (film): n˜ =3317 (s), 3062 (w), 2923 (w),
1596 (s), 1578 (m), 1457 (s), 1054 cmꢀ1 (s); LRMS (ESI): m/z: 236
[M+Na+]; elemental analysis calcd (%) for C14H15NO (213.3): C 78.84,
H 7.09, N 6.57; found: C 78.88, H 6.96, N 6.41.
(Daicel Chiralcel AD-H, 208C, n-heptane/iPrOH 95:5, flow rate
1.0 mLminꢀ1
, l=230 nm): tR =15.7 min [(S)-37], 19.9 min [(R)-37];
1H NMR (400 MHz, CDCl3): d=2.04 (s, 3H), 2.19 (s, 3H), 3.02 (d, J=
6.5 Hz, 2H), 4.23 (brs, 1H), 5.15 (t, J=6.5 Hz, 1H), 7.25–7.31 (m, 1H),
7.32–7.38 (m, 2H), 7.39–7.43 ppm (m, 2H); 13C NMR (100 MHz, CDCl3):
d=9.9, 11.0, 37.6, 71.4, 125.8, 127.7, 128.5, 130.1, 142.9, 143.1, 159.9 ppm;
LRMS (CI/NH3): m/z: 218 [M+H+]; elemental analysis calcd (%) for
C13H15NO2 (217.3): C 71.87, H 6.96, N 6.45; found: C 71.72, H 6.98, N
6.34.
A
A
flame-dried
Schlenk tube equipped with a magnetic stirring bar was charged with a
mixture of CuCl (5.8 mg, 0.059 mmol, 0.10 equiv) and L1d (40.9 mg,
0.118 mmol, 0.200 equiv) under argon atmosphere. The solids were sus-
pended in degassed, anhydrous toluene (4.0 mL). After the additon of
NaOtBu (5.7 mg, 0.059 mmol, 0.10 equiv), a pale yellow solution formed
within 1 min at room temperature. Subsequently, a solution of silane
(SiR)-4a (120 mg, 0.590 mmol, 1.00 equiv) in toluene (2.0 mL) followed
by dry degassed MeOH (240 mL, 189 mg, 5.90 mmol, 10.0 equiv) were se-
quentially added via syringe. The resulting bright yellow solution showed
gas evolution and was maintained at room temperature for 20 h, after
which GLC analysis revealed approximately 80% conversion. After addi-
tion of tert-butyl methyl ether (10 mL) and a small portion of silica gel,
the solvents were evaporated. Purification by flash column chromatogra-
phy on silica gel (cyclohexane/tert-butyl methyl ether 97:3) afforded silyl
ether (SiS)-52a (95 mg, 70%) along with unreacted silane (SiR)-4a
(13 mg, 11%, 98% ee) as colorless oils. (SiS)-52a: Rf =0.24 (cyclohexane/
tert-butyl methyl ether 90:10); 1H NMR (500 MHz, CDCl3): d=0.95
(ddd, J=15.2, 12.4, 5.6 Hz, 1H), 0.99 (s, 9H), 1.08 (dddd, J=15.2, 5.2,
4.4, 1.6 Hz, 1H), 1.76 (ddddd, J=13.2, 12.4, 11.4, 4.2, 2.6 Hz, 1H), 2.15
(ddddd, J=13.3, J=J=J=5.6, 2.7 Hz, 1H), 2.63 (ddd, J=15.8, 1.3,
2.6 Hz, 1H), 2.80 (dddd, J=15.6, 6.0, J=J=2.0 Hz, 1H), 3.43 (s, 3H),
7.17 (d, J=7.6 Hz, 1H), 7.24 (dd, J=J=7.2 Hz, 1H), 7.32 (ddd, J=J=
7.4, 1.3 Hz, 1H), 7.63 ppm (dd, J=7.2, 1.5 Hz, 1H); 13C NMR (125 MHz,
CDCl3): d=8.8, 18.5, 23.4, 26.1, 35.7, 51.4, 125.2, 128.6, 129.5, 130.9,
134.9, 150.7 ppm; IR (ATR): n˜ =3055 (w), 3001 (w), 2928 (s), 2856 (s),
2830 (m), 1591 (w), 1466 (m), 1434 (m), 1360 (w), 1293 (w), 1266 (w),
1187 (m), 1142 (m), 1077 (s), 1007 (m), 971 (m), 917 (m), 864 (w), 825
(s), 790 (s), 768 (s), 749 (s), 729 (s), 692 (s), 657 (m), 612 cmꢀ1 (s); HRMS
(ESI): m/z: calcd for C14H22OSiNa [M+Na+]: 257.1332, found: 257.1315;
elemental analysis calcd (%) for C14H22OSi (234.4): C 71.73, H 9.46;
found: C 72.11, H 9.67.
Enantiospecific reductive cleavage of (SiS)-52a: A 25 mL round-bottom
flask equipped with a magnetic stirring bar, a reflux condenser and an
argon-inlet was charged with a solution of (SiS)-53a (95 mg, 0.41 mmol,
1.0 equiv) in n-heptane (5 mL). DIBAL-H (2.0 mL, 2.0 mmol, 4.9 equiv,
1.0m in cyclohexane) was added, the reaction mixture was subsequently
heated to reflux and maintained at this temperature for 12 h. The reac-
tion was quenched at ambient temperature by careful addition of water
(10 mL) followed by 2m HCl (5 mL). The mixture was extracted with
tert-butyl methyl ether (4ꢅ10 mL) and the combined organic layers were
washed with brine (10 mL), dried (Na2SO4) and filtered. Removal of vol-
atiles under reduced pressure followed by flash chromatography on silica
gel (cyclohexane) furnished silane (SiR)-4a (59 mg, 71%; 49% over two
steps, 97% ee) as a colorless oil. Analytical data agreed with those previ-
ously reported.[20b] HPLC (Daicel Chiralcel OJ-RH, 128C, MeCN:wa-
ter=75:25, flow rate 0.50 mLminꢀ1, l=230 nm): tR =10.6 min [(SiR)-4a],
tR =11.9 min [(SiS)-4a].
(S)-1-Phenyl-2-(quinolin-2-yl)ethanol [(S)-35] (Table 6, entry 2): Accord-
ing to GP 1, starting from rac-35 (99.7 mg, 0.400 mmol, 1.00 equiv), (SiS)-
4a (50.8 mg, 0.250 mmol, 0.625 equiv, 99% ee), CuCl (2.0 mg,
0.020 mmol, 0.050 equiv), L1d (13.9 mg, 0.0400 mmol, 0.100 equiv) and
NaOtBu (1.9 mg, 0.020 mmol, 0.050 equiv) in toluene (4.0 mL), silyl ether
(SiR,R)-42a (95 mg, 52%, d.r. 81:19) and alcohol (S)-35 (36 mg, 36%,
82% ee) were isolated as analytically pure materials. (S)-35: M.p. 115–
1178C (cyclohexane/tert-butyl methyl ether); Rf =0.46 (cyclohexane/tert-
butyl methyl ether 1:1); [a]2D0 =ꢀ59.0 (c=0.990, CHCl3), [a]25708 =ꢀ61.5,
[a]25046 =ꢀ69.4, [a]42306 =ꢀ109; HPLC (Daicel Chiralcel OD-H, 208C, n-
heptane/iPrOH 90:10, flow rate 0.80 mLminꢀ1, l=230 nm): tR =15.9 min
1
[(R)-35], 20.5 min [(S)-35]; H NMR (400 MHz, CDCl3): d=3.31 (dd, J=
15.6, 4.1 Hz, 1H), 3.34 (dd, J=15.6, 8.1 Hz, 1H), 5.34 (dd, J=8.1, 4.1 Hz,
1H), 6.17 (brs, 1H), 7.23 (d, J=8.4 Hz, 1H), 7.26–7.31 (m, 1H), 7.34–
7.40 (m, 2H), 7.47–7.51 (m, 2H), 7.54 (ddd, J=8.1, 6.9, 1.2 Hz, 1H), 7.73
(ddd, J=8.5, 6.9, 1.5 Hz, 1H), 7.81 (dd, J=8.1, 1.4 Hz, 1H), 8.06–
8.11 ppm (m, 2H); 13C NMR (100 MHz, CDCl3): d=46.2, 73.0, 122.2,
126.0, 126.3, 126.9, 127.4, 127.7, 128.4, 128.8, 129.9, 136.9, 144.1, 147.1,
160.6 ppm; IR (cuvette/CDCl3): n˜ =3308 (m), 1600 (s), 1505 (s), 1426 (s),
1057 cmꢀ1 (s); LRMS (CI/NH3): m/z: 250 [M+H+]; elemental analysis
calcd (%) for C17H15NO (249.3): C 81.90, H 6.06, N 5.62; found: C 81.67,
H 6.26, N 5.70.
(S)-2-(Isoquinolin-1-yl)-1-phenylethanol [(S)-36] (Table 6, entry 3): Ac-
cording to GP 1, starting from rac-36 (99.7 mg, 0.400 mmol, 1.00 equiv),
(SiS)-4a (45.0 mg, 0.220 mmol, 0.550 equiv, 97% ee), CuCl (2.0 mg,
0.020 mmol, 0.050 equiv), L1d (13.9 mg, 0.0400 mmol, 0.100 equiv) and
NaOtBu (1.9 mg, 0.020 mmol, 0.050 equiv) in toluene (4.0 mL), silyl ether
(SiR,R)-43a (83 mg, 46%, d.r. 83:17) and alcohol (S)-36 (47 mg, 47%,
54% ee) were isolated as analytically pure materials. (S)-36: M.p. 1158C
(tert-butyl methyl ether/CH2Cl2); Rf =0.31 (cyclohexane/tert-butyl methyl
ether 1:1); [a]D20 =ꢀ45.9 (c=0.810, CHCl3), [a]25708 =ꢀ48.0, [a]52406 =ꢀ53.8,
[a]24036 =ꢀ86.9; HPLC (Daicel Chiralpak IB, 208C, n-heptane/iPrOH
90:10, flow rate 0.80 mLminꢀ1
, l=230 nm): tR =15.9 min [(R)-36],
19.1 min [(S)-36]; 1H NMR (600 MHz, CDCl3): d=3.57 (dd, J=16.4,
9.6 Hz, 1H), 3.72 (dd, J=16.4, 2.2 Hz, 1H), 5.45 (dd, J=9.9, 2.2 Hz, 1H),
6.19 (brs, 1H), 7.30–7.34 (m, 1H), 7.38–7.44 (m, 2H), 7.54–7.57 (m, 2H),
7.58 (ddd, J=8.2, 6.9, 1.2 Hz, 1H), 7.60 (d, J=6.0 Hz, 1H), 7.70 (ddd, J=
8.2, 6.8, 1.0 Hz, 1H), 7.84 (brd, J=8.2 Hz, 1H), 8.05 (dd, J=8.4, 0.8 Hz,
1H), 8.45 ppm (d, J=5.8 Hz, 1H); 13C NMR (150 MHz, CDCl3): d=41.9,
72.4, 120.0, 124.8, 126.1, 127.3, 127.4, 127.5, 127.6, 128.5, 130.5, 136.2,
140.6, 144.1, 160.0 ppm; IR (cuvette/CDCl3): n˜ =3300 (s), 3028 (s), 2923
(m), 2864 (s), 1603 (w), 1496 (s), 1454 (s), 1410 (m), 1317 (m), 1274 (m),
1209 (w), 1151 (w), 1072 (s), 1040 (s), 1027 (s), 953 (m), 917 (m), 831 (w),
778 (w), 761 (m), 742 (s), 995 cmꢀ1 (s); HRMS (ESI): m/z: calcd for
C17H16NO [M+H+]: 250.1226, found: 250.1216; elemental analysis calcd
(%) for C17H15NO (249.3): C 81.90, H 6.06, N 5.62 ; found: C 81.64, H
5.82, N 5.44.
Stereospecific reductive cleavage of (SiR,R)-7a: A 10 mL Schlenk tube
equipped with a magnetic stirring bar was charged with a solution of
(SiR,R)-7a (88.5 mg, 0.220 mmol, 1.00 equiv, d.r. 86:14, ꢂ96% ee) in
CH2Cl2 (1.5 mL). DIBAL-H (0.44 mL, 0.44 mmol, 2.00 equiv, 1.0m in n-
hexane) was added and the reaction mixture was subsequently main-
tained at ambient temperature for 24 h. The reaction was carefully
quenched with tert-butyl methyl ether (20 mL), water (10 mL) followed
by 2m HCl (5 mL) until pH 7 was reached. After separation of the organ-
ic phase, the aqueous phase was extracted with tert-butyl methyl ether
(4ꢅ10 mL). The combined organic layers were washed with brine
(10 mL), dried (Na2SO4), filtered and, after addition of a small portion of
(S)-2-(4,5-Dimethyloxazol-2-yl)-1-phenylethanol
[(S)-37]
(Table 6,
entry 4): According to GP 1, starting from rac-37 (86.9 mg, 0.400 mmol,
1.00 equiv), (SiS)-4a (46.0 mg, 0.225 mmol, 0.560 equiv, 99% ee), CuCl
(2.0 mg, 0.020 mmol, 0.050 equiv), L1d (13.9 mg, 0.0400 mmol,
0.100 equiv) and NaOtBu (1.9 mg, 0.020 mmol, 0.050 equiv) in toluene
(4.0 mL), silyl ether (SiR,R)-44a (83 mg, 50%, d.r. 79:21) and alcohol (S)-
37 (36 mg, 41%, 65% ee) were isolated as analytically pure materials.
(S)-37: Rf =0.25 (cyclohexane/tert-butyl methyl ether 1:1); [a]2D0 =ꢀ19.4
(c=0.530, CHCl3), [a]25708 =ꢀ20.4, [a]25046 =ꢀ22.9, [a]24306 =ꢀ38.1; HPLC
11526
www.chemeurj.org
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 11512 – 11528

Stereoselective Alcohol Silylation
FULL PAPER
silica gel, volatiles were evaporated under reduced pressure. The crude
product was purified by flash chromatography on silica gel with cyclohex-
ane/tert-butyl methyl ether (95:5!1:1) furnishing (SiS)-4a (44 mg, 98%,
95% ee) as a colorless oil and (R)-6 (34 mg, 78%, 70% ee) as a white
solid.
heim, 2008, pp. 183–207; b) O. Riant, N. Mostefaꢆ, J. Courmarcel,
Synthesis 2004, 2943–2958; c) T. Ohkuma, R. Noyori in Comprehen-
sive Asymmetric Catalysis—Supplement 1 (Eds.: E. N. Jacobsen, A.
Pfaltz, H. Yamamoto), Springer, Berlin, 2004, pp. 55–71; d) H. Nish-
iyama in Comprehensive Asymmetric Catalysis, Vol. 1 (Eds.: E. N.
Jacobsen, A. Pfaltz, H. Yamamoto), Springer, Berlin, 1999, pp. 267–
288.
(S)-8-Hydroxy-8-phenyl-octane-2,6-dione [(S)-53]: To
a 100 mL flask
charged with (S)-34 (0.500 g, 2.34 mmol, 1.00 equiv), THF (5.0 mL), abso-
lute ethanol (0.50 mL) and liquid ammonia (65 mL) was added at ꢀ788C
sodium metal (124 mg, 5.38 mmol, 2.30 equiv). The reaction was main-
tained at this temperature for 1 h. Ammonia was removed under a
stream of argon. To the residue was added THF (5.0 mL) and aqueous
10% H2SO4 (3.5 mL), and the solution was stirred at room temperature
for 15 min. The reaction mixture was diluted with water (5.0 mL) and ex-
tracted with CH2Cl2 (3ꢅ25 mL). The organic phase was dried over
MgSO4 and the solvents were removed under reduced pressure. Purifica-
tion by flash column chromatography on silica gel with cyclohexane/t-
butyl methyl ether (1:1) as eluent gave analytically pure (S)-53 (120 mg,
[3] T. Isobe, K. Fukuda, Y. Araki, T. Ishikawa, Chem. Commun. 2001,
243–244.
[4] For a brief summary, see: S. Rendler, M. Oestreich, Angew. Chem.
2008, 120, 254–257; Angew. Chem. Int. Ed. 2008, 47, 248–250.
[5] a) Cu–H-catalyzed silylation of secondary alcohols: S. Rendler, G.
Auer, M. Oestreich, Angew. Chem. 2005, 117, 7793–7797; Angew.
Chem. Int. Ed. 2005, 44, 7620–7624; b) Cu–H-catalyzed silylation of
tertiary alcohols: B. Karatas, S. Rendler, R. Frçhlich, M. Oestreich,
Org. Biomol. Chem. 2008, 6, 1435–1440; c) Rh–H-catalyzed silyla-
tion of secondary alcohols: H. F. T. Klare, M. Oestreich, Angew.
Chem. 2007, 119, 9496–9499; Angew. Chem. Int. Ed. 2007, 46, 9335–
9338.
[6] a) Catalyst-controlled silylation of meso diols: Y. Zhao, J. Rodrigo,
A. H. Hoveyda, M. L. Snapper, Nature 2006, 443, 67–70; b) catalyst-
controlled kinetic resolution of racemic diols: Y. Zhao, A. W. Mitra,
A. H. Hoveyda, M. L. Snapper, Angew. Chem. 2007, 119, 8623–
8626; Angew. Chem. Int. Ed. 2007, 46, 8471–8474.
[7] a) E. Vedejs, M. Jure, Angew. Chem. 2005, 117, 4040–4069; Angew.
Chem. Int. Ed. 2005, 44, 3974–4001; b) H. B. Kagan, J. C. Fiaud,
Top. Stereochem. 1988, 18, 249–330.
[8] a) T. Rovis in New Frontiers in Asymmetric Catalysis (Eds.: K.
Mikami, M. Lautens), Wiley, New York, 2007, pp. 275–311; b) M. C.
Willis, J. Chem. Soc. Perkin Trans. 1 1999, 1765–1784.
[9] E. J. Jarvo, S. J. Miller in Comprehensive Asymmetric Catalysis—
Supplement 1 (Eds.: E. N. Jacobsen, A. Pfaltz, H. Yamamoto),
Springer, Berlin, 2004, pp. 189–206.
[10] a) J. Y. Corey, Adv. Silicon Chem. 1991, 1, 327–387; b) E. Lukevics,
M. Dzintara, J. Organomet. Chem. 1985, 295, 265–315.
22%, 69% ee) as a yellow oil. (S)-53: Rf =0.18 (cyclohexane/tert-butyl
20
546
methyl ether 1:2); [a]D20 =ꢀ29.4 (c=0.425, CHCl3), [a]25708 =ꢀ31.2, [a]
=
ꢀ35.0, [a]24306 =ꢀ60.0, [a]32605 =ꢀ94.8; HPLC (Daicel Chiralpak AS-H,
208C, n-heptane/iPrOH 90:10, flow rate 0.70 mLminꢀ1, l=230 nm): tR =
54.3 min [(R)-53], 57.4 min [(S)-53]; 1H NMR (400 MHz, CDCl3): d=
1.83 (quin, J=7.1 Hz, 2H), 2.12 (s, 3H), 2.47 (m, 4H), 2.75 (ddd, J=17.0,
8.0, 3.5 Hz, 1H), 2.87 (dd, J=17.0, 9.2 Hz, 1H), 3.62 (brs, 1H), 5.16 (dd,
J=9.2, 3.5 Hz, 1H), 7.27–7.36 ppm (m, 5H); 13C NMR (100 MHz,
CDCl3): d=17.3, 29.8, 42.2, 42.4, 51.1, 69.9, 125.6, 127.6, 128.4, 143.0,
208.4, 210.4 ppm; IR (film): n˜ =3445 (s), 2938 (w), 2893 (w), 1712 (s),
1695 (s), 1494 (w), 750.3 cmꢀ1 (s); HRMS (ESI+): m/z calcd for
C14H18O3Na: 257.1148; found: 257.1138 [M+Na]+.
Crystallographic data: Data set was collected with a Nonius KappaCCD
diffractometer. Programs used: data collection COLLECT (Nonius B.V.,
1998), data reduction Denzo-SMN,[43] absorption correction Denzo,[44]
structure solution SHELXS-97,[45] structure refinement SHELXL-97,[46]
graphics SCHAKAL.[47]
CCDC 693437 [(SiR*,R)-43a] contains the supplementary crystallographic
data for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif
[11] Selected recent examples: a) Pd/C and RhII: C. N. Scott, C. S.
Wilcox, Synthesis 2004, 2273–2276; b) RuI: R. L. Miller, S. V. Mai-
feld, D. Lee, Org. Lett. 2004, 6, 2773–2776; c) CuI: H. Ito, A. Wata-
nabe, M. Sawamura, Org. Lett. 2005, 7, 1869–1871; d) AuI: H. Ito,
K. Takagi, T. Miyahara, M. Sawamura, Org. Lett. 2005, 7, 3001–
3004; e) CuII: Y. Gunji, Y. Yamashita, T. Ikeno, T. Yamada, Chem.
Lett. 2006, 714–715.
[12] Review on the related Cu–H-catalyzed carbonyl reduction: S. Ren-
dler, M. Oestreich, Angew. Chem. 2007, 119, 504–510; Angew.
Chem. Int. Ed. 2007, 46, 498–504.
X-ray crystal structure analysis for (SiR*,R)-43a: Formula C30H33NOSi,
M=451.66, colorless crystal 0.25ꢅ0.25ꢅ0.12 mm, a=16.3282(7), b=
28.8702(12), c=11.0496(4) ꢄ, b=92.553(2)8, V=5203.6(4) ꢄ3, 1calcd
=
1.153 gcmꢀ3, m=0.948 mmꢀ1, empirical absorption correction (0.798 ꢁ T
ꢁ 0.895), Z=8, monoclinic, space group P21/c (No. 14), l=1.54178 ꢄ,
T=223(2) K, w and f scans, 43350 reflections collected (ꢃh,ꢃk,ꢃl),
[(sinq)/l]=0.60 ꢄꢀ1, 9217 independent (Rint =0.066) and 6871 observed
reflections [I ꢄ 2s(I)], 601 refined parameters, R=0.064, wR2 =0.164,
max. (min.) residual electron density 0.36 (ꢀ0.230) eꢄꢀ3, two almost
identical molecules in the asymmetric unit, hydrogen atoms calculated
and refined as riding atoms.
[13] C. Lorenz, U. Schubert, Chem. Ber. 1995, 128, 1267–1269.
[14] A closely related s-bond metathesis has been suggested earlier:
D. H. Appella, Y. Moritani, R. Shintani, E. M. Ferreira, S. L. Buch-
wald, J. Am. Chem. Soc. 1999, 121, 9473–9474.
[15] a) M. Oestreich, Chem. Eur. J. 2006, 12, 30–37; b) M. Oestreich,
Synlett 2007, 1629–1643.
[16] a) M. Oestreich, S. Rendler, Angew. Chem. 2005, 117, 1688–1691;
Angew. Chem. Int. Ed. 2005, 44, 1661–1664; b) S. Rendler, M. Oes-
treich, C. P. Butts, G. C. Lloyd-Jones, J. Am. Chem. Soc. 2007, 129,
502–503; c) S. Rendler, M. Oestreich, Beilstein J. Org. Chem. 2007,
3, 9; d) S. Rendler, R. Frçhlich, M. Keller, M. Oestreich, Eur. J.
Org. Chem. 2008, 2582–2591.
[17] a) L. H. Sommer, Stereochemistry, Mechanism, and Silicon,
McGraw-Hill, New York, 1965; b) L. H. Sommer, Intra-Sci. Chem.
Rep. 1973, 7, 1–44; c) R. J. P. Corriu, C. Guꢇrin, J. J. E. Moreau,
Top. Stereochem. 1984, 15, 43–198.
Acknowledgement
This research was funded by the Deutsche Forschungsgemeinschaft (Oe
249/4-1) and an Emmy Noether-Nachwuchsgruppe to M.O. (Oe 249/2-3
and Oe 249/2-4), as well as the Aventis Foundation (Karl-Winnacker-Sti-
pendium to M.O., 2006–2008) and the Fonds der Chemischen Industrie
(predoctoral fellowship to S.R., 2005–2007). The authors thank Gerd
Fehrenbach (Universitꢀt Freiburg) for performing HPLC analyses.
[18] a) L. H. Sommer, C. L. Frye, J. Am. Chem. Soc. 1959, 81, 1013;
b) L. H. Sommer, C. L. Frye, G. A. Parker, K. W. Michael, J. Am.
Chem. Soc. 1964, 86, 3271–3279.
[19] a) G. Bertrand, J. Dubac, P. Mazerolles, J. Ancelle, Nouv. J. Chim.
1982, 6, 381–386; b) G. L. Larson, E. Torres, J. Organomet. Chem.
1985, 293, 19–27; c) P. Jankowski, E. Schaumann, J. Wicha, A. Zar-
ecki, G. Adiwidjaja, Tetrahedron: Asymmetry 1999, 10, 519–526;
[1] a) P. G. M. Wuts, T. W. Greene, Greeneꢀs Protective Groups in Or-
ganic Synthesis, 4th ed., Wiley, New York, 2007, pp. 165–221; b) P. J.
KocieÇsky, Protecting Groups, 3rd ed., Thieme, Stuttgart, 2004,
pp. 188–230.
[2] For reviews, see: a) S. Rendler, M. Oestreich in Modern Reduction
Methods (Eds.: P. G. Andersson, I. J. Munslow), Wiley-VCH, Wein-
Chem. Eur. J. 2008, 14, 11512 – 11528
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
11527

M. Oestreich et al.
d) P. Jankowski, E. Schaumann, J. Wicha, A. Zarecki, G. Adiwidjaja,
M. Asztemborska, Chem. Commun. 2000, 1029–1030.
[20] a) M. Oestreich, U. K. Schmid, G. Auer, M. Keller, Synthesis 2003,
2725–2739; b) S. Rendler, G. Auer, M. Keller, M. Oestreich, Adv.
Synth. Catal. 2006, 348, 1171–1182.
[32] C. Mazet, S. Roseblade, V. Kçhler, A. Pfaltz, Org. Lett. 2006, 8,
1879–1882.
[33] S. Danishefsky, A. Zimmer, J. Org. Chem. 1976, 41, 4059–4064.
[34] TURBOMOLE (Ver. 5.9); University of Karlsruhe, 2006 (http://
www.turbomole.com).
[21] D. R. Schmidt, S. J. OꢃMalley, J. L. Leighton, J. Am. Chem. Soc.
2003, 125, 1190–1191.
[35] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77,
3865–3868.
[22] It must be noted that this correlation only holds true if both reac-
tions—with racemic as well as enantiopure silane—follow identical
kinetics.
[36] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652; b) P. J. Ste-
phens, F. J. Devlin, C. F. Chabalowski, M. J. Frisch, J. Phys. Chem.
1994, 98, 11623–11627.
[23] Synthesis and resolution of 1-silaindane 5a was conducted analogous
to a previously reported procedure:[20b] see Supporting Information
for details.
[24] a) S. E. Denmark, R. T. Jacobs, G. Dai-Ho, S. Wilson, Organometal-
lics 1990, 9, 3015–3019; b) A. G. Myers, S. E. Kephart, H. Chen, J.
Am. Chem. Soc. 1992, 114, 7922–7923; c) S. E. Denmark, B. D.
Griedel, D. M. Coe, J. Org. Chem. 1993, 58, 988–990; d) K. Matsu-
moto, K. Oshima, K. Utimoto, J. Org. Chem. 1994, 59, 7152–7155;
e) X. Zhang, K. N. Houk, J. L. Leighton, Angew. Chem. 2005, 117,
960–963; Angew. Chem. Int. Ed. 2005, 44, 938–941.
[25] In this reversed scenario for kinetic resolution, racemic silane rac-4a
(1.0 equiv) was subjected to Cu–H-catalyzed dehydrogenative Si–O
coupling using an enantiopure pyridyl alcohol (0.50 equiv).[20b]
[26] J. F. Daeuble, C. McGettigan, J. M. Stryker, Tetrahedron Lett. 1990,
31, 2397–2400.
[37] S. Grimme, J. Chem. Phys. 2003, 118, 9095–9102.
[38] A. Schꢀfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100, 5829–
5835. For TZVPP, additional functions of higher angular momentum
are added. Contradictions: Cu: [6s4p3d1f], P: [5s5p2d1f], O:
[5s3p2d1f], H: [3s2p1d], C: [5s3p2d1f], Si: [5s5p2d1f] (basis set
from ftp.chemie.uni-karlsruhe.de/pub/).
[39] S. Dꢈez-Gonzꢉlez, E. D. Stevens, N. M. Scott, J. L. Petersen, S. P.
Nolan, Chem. Eur. J. 2008, 14, 158–168.
[40] a) H. Brunner, W. Miehling, J. Organomet. Chem. 1984, 275, C17–
C21; b) J. M. Stryker, D. M. Bestrensky in Catalysis of Organic Re-
actions (Ed.: W. E. Pascoe), Marcel Dekker, New York, 1992,
pp. 29–44.
[41] a) L1b: T. Allman, R. G. Goel, Can. J. Chem. 1982, 60, 716–722;
b) L1c, L1e: H. A. Brune, M. Falck, R. Hemmer, G. Schmidtberg,
H. G. Alt, Chem. Ber. 1984, 117, 2791–2802; c) L1d: G. P. Schie-
menz, K. Rçhlk, Chem. Ber. 1971, 104, 1219–1233; d) L1 f: J. Porwi-
siak, M. Schlosser, Chem. Ber. 1996, 129, 233–235.
[42] L. Jafarpour, E. D. Stevens, S. P. Nolan, J. Organomet. Chem. 2000,
606, 49–54.
[43] Z. Otwinowski, W. Minor, Methods Enzymol. 1997, 276, 307–326.
[44] Z. Otwinowski, D. Borek, W. Majewski, W. Minor, Acta Crystallogr.
Sect. A 2003, 59, 228–234.
[27] T. O. Luukas, C. Girard, D. R. Fenwick, H. B. Kagan, J. Am. Chem.
Soc. 1999, 121, 9299–9306.
1
[28] In our hands, the determination of conversion by H NMR spectros-
copy from the crude reaction mixture emerged as a critical point.
The deviation from the given values is about ꢃ3% and might ac-
count for occasional inconsistencies in Tables 4–7.
[29] Reductive cleavage of silyl ethers proceeds under retention of con-
figuration at silicon: M. Oestreich, G. Auer, M. Keller, Eur. J. Org.
Chem. 2005, 184–195.
[30] C. Yu, O. Meth-Cohn, Tetrahedron Lett. 1999, 40, 6665–6668.
[31] Selected examples for enzymatic kinetic resolutions of pyridyl meth-
anols: a) By oxidation: M. Takemoto, K. Achiwa, Tetrahedron:
Asymmetry 1995, 6, 2925–2928; b) by acylation: J. Uenishi, T. Hir-
aoka, S. Hata, K. Nishikawi, O. Yonemitsu, K. Nakamura, H. Tsu-
kube, J. Org. Chem. 1998, 63, 2481–2487.
[45] G. M. Sheldrick, Acta Crystallogr. Sect. A 1990, 46, 467–473.
[46] G. M. Sheldrick, Acta Crystallogr. Sect. A 2008, 64, 112–122.
[47] E. Keller, SCHAKAL, University of Freiburg (Germany), 1997.
Received: July 7, 2008
Published online: November 19, 2008
11528
www.chemeurj.org
ꢂ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2008, 14, 11512 – 11528