A comparison of electron transfer reagents in the reductive opening of epoxides: Reasons for the superiority of titanocene based complexes

Article abstract of DOI:10.1016/S0040-4020(02)00697-X

Several commonly used electron transfer reagents were compared in their reactivity towards three epoxides. These substrates were designed to allow for a distinction between competing courses of the reaction. It was found that titanocene reagents were clearly superior due to their low Lewis acidity, high reduction tendency towards epoxides, and low reduction tendency towards the pivotal β-metal oxy radical intermediates.

Full text of DOI:10.1016/S0040-4020(02)00697-X

Tetrahedron 58 (2002) 7017–7026
A comparison of electron transfer reagents in the reductive
opening of epoxides: reasons for the superiority of titanocene
based complexes
*
Andreas Gansauer and Bjorn Rinker
¨
¨
´
¨ ¨
Kekule Institut fur Organische Chemie und Biochemie der Universitat Bonn, Gerhard-Domagk-Str. 1, 53121 Bonn, Germany
Received 2 April 2002; accepted 9 May 2002
Abstract—Several commonly used electron transfer reagents were compared in their reactivity towards three epoxides. These substrates
were designed to allow for a distinction between competing courses of the reaction. It was found that titanocene reagents were clearly
superior due to their low Lewis acidity, high reduction tendency towards epoxides, and low reduction tendency towards the pivotal b-metal
oxy radical intermediates. q 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
Epoxides are amongst the most frequently employed
substrates in organic synthesis. This is due to the ease of
their preparation from readily available precursors, e.g.
olefins and carbonyl compounds, and their high reactivity.1
The latter point arises mainly from the strain inherent in the
three-membered ring that is released during ring opening.
Epoxides, especially when prepared in high enantiomeric
excess, have been very useful in SN2 reactions in this
respect. An alternative approach to exploiting the high
reactivity of the strained epoxide is constituted by ring
opening reactions utilizing electron transfer reagents.2 In the
light of the success of the Birch conditions for reducing
organic compounds it is not surprising that epoxides can be
opened by solvated electrons.3 The initially formed radical
is then further reduced to give carbanionic species. This
concept has been extended by Bartmann,4 Cohen,5 and Yus6
who have employed aromatic radical anions as the reducing
agents in many synthetically useful applications.
Figure 1. Analogy of epoxide opening and cyclopropylcarbinyl radical
opening.
Figure 2. Chakraborty’s epoxide opening for the synthesis of 1,3-diols.
employed in transformations that do not lead to deoxygena-
tions of epoxides. An example is shown in Fig. 2.2,9
We have recently developed a catalytic system for
conducting these reactions that has also allowed for the
first enantioselective generation of radicals.10
With respect to the utilization of the intermediate radicals
for organic synthesis the use of low valent metal complexes
is more promising. The general idea of this concept was first
outlined by Nugent and RajanBabu7 as shown in Fig. 1 and
constitutes an analogue of the well established opening of a
cyclopropylcarbinyl radical.7,8
Although the use of titanocene based reagents has a number
of advantages it would clearly be attractive to use other
metal complexes in order to establish novel reactivity
patterns. In this contribution we describe our results in this
area. We have concentrated on the use of single electron
transfer reagents to avoid two electron processes via
oxidative insertions or SN2 type reactions with metals.11
So far, only titanocene(III) reagents that were introduced by
Nugent and RajanBabu in this field have been successfully
Keywords: chromium; cyclization; epoxides; radical reaction; samarium;
titanium; vanadium.
2. Results
*
Corresponding author. Tel.: þ49-228-732800; fax: þ49-228-734760;
e-mail: andreas@gansaeuer.de
It is generally agreed that with the SET reagents the pivotal
0040–4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)00697-X

¨
A. Gansauer, B. Rinker / Tetrahedron 58 (2002) 7017–7026
7018
Table 1. Deoxygenation reactions of 1 in the presence of single electron
transfer reagents
Entry
1
ET-reagent
Product
Yield (%)
19a
CrCl2
16
73
2
VCl3(THF)3/Zn
Figure 3. Deoxygenation of epoxides via electron transfer.
intermediates are indeed formed if the deoxygenation
reaction of epoxides proceeds with low stereoselectivity to
mixtures of the corresponding (E ) and (Z ) olefins as shown
in Fig. 3.7a,10
3
4
SmI2
53
19
Cp2TiCl
This argument is based on the unselective trapping of the
configurationally labile radical intermediate by the second
equivalent of the SET reagent. The resulting diastereomeric
mixture eliminates to give the depicted mixture of olefins.
47
The deoxygenation of simple unfunctionalized epoxides has
already been investigated with titanocene7 and samarium12
reagents. Usually both complexes give mixtures of the
isomers with low selectivity. The epoxide 1 investigated
here is mechanistically more interesting because the
organometallic intermediate formed after reductive trapping
with a second equivalent of the low valent metal complex
can give two different elimination products. Consequently,
the issue of regioselectivity of the overall transformation
and the factors controlling it are raised as shown in Fig. 4.
a
49% Substrate reisolated.
introduced by Pedersen.15 CrCl2 has, to the best of our
knowledge, only been used in deoxygenation reactions of
cyclohexene and styrene oxide where no problems of
selectivity can occur.16 Our results are summarized in
Table 1.
Interestingly, [V2Cl3(THF)6]2[Zn2Cl6] and SmI2 react with
complete albeit opposite selectivity. Whereas mixtures of
the (E) and (Z ) isomers (1:1) of 2 are formed in 73% yield
with [V2Cl3(THF)6]2[Zn2Cl6], SmI2 gives only the product
of propoxide elimination 3 in 53% yield. No other products
were obtained. The rather low mass balance could be due to
the noticeable volatility and water solubility of 3. Surpris-
ingly, the chromium reagent exhibited distinctly lower
reactivity even when employed in DMF and in the presence
of the diamine ligand ethylene diamine. Besides 49% of
reisolated starting material, the deoxygenation products 2
and 3 were obtained in low yields. The titanocene reagent
gave 2 in 19% yield as 58:42 mixture of (E) and (Z ) isomers
and 3 in 47% yield.
We have decided to use SmI2, CrCl2, [V2Cl3(THF)6]2-
13
[Zn2Cl6] obtained by reduction of VCl3(THF)3 with Zn
dust in THF, and Cp2TiCl14 as the actual reducing agents.
The vanadium based reagent has to the best of our
knowledge not been used in epoxide openings as yet. It
has proven to give excellent results in pinacol type reactions
Although the reason for this different behavior is unclear at
present it seems that the higher Lewis acidity of samarium
could lead to a more efficient complexation of the propoxy
group after reductive trapping. Elimination then occurs to
yield the allylic alcohol. This experiment does, however,
give no indication if the epoxide is opened via electron
transfer directly or opened by nucleophilic substitution with
iodide and ensuing reduction of the iodo compound.
The less Lewis acid vanadium species could react via ligand
exchange to give a metalla oxetane that eliminates a
vanadium oxo species to yield the allylic ether. An
alternative explanation for this mode of deoxygenation
could be the dimeric nature of the vanadium complex. The
Figure 4. Competing modes of deoxygenation for epoxide 1.

¨
A. Gansauer, B. Rinker / Tetrahedron 58 (2002) 7017–7026
7019
second electron could then be transferred to the radical in an
intramolecular manner. To justify this speculative thought a
rigorous structural determination of the vanadium complex
in solution would have to be carried out.
Titanocene chloride displays intermediate behavior with the
product of propoxide elimination 3 being the major product.
Thus, this reagent is expectedly less Lewis acidic than
samarium diiodide.
The deoxygenation of epoxides with the metal complexes
mentioned above all seem to proceed via intermediate
b-metal oxy radicals. The reaction path after their trapping
seems, however, to depend on the Lewis acidity of the
electron transfer reagent.
Figure 6. 5-exo trig radical cyclization after reductive opening of 4.
Table 2. Reactions of 4 with single electron transfer reagents
With these results in hands we turned our attention to the
decisive question for the desired use of the pivotal b-metal
oxy radicals, the persistence of carbon-centered radicals in a
reductive medium. Radical reactivity for CH and CC bond
formation can only be exploited if the reduction of the
radical is slower than the attempted ensuing radical
transformation, e.g. a 5-exo cyclization.17 For SmI2 it is
known that primary alkyl radicals are reduced with rate
Entry
1
ET-reagent
SmI2
Product
Yield (%)
31, dr¼61:39
constants of about 6£106 M21 s21 18
.
However, 5-exo
55
cyclizations of radicals derived from SmI2 mediated halide
abstraction with alkenes and alkynes are well documented
and must therefore be considered faster than reductive
trapping.17
Therefore, we decided to investigate SmI2 and the other
electron transfer reagents mentioned above with suitably
unsaturated epoxides. We chose epoxide 4 as substrate in
these reactions for four reasons shown in Fig. 5.
2
VCl3(THF)3/Zn
59
Firstly, the ester groups can act as internal nucleophiles for
Lewis acid assisted epoxide opening, secondly the mono-
substituted epoxide is readily attacked by external nucleo-
philes, e.g. iodide, and thirdly the trisubstituted olefin is
known to accelerate 5-exo cyclizations and should intercept
radicals efficiently,19 and last but not least the ester groups
can trap alkoxides formed by SN2 opening of the epoxide as
acylating reagents.
7, dr¼80:20
3
4
5
VCl3/Zn
48, dr¼60:40
25, dr¼40:60
57
Thus, substrate 4 should allow for an analysis of all effects
especially when they occur with similar rates. Also the
question of direct epoxide reduction and SN2 opening
followed by reduction of metallated iodohydrins should be
resolved by analysis of the opening products.
VCl3(THF)3
Cp2TiCl
Figure 5. Possible side reactions during the opening of epoxide 4.

¨
A. Gansauer, B. Rinker / Tetrahedron 58 (2002) 7017–7026
7020
The results of our investigations based on the opening
shown in Fig. 6 are summarized in Table 2.
SmI2 as electron transfer reagent resulted in the formation of
two products in good combined yield. Lactone 5 is formed
by epoxide opening with iodide via SN2 and ensuing fast
lactonization. The deoxygenation product 6 can only be
formed by reduction of 5 with 2 equiv. of SmI2 and
b-elimination of a carboxylate. Thus, the epoxide ring is not
opened via electron transfer at all and the SN2 reaction with
iodide must be considered much faster. In the case of the
vanadium reagent different results were obtained. Expect-
edly, no chlorohydrins from external nucleophilic attack
could be observed. Chloride is only a weak nucleophile.
However, lactone 8 was isolated in low yield (7%). Thus,
some vanadium or zinc species was Lewis acidic enough to
promote an intramolecular nucleophilic epoxide opening.
The main product is, however, constituted by the deoxy-
genation product 7 that could be isolated in 59% yield. This
result was rather surprising for us because we expected a
5-exo cyclization to be faster than any second supposedly
intermolecular trapping of the radical by vanadium.
Because of the dimeric nature of the vanadium reagent in
the solid state it could be that the second electron is
transferred to the radical center in an intramolecular
manner. This intramolecular process should be able to
efficiently compete with the radical cyclization. This
interpretation is in line with the 1:1 mixture of olefins
obtained in the deoxygenation of 1 with vanadium via a
radical intermediate (Table 1, entry 2). A rigorous
confirmation of this presently speculative hypothesis could
be obtained from solution studies of the vanadium reagent’s
structure. It should be noted, however, that preliminary
kinetic investigations of the reaction of [V2Cl3(THF)6]2-
[Zn2Cl6] with benzaldehyde have indicated that the reactive
species in solution is VCl2.20
Figure 7. 5-exo trig radical cyclization after reductive opening of 10.
Table 3. Reactions of 10 with single electron transfer reagents
Entry
1
ET-reagent
Product
Yield (%)
57
VCl3(THF)3/Zn
2
SmI2
36
45
3
4
Cp2TiCl
69–82
Entries 4 and 5 readdress the issue of the intramolecular
nucleophilic epoxide opening leading to the formation of 8.
The precursor to the vanadium(II) reagent, VCl3(THF)3 as
well as VCl3/Zn in THF result in the formation of 8 but in
higher yields than with [V2Cl3(THF)6]2[Zn2Cl6]. Since both
reagents constitute stronger Lewis acids than [V2Cl3-
(THF)6]2[Zn2Cl6] vanadium(III) species are likely to be
responsible for the undesired side reaction and not the
vanadium(II) complex.
CrCl2
38a
a
37% Starting material reisolated.
e.g. SmI2.22 Therefore, we expected 10 to be a better
substrate for cyclization than 4. However, neither [V2Cl3-
(THF)6]2[Zn2Cl6] nor SmI2 gave any of the desired product.
Utilizing vanadium, only the product of deoxygenation 11
could be obtained in significant amounts. Trace amounts of
the chlorohydrins were detected in the crude reaction
mixture. Thus, [V2Cl3(THF)6]2[Zn2Cl6] constitutes a highly
selective reagent for the deoxygenation of epoxides and
does not allow for other synthetic applications of the pivotal
b-metal oxy radical formed during reductive epoxide
opening. SmI2 gave a the primary iodohydrins in 45%
yield and the deoxygenation product in 36% yield. As in the
reactions of 4 nucleophilic opening of the epoxide is the
main course of events. The Sm salt of iodohydrin 12 can be
further reduced by SmI2. The result obtained with CrCl2 was
rather surprising. As in the deoxygenation reaction the
substrate 10 could be reisolated in substantial amounts.
Titanocene chloride gave product 9 in 57% yield that arises
through epoxide opening via electron transfer and the
ensuing 5-exo cyclization. The second ring is closed through
a radical substitution reaction. We are currently investi-
gating the scope of the unprecedented and highly interesting
reaction.21
Finally, we investigated the behavior of epoxide 10 under
electron transfer conditions. Here, a tertiary radical would
be formed after reductive opening that is more stable than
the secondary radical obtained from 4 as depicted in Fig. 7.
The results of the opening reactions are summarized in
Table 3.
It is also well documented that tertiary radicals are only very
slowly reduced even by potent electron transfer reagents,

¨
A. Gansauer, B. Rinker / Tetrahedron 58 (2002) 7017–7026
7021
However, the product of the radical cyclization 13 was also
obtained in noticeable amounts (38%). Thus, CrCl2, albeit
being rather unreactive, is the only other reagent than
Cp2TiCl that allows for the exploitation of the b-metal oxy
radical in a carbon–carbon bond forming reaction. Our
titanocene based protocol, on the other hand, gave the
desired cyclization product 13 in good yield (69–82%) and
once again demonstrates the superiority of the
titanocene(III) reagents in reductive epoxide openings.
Isomer ratios were determined from suitable 1H NMR
integrals of cleanly separated signals.
1H and 13C NMR spectra were recorded on a Brucker AM
400 or DRX 300 instrument. For the 1H NMR spectra
chemical shifts are expressed relative to trichloromethane
(CHCl3: 7.26 ppm) and penta-deuterated benzene (C6H D5:
7.16 ppm) as internal standards in the corresponding fully
deuterated solvents. For the 13C NMR spectra the central
signals of deuterotrichloromethane (C DCl3: 77.16 ppm)
and of deuterated benzene (C6D6: 128.06 ppm) were
1
employed as internal standards. The integrals of the H
3. Conclusion
NMR signals are in accord with the assignment. Coupling
constants are measured in Hz. Mass spectra were
obtained on a Kratos MS 50 or a Thermoquest MAT
95XL operating at 70 eV. Infrared spectra were recorded on
a Perkin–Elmer FT-IR spectrometer 1620. All substances
were measured as neat film between KBr plates. Com-
bustion analyses were performed by Mrs Martens at the
In summary our studies have revealed that SmI2 is not a
suitable reagent for the reductive opening of epoxides. The
high Lewis acidity of this metal combined with the high
nucleophilicity of the iodide ions leads to the formation of
iodohydrins that are further reduced by samarium. The
typical radical reactivity of b-metal oxy radicals as shown in
Figs. 3, 6, and 7 could not be observed in the cases
investigated here. In the case of [V2Cl3(THF)6]2[Zn2Cl6] no
other products than those of epoxide deoxygenation could
be observed. Although it seems as if this reagent opens
epoxides via electron transfer the resulting b-metal oxy
radicals could not be intercepted by C–C bond forming
reactions. This unexpected result can be tentatively
rationalized by assuming a dimeric structure of the
vanadium reagent in solution that would result in an
intramolecular second electron transfer. These findings
suggest that the reason for the superiority of Cp2TiCl
reagents stems from their unique combination of low Lewis
acidity preventing epoxide opening via SN2 or SN1 and low
reducing power towards the b-metal oxy radical. The only
other reagent allowing carbon carbon bond formation,
CrCl2, is unfortunately severely limited by its low reactivity.
We are currently working on developing more reactive
chromium reagents for these purposes.
´
Kekule-Institut fu¨r Organische Chemie und Biochemie,
Universitat Bonn.
¨
4.2. Preparation of starting materials: synthesis of the
tris(tetrahydrofuran)vanadiumtrichloride
Under an argon atmosphere vanadium(III) chloride (5.04 g,
32 mmol) was extracted from a Soxhlet extractor with
150 ml dry THF until the extract becomes colorless. After
reducing the solution volume to 40 ml the tris(tetrahydro-
furan)vanadiumtrichloride was crystallized in the refriger-
ator by 238 8C over night. The complex was filtrated under
argon and dried to obtain the pale red tris(tetrahydro-
furan)vanadiumtrichloride (6.96 g, 58%). The complex was
stored under argon in the refrigerator and used without
further purification.
4.2.1. (Z )-1,4-Dipropoxy-but-2-ene-oxide (1). (Z)-1,4-
Dipropoxy-but-2-ene24 (12.7 g, 80 mmol), methyltrioxo-
rhenium (100 mg, 0.4 mmol, 0.005 equiv.), H2O2 (16.2 ml,
30% solution in H2O, 160 mmol, 2 equiv.) and 3-cyano-
pyridine (833 mg, 8 mmol, 0.1 equiv.) in CH2Cl2 (48 ml)
were stirred 65 h at rt. The reaction was quenched with ice
(8 g) and MnO2 (10 mg), stirred for 1 h, poured into H2O
(50 ml) and extracted with CH2Cl2 (3£50 ml). The organic
layer was dried over MgSO4 and evaporated under reduced
pressure. The oil obtained was diluted with pentane
(100 ml) and the white precipitate was filtered off. The
solvent was evaporated and the residue distilled (b.p. 52–
57 8C, 0.1 mbar). (Z )-1,4-Dipropoxy-but-2-ene-oxide (1)
(11.3 g, 60 mmol) was obtained as colorless oil (75%).
4. Experimental
4.1. General methods
Unless otherwise indicated all reactions were performed in
oven-dried (100 8C) glassware under argon. Solutions were
transferred by means of teflon tubes. Liquid substances and
solvents were added by means of plastic syringes. THF was
dried over potassium and freshly distilled before use. DMF
was used as received from Merck, just as ethylenediamine
(p.a., abs.) obtained from Fluka. Chromium(II) chloride
(anhydrous) was purchased from Alfa Aesar and stored
under Argon. Samarium(II) iodide was prepared from
samarium powder (40 mesh) with 1,2-diiodoethane.21
Bis(cylopentadienyl)titanocenechloride was prepared in
situ from bis(cyclopentadienyl)titanocenedichloride pur-
chased from Aldrich and zinc or manganese dust. The
reduction of the Vanadium–THF-complex was carried out
in situ using zinc as reducing agent. Products were purified
by flash chromatography on silica gel from Macherey-Nagel
or Merck (particle size 43–60 mm, 230–400 mesh, ASTM).
Eluents are given (MTBE refers to tert-butyl methyl ether,
Cy refers to cyclohexane, EtOAc to ethyl acetate, Et2O to
diethyl ether and PE to petrol ether, 40–60 8C fractions).
(Z )-1,4-Dipropoxy-but-2-ene-oxide (1): C10H20O3
1
(188.27 g/mol); H NMR (400 MHz, C6D6): d¼3.42 (dd,
J¼11.3, 3.9 Hz, 2H, A-part of an AB-system), 3.31 (dd,
J¼11.3, 6.4 Hz, 2H, B-part of an AB-system), 3.27 (dt,
J¼9.1, 6.6 Hz, 2H, A-part of an AB-system), 3.18
(dt, J¼9.1, 6.6 Hz, 2H, B-part of an AB-system), 3.01 (m,
2H), 1.50 (qt, J¼7.4, 6.6 Hz, 4H), 0.91 (t, J¼7.3 Hz, 6H).
13C NMR (100 MHz, C6D6): d¼73.0, 68.22, 54.4, 23.4,
10.8. Anal. calcd for C10H20O3: C, 63.80; H, 10.71. Found:
C, 63.67; H, 10.86. IR (neat): n¼2965, 2875, 1690, 1465,
1385, 1355, 1325, 1255, 1110, 1050, 955, 910, 845,
780 cm21

¨
A. Gansauer, B. Rinker / Tetrahedron 58 (2002) 7017–7026
7022
4.2.2. 2-(3-Methyl-but-2-enyl)-2-oxiranylmethylmalonic
acid diethylester (4). 2-Oxiranylmethylmalonic acid
diethylester (2.16 g, 10 mmol),10c NaH (288 mg, 12 mmol,
1.2 equiv.) and 1-bromo-3-methyl-2-butene (1.28 ml,
11 mmol, 1.1 equiv.) were stirred in THF (100 ml) at 0 8C
for 2 h. MTBE was added and the solution washed with
water (4£50 ml). The organic layer was dried over MgSO4
and evaporated under reduced pressure. The purification of
the residual oil by SiO2 flash chromatography (PE/MTBE;
80:20) yielded 2-(3-methyl-but-2-enyl)-2-oxiranylmethyl-
malonic acid diethylester (4) (2.25 g, 7.9 mmol) as colorless
oil (79%).
2.2 equiv.) in dry THF (20 ml) was added zinc or
manganese powder (288 mg or 242 mg, 4.4 mmol,
4.4 equiv.) and the reaction mixture was stirred for 2 h to
enable complete conversion to Cp2TiCl. After the addition
of the epoxide (1.0 mmol, 1.0 equiv.) the suspension was
stirred for further 16 h at rt. The reaction was quenched by
addition of HCl (2 N, 30 ml), then CH2Cl2 (30 ml) was
added and the organic layer was separated. After extraction
of the aqueous layer with CH2Cl2 (2£30 ml) the combined
organic layers were washed with saturated aq. NaHCO3
(30 ml), H2O (2£30 ml) and saturated aq. NaCl (30 ml) and
dried over MgSO4. The solvent was removed under reduced
pressure and the residue purified by SiO2 flash
chromatography.
2-(3-Methyl-but-2-enyl)-2-oxiranylmethylmalonic acid
diethylester (4): C15H24O5 (284.35 g/mol); Rf (89% Cy,
11% EE): 0.26; 1H NMR (400 MHz, CDCl3): d¼4.91 (tsept,
J¼7.4, 1.5 Hz, 1H), 4.14 (dq, J¼14.1, 7.1 Hz, 1H, A-part of
an AB-system), 4.13 (q, J¼7.1 Hz, 2H), 4.11 (dq, J¼14.2,
7.1 Hz, 1H, B-part of an AB-system), 2.90 (dddd, J¼6.6,
5.1, 4.0, 2.6 Hz, 1H), 2.80–2.60 (m, 2H ), 2.66 (dd, J¼4.9,
4.2 Hz, 1H, A-part of an AB-system), 2.36 (dd, J¼5.2,
2.7 Hz, 1H, B-part of an AB-system), 2.04 (dd, J¼14.8,
5.2 Hz, 1H, A-part of an AB-system), 1.95 (dd, J¼14.6,
6.5 Hz, 1H, B-part of an AB-system), 1.62 (d, J¼1.0 Hz,
3H), 1.56 (s, 3H), 1.19 (t, J¼7.1 Hz, 3H), 1.19 (t, J¼7.1 Hz,
3H). 13C NMR (100 MHz, CDCl3): d¼171.1 (2C), 135.9,
117.5, 56.4, 61.4, 61.3, 48.6, 46.9, 36.0, 32.1, 26.0, 18.0,
14.2, 14.0. HRMS (EI/70 eV) calcd for Mþz: 284.1624;
found: 284.1634; Anal. calcd for C15H24O5: C, 63.36; H,
8.51. Found: C, 63.26; H, 8.74. IR (neat): n¼2980, 2930,
4.4. General procedure 2 (GP2): epoxide openings with
chromium(II) chloride
A solution of ethylenediamine (0.5 ml) in DMF (40 ml) was
degassed three times under vacuum in an ultrasonic bath.
The solution was then added dropwise to a flask containing
chromium(II) chloride (492 mg, 4 mmol, 4 equiv.) generat-
ing a blue-purple suspension. After addition of the epoxide
(1.0 mmol, 1.0 equiv.) the reaction mixture was stirred for
70 h at rt. The reaction was quenched by addition of H2O
(50 ml) and HCl (2 N, 20 ml) and the aqueous layer was
extracted with CH2Cl2 (4£30 ml). The combined organic
layers were washed with saturated aq. NaHCO3 (30 ml),
H2O (4£30 ml) and dried over MgSO4. The solvent was
removed under reduced pressure and the residue purified by
SiO2 flash chromatography.
1730, 1445, 1365, 1285, 1225, 1095, 1025, 935, 860 cm21
.
4.2.3. 2-(1-Ethynyl-cyclohexyloxymethyl)-2-methyloxir-
ane (10).10g 1-Ethynyl-1-(2-methylallyloxy)cyclohexane
(2.27 g, 12.7 mmol), methyltrioxorhenium (30 mg,
0.1 mmol, 0.008 equiv.), H2O2 (6 ml, 30% solution in
H2O, 50 mmol, 3.9 equiv.), and 3-cyanopyridine (260 mg,
2.5 mmol, 0.2 equiv.) in CH2Cl2 (10 ml) were stirred at rt
for 3 days. The reaction was quenched with ice and MnO2 (a
few mg), stirred for 1 h, poured into H2O (50 ml) and
extracted with CH2Cl2 (3£50 ml). The organic layer was
dried over MgSO4 and evapurated under reduced pressure.
The oil obtained was diluted with pentane and the white
precipitate was filtered off. The solvent was evaporated and
the residue purified by SiO2 flash chromatography (Cy/EE
98:2) to afford 2-(1-ethynyl-cyclohexyloxymethyl)-2-
methyloxirane (10) (2.41 g, 12.4 mmol) as a colorless oil
(97%).
4.5. General procedure 3 (GP3): epoxide openings with
samarium(II) iodide
A deep blue solution of samarium(II) iodide (849 mg,
2.1 mmol, 2.1 equiv.) in dry THF (25 ml) was prepared
according to Kagan et al.23 from samarium (361 mg,
2.4 mmol, 2.4 equiv.) and 1,2-diiodoethane (592 mg,
2.1 mmol, 2.1 equiv.). After addition of the epoxide
(1.0 mmol, 1.0 equiv.) the solution was stirred 60 h at rt.
The reaction was quenched by addition of HCl (1 N, 30 ml)
and the aqueous layer was extracted with CH2Cl2
(3£30 ml). The combined organic layers were washed
with saturated aq. Na2S2O3 (30 ml), H2O (3£30 ml) and
saturated aq. NaCl (30 ml) and dried over MgSO4. The
solvent was removed under reduced pressure and the residue
purified by SiO2 flash chromatography.
2-(1-Ethynyl-cyclohexyloxymethyl)-2-methyloxirane (10):
C12H18O2 (194.27 g/mol); 1H NMR (300 MHz, CDCl3):
d¼3.58 (s, 2H), 2.78 (d, J¼4.9 Hz, 1H), 2.64 (d, J¼4.9 Hz,
1H), 2.46 (d, J¼1.0 Hz, 1H), 1.91–1.84 (m, 2H), 1.71–1.28
(m, 8H), 1.39 (s, 3H). 13C NMR (100 MHz, CDCl3):
d¼85.1, 73.8, 73.7, 67.0, 56.1, 52.3, 37.0 (2C), 25.4, 22.6,
18.8. Anal. calcd for C12H18O2: C, 74.19; H, 9.34. Found: C,
73.97; H, 9.36. IR (neat): n¼3290, 3045, 2935, 2860, 2100,
4.6. General procedure 4 (GP4): epoxide openings with
vanadium complexes
Method A: VCl3/Zn: A solution of vanadiumtrichloride
(300 mg, 2 mmol, 2 equiv.) in dry THF (10 ml) was reduced
in situ with zinc dust (261 mg, 4 mmol, 4 equiv.) by stirring
for 2 h to obtain a green-blue solution of the vanadium(II)
complex ([V2Cl3(THF)6][Zn2Cl6]). After addition of the
epoxide (1 mmol, 1 equiv.) the solution was stirred for 16 h
at rt. The reaction was quenched by addition of H2O (20 ml)
and HCl (2 N, 10 ml) and the aqueous layer was extracted
with CH2Cl2 (3£30 ml). The combined organic layers were
washed with H2O (3£20 ml) and saturated aq. NaCl (30 ml)
and dried over MgSO4. The solvent was removed under
1450, 1150, 1090, 1030, 950, 905 cm21
.
4.3. General synthetic procedures—general procedure 1
(GP1): epoxide openings with titanocene chloride
To a suspension of Cp2TiCl2 (548 mg, 2.2 mmol,

¨
A. Gansauer, B. Rinker / Tetrahedron 58 (2002) 7017–7026
7023
reduced pressure and the residue purified by SiO2 flash
chromatograph.
AB-system), 3.42 (dt, J¼9.4, 6.7 Hz, 1H, B-part of an
AB-system), 3.30 (dd, J¼9.7, 8.0 Hz, 1H, B-part of an AB-
system), 1.60 (sext, J¼7.1 Hz, 2H), 0.92 (t, J¼7.4 Hz, 3H).
13C NMR (100 MHz, CDCl3): d¼136.9, 116.4, 74.6, 73.2,
71.6, 22.9, 10.6. IR (neat): n¼3440, 3080, 2960, 2875, 1645,
Method B: V(THF)3Cl3/Zn: A solution of tris(tetrahydro-
furan)vanadiumtrichloride (747 mg, 2 mmol, 2 equiv.) in
dry THF (10 ml) was reduced in situ with zinc dust (261 mg,
4 mmol, 4 equiv.) by stirring for 2 h to obtain a green-blue
solution of the vanadium(II) complex ([V2Cl3(THF)6]2-
[Zn2Cl6]). After addition of the epoxide (1 mmol, 1 equiv.)
the solution was stirred for 16 h at rt. The reaction was
quenched by addition of H2O (20 ml) and HCl (2 N, 10 ml)
and the aqueous layer was extracted with CH2Cl2
(3£30 ml). The combined organic layers were washed
with H2O (3£20 ml) and saturated aq. NaCl (30 ml) and
dried over MgSO4. The solvent was removed under reduced
pressure and the residue purified by SiO2 flash
chromatography.
1460, 1380, 1315, 1115, 990, 925, 670 cm21
.
4.8. Opening of epoxide 1 in the presence of CrCl2
(Table 1, entry 1)
According to GP2: CrCl2 (492 mg, 4 mmol), ethylene-
diamine (0.5 ml) and (Z)-1,4-dipropoxy-but-2-ene-oxide
(188 mg, 1 mmol) in DMF (40 ml) were reacted for 70 h
at rt. After work-up and chromatography (SiO2; PE/Et2O,
90:10) (E/Z )-1,4-dipropoxy-but-2-ene (2) (33 mg,
0.19 mmol) was obtained in 19% yield as colorless oil as
a mixture of diastereoisomers (Z/E 59:41). 1-Propoxy-but-
3-ene-2-ol (3) (21 mg, 0.16 mmol) was obtained in 16%
yield as colorless oil. Starting material was also recovered
(92 mg, 0.49 mmol).
Method C: V(THF)3Cl3: To a red solution of tris(tetra
hydrofuran)vanadiumtrichloride (747 mg, 2 mmol,
2 equiv.) in dry THF (10 ml) the epoxide (1 mmol,
1 equiv.) was added and the reaction mixture was stirred
for 96 h at rt. The reaction was quenched by addition of H2O
(30 ml) and the aqueous layer was extracted with MTBE
(3£30 ml). The combined organic layers were washed with
H2O (3£20 ml) and saturated aq. NaCl (30 ml) and dried
over MgSO4. The solvent was removed under reduced
pressure and the residue purified by SiO2 flash
chromatography.
4.9. Opening of epoxide 1 in the presence of
VCl3(THF)3/Zn (Table 1, entry 2)
According to GP4, Method B: VCl3(THF)3 (747 mg,
2 mmol), Zn (261 mg, 4 mmol) and (Z)-1,4-dipropoxy-
but-2-ene-oxide (188 mg, 1 mmol) in dry THF (10 ml)
were reacted for 16 h at rt. After work-up and chromato-
graphy (SiO2; PE/Et2O, 93:7) (E/Z)-1,4-dipropoxy-but-2-
ene (2) (125 mg, 0.73 mmol) was obtained in 73% yield
as colorless oil as a mixture of diastereoisomers (Z/E
50:50).
4.7. Opening of epoxide 1 in the presence of Cp2TiCl
(Table 1, entry 4)
According to GP1: Cp2TiCl2 (548 mg, 2.2 mmol), Mn
(242 mg, 4.4 mmol) and (Z )-1,4-dipropoxy-but-2-ene-
oxide (188 mg, 1 mmol) in dry THF (20 ml) were reacted
for 16 h at rt. After work-up and chromatography (SiO2;
PE/Et2O, 93:7) (E/Z )-1,4-dipropoxy-but-2-ene23 (2)
(34 mg, 0.19 mmol) was obtained in 19% yield as colorless
oil as a mixture of diastereoisomers (Z/E 42:58) and
1-propoxy-but-3-ene-2-ol25 (3) (61 mg, 0.47 mmol) was
obtained in 47% yield as colorless oil.
4.10. Opening of epoxide 1 in the presence of SmI2
(Table 1, entry 3)
According to GP3: Sm (361 mg, 2.4 mmol), 1,2-diiodo-
ethane (592 mg, 2.1 mmol) and (Z)-1,4-dipropoxy-but-2-
ene-oxide (188 mg, 1 mmol) in dry THF (25 ml) were
reacted for 16 h at rt. After work-up and chromatography
(SiO2; PE/Et2O, 90:10) 1-propoxy-but-3-ene-2-ol (3)
(69 mg, 0.53 mmol) was obtained in 53% yield as pale
yellow oil.
4.7.1. (E/Z )-1,4-dipropoxy-but-2-ene (2).24 C10H20O2
(172.27 g/mol); Rf (89% Cy, 11% EE): 0.38; 1H NMR
(400 MHz, CDCl3): d¼5.80 (tt, J¼2.9 Hz, 1.4, 2H)a, 5.70
(ddd, J¼4.7, 3.7, 0.9 Hz, 2H)b, 4.03 (dm, J¼4.8 Hz, 4H)b,
3.96 (dd, J¼2.9, 1.5 Hz, 4H)a, 3.37 (t, J¼6.8 Hz, 4H)a, 3.37
(t, J¼6.7 Hz, 4H)b, 1.59 (sext, J¼6.9 Hz, 4H)a, 1.59 (sext,
4.11. Opening of epoxide 4 in the presence of SmI2
(Table 2, entry 1)
According to GP3: Sm (361 mg, 2.4 mmol), 1,2-diiodo-
ethane (592 mg, 2.1 mmol) and 2-(3-methyl-but-2-enyl)-2-
oxiranylmethylmalonic acid diethylester (284 mg, 1 mmol)
in dry THF (25 ml) were reacted for 50 h at rt. After work-
up and chromatography (SiO2; Cy/EE, 90:10) 5-iodo-
methyl-3-(ethoxycarbonyl)-3-(3-methyl-but-2-enyl)-2-oxo-
tetrahydrofuran (116 mg, 0.31 mmol) (5) was obtained in
31% yield as pale yellow oil as a mixture of diasteroisomers
(dr¼61: 39). 2-Allyl-2-(3-methyl-but-2-enyl)-malonic acid
monoethyl ester (133 mg, 0.55 mmol) was also obtained in
55% yield as colorless oil.
J¼7.5 Hz, 4H)b, 0.91 (t, J¼7.5 Hz, 12H)a,b a,b
.
Signals of the
two isomers. 13C NMR (100 MHz, CDCl3): d¼129.6a,p
,
.
129.6b,p, 72.3a,pp, 72.3b,pp, 70.9a, 66.6b, 23.1a,b, 10.7a,b
a,bSignals of the two isomers. p,ppAssignment interchange-
able. HRMS (EI/70 eV) calcd for Mþz: 172.1463; found:
172.1459; IR (neat): n¼2960, 2855, 1465, 1360, 1260,
1100, 1040, 805 cm21
.
4.7.2.
1-Propoxy-but-3-ene-2-ol
(3).25
C7H14O2
(130.19 g/mol); Rf (89% Cy, 11% EE): 0.16; 1H NMR
(400 MHz, CDCl3): d¼5.83 (ddd, J¼17.3, 10.6, 5.7 Hz,
1H), 5.35 (dt, J¼17.3, 1.6 Hz, 1H), 5.18 (dt, J¼10.6, 1.5 Hz,
1H), 4.30 (m, 1H), 3.47 (dd, J¼9.7, 3.5 Hz, 1H, A-part of an
AB-system), 3.46 (dt, J¼9.4, 6.7 Hz, 1H, A-part of an
4.11.1. 5-Iodomethyl-3-(ethoxycarbonyl)-3-(3-methyl-
C13H19IO4
but-2-enyl)-2-oxo-tetrahydrofuran
(5).
(366.20 g/mol); Rf (80% Cy, 20% EE): 0.45; 1H NMR
(400 MHz, CDCl3): d¼5.05–4.98 (m, 2H)a,b, 4.52 (dddd,

¨
A. Gansauer, B. Rinker / Tetrahedron 58 (2002) 7017–7026
7024
J¼8.4, 7.6, 6.6, 4.9 Hz, 1H)a, 4.48 (dddd, J¼9.6, 6.6, 6.5,
4.3 Hz, 1H)b, 4.27–4.16 (m, 4H)a,b, 3.41 (dd, J¼10.1,
4.9 Hz, 1H)a, 3.39 (dd, J¼10.6, 4.4 Hz, 1H)b, 3.29 (dd,
J¼10.6, 6.8 Hz, 1H)b, 3.27 (dd, J¼9.9, 8.6 Hz, 1H)a, 2.80–
2.67 (m, 2H)a,b, 2.78 (dd, J¼13.3, 6.3 Hz, 1H)b, 2.62–2.53
(m, 2H)a,b, 2.56 (dd, J¼13.6, 6.7 Hz, 1H)a, 2.43 (dd,
J¼13.8, 7.7 Hz, 1H)a, 1.93 (dd, J¼13.3, 9.5 Hz, 1H)b, 1.71
(m, 6H)a,b, 1.65 (m, 6H)a,b, 1.29 (t, J¼7.1 Hz, 3H)a, 1.28 (t,
J¼7.1 Hz, 3H)b. a,bSignals of the two isomers. 13C NMR
(100 MHz, CDCl3): d¼174.1a, 173.4b, 169.8a, 169.5b,
137.8a, 137.4b, 117.5b, 117.0a, 76.9a, 76.4b, 62.5a,b, 56.8b,
56.0a, 37.8b, 36.6a, 33.2a, 32.5b, 26.1a,b, 18.2a,b, 14.1a,b, 6.8b,
6.3b. a,bSignals of the two isomers. HRMS (EI/70 eV) calcd
for Mþz: 366.0328; found: 366.0329; IR (neat): n¼2965,
1780, 1730, 1445, 1365, 1340, 1160, 1095, 1010, 860,
4.12.2. 5-Hydroxymethyl-3-(3-ethoxycarbonyl)-3-(3-
methyl-but-2-enyl)-2-oxo-tetrahydro-furan (8). C13H20O5
(256.30 g/mol); Rf (50% Cy, 50% EE): 0.35; 1H NMR
(400 MHz, CDCl3) (contains cyclohexan): d¼5.07 (ddsept,
J¼9.3, 5.8, 1.5 Hz, 1H)a, 5.01 (ddsept, J¼8.2, 6.8, 1.4 Hz,
1H)b, 4.64 (dddd, J¼10.0, 6.4, 4.8, 2.8 Hz, 1H)b, 4.53
(dddd, J¼8.0, 7.1, 5.5, 3.2 Hz, 1H)a, 4.23 (q, J¼7.1 Hz,
3H)a, 4.22 (qd, J¼7.1, 0.3 Hz, 3H)b, 3.92 (ddd, J¼12.6, 5.8,
2.9 Hz, 1H)b, 3.83 (ddd, J¼12.5, 6.6, 3.2 Hz, 1H)a, 3.69 (dt,
J¼12.2, 5.7 Hz, 1H)a, 3.61 (ddd, J¼12.5, 6.7, 4.9 Hz, 1H)b,
2.78 (ddm, J¼14.7, 8.2 Hz, 1H)b, 2.72 (ddm, J¼14.6,
8.1 Hz, 1H)a, 2.65–2.51 (m, 2H)a,b, 2.62 (dd, J¼13.5,
7.2 Hz, 1H)a, 2.56 (dd, J¼13.1, 6.4 Hz, 1H)b, 2.26 (dd,
J¼13.5, 8.0 Hz, 1H)a, 2.15 (dd, J¼13.1, 10.0 Hz, 1H)b, 2.04
(m, OH)a, 1.92 (m, OH)b, 1.73 (q, J¼1.2 Hz, 3H)a, 1.70 (q,
J¼1.2 Hz, 3H)b, 1.66 (d, J¼1.3 Hz, 3H)a, 1.64 (d,
J¼0.8 Hz, 3H)b, 1.29 (t, J¼7.1 Hz, 3H)a, 1.28 (t,
J¼7.1 Hz, 3H)b. a,bSignals of the two isomers. 13C NMR
(100 MHz, CDCl3): d¼174.7a, 174.2b, 170.3a, 169.7b,
137.7a, 137.1b, 117.5b, 117.1a, 78.8b, 78.4a, 64.4a, 63.5b,
62.5b, 62.4a, 56.3b, 55.5a, 33.4a, 32.7b, 32.1b, 32.0a, 26.1a,
800 cm21
.
4.11.2. 2-Allyl-2-(3-methyl-but-2-enyl)-malonic acid
monoethyl ester (6). C13H20O4 (240.30 g/mol); Rf (80%
1
Cy, 20% EE): 0.05; H NMR (400 MHz, CDCl3): d¼5.67
(dddd, J¼17.0, 10.1, 7.6, 7.0 Hz, 1H), 5.11 (dm, J¼17.0 Hz,
1H), 5.09 (dm, J¼10.2 Hz, 1H), 4.98 (tsep, J¼7.5, 1.4 Hz,
1H), 4.24 (dq, J¼10.8, 7.1 Hz, 1H, A-part of an
AB-system), 4.21 (dq, J¼10.8, 7.1 Hz, 1H, B-part of an
AB-system), 2.73 (ddt, J¼14.0, 7.0, 1.1 Hz, 1H), 2.68–2.57
(m, 2H), 2.61 (ddt, J¼13.8, 7.8, 1.0 Hz, 1H), 1.69 (d,
J¼1.0 Hz, 3H), 1.61 (d, J¼0.8 Hz, 3H), 1.29 (t, J¼7.1 Hz,
3H). 13C NMR (100 MHz, CDCl3): d¼174.6, 173.5, 136.5,
132.3, 119.4, 117.3, 62.2, 57.9, 39.0, 33.7, 26.1, 18.1, 14.1.
HRMS (EI/70 eV) calcd for Mþz: 240.1362; found:
240.1360; IR (neat): n¼2980, 1710, 1445, 1380, 1220,
26.0b, 18.2a,b, 14.1a,b a,bSignals of the two isomers. HRMS
.
(EI/70 eV) calcd for Mþz: 256.1311; found: 256.1306; IR
(neat): n¼3450, 2935, 1775, 1450, 1365, 1180, 1035, 915,
860 cm21
.
4.13. Opening of epoxide 4 in the presence of a vanadium
complex (Table 2, entry 3)
According to GP4, Method A: VCl3 (150 mg, 1 mmol), Zn
(130 mg, 2 mmol) and 2-(3-methyl-but-2-enyl)-2-oxiranyl-
methylmalonic acid diethylester (142 mg, 0.5 mmol) in dry
THF (10 ml) were reacted for 20 h at rt. After work-up and
chromatography (SiO2; Cy/EE, 60:40) 5-hydroxymethyl-3-
(3-ethoxycarbonyl)-3-(3-methyl-but-2-enyl)-2-oxo-tetra-
hydro-furan (8) (62 mg, 0.24 mmol) was obtained in 48%
yield as colorless oil as a mixture of diastereoisomers
(dr¼60:40).
1065, 920, 860, 655 cm21
.
4.12. Opening of epoxide 4 in the presence of
VCl3(THF)3/Zn (Table 2, entry 2)
According to GP4, Method B: VCl3(THF)3 (747 mg,
2 mmol), Zn (261 mg, 4 mmol) and 2-(3-methyl-but-2-
enyl)-2-oxiranylmethylmalonic acid diethylester (284 mg,
1 mmol) in dry THF (10 ml) were reacted for 60 h at rt.
After work-up and chromatography (SiO2; Cy/EE, 93:7)
2-allyl-2-(3-methyl-but-2-enyl)-malonic acid diethyl ester
(7)26 (160 mg, 0.59 mmol) was obtained in 59% yield as
colorless oil. 5-Hydroxymethyl-3-(3-ethoxycarbonyl)-3-(3-
methyl-but-2-enyl)-2-oxo-tetrahydrofuran (8) (18 mg,
0.07 mmol) was obtained in 7% yield as colorless oil as a
mixture of diastereoisomers (dr¼80:20).
4.14. Opening of epoxide 4 in the presence of a vanadium
complex (Table 2, entry 4)
According to GP4, Method C (Table 2, entry 5): VCl3-
(THF)3 (747 mg, 2 mmol), Zn (261 mg, 4 mmol) and 2-(3-
methyl-but-2-enyl)-2-oxiranylmethylmalonic acid diethyl-
ester (284 mg, 1 mmol) in dry THF (10 ml) were reacted for
60 h at rt. After work-up and chromatography (SiO2; Cy/EE,
60:40)
5-hydroxymethyl-3-(3-ethoxycarbonyl)-3-(3-
4.12.1. 2-Allyl-2-(3-methyl-but-2-enyl)-malonic acid
diethyl ester (7). C15H24O4 (268.35 g/mol); Rf (89% Cy,
methyl-but-2-enyl)-2-oxo-tetrahydrofuran (8) (64 mg,
0.25 mmol) was obtained in 25% yield as colorless oil as
a mixture of diastereoisomers (dr¼40:60).
1
11% EE): 0.5; H NMR (400 MHz, CDCl3): d¼5.66 (ddt,
J¼16.3, 10.8, 7.4 Hz, 1H), 5.08 (dm, J¼17.1 Hz, 1H), 5.07
(dm, J¼10.4 Hz, 1H), 4.97 (tsep, J¼7.4, 1.4 Hz, 1H), 4.18
(dq, J¼10.9, 7.1 Hz, 1H, A-part of an AB-system), 4.15
(dq, J¼10.8, 7.1 Hz, 1H, B-part of an AB-system), 2.61
(dt, J¼7.4, 1.2 Hz, 2H), 2.59 (dm, J¼7.6 Hz, 2H), 1.68
(d, J¼1.1 Hz, 3H), 1.60 (s, 3H), 1.23 (t, J¼7.1 Hz, 3H).
13C NMR (100 MHz, CDCl3): d¼171.3 (2C), 135.6,
132.9, 119.0, 117.8, 61.2 (2C), 57.8, 36.9, 31.1, 26.1,
18.1, 14.2 (2C). HRMS (EI/70 eV) calcd for Mþz:
268.1675; found: 268.1674; IR (neat): n¼2980, 2930,
1730, 1640, 1445, 1365, 1285, 1220, 1095, 1065, 920,
4.15. Opening of epoxide 4 in the presence of Cp2TiCl
(Table 2, entry 5)
According to GP1: Cp2TiCl2 (249 mg, 1 mmol), Mn
(110 mg, 2 mmol) 2-(3-methyl-but-2-enyl)-2-oxiranyl-
methylmalonic acid diethylester (284 mg, 1 mmol) in dry
THF (10 ml) were reacted for 16 h at rt. After work-up and
chromatography (SiO2; Cy/EE, 80:20) (4,4-dimethyl-3-oxa-
bicyclo[3.3.0]octane)-7,7-dicarboxylic acid diethyl ester (9)
(162 mg, 0.57 mmol) was obtained in 57% yield as colorless
oil.
860 cm21
.

¨
A. Gansauer, B. Rinker / Tetrahedron 58 (2002) 7017–7026
7025
4.15.1. (4,4-Dimethyl-3-oxa-bicyclo[3.3.0]octane)-7,7-
dicarboxylic acid diethyl ester (9). C15H24O5
(284.35 g/mol); Rf (89% Cy, 11% EE): 0.14; 1H NMR
(400 MHz, CDCl3): d¼4.19–4.08 (m, 4H), 3.89 (dd, J¼9.4,
7.9 Hz, 1H, A-part of an AB-system), 3.51 (dd, J¼9.1,
3.5 Hz, 1H, B-part of an AB-system), 2.84 (quintd, J¼8.2,
3.3 Hz, 1H), 2.58 (ddd, J¼13.3, 8.6, 2.0 Hz, 1H), 2.36 (ddd,
J¼10.6, 8.9, 7.8 Hz, 1H), 2.22 (ddd, J¼12.8, 7.5, 2.1 Hz,
1H), 1.95 (dd, J¼12.8, 11.1 Hz, 1H),), 1.82 (dd, J¼13.4,
7.3 Hz, 1H), 1.20 (t, J¼7.0 Hz, 3H), 1.20 (s, 3H), 1.18 (t,
J¼7.0 Hz, 3H), 1.09 (s, 3H). 13C NMR (100 MHz, CDCl3):
d¼171.9, 171.2, 81.5, 71.9, 63.0, 61.4 (2C), 52.7, 43.4, 40.4,
36.0, 26.6, 23.7, 14.1 (2C). HRMS (EI/70 eV) calcd for
Mþz: 284.1624; found: 284.1625; Anal. calcd for C15H24O5:
C, 63.36; H, 8.51. Found: C, 63.17; H, 8.40. IR (neat):
n¼2975, 2870, 1730, 1465, 1445, 1365, 1295, 1260, 1180,
3295, 2935, 2860, 1450, 1380, 1330, 1257, 1195, 1150,
1125, 1085, 950, 905, 850, 785, 625 cm21
.
4.18. Opening of epoxide 7 in the presence of Cp2TiCl
(Table 3, entry 3)
Collidine hydrochloride (394 mg, 2.5 mmol, 2.5 equiv.) was
dried in vacuo before adding 2-(1-ethynyl-cyclohexyloxy-
methyl)-2-methyloxirane (194 mg, 1 mmol), Mn powder
(110 mg, 2 mmol, 2 equiv.), Cp2TiCl2 (24.9 mg, 0.1 mmol,
0.1 equiv.) and dry THF (10 ml). After stirring for 24 h at rt
the reaction was quenched by addition of HCl (40 ml, 2 N),
then MTBE was added (60 ml) and the organic layer was
separated. It was further washed with sat. aq. NaHCO3
(40 ml), H2O (40 ml), dried over MgSO4 and evaporated
under reduced pressure. Purification by SiO2 flash chroma-
tography (Cy/EE 92:8) afforded (3-methyl-4-methylene-1-
oxa-spiro[4.5]dec-3-yl)-methanol (13)10g (135–162 mg,
0.69–0.82 mmol) as a colorless oil (69–82%).
1100 cm21
.
4.16. Opening of epoxide 10 in the presence of
VCl3(THF)3/Zn (Table 3, entry 1)
4.18.1. (3-Methyl-4-methylene-1-oxa-spiro[4.5]dec-3-yl)-
methanol (13).10g C12H20O2 (196.29 g/mol); Rf (80% Cy,
According to GP4, Method B: VCl3(THF)3 (747 mg,
2 mmol), Zn (261 mg, 4 mmol) 2-(1-ethynyl-
cyclohexyloxymethyl)-2-methyloxirane (194 mg, 1 mmol)
in dry THF (15 ml) were reacted for 14 h at rt. After work-
up and chromatography (SiO2; PE) 1-ethynyl-1-(2-methyl-
allyloxy)cyclohexane (11) (102 mg, 0.57 mmol) was
obtained in 57% yield as colorless oil.
1
20% EE): 0.25; H NMR (300 MHz, CDCl3): d¼4.89 (s,
1H), 4.84 (s, 1H), 3.87 (d, J¼9.0 Hz, 1H, A-part of an AB-
system), 3.53 (d, J¼9.0 Hz, 1H, B-part of an AB-system),
3.49 (dd, J¼10.8, 7.0 Hz, 1H, A-part of an AB-system);
3.38 (dd, J¼10.8, 5.6 Hz, 1H, B-part of an AB-system);
1.80–1.55, (m, 8H), 1.45–1.16 (m, 3H), 1.13 (s, 3H). 13C
NMR (75 MHz, CDCl3): d¼162.4, 104.1, 84.6, 73.0, 68.3,
48.9, 37.1, 36.1, 25.6, 22.6 (2C), 21.3. HRMS (EI/70 eV)
calcd for Mþz: 196.1463; found: 196.1463; Anal. calcd for
C12H20O2: C, 73.43; H, 10.27. Found: C, 73.12; H, 10.06. IR
(neat): n¼3410, 2930, 2855, 1735, 1655, 1445, 1145, 1045,
4.16.1. Compound (11). C12H18O (178.27 g/mol); Rf
1
(100% PE): 0.05; H NMR (400 MHz, CDCl3): d¼5.02–
5.00 (m, 1H), 4.86–4.84 (m, 1H), 4.01 (s, 2H), 2.45 (s, 1H),
1.93–1.88 (m, 2H), 1.77–1.76 (m, 3H), 1.71–1.61 (m, 4H),
1.58–1.46 (m, 3H), 1.36–1.26 (m, 1H). 13C NMR
(100 MHz, CDCl3): d¼143.2, 111.5, 85.7, 73.8, 67.5,
37.4, 25.7, 22.9, 20.0. HRMS (EI/70 eV) calcd for Mþz:
178.1358; found: 178.1358; IR (neat): n¼3305, 2935, 2860,
935, 890 cm21
.
4.19. Opening of epoxide 10 in the presence of CrCl2
(Table 3, entry 4)
1655, 1450, 1260, 1090, 895, 800, 655 cm21
.
According to GP2: CrCl2 (492 mg, 4 mmol), ethylene-
diamine (0.5 ml) and 2-(1-ethynyl-cyclohexyloxymethyl)-
2-methyloxirane (194 mg, 1 mmol) in DMF (40 ml) were
reacted for 70 h at rt. After work-up and chromatography
(SiO2; PE/Et2O, 60:40) (3-methyl-4-methylene-1-oxa-
spiro[4.5]dec-3-yl)-methanol (13) (75 mg, 0.38 mmol) was
obtained in 38% yield as colorless oil. Starting material was
also recovered (73 mg, 37%).
4.17. Opening of epoxide 10 in the presence of SmI2
(Table 3, entry 2)
According to GP3: Sm (361 mg, 2.4 mmol), 1,2-diiodo-
ethane (592 mg, 2.1 mmol) and 2-(1-ethynyl-cyclo-
hexyloxymethyl)-2-methyloxirane (194 mg, 1 mmol) in
dry THF (25 ml) were reacted for 50 h at rt. After work-up
and chromatography (SiO2; PE/Et2O, 98:2) 1-ethynyl-1-(2-
methyl-allyloxy)cyclohexane (11)10g (64 mg, 0.36 mmol)
was obtained in 36% yield as colorless oil. 1-(1-Ethynyl
cyclohexyloxy)-3-iodo-2-methyl-propan-2-ol (12) (147 mg,
0.45 mmol) was obtained in 45% yield as pale yellow
oil.
Acknowledgments
We thank the Deutsche Forschungsgemeinschaft (Gerhard
Hess-Programm, Ga 619/2-1) for generous support of our
work.
4.17.1. 1-(1-Ethynyl cyclohexyloxy)-3-iodo-2-methyl-
propan-2-ol (12). C12H19IO2 (322.19 g/mol); Rf (89% Cy,
1
11% EE): 0.35; H NMR (400 MHz, CDCl3): d¼3.68 (d,
J¼8.8 Hz, 1H, A-part of an AB-system), 3.53 (d, J¼8.8 Hz,
1H, B-part of an AB-system), 3.35 (s, 2H), 2.52 (s, 1H), 2.48
(s, 1H), 1.89–1.82 (m, 2H), 1.69–1.60 (m, 4H), 1.58–1.44
(m, 3H), 1.40–1.28 (m, 1H), 1.36 (s, 3H). 13C NMR
(100 MHz, CDCl3): d¼85.0, 74.3, 73.7, 70.7, 68.0, 37.2,
36.9, 25.4, 24.0, 22.6 (2C), 16.9. HRMS (EI/70 eV) calcd
for Mþz: 322.0430; found: 322.0414; IR (neat): n¼3440,
References
1. For recent reviews on the generation of epoxides see: (a) Rao,
A. S. Comprehensive Organic Synthesis; Trost, B. M., Ed.;
Pergamon: Oxford, 1991; Vol. 7, pp 357–387. (b) Meunier, B.
¨
Chem. Rev. 1992, 92, 1411. (c) Gansauer, A. Angew. Chem.
1997, 109, 2701. Angew. Chem. Int. Ed., 1997, 36, 2591. For

¨
A. Gansauer, B. Rinker / Tetrahedron 58 (2002) 7017–7026
7026
¨
(f) Gansauer, A.; Bluhm, H.; Pierobon, M.; Keller, M.
¨
Organometallics 2001, 20, 914. (g) Gansauer, A.; Pierobon,
recent reviews on enantioselective epoxidations see:
(d) Katsuki, T. Comprehensive Asymmetric Catalysis;
Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer:
Berlin, 1999; Vol. 2, pp 513–538. (e) Jacobsen, E. N.; Wu,
M. H. Comprehensive Asymmetric Catalysis; Jacobsen, E. N.,
Pfaltz, A., Yamamoto, H., Eds.; Springer: Berlin, 1999; Vol. 2,
pp 649–678. (f) Aggarwal, V. K. Comprehensive Asymmetric
Catalysis; Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.;
Springer: Berlin, 1999; Vol. 2, pp 679–696.
¨
M.; Bluhm, H. Synthesis 2001, 2500. (h) Gansauer, A.; Bluhm,
H.; Lauterbach, T. Adv. Synth. Catal. 2001, 343, 785.
11. (a) Sharpless, K. B.; Umbreit, M. A.; Nieh, M. T.; Flood, T. C.
J. Am. Chem. Soc. 1972, 94, 6538. (b) Su, H.; Walder, L.;
Zhang, Z.-d.; Scheffold, R. Helv. Chim. Acta 1988, 71, 1073.
(c) Bonhote, P.; Scheffold, R. Helv. Chim. Acta 1991, 74,
1425.
¨
2. Gansauer, A.; Bluhm, H. Chem. Rev. 2000, 100, 2771.
3. (a) Birch, A. J. J. Proc. R. Soc. N.S.W. 1950, 83, 245.
(b) Hallsworth, A. S.; Henbest, H. B. J. Chem. Soc. 1957,
4604. (c) Hallsworth, A. S.; Henbest, H. B. J. Chem. Soc.
1960, 3571. (d) Brown, H. C.; Ikegami, S.; Kawakami, J. H.
J. Org. Chem. 1970, 35, 3243. (e) Benkeser, R. A.; Rappa, A.;
Wolsieffer, L. A. J. Org. Chem. 1986, 51, 3391.
12. Matsukawa, M.; Tabuchi, T.; Inanaga, J.; Yamaguchi, M.
Chem. Lett. 1987, 2101.
13. Manzer, L. E. Inorg. Synth. 1982, 21, 135.
14. For a discussion of the structure of these reagents in solution
see: Enemærke, R. J.; Hjøllund, G. H.; Daasbjerg, K.;
Skrydstrup, T. C. R. Acad. Sci. Paris, Chim./Chem. 2001, 4,
435.
4. Bartmann, E. Angew. Chem. 1986, 98, 629. Angew. Chem. Int.
Ed. Engl., 1986, 25, 855.
15. (a) Freudenberger, J. H.; Konradi, A. W.; Pedersen, S. F. J. Am.
Chem. Soc. 1989, 111, 8014. (b) Konradi, A. W.; Kemp, S. J.;
Pedersen, S. F. J. Am. Chem. Soc. 1994, 116, 1316.
(c) Konradi, A. W.; Pedersen, S. F. J. Org. Chem. 1992, 57,
28. (d) Reetz, M. T.; Griebenow, N. Liebigs Ann. 1996, 335.
(e) Kammermeier, B.; Beck, G.; Holla, W.; Jacobi, D.;
Napierski, B.; Jendralla, H. Chem. Eur. J. 1996, 2, 307.
16. Kochi, J. K.; Singleton, D. M.; Andrews, L. J. Tetrahedron
1968, 24, 3503.
5. Cohen, T.; Jeong, I.-H.; Mudryk, B.; Bhupathy, M.; Awad,
M. M. A. J. Org. Chem. 1990, 55, 1528.
6. (a) Yus, M. Synlett 2001, 1197. (b) Bachki, A.; Foubelo, F.;
Yus, M. Tetrahedron: Asymmetry 1995, 6, 1907. (c) Bachki,
A.; Foubelo, F.; Yus, M. Tetrahedron: Asymmetry 1996, 7,
2997.
7. (a) Nugent, W. A.; RajanBabu, T. V. J. Am. Chem. Soc. 1988,
110, 8561. (b) RajanBabu, T. V.; Nugent, W. A. J. Am. Chem.
Soc. 1989, 111, 4525. (c) RajanBabu, T. V.; Nugent, W. A.;
Beattie, M. S. J. Am. Chem. Soc. 1990, 112, 6408.
(d) RajanBabu, T. V.; Nugent, W. A. J. Am. Chem. Soc.
1994, 116, 986.
17. Molander, G. A. Radicals in Organic Synthesis; Renaud, P.,
Sibi, M. P., Eds.; Wiley-VCH: Weinheim, 2001; Vol. 1,
pp 153–182 and references cited therein.
18. (a) Hasegawa, E.; Curran, D. P. Tetrahedron Lett. 1993, 34,
1717. (b) Enemærke, R. J.; Daasbjerg, K.; Skrydstrup, T.
J. Chem. Soc., Chem. Commun. 1999, 343. (c) Shabangi, M.;
Kuhlman, M. L.; Flowers, II R. A. Org. Lett. 1999, 1, 2133.
19. Beckwith, A. L. J.; Schiesser, C. H. Tetrahedron 1985, 41,
3925.
¨
8. For recent review see:Gansauer, A. Radicals in Organic
Synthesis; Renaud, P., Sibi, M. P., Eds.; Wiley-VCH:
Weinheim, 2001; Vol. 2, pp 207–220 and references cited
therein.
9. For the most recent example see: (a) Chakraborty, T. K.; Das,
S. Tetrahedron Lett. 2002, 43, 2313. For a review on
titanocene(III) chemistry see: (b) Spencer, R. P.; Schwartz,
J. Tetrahedron 2000, 56, 2103. For a review on radical
chemistry with epoxides see: (c) Li, J. J. Tetrahedron 2001, 57,
1.
20. Daasbjerg, K.; private communication.
21. (a) Schiesser, C. H.; Wild, L. M. Tetrahedron 1996, 42, 13265.
(b) Walton, J. C. Acc. Chem. Res. 1998, 31, 99.
22. (a) Curran, D. P.; Fevig, T. L.; Jasperse, C. P.; Totleben, M. J.
Synlett 1992, 943. (b) Krief, A.; Laval, A.-M. Chem. Rev.
1999, 99, 745.
¨
10. (a) Gansauer, A.; Pierobon, M.; Bluhm, H. Angew. Chem.
1998, 110, 107. Angew. Chem. Int. Ed. Engl., 1998, 37, 101.
23. Girard, P.; Namy, J.-L.; Kagan, H. B. J. Am. Chem. Soc. 1980,
102, 2693.
¨
(b) Gansauer, A.; Bluhm, H. J. Chem. Soc., Chem. Commun.
¨
1998, 2143. (c) Gansauer, A.; Bluhm, H.; Pierobon, M. J. Am.
¨
Chem. Soc. 1998, 120, 12849. (d) Gansauer, A.; Lauterbach,
T.; Bluhm, H.; Noltemeyer, M. Angew. Chem. 1999, 111,
3112. Angew. Chem. Int. Ed. Engl., 1999, 38, 2909.
24. Mkryan, G. M.; Pogosyan, A. A. J. Org. Chem. USSR 1969, 5,
1693.
25. Mikhantev, V. B.; Negrobov, A. F. Sb. Nauch. Rab. Aspir.
Voronezh. Gos. Univ. 1965, 2, 88.
26. Staudinger, H.; Ruzicka, L. Helv. Chim. Acta 1924, 7, 377.
¨
(e) Gansauer, A.; Pierobon, M. Synlett 2000, 1357.