Titanocene-catalyzed pinacol couplings: Reagent-controlled transition-metal-catalyzed radical reactions

Article abstract of DOI:10.1002/(SICI)1099-0690(199809)1998:9<1923::AID-EJOC1923>3.0.CO;2-B

The titanocene-catalyzed pinacol coupling of aromatic aldehydes proceeding in good yields and with high diastereoselectivity under reagent control is described.

Full text of DOI:10.1002/(SICI)1099-0690(199809)1998:9<1923::AID-EJOC1923>3.0.CO;2-B

FULL PAPER
Titanocene-Catalyzed Pinacol Couplings: Reagent-Controlled
Transition-Metal-Catalyzed Radical Reactions
Andreas Gansäuer*, Monica Moschioni, and Daniel Bauer
Institut für Organische Chemie der Georg-August-Universität,
Tammannstraße 2, D-37077 Göttingen, Germany
Fax: (internat.) ϩ 49(0)551/393244
E-mail: agansae@gwdg.de
Received April 3, 1998
Keywords: Aldehydes / Diastereoselectivity / 1,2-Diols / Radical reactions / Reagent control
The titanocene-catalyzed pinacol coupling of aromatic aldehydes proceeding in good yields and with high diastereo-
selectivity under reagent control is described.
Introduction
selectivity. Clearly, a reagent-controlled catalytic pinacol
coupling would circumvent these problems.
The reductive coupling of carbonyl compounds, the pina-
col coupling, is the most direct and efficient way to synthe-
size 1,2-diols by formation of the functionalized car-
bonϪcarbon bond.[1] The pinacol coupling is not only a
shortcut of the alternative two-step sequence, i.e. McMurry
coupling[2] and syn,vic-dihydroxylation.[3] More impor-
tantly, functional groups being sensitive to either or both of
these two reactions can be tolerated if the pinacolization
reagent is chosen carefully. Of the many stoichiometric re-
agents known to induce intermolecular dimerizations of al-
dehydes and ketones only few allow for highly dia-
stereoselective coupling.[1] An enantioselective coupling has
remained elusive so far. The reason for this is to be sought
in the mechanism of the transformation. The reaction is
generally thought to proceed via intermediate ketyl radicals
formed by single electron transfer (SET) from low-valent
transition-metal complexes, metal powders, or electrodes.
The ketyl radicals can either dimerize,[4a][4b][4c] or add to a
second equivalent of carbonyl compound with formation of
an oxo radical that is subsequently reduced.[4d][4e][4f] Most
of the SET reagents do, however, not allow for binding of
both carbonyl partners in a stereodefined manner, thus not
enabling effcient control of diastereoselectivity or enantio-
selectivity.
An important recent development in this area has been
the emergence of catalytic reactions from stoichiometric
processes. Metal oxides or alkoxides, the products of ad-
ditions to carbonyl groups are converted to metal chlorides
by silylation with Me3SiCl. The low-valent redox-active
metal compound is then regenerated by in situ reduction
with a readily available low-cost stoichiometric reductant,
usually a metal powder. In this way the former stoichio-
metric reagent is essentially replaced by the readily available
and cheap Me3SiCl. The first succesful realization of this
concept was a McMurry reaction catalytic in titanium.[6]
Other examples are Nozaki-Hiyama couplings catalytic in
chromium,[7] catalytic pinacol couplings,[8] and reactions
catalytic in samarium diiodide.[9] In this full paper we wish
to disclose our own results on the titanocene-catalyzed pin-
acol coupling. Our reaction shown in Figure 1 represents
the so far only example of a catalytic pinacol coupling pro-
ceeding with high diastereoselectivity (> 95:5).
Figure 1. Titanocene-catalyzed pinacol coupling with Me3SiCl
Among the rare examples of highly diastereoselective
stoichiometric reagents are titanium(III) complexes, e.g. ti-
tanium trichloride and titanocene chloride.[5] This can be
attributed to the dinuclear nature of these reagents and al-
lows for binding of both carbonyl compounds and thus
both formed ketyl radicals in the desired stereodefined
Results and Discussion
Titanocene Dichloride as SET Catalyst
Based on the ability of cyclopentadienyl-bonded ti-
manner. The repulsive steric interactions of the residues R tanium(III) complexes[5b] to induce stoichiometric pinacol
of both aldehydes results in high diastereoselectivity in the couplings we anticipated the catalytic cycle depicted in Fig-
formation of the 1,2-diols. However, the disadvantage of ure 1 to work.
this otherwise efficient reaction is the inevitable use of stoi-
There are, however, three points of concern that have to
chiometric amounts of the titanium complex. This is es- be carefully addressed: 1) reduction of titanocene dichloride
pecially disadvantageous when more elaborate complexes has to be fast and quantitative to allow for efficient catalytic
are to be employed for achieving high enantio- or diastereo- turn-over, 2) Me3SiCl-induced pinacolization has to be sup-
Eur. J. Org. Chem. 1998, 1923Ϫ1927
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 1998
1434Ϫ193X/98/0909Ϫ1923 $ 17.50ϩ.50/0
1923

A. Gansäuer, M. Moschioni, D. Bauer
FULL PAPER
Figure 2. Catalytic cycle for titanocene-catalyzed pinacol coupling
Mg turnings. Homogeneous solutions of the salt could only
be obtained when about 10% of benzene was added to the
mixture.[13] Otherwise phase separation was observed. In
this manner 2.5Ϫ3.0  solutions of magnesium bromide in
diethyl ether could be readily obtained. A mechanistically
interesting feature of the reaction is that lowering the tem-
perature led to a dramatically reduced diastereoselectivity
of the reaction and pronouncedly longer reaction times. A
reasonable explanation is that one step in the catalytic cycle
was slow and that a less selective pathway was operating.
Since both reduction of titanocene dichloride and the coup-
ling are fast at and below room temperature it seems that
silylation of the titanocene alkoxide to regenerate titano-
cene dichloride was slow.[7] Low diastereoselectivity can
then be readily explained by uncatalyzed Me3SiCl initiated
coupling. At higher temperatures diastereoselectivity also
decreased. Thus, 25° C seems to be the ideal temperature
for conducting our titanocene-catalyzed pinacol coupling
because all kinetic requirements for efficient catalysis are
met. Clearly, this is preparatively convenient since no heat-
ing or cooling is required. It should be noted that only Me3-
SiCl and the even more reactive silylating reagent Me2SiCl2
gave catalytic turn-over. Me2tBuSiCl and less reactive re-
agents failed to yield substantial amounts of products. In
other less polar solvent systems, e.g. Et2O, CH2Cl2, or mix-
tures of the two, the catalyst and the zinc chloride formed
in situ were insoluble. DMF or DMSO as solvents or cosol-
vents resulted in formation of substantial amounts of
benzyl alcohol as by-product.
pressed,[10] and 3) the right metal ion for the trimeric com-
plex has to be utilized to achieve high diastereoselec-
tivity.[11] We started our investigations with Zn powder, be-
ing known to efficiently reduce titanocene dichloride to the
corresponding titanium(III) derivative over a wide range of
temperatures as stoichiometric reductant[12] and Me2SiCl2
as highly active scavenger for the titanocene alkoxide in the
presence of 10 mol-% of titanocene dichloride. Other met-
als, e.g. Mg, Mn, Al, turned out to be vastly inferior due
to slow reduction and concomittant formation of benzyl
alcohol. Further optimization studies are summarized in
Table 1.
Table 1. Coupling of benzaldehyde (0.1  in THF) in the presence
of Me2SiCl2 or Me3SiCl
Entry
Amount of
t
T
[°C]
Yield
[%]
1/2[a]
catalyst [%] [h]
1a
1b
1c
1d
1e
10
10
3
1.2
22
0
93
92
96
88
90
86:14
Ϫ40
25
70:30
Under the optimized conditions a variety of symmetrical
diols were synthesized in good yields and with high dia-
stereoselectivities. The results of these investigations are
summarized in Table 2. Functional groups interfering with
McMurry couplings or dihydroxylation reactions such as
chlorides, double bonds, phenolate esters and α,β-unsatu-
rated esters are readily tolerated. Although initial experi-
0.25
84:16[b]
89:11[b][c]
95:5[b][d]
3
3
3
25
25
3
[a]
1
[b]
Determined by H NMR of the crude mixture. Ϫ
1.5 equiv.
of Me3SiCl. Ϫ [c] Slow addition of PhCHO and Me3SiCl. Ϫ [d] Slow
addition in the presence of MgBr2.
Benzaldehyde coupling in the presence of Me2SiCl2 and ments were performed on a small scale, the reaction can be
10 mol-% of titanocene dichloride gave complete conver- readily scaled up and gives essentially the same results for
sion to products in less than 10 min. Me3SiCl as additive 0.5, 5, and 50 mmol of substrate. In the latter case 374 mg
also leads to clean formation of products. Me3SiCl and Zn of the catalyst were used. This compares favourably with
react to give the pinacols over 20 h ableit with low dia- the use of 24.9 g of titanocene dichloride in the stoichio-
stereoselectivity.[10] The catalytic acceleration of the reac- metric reaction.[5b]
tion was thus significant as was the increase in diastereose-
lectivity. However, the upper limit for the formation of 1
and 2 in the stoichiometric reaction with zinc is 92:8[5b] in-
dicating that some of the uncatalyzed pathway was still op-
erating. To achieve useful levels of selectivity that are com-
parable to the stoichiometric parent system the following
variations were investigated. To reduce interference of the
noncatalyzed pathway to less significant levels both the al-
dehyde and Me3SiCl were added slowly to a mixture of tit-
anocene dichloride and Zn in THF. Indeed, selectivity in-
creased and the yield remained high. Further improvement
could be achieved by adding 1 equivalent of MgBr2 to the
benzaldehyde solution. This gave a tighter dimeric titanium
complex by replacing Zn with Mg. A similar effect is known
from stoichiometric reactions with titanocene(III) deriva-
Table 2. Pinacol coupling of substituted aldehydes with 3 mol-% of
Cp2TiCl2 under optimized conditions
Entry
Aryl
t [h]
Yield [%]
1/2[a]
2a
2b
2c
2d
2e
2f
4-CH3-C6H4
4-Cl-C6H4
2.5
3
82
78
87
91
83
79
88
92:8
93:7
93:7[b]
92:8
92:8
92:8
92:8
3-Cl-C6H4
3
4-Vinyl-C6H4
4-AcO-C6H4
4-Crotonyloxy-C6H4
2-Furyl
2
3
2
2g
1.5
[a]
1
Determined by H NMR of the crude mixture after hydrolysis.
[b]
Ϫ
Reaction also performed on a 50-mmol scale.
Mechanistically, it seems reasonable to assume that the
tives.[5b] MgBr2 was prepared from 1,2-dibromoethane with catalytically active species is a dimeric titanium complex
1924
Eur. J. Org. Chem. 1998, 1923Ϫ1927

Titanocene-Catalyzed Pinacol Couplings
FULL PAPER
binding both ketyl radicals as indicated in Figure 3. The for intermolecular pinacol couplings and is superior to all
observed syn selectivity can then be readily explained by reagents known so far. The reaction represents a rare ex-
minimization of steric interference through anti orientation ample of a transition-metal-catalyzed radical reaction pro-
of the R groups in the complex.
ceeding with reagent control.[8c][18d][16] From these studies it
has become clear that the reagent combination Zn/Me3SiCl,
when properly used, is a valuable stoichiometric reductive
system for conducting titanocene-catalyzed redox reactions.
Figure 3. Possible catalytically active species binding both ketyl
radicals
We thank the Fonds der Chemischen Industrie, the Erasmus Pro-
gram, and the Deutsche Forschungsgemeinschaft for financial sup-
port. Prof. R. BrücknerЈs constant encouragement and generous
support is gratefully acknowledged.
Experimental Section
rac-Ethylenebis(η5-tetrahydroindenyl)titanium Dichloride
(EBTHITiCl2) as SET Catalyst
General: All reactions were performed in oven-dried (100°C)
glassware under N2. THF was freshly distilled from LiAlH4. Me2Si-
Cl2 and Me3SiCl were distilled under N2 immediately before use
from CaH2. Products were purified by flash chromatography[17] on
Merck silica gel 50 (eluents given in brackets). Yields refer to ana-
lytically pure samples. Isomer ratios were determined from suitable
1H-NMR integrals of cleanly separated signals. Ϫ 1H NMR: [tetra-
methylsilane (δ ϭ 0.00) in the indicated solvent, CHCl3 (δ ϭ 7.26)
as internal standard in the same solvent] and 13C NMR [tetrameth-
ylsilane (δ ϭ 0.00) in the indicated solvent or CDCl3 (δ ϭ 77.00)
and [D6]DMSO (δ ϭ 39.52) as internal standards in the same sol-
vent]; Bruker AMX 300 and Varian; integrals in accord with as-
signments, coupling constants are measured in Hz. Ϫ Combustion
analyses: F. Hambloch, Institute of Organic Chemistry, University
of Göttingen. Ϫ Mass spectra: G. Remberg, Institute of Organic
Chemistry, University of Göttingen. Ϫ IR spectra: Perkin Elmer
1600 series FTIR as KBr pellets.
Although diastereoselectivity was high with titanocene
dichloride from a synthetic point of view an improvement
is still desirable. Also, it is worth knowing if structurally
more demanding titanocene complexes can act as catalysts
to induce enantioselectivity in future investigations. We thus
decided to investigate the readily available EBTHITiCl2 as
catalyst shown in Figure 4.[14][15] Table 3 shows a summary
of reactions performed under the optimized reaction con-
ditions.
Figure 4. rac-EBTHITiCl2 catalyst
Table 1, Entry 1e: To titanocene dichloride (22.4 mg, 90 µmol)
in THF (30 ml) was added Zn (196 mg, 3.0 mmol) and the mixture
was stirred for 5 min. Benzaldehyde (300 µl, 3.0 mmol), MgBr2 [1.8
ml of a 2.2  solution in Et2O containing 10% benzene (v/v), 4.0
mmol], and Me3SiCl (570 µl, 4.5 mmol) were added in THF (15
ml) over 1.5 h and the mixture was stirred for 2 h. After addition
of 2  HCl (10 ml), stirring was continued for 15 min and the
mixture extracted with tBuOMe (2ϫ 50 ml). After washing with
H2O (20 ml), sat. NaHCO3 (40 ml), H2O (20 ml), and drying
(MgSO4), the crude product was purified by chromatography on
silica gel (50% tBuOMe, 50% petroleum ether) to give 290 mg of
1,2-diphenyl-1,2-ethanediol as a 95:5 mixture of 1/2 (90%).
Table 3. Pinacol coupling of substitued aldehydes with 3 mol-% of
rac-EBTHITiCl2 under optimized conditions
Entry
Aryl
Yield [%]
1/2[a]
3a
3b
3c
3d
3e
3f
C6H5
4-C6H5-C6H4
2-thienyl
88
79
79
82
83
87
78
81
97:3
98:2
98:2
96:4
96:4
96:4
97:3
97:3
4-Br-C6H4
4-Cl-C6H4
3-Cl-C6H4
4-Vinyl-C6H4
4-Crotonyloxy-C6H4
3g
3h
General Procedure for the Coupling in the Presence of Titanocene
Dichloride: To a solution of titanocene dichloride (22.4 mg, 90
µmol) in dry THF (15 ml) under N2 was added Zn (196 mg, 3.0
mmol) and the mixture was stirred until the red colour had faded
and turned into lime green (typically 2Ϫ5 min). Then the aldehyde
(3.0 mmol), Me3SiCl (570 µl, 4.5 mmol) and MgBr2 [1.8 ml of a
2.2  solution in Et2O containing 10% benzene (v/v), 4.0 mmol] in
THF (15 ml) were added dropwise over 2 h. Stirring was continued
for the indicated time and 2  HCl (10 ml) was added). After 0.5
h, the mixture was extracted with tBuOMe (2 ϫ 50 ml), washed
with H2O (20 ml), sat. NaHCO3 (40 ml), H2O (20 ml), and dried
(MgSO4). After evaporation of the solvent, the product was puri-
fied by chromatography on silica gel.
[a]
1
Determined by H NMR of the crude mixture after hydrolysis.
Adding the reagents all at once resulted in inferior dia-
stereoselectivity (1/2 ϭ 70:30) compared to the parent sys-
tem titanocene dichloride. This indicates that silylation is
slower using the sterically more demanding EBTHITiCl2.
However, once this observation was accounted for by care-
ful slow addition, diastereoselectivity was significantly
higher compared to titanocene dichloride. This demon-
strates clearly that even though silylation is slowed down by
steric congestion around titanium, the uncatalyzed pathway
is not interfering to an important extent when proper reac-
tion conditions are used. Here the optimal temperature for
conducting the reaction is also 25° C. Thus, EBTHITiCl2
(R*,R*)-1,2-Bis(4-methylphenyl)-1,2-ethanediol (Table 2, entry
2a)[18]: After stirring for 2.5 h and silica gel chromatography (25%
tBuOMe Ǟ 50% tBuOMe, 75% petroleum ether Ǟ 50% petroleum
ether), 298 mg of 1,2-bis(4-methylphenyl)-1,2-ethanediol was ob-
constitutes a highly effective and diastereoselective catalyst tained as a 92:8 mixture of 1/2 (82%).
Eur. J. Org. Chem. 1998, 1923Ϫ1927
1925

A. Gansäuer, M. Moschioni, D. Bauer
FULL PAPER
(R*,R*)-1,2-Bis(4-chlorophenyl)-1,2-ethanediol (Table 2, entry
at δ ϭ 6.98, AAЈ of AAЈXXЈ centred at δ ϭ 7.15. Ϫ 13C NMR
2b)[5a]:After stirring for 3 h and silica gel chromatography (40% ([D6] DMSO) δ ϭ 20.74 (CH3), 76.64 (CHOH), 120.44 (Ar-3-C),
tBuOMe Ǟ 60% tBuOMe, 60% petroleum ether Ǟ 40% petroleum 127.95 (Ar-2-C), 139.77 (Ar-1-C), 149.12(Ar-4-C), 168.99 (CO). Ϫ
ether), 331 mg of 1,2-bis(4-chlorophenyl)-1,2-ethanediol was ob-
tained as a 93:7 mixture of 1/2 (78%).
C18H18O6 (330.3) calcd. C 65.45, H 5.49; found C 65.40, H 5.75.
For the minor diastereoisomer the following 1H-NMR data are
available from the spectrum of the mixture;[5b] other signals are
overlapping. Ϫ 1H NMR (CDCl3): δ ϭ 2.29 (s, CH3), 4.72 (s,
CHOH).
(R*,R*)-1,2-Bis(3-chlorophenyl)-1,2-ethanediol (Table 2, entry
2c): After stirring for 3 h and silica gel chromatography (30% tBu-
OMe Ǟ 50% tBuOMe, 70% petroleum ether Ǟ 60% petroleum
ether), 370 mg of 1,2-bis(3-chlorophenyl)-1,2-ethanediol was ob-
tained as a 93:7 mixture of 1/2 (87%). Colorless crystals, m.p. 80°C.
(R*,R*)-1,2-Bis(4-crotonyloxyphenyl)-1,2-ethanediol (Table 2,
entry 2f): After stirring for 2 h and silica gel chromatography (25%
tBuOMe Ǟ 50% tBuOMe, 75% petroleum ether Ǟ 50% petroleum
ether), 453 mg of 1,2-bis(4-crotonyloxyphenyl)-1,2-ethanediol was
obtained as a 92:8 mixture of 1/2 (79%). Colorless crystals m.p.
118°C. Ϫ IR (KBr): ν˜ 3465, 1730, 1655, 1505, 1205, 910 cmϪ1. Ϫ
1H NMR (CDCl3): δ ϭ 1.96 (dd, J ϭ 7.6, 1.5, CH3), 2.78 (br. s,
OH), 4.67 (s, CHOH), 6.51 (dq, J ϭ 15.4, 1.6 Hz, COCHCH, 2
H), BBЈ of AAЈBBЈ centred at δ ϭ 7.00 (4 H), 7.11Ϫ7.21 (m, 6 H).
1
Ϫ IR (KBr): ν˜ ϭ 3380, 1575, 1475, 1200, 1080, 1000 cmϪ1. Ϫ H
NMR (CDCl3): δ ϭ 2.82 (br. s, OH), 4.66 (s, CHOH), 6.92 (d, J ϭ
7.6 Hz, 6-H, 2 H), 7.17 (dd, J ϭ 7.5, 7.5 Hz, 5-H, 2 H), 7.24Ϫ7.21
(m, 2-H, 4-H, 4 H). Ϫ 13C NMR (CDCl3): δ ϭ 78.23 (CHOH),
125.15 (ArϪC-6), 126.89 (Ar-C-2), 128.26 (Ar-C-4), 129.39 (Ar-C-
5), 134.19 (Ar-C-3), 141.49 (Ar-C-1). Ϫ C14H12Cl2O2 (283.2): calcd.
C 59.39, H 4.27; found C 59.31, H 4.43. Ϫ For the minor dia-
stereoisomer the following 1H-NMR data are available from the
spectrum of the mixture;[5b] other signals are overlapping. Ϫ 1H
NMR (CDCl3): δ ϭ 2.60 (br. s, OH), 4.82 (s, CHOH).
Ϫ
13C NMR (CDCl3) δ ϭ 18.21 (CH3), 78.34 (CHOH), 121.22
(Ar-C-2), 121. 89 (CHCHCO), 127. 92 (Ar-C-3), 137.25 (Ar-C-1),
147.07 (Ar-C-4), 150.17 (CHCHCO), 164.76 (CO). Ϫ C22H22O6
calcd. 382.1416; found 382.1416. Ϫ For the minor diastereoisomer
the following 1H-NMR data are available from the spectrum of the
mixture;[5b] other signals are overlapping. Ϫ 1H NMR (CDCl3):
δ ϭ 4.81 (s, CHOH).
Reaction on 50-mmol Scale: To a solution of titanocene dichlor-
ide (374 mg, 1.5 mmol) in dry THF (250 ml) under N2 was added
Zn (3.272 g, 50 mmol) and the mixture was stirred until the red
colour had faded and turned into lime green. Then the 3-chloro-
benzaldehyde (5.63 ml, 50 mmol), Me3SiCl (9.50 ml, 75 mmol), and
MgBr2 [30.3 ml of a 2.2  solution in Et2O containing 10% benzene
(v/v), 67 mmol] were added dropwise in THF (250 ml) over 4 h.
Stirring was continued for 4 h and 2  HCl (100 ml) was added.
After 0.5 h, the mixture was extracted with tBuOMe (2 ϫ 500 ml),
washed with H2O (500 ml), sat. NaHCO3 (100 ml), H2O (500 ml),
and dried (MgSO4). After evaporation of the solvent, the product
was purified by chromatography on silica gel (30% tBuOMe Ǟ
50% tBuOMe, 70% petroleum ether Ǟ 60% petroleum ether) to
give 5.80 g of 1,2-bis(3-chlorophenyl)-1,2-ethanediol as a 92:8 mix-
ture of 1/2 (82%).
(R*,R*)-1,2-Bis(2-furyl)-1,2-ethanediol (Table 2, entry 2g)[18]
:
After stirring for 1.5 h and silica gel chromatography (25% tBu-
OMe Ǟ 50% tBuOMe, 75% petroleum ether Ǟ 50% petroleum
ether), 256 mg of 1,2-bis(2-furyl)-1,2-ethanediol was obtained as a
92:8 mixture of 1/2 (88%).
General Procedure for the Coupling in the Presence of rac-EBTHI-
TiCl2: To a solution of EBTHITiCl2 (34.2 mg, 90 µmol) in dry
THF (15 ml) under N2 was added Zn (196 mg, 3.0 mmol) and the
mixture was stirred until the red colour had faded and turned into
lime green. Then the aldehyde (3.0 mmol), Me3SiCl (570 ml, 4.5
mmol), and MgBr2 [1.8 ml of a 2.2  solution in Et2O containing
10% benzene (v/v), 4.0 mmol] were added dropwise over 2 h. Stir-
ring was continued for 3 h and 2  HCl (5 ml) was added. After
0.5 h, the mixture was extracted with tBuOMe (2 ϫ 20 ml), washed
with H2O (20 ml), sat. NaHCO3 (20 ml), H2O (20 ml), and dried
(MgSO4). After evaporation of the solvent, the product was puri-
fied by chromatography on silica gel.
(R*,R*)-1,2-Bis(4-vinylphenyl)-1,2-ethanediol (Table 2, entry
2d): After stirring for 2 h and silica gel chromatography (25% tBu-
OMe Ǟ 40% tBuOMe, 75% petroleum ether Ǟ 60% petroleum
ether), 363 mg of 1,2-bis(4-vinylphenyl)-1,2-ethanediol was ob-
tained as a 92:8 mixture of 1/2 (91%). Colorless crystals, m.p.
109°C. Ϫ IR (KBr): ν˜ ϭ 3380, 2350, 1630, 1405, 1035, 910 cmϪ1
.
1
Ϫ H NMR (CDCl3): δ ϭ 2.78 (br s, OH), 4.52 (s, CHOH), 5.08
(dd, J ϭ 10.6, 1.1 Hz, CHZ2CH, 2 H), 5.56 (dd, J ϭ 17.4, 1.1,
CHE2CH, 2 H), 6.51 (dd, J ϭ 17.4, 10.6 Hz, CH2CH, 2 H), XXЈ
of AAЈXXЈ centred at 6.92, AAЈ of AAЈXXЈ centred at 7.13. Ϫ
13C NMR (CDCl3): δ ϭ 78.66 (CHOH), 113.80 (CHCH2), 125.88,
127. 13, 136.39, 136.99, 139.90 (Ar-1-C). Ϫ C18H18O2 (266.2) calcd.
C 81.17, H 6.81; found C 81.08, H 6.82. Ϫ For the minor dia-
stereoisomer the following 1H-NMR data are available from the
spectrum of the mixture;[5b] other signals are overlapping. Ϫ 1H
NMR (CDCl3): δ ϭ 2.10 (br. s, OH), 4.68 (s, CHOH), 5.09 (dd,
J ϭ 10.2, 1.1 Hz, CHZ2CH, 2 H), 5.59 (dd, J ϭ 17.0, 1.1 Hz,
CHE2CH, 2 H), 6.56 (dd, J ϭ 17.0, 10.2 Hz, CH2CH, 2 H), AAЈ
of AAЈXXЈ centred at δ ϭ 7.20.
(R*,R*)-1,2-Diphenyl-1,2-ethanediol (Table 3, entry 3a): After
chromatography on silica gel (50% tBuOMe, 50% petroleum ether),
283 mg of 1,2-diphenyl-1,2-ethanediol was obtained as a 97:3 mix-
ture of 1/2 (88%).
(R*,R*)-1,2-Bis(4Ј-biphenyl)-1,2-ethanediol (Table 3, entry
3b)[19]: After chromatography on silica gel (40% tBuOMe Ǟ 70%
tBuOMe, 60% petroleum ether Ǟ 30% petroleum ether), 434 mg
of 1,2-bis(4Ј-biphenyl)-1,2-ethanediol was obtained as a 98:2 mix-
ture of 1/2 (79%).
(R*,R*)-1,2-Bis(2-thienyl)-1,2-ethanediol (Table 3, entry 3c)[18]
:
After silica gel chromatography (40% tBuOMe Ǟ 60% tBuOMe,
60% petroleum ether Ǟ 40% petroleum ether), 268 mg of 1,2-bis(2-
thienyl)-1,2-ethanediol was obtained as a 98:2 mixture of 1/2
(79%).
(R*,R*)-1,2-Bis(4-acetoxyphenyl)-1,2-ethanediol (Table 2, entry
2e): After stirring for 3 h and silica gel chromatography (100 ml of
50% tBuOMe, 50% petroleum ether; 100 ml of 90% tBuOMe, 10%
EtOH, 100 ml of EtOH), 411 mg of 1,2-bis(4-acetoxyphenyl)-1,2-
(R*,R*)-1,2-Bis(4-bromophenyl)-1,2-ethanediol (Table 3, entry
ethanediol was obtained as a 92:8 mixture of 1/2 (83%). Colorless 3d)[20]: After silica gel chromatography (50% tBuOMe Ǟ 70% tBu-
crystals m.p. 190Ϫ192°C. Ϫ IR (KBr): ν˜ 3145, 1750, 1605, 1510,
OMe, 50% petroleum ether Ǟ 30% petroleum ether), 458 mg of
1400, 1370, 1215, 1060 cmϪ1. Ϫ 1H NMR (CDCl3): δ ϭ 2.28 (s, 1,2-bis(4-bromophenyl)-1,2-ethanediol was obtained as a 96:4 mix-
CH3), 2.82 (br. s, OH), 4.70 (s, CHOH), XXЈ of AAЈXXЈ centred ture of 1/2 (82%).
1926
Eur. J. Org. Chem. 1998, 1923Ϫ1927

Titanocene-Catalyzed Pinacol Couplings
FULL PAPER
[4e]
Am. Chem. Soc. 1995, 117, 7283. Ϫ
T. Naito, K. Tajiri, T.
(R*,R*)-1,2-Bis(4-chlorophenyl)-1,2-ethanediol (Table 3, entry
3e): After silica gel chromatography (50% tBuOMe, 50% tBuOMe),
352 mg of 1,2-bis(4-chlorophenyl)-1,2-ethanediol was obtained as
a 96:4 mixture of 1/2 (83%).
Harimoto, I. Ninomiya, Tetrahedron Lett. 1994, 35, 2205. Ϫ [4f]
T. Kiguchi, K. Tajiri, I. Ninomiya, T. Naito, H. Hiramatsu,
Tetrahedron Lett. 1995, 36, 253.
[5] [5a]
H. G. Raubenheimer, D. Seebach, Chimia 1986, 40, 12. Ϫ
Y. Handa, J. Inanaga, Tetrahedron Lett. 1987, 28, 5717. Ϫ
[5b]
[5c]
(R*,R*)-1,2-Bis(3-chlorophenyl)-1,2-ethanediol (Table 3, entry
3f): After silica gel chromatography (30% tBuOMe Ǟ 50% tBu-
OMe, 70% petroleum ether Ǟ 50% petroleum ether), 370 mg of
1,2-bis(3-chlorophenyl)-1,2-ethanediol was obtained as a 96:4 mix-
ture of 1/2 (87%).
M. C. Barden, J. Schwartz, J. Am. Chem. Soc. 1996, 118,
[5d]
5484. Ϫ
A. Clerici, L. Clerici, O.Porta, Tetrahedron Lett.
1996, 37, 3035.
[6]
A. Fürstner, A. Hupperts, J. Am. Chem. Soc. 1995, 117, 4468.
[7] [7a]
[7b]
A. Fürstner, N. Shi, J. Am. Chem. Soc. 1996, 118, 2533. Ϫ
A. Fürstner, N. Shi, J. Am. Chem. Soc. 1996, 118, 12349.
[8a] T. Hirao, T. Hasegawa, Y. Muguruma, I. Ikeda, J. Org. Chem
[8]
[8b]
(R*,R*)-1,2-Bis(4-vinylphenyl)-1,2-ethandiol (Table 3, entry 3g):
After silica gel chromatography (30% tBuOMe Ǟ 50% tBuOMe,
70% petroleum ether Ǟ 50% petroleum ether), 312 mg of 1,2-bis(4-
vinylphenyl)-1,2-ethanediol was obtained as a 97:3 mixture of 1/
2 (78%).
1996, 61, 366. Ϫ
R. Nomura, T. Matsuno, T. Endo, J. Am.
Chem. Soc. 1996, 118, 11666. Ϫ [8c] A. Gansäuer, J. Chem. Soc.,
[8d]
Chem. Commun. 1997, 457. Ϫ
A. Gansäuer, Synlett 1997,
[8e]
363. Ϫ
T. A. Lipski, M. A. Hilfiker, S. G. Nelson, J. Org.
Chem. 1997, 62, 4566.
[9]
E. J. Corey, G. Z. Zheng, Tetrahedron Lett. 1997, 38, 1045.
J. H. So, M.-K. Park, P. Boudjouk, J. Org. Chem. 1988, 53,
5871.
[10]
(R*,R*)-1,2-Bis(4-crotonyloxyphenyl)-1,2-ethanediol (Table 3,
entry 3 h): After silica gel chromatography (30% tBuOMe Ǟ 50%
tBuOMe, 70% petroleum ether Ǟ 50% petroleum ether), 464 mg
of 1,2-bis(4-crotonyloxyphenyl)-1,2-ethanediol was obtained as a
97:3 mixture of 1/2 (81%).
[11] [11a] D. Sekutowski, R. Jungst, G. D. Stucky, Inorg. Chem. 1978,
[11b]
17, 1848. Ϫ
D. W. Stephan, Organometallics 1992, 11, 996.
[12] [12a] M. L. H. Green,, C. R. Lucas, J.Chem. Soc., Dalton Trans.
[12b]
1972, 1000. Ϫ
T. V. Rajan Babu, W. A. Nugent, J. Am.
Chem. Soc. 1994, 116, 986.
[13]
See reference [43] in S. J. Danishefsky, W. H. Pearson, D. F.
Harvey, C. J. Maring, J. P. Springer, J. Am. Chem. Soc. 1985,
107, 1256.
[1]
[1a] G. M. Robertson, in Comprehensive Organic Synthesis (Eds.:
B. M. Trost, I. Fleming, G. Pattenden), Pergamon Press, Ox- [14] [14a] F. R. W. P. Wild, J. Zsolnai, G. Huttner, H. H. Brintzinger,
[1b]
[14b]
ford, 1991, vol. 3, p. 563. Ϫ
T. Wirth, Angew. Chem. 1996,
J. Organomet. Chem 1982, 232, 233. Ϫ
S. Collins, B. A.
108, 65; Angew. Chem Int. Ed. Engl. 1996, 35, 61.
Kuntz, N. J. Taylor, D. G. Ward, J. Organomet. Chem 1988,
342, 21.
[2] [2a]
[2b]
J. E. McMurry, Chem. Rev. 1989, 89, 1513. Ϫ
A.
[15]
[16]
Fürstner, Angew. Chem. 1993, 105, 171; Angew. Chem. Int. Ed.
Engl. 1993, 32, 164.
A. H. Hoveyda, J. P. Morken, Angew. Chem. 1996, 108, 1378;
Angew. Chem. Int. Ed. Engl. 1996, 35, 1262.
A. Gansäuer, M. Pierobon, H. Bluhm, Angew. Chem. 1998, 110,
107; Angew. Chem. Int. Ed. Engl. 1998, 35, 101.
W. C. Still, M. Kahn, A. Mitra, J. Org. Chem. 1978, 43, 2923.
K. R. K. Prasad, N. N. Joshi, J. Org.Chem. 1996, 61, 388.
E. Juaristi, A. Martinez-Richa, A. Garcia-Rivera, J. S. Cruz-
Sanchez, J. Org. Chem. 1983, 48, 2603.
[3] [3a]
R. A. Johnson, K. B. Sharpless, in Catalytic Asymmetric
[3b]
Synthesis (Ed.: I. Ojima), VCH, Weinheim, 1993, p. 103. Ϫ
[17]
[18]
[19]
H. C. Kolb, M. S. Vannieuwenhze, K. B. Sharpless, Chem. Rev.
1994, 94, 2483.
[4]
For some examples of mechanistically interesting pinacol coup-
[4a]
lings see:
A. W. Konradi, S. J. Kemp. S. F. Pedersen, J. Am.
[4b]
[20]
Chem. Soc. 1994, 116, 1316. Ϫ
B. Kammermeier, G. Beck,
W. S. Trahanovsky, J. R. Gilmore, P. C. Heaton, J. Org. Chem.
1983, 48, 2603.
D. Jacobi, H. Jendrella, Angew. Chem. 1994, 106, 719; Angew.
[4c]
Chem Int. Ed. Engl. 1994, 33, 685. Ϫ
M. T. Reetz, N. Grie-
D. S. Hays, G. C. Fu, J.
[98145]
[4d]
benow, Liebigs Ann. 1996, 335. Ϫ
Eur. J. Org. Chem. 1998, 1923Ϫ1927
1927