Oxidation of organometallic compounds (RM, M = Li, MgBr, ZnBr, CuCNLi, Cu(R)CNLi2) with tBuOOLi, Ti(OiPr)4-mediated with tBuOOH, and with O2, to give alcohols (ROH). Are radicals R? involved?

Article abstract of DOI:10.1016/S0022-328X(00)00596-9

Organometallic compounds (RM, M = Li, MgBr, ZnBr, Cu(CN)Li, CuR(CN)Li₂) are oxidized with ^tBuOOLi (or PhCMe₂OOLi) (method A) to the corresponding alcohols (ROH), in good to very good yields. This oxidation works also well with the protic system Ti(OⁱPr)₄ + ^tBuOOH (method B) in the case of RM, M = Li, MgBr and ZnBr. Stereochemical studies with the configurationally stable cis- and trans-cyclopropyl metal compounds, cis- and trans-1-M, respectively, show that the reactions of RM, M = Li, MgBr, with ^tBuOOLi (method A) and Ti(OⁱPr)₄ + ^tBuOOH (method B) follow a S_N2-type pathway of the nucleophile RM at the electrophilic (although anionic!) oxygen atom of ^tBuOOLi(TiX₃) to give ROM (isolated as the esters 2) with retention of configuration at R. In contrast, the oxidations with dioxygen O₂ (method C) occur with loss of stereochemistry in the cyclopropyl alcohols ROH (RM, M = Li, MgBr), and additional formation of cyclopropyl dimers R-R 3 in the case of RM, M = Cu(CN)Li, CuR(CN)Li₂. This is due to facile electron transfer from RM to O₂ and fast isomerization (dimerization) of the intermediate radicals R^?. The high tendency of RM cuprates, M = Cu(CN)Li, CuR(CN)Li₂, for electron transfer reactions is also indicated by some loss of stereospecificity in the formation of ROH, and some dimer 3 formation, even with the oxidant ^tBuOOLi (method A) .

Full text of DOI:10.1016/S0022-328X(00)00596-9

Journal of Organometallic Chemistry 624 (2001) 47–52
www.elsevier.nl/locate/jorganchem
Oxidation of organometallic compounds (RM, M=Li, MgBr,
t
i
ZnBr, CuCNLi, Cu(R)CNLi ) with BuOOLi, Ti(O Pr) -mediated
2
4
t
with BuOOH, and with O , to give alcohols (ROH). Are radicals
2
’
R involved?
Michael M o¨ ller, Marc Husemann, Gernot Boche *
Fachbereich Chemie der Philipps-Uni6ersit a¨ t Marburg, Hans-Meerwein Strasse, D-35032 Marburg, Germany
Received 7 July 2000; accepted 23 August 2000
This work is dedicated to Professor Dr. Jean Normant on the occasion of his 65th birthday
Abstract
t
Organometallic compounds (RM, M=Li, MgBr, ZnBr, Cu(CN)Li, CuR(CN)Li ) are oxidized with BuOOLi (or
2
PhCMe OOLi) (method A) to the corresponding alcohols (ROH), in good to very good yields. This oxidation works also well
2
i
t
with the protic system Ti(O Pr) + BuOOH (method B) in the case of RM, M=Li, MgBr and ZnBr. Stereochemical studies with
4
the configurationally stable cis- and trans-cyclopropyl metal compounds, cis- and trans-1-M, respectively, show that the reactions
t
i
t
of RM, M=Li, MgBr, with BuOOLi (method A) and Ti(O Pr) + BuOOH (method B) follow a S 2-type pathway of the
4
N
t
nucleophile RM at the electrophilic (although anionic!) oxygen atom of BuOOLi(TiX ) to give ROM (isolated as the esters 2)
3
with retention of configuration at R. In contrast, the oxidations with dioxygen O (method C) occur with loss of stereochemistry
2
in the cyclopropyl alcohols ROH (RM, M=Li, MgBr), and additional formation of cyclopropyl dimers R–R 3 in the case of
RM, M=Cu(CN)Li, CuR(CN)Li . This is due to facile electron transfer from RM to O and fast isomerization (dimerization)
2
2
’
of the intermediate radicals R . The high tendency of RM cuprates, M=Cu(CN)Li, CuR(CN)Li , for electron transfer reactions
2
is also indicated by some loss of stereospecificity in the formation of ROH, and some dimer 3 formation, even with the oxidant
t
BuOOLi (method A) . © 2001 Elsevier Science B.V. All rights reserved.
Keywords: Organometallic compounds; Oxidations; Alcohols
to sulfoxides, the a-hydroxylation of (RO) Ti-enolates
1
. Introduction
3
by Schulz [4], and the epoxidation of hydroperoxy-
olefins by Adam [5]. In the following we show that the
TiX -catalyzed(mediated) oxidation reactions deserve
4
i
t
IV
protic system Ti(O Pr) + BuOOH can also be used,
special interest since a heterogeneous Ti /SiO catalyst
4
2
had been developed by Shell [1], and since Sharpless
had published the catalytic enantioselective epoxidation
of allylic alcohols with titanium tetra-isopropoxide
besides the oxenoids [6a,b] lithium t-butylhydroperox-
t
ide ( BuOOLi) or lithium cumylhydroperoxide
(PhCMe OOLi) [6], for the oxidation of organometallic
2
i
t
t
Ti(O Pr) , butylhydroperoxide BuOOH and (+)- or
compounds (RM, M=Li) [6], MgBr and ZnBr; in the
4
(
−)-diethyl tartrate [2]. Related systems were used by
case of M=CuCNLi and Cu(R)CNLi the oxidation
works only with BuOOLi in good yields. Furthermore,
2
t
Kagan [3] for the enantioselective oxidation of sulfides
the stereochemistry of some of these oxidation reactions
is compared with the stereochemistry of the oxidations
of the corresponding organometallic compounds (RM)
with dioxygen O2.
*
Corresponding author. Fax: +49-(0)6421-282-8917.
E-mail address: boche@chemie.uni-marburg.de (G. Boche).
0
022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(00)00596-9

4
8
M. M o¨ ller et al. / Journal of Organometallic Chemistry 624 (2001) 47–52
Table 1
Oxidation reactions of RLi with BuOOLi (PhCMe OOLi) (method
2
. Results and discussion
t
2
i
t
A), and Ti(O Pr) + BuOOLi (method B), respectively, to give, after
4
In Table 1, oxidations of RLi compounds with
acid work-up, ROH; yields of ROH (%)
t
BuOOLi (PhCMe OOLi) [6] (method A) and
2
i
t
Ti(O Pr) + BuOOH (method B), respectively, are com-
4
pared with each other. In general, both methods lead to
t
similar results. Exceptions: BuLi (entry 3) is only
i
t
poorly oxidized by Ti(O Pr) + BuOOH (method B),
4
while the vinyllithium species (entry 5) gives higher
yields with this method.
Table 2 summarizes different organometallic com-
pounds (RM) and their oxidation reactions with
t
i
t
BuOOLi (method A) and/or Ti(O Pr) + BuOOH
4
(
method B).
Entries 1–5 show the facile transformation of aryl-
lithium compounds via method B into phenols al-
though the solution contains a proton [7–10]. Similarly,
the oxidation of Grignard compounds [11] succeeds in
t
high yields except for the Butyl case (entries 7–10).
Table 2
t
Oxidation of RM with BuOOLi (method A) and /or Ti(Oiprop) +
Benzyl-Li (entry 6) and benzyl-MgBr (entry 11) are
oxidized to benzyl alcohol without the formation of
bibenzyl [12]. Entries 12 and 13 are representative for
the preparation of ROH from Zn compounds [13]. The
preferential oxidation reactions of the cuprates to the
alcohols (ROH) (entries 14–19) are of special interest,
since cuprates are known for their facile formation of
coupling products R–R in oxidation reactions, at least
4
t
BuOOH (method B) to give alcohols ROH; yields of ROH (%)
with O [14]. The coupling product bibenzyl indeed is
2
observed in the case of the benzyl cuprates (entries 20
and 21): apart from benzyl alcohol, these easily oxi-
dized species lead to 19% and 14%, respectively, of
bibenzyl. In the case of the cuprates the oxidations to
i
t
the alcohols ROH with Ti(O Pr) + BuOOH (method
4
t
B) give much poorer yields than with BuOOLi (method
A). Therefore only the latter ones are listed in Table 2.
Earlier studies of the oxidation of diastereomeric,
configurationally stable cyclopropyl- [6d] and vinyl-
lithium species [6e] showed the intermediate formation
’
’ ’
of radicals R (or radical pairs R OOLi) in which
stereoisomerization at the former carbanionic carbon
atom takes place, if the oxidation is performed with O2.
In contrast, a S 2-type reaction of the nucleophile RLi
N
at the anionic, electrophilic oxenoid oxygen atom of
t
BuOOLi should be responsible for the retention of
configuration at the carbanionic carbon atom in this
oxidation reaction [6d,e], which is in agreement with
calculations [15]. In order to find out whether the
oxidations described in Tables 1 and 2 follow an elec-
’
tron transfer pathway via intermediate radicals R , or
via a S 2-type oxygen transfer of the oxenoid oxygen
N
t
atom of BuOOM (M=Li, TiX ) onto RM, stereo-
3
chemical studies were performed with the cyclopropyl
metal compounds cis-1-M and trans-1-M, respectively
[20], see Scheme 1. Depending on the mechanism, the
oxidation leads to one of the esters 2 (after acylation of
the first formed alcoholate) with retention of configura-

M. M o¨ ller et al. / Journal of Organometallic Chemistry 624 (2001) 47–52
49
formed stereospecifically in the respective alcohols
ROH by oxidation reactions with oxenoids ROOM.
The oxidation of the cuprates with O2 (method C)
(
entries 5, 7, 9 and 11) is rather different from the
above discussed results: both isomeric esters 2 together
with significant amounts of the dimers 3 indicate a
mechanism of the O -oxidation which is not the same
2
as in the case of the O -oxidations of RLi and RMgX,
2
I
II
and in which an electron transfer Cu “Cu might be
involved [14]. In the oxidation reactions of the same
cuprates with BuOOLi (method A, see entries 6, 8, 10
and 12) the esters 2, with retention of configuration, are
generally the preferred products. However, in entry 8
t
Scheme 1. cis- and trans-1-M, the ester oxidation (and acylation)
products cis- and trans-2, and the radical dimers 3 (mixture of
isomers).
5
% of cis-2 are also found together with 3% of the
tion at carbon (S 2-type), or, in the case of electron
N
dimers 3. Three percent of the dimers 3 are likewise
formed in entry 12. Thus, BuOOLi (method A) has, as
in the case of cuprates, a strong tendency for S 2-like
transfer of oxygen. However, an electron transfer path-
way seems also accessible. The formation of bibenzyl
transfer and radical intermediates, to both stereoiso-
meric esters 2, and possibly to the dimers 3 (mixture of
isomers), see Scheme 1. The results of the oxidations of
t
N
t
i
1
-M with BuOOLi (method A) and Ti(O Pr) +
4
t
BuOOH (method B) are listed in Table 3 together with
(
entries 21 and 22, Table 2) support these observations.
those of the oxidations of 1-M with O (method C).
2
The oxidations of the Li-compounds cis-1-Li and
t
trans-1-Li with O (method C) and BuOOLi (method
2
3
. Conclusion
A) in entries 1 and 2 agree with earlier results [6d,e]: O2
(
method C) leads to stereoisomeric esters 2, and
RLi, RMgBr, RCuCNLi and R CuCNLi are oxi-
t
2
2
BuOOLi (method A) oxidizes only to one stereoisomer
with retention of configuration. The latter is also true
for Ti(O Pr) + BuOOH (method B) as the oxidant.
Dimers 3 are not formed in all these cases, not even
with O (method C). The Grignard compounds cis-1-
MgBr and trans-1-MgBr (entries 3 and 4) are also
oxidized by Ti(O Pr) + BuOOH (method B) with re-
tention of configuration to give the esters 2. Again
dimers 3 are not formed. This is similarly the case for
dized by O (method C) to the corresponding alcohols
2
ROH via radical intermediates as shown by loss of
stereochemistry at the nucleophilic carbon atom of
configurationally stable cyclopropyl metal compounds.
Dimers R–R are only formed in the oxidation of
i
t
4
2
t
cuprates. In contrast, by the oxenoid systems BuOOLi
i
t
i
t
4
(method A) and Ti(O Pr) + BuOOH (method B), RLi
4
and RMgBr [16] are oxidized to ROH with retention of
configuration at R, probably via a S 2-type mecha-
N
t
oxidations with BuOOLi (method A) although the
yields are much smaller. Grignard species RMgX, like
organolithium compounds RLi, can thus be trans-
nism. Some stereoisomerization as well as dimer forma-
tion is observed in some cases of the oxenoid oxidation
reactions (method A) of cuprates of the type RCuCNLi
Table 3
Stereochemistry of the oxidations of cis- and trans-1-M, respectively, to give cis- and trans-2 and the dimers 3, respectively, with different
oxidants; yields of 2 and 3 (%)
t
t
Entry
1-M
Method A BuOOLi
Method B Ti(Oiprop) + BuOOH
Method C O2
3
4
cis-2
trans-2
cis-2
trans-2
cis-2
trans-2
1
2
3
4
5
6
7
8
9
cis-1-Li
47
0
25
0
–
77
–
5
–
86
–
0
0
37
0
12
–
0
–
58
–
0
68
0
79
0
–
–
–
–
–
–
0
43
0
60
–
–
–
–
–
39
8
–
–
19
–
9
–
16
–
15
–
38
72
–
–
8
–
33
–
38
–
18
–
0
0
0
0
60
0
26
3
35
0
trans-1-Li
cis-1-MgBr
trans-1-MgBr
(cis-1) -CuCNLi
2
2
(cis-1) -CuCNLi
2
2
(trans-1) -CuCNLi
(trans-1) -CuCNLi
cis-1-CuCNLi
cis-1-CuCNLi
trans-1-CuCNLi
trans-1-CuCNLi
2
2
2
2
1
1
1
0
1
2
–
–
–
–
76
–
–
25
3

5
0
M. M o¨ ller et al. / Journal of Organometallic Chemistry 624 (2001) 47–52
13
and R CuCNLi . Although such cuprates have a strong
1 H, CHBr), 7.17–7.42 (m, 5 H, Ph-H); C-NMR
2
2
tendency for oxidation with electron transfer they are
oxidized by BuOOLi (method A) preferentially to the
(CDCl ): l=22.1 (C8), 27.0 (C2), 27.3 (C1), 27.6 (C3),
126.8, 127.1, 129.4, 142.2 (Ph-C).
3
t
alcohols ROH with a high degree of retention of
configuration, and in high yields.
4.4. trans-1-Bromo-2-methyl-2-phenyl-cyclopropane
[19]
4
4
4
. Experimental
1
H-NMR (CDCl ): l=1.00 (dd, Jtrans=4.74 Hz,
3
Jgem=6.44 Hz, 1 H, CH (cis)), 1.54 (s, 3 H, CH ), 1.58
pseudo t, J=7.2 Hz, 1 H, CH (trans)), 3.15 (dd,
Jtrans=4.67 Hz, J =7.97 Hz, 1 H, CHBr), 7.11–7.23
m, 5 H, Ph-H); C-NMR (CDCl ): l=23.3 (C8), 23.9
C3), 25.8 (C2), 30.5 (C1), 126.5, 127.0, 127.5, 144.5
Ph-C).
2
3
.1. General remarks
(
2
cis
13
.1.1. Analytical GC
Sichromat (Siemens) with an Integrator Chromato-
(
(
(
3
pac C-R3A (Shimadzu). Column: glass capillary
column SE 52 (0.3 mm×30 m), helium (0.85 bar).
4
.5. cis-1-Lithio-2-methyl-2-phenyl-cyclopropane
4
.1.2. Preparati6e GC
Aerograph A-90-P3 (Wilkens Instruments & Re-
(cis-1-Li )
search Inc.). Steel column, 5% carbowax 20 M on
chromosorb G AW-DMCS 80–100 mesh, helium.
To a solution of 0.42 g (2.00 mmol) cis-1-bromo-2-
methyl-2-phenyl-cyclopropane in 15 ml Et O was added
2
t
at −78°C, 2.75 mL (4.40 mmol) of BuLi (1.6 M) in
4
4
.1.3. Column chromatography
Silicagel 60 (Merck), 0.063–0.200 mm.
hexane, and then stirred for 60 min at that temperature.
Cis-1-Li is formed with complete retention of configu-
ration [20]. This solution was used for oxidation reac-
tions with methods A, B and C, respectively.
.1.4. Chromatotron
Modell 7924T (Harrison Research), silicagel 60 PF256
(
Merck).
4.6. trans-1-Lithio-2-methyl-2-phenyl-cyclopropane
(trans-1-Li )
4.1.5. Preparation of RM (Tables 1 and 2)
The organometallic compounds (RM) outlined in
Tables 1 and 2 were prepared according to standard
trans-1-Li [20] was prepared exactly as cis-1-Li.
procedures in Et O or THF.
2
4.7. cis- and
trans-1-MgBr-2-methyl-2-phenyl-cyclopropane
4
.2. cis- and
(cis-1-MgBr and trans-1-MgBr, respecti6ely)
trans-1-Bromo-2-methyl-2-phenyl-cyclopropanes
(prepared according to Schneider [17a])
cis- and trans-1-MgBr, respectively, were prepared
from cis-1-Li and trans-1-Li, respectively. The solutions
of cis(trans)-1-Li were added at −78°C to 0.37 g (2.00
To 145 g (0.50 mol) of 1,1-dibromo-2-methyl-2-
phenyl-cyclopropane (prepared according to Dehmlow
18]) in 600 ml acetic acid were added at 60°C, 200 g
3.00 mol) of Zn powder. The solution was stirred at
0°C for 10 h and then poured into 500 ml of cold
mmol) of MgBr in 15 ml of THF and stirred for 60
[
(
6
2
min at that temperature [20].
water. The mixture was extracted four times with 200
ml of hexane, the hexane solution dried with MgSO4,
hexane removed, and the mono-bromides distilled in
vacuo (b.p. 55°C, 0.05 Torr). Yield: 70.7 g (0.34 mol),
4.8. Lithium [cis-(1-(2-methyl-2-phenyl-2-
cyclopropyl))-(cyano)-cuprate] (cis-1-CuCNLi ) [21]
A total of 0.18 g (2.00 mmol) of CuCN and 15 ml of
THF were cooled to −78°C, and then 2.00 mmol of a
solution of cis-1-Li added to the suspension of CuCN.
After solution of CuCN (30 min at −78°C and 30 min
at −45°C) the cuprate solutions are ready for the
oxidation reactions [21].
67% cis:trans mixture with a 1.3:1 ratio (by NMR).
Cis- and trans-1-bromo-2-methyl-2-phenyl-cyclo-
propane were separated on a silicagel column with a
petrol ether/CHCl3 mixture (25:1); R (cis) 0.31; R
f
f
(
trans) 0.22.
4.3. cis-1-Bromo-2-methyl-2-phenyl-cyclopropane [19]
4.9. trans-1-CuCNLi [21]
1
H-NMR(CDCl ): l=1.27–1.34 (m, 2 H, CH ), 1.46
The preparation of trans-1-CuCNLi followed exactly
the preparation of cis-1-CuCNLi.
3
2
(
s, 3 H, CH ), 3.03 (dd, J =7.08 Hz, Jtrans=4.69 Hz,
3 cis

M. M o¨ ller et al. / Journal of Organometallic Chemistry 624 (2001) 47–52
51
4.10. Lithium [bis(cis-1-(2-methyl-2-phenyl-
4.14. Stereoisomeric coupling products 3
cyclopropyl)-cuprate]·LiCN ((cis-1) -CuCNLi ) [21]
2
2
Into a solution of cis-1-CuCNLi in Et O–THF was
2
The preparation of (cis-1) -CuCNLi2 followed the
bubbled O first at −78°C for 1 h and then for 30 min
2
2
preparation of cis-1-CuCNLi except that 0.09 g (1.00
mmol) CuCN were used.
on warming to 0°C. Addition of acetic anhydride (one
molecular equivalent) was added to trap the alcoho-
lates. Addition of an aqueous NH Cl-solution and ex-
4
traction with Et O (see the preparation of cis-3) led to
2
4
.11. (trans-1) -CuCNLi [21]
a mixture of cis-2, trans-2 and 3, which was separated
on a Chromatotron (PE 40/60 ethyl acetate gradient). 3
2
2
The preparation is analogous to that of (cis-1)2-
CuCNLi2.
was isolated in 35% yield.
1
H-NMR(CDCl ): l=0.56–0.61 (m, 2 H), 0.74–0.92
3
(
(
m, 1 H), 1.13–1.15 (m, 1 H), 1.17–1.26 (m, 2 H), 1.54
13
s, 6 H, CH ), 7.13–7.28 (m, 10 H, Ph–H); C-NMR
3
4.12. cis-2-Methyl-2-phenyl-cyclopropyl acetate (cis-2)
(CDCl ): l=20.9 (C3), 21.2 (C1), 24.1 (C2), 25.8 (C8),
3
1
25.3, 126.5, 128.1, 148.4 (Ph-C).
To a solution of cis-1-Li in Et O was added at
2
t
−
78°C, 1.5 molecular equivalents of BuOOLi (or
t
PhCMe OOLi), and then stirred at −78°C for 1 h.
4.15. Oxidation with BuOOLi (method A)
2
After addition of 1.2 molecular equivalents of acetic
acid anhydride, the solution was warmed to 20°C.
t
The oxidations with BuOOLi (PhCMe OOLi) were
2
Addition of an aqueous solution of NH Cl and extrac-
performed as detailed in the preparation of cis-2 (trap-
ping with acetic acid anhydride to give the esters). For
the stereochemical studies summarized in Table 3 the
reactions were stopped after 1 h at −78°C. The syn-
thetically oriented data of Tables 1 and 2 were received
after 12 h at −78°C and warming up to room temper-
ature (r.t.). Hydrolysis, to give the alcohols (phenols)
ROH, was performed with 2 N HCl. The alcohols
(phenols) were extracted with Et O, dried, Et O re-
4
tion with Et O, drying of the etheral solution with
2
MgSO and removal of Et O, allowed the isolation of
4
2
cis-2 after chromatography with a PE 40/60 ethyl ace-
tate gradient in 47% yield. If the oxidation is performed
at −78°C for 3 h, the yield of cis-2 is 90%. At 0°C,
91% cis-2 is isolated after 1 h. trans-2 is not found in
any of these reactions (analysis by analytical GC) [22].
1
H-NMR(CDCl ): l=0.99 (t, J=6.6 Hz, 1 H, CH2
3
2
2
(
trans)), 1.26 (dd, Jgem=6.04 Hz, Jtrans=3.04 Hz, 1 H,
moved, and the alcohol (phenol) analysed by NMR and
GC [22] (comparison with authentic sample if possible).
CH2 (cis)), 1.33 (s, 3 H, CCH ), 1.70 (s, 3 H,
3
C(O)CH ), 4.09 (dd, J =6.8 Hz, J =3.4 Hz, 1 H,
trans
3
cis
13
CH(O)), 7.15–7.24 (m, 5 H, Ph–H);
C-NMR
i
t
(
CDCl ): l=17.6 (C10), 20.6 (C9), 25.4 (C2), 26.5
4.16. Oxidation with Ti(O Pr) + BuOOH (method B)
3
4
(
C3), 57.4 (C1), 126.4, 127.1, 129.0, 141.0 (Ph–C),
171.8 (C8).
The RM (Tables 1–3) were reacted with two molecu-
lar equivalents of Ti(O Pr)4 at −78°C and then
i
warmed to 0°C for 15 min. After cooling to −78°C,
t
4.13. trans-2-Methyl-2-phenyl-cyclopropyl acetate
one molecular equivalent of BuOOH was added. The
(
trans-2)
results of Table 3 were received after 1 h reaction time
at −78°C and trapping of the alcoholate with acetic
anhydride. The synthetically oriented reactions of Ta-
bles 1 and 2 were run for 12 h at −78°C and then
warmed to r.t. For the isolation of the alcohols (phe-
nols) the reaction mixture was hydrolysed with 2 N
The preparation followed the preparation of cis-2
except that trans-1-Li was used. Isolated yields after
column chromatography with a PE 40/60 ethyl acetate
gradient: at −78°C, 1 h: 37%; at −78°C, 3 h: 84%; at
0°C, 1 h: 82%. Cis-2 is not formed in any of these
HCl, the alcohols (phenols) extracted with Et O, the
2
reactions (analytical GC) [22].
solution dried with MgSO , Et O removed, and the
alcohol (phenol) analysed by NMR and GC [22] (com-
parison with authentic sample, if possible).
4
2
1
H-NMR(CDCl ): l=0.89 (dd, Jtrans=3.07 Hz,
3
Jgem=6.03 Hz, 1 H, CH (cis)), 1.23 (bt, J=6.8 Hz, 1
2
H, CH (trans)), 1.32 (s, 3 H, CCH ), 2.07 (s, 3 H,
2
3
C(O)CH ); 4.11 (dd, J
CH(O)), 7.11–7.34 (m, 5 H, Ph–H);
=3.7 Hz, J =7.3 Hz, 1 H,
4.17. Oxidation with dioxygen O (method C)
3
trans
cis
2
13
C-NMR
(
CDCl ): l=17.7 (C10), 20.0 (C9), 20.7 (C3), 26.0
Dioxygen O was bubbled through the solutions of
3
2
(
C2), 59.0 (C1), 126.4, 127.5, 127.8, 144.8 (Ph–C),
the RM listed in Table 3 for 1 h at −78°C. Then the
171.8 (C8).
reactions were stopped with acetic acid anhydride and

5
2
M. M o¨ ller et al. / Journal of Organometallic Chemistry 624 (2001) 47–52
5
302. (f) H. Neumann, D. Seebach, Chem. Ber. 111 (1978) 2782
worked up as outlined in the preparation of cis-2 and
analysed by NMR and GC [22].
also used oxenoids LiOOR to oxidize vinyllithium compounds.
t
[
7] BuOOH should react with Ti(OiPr) to give tBuOOTi(OiPr) (or
4
3
a dimer, thereof) and isopropanol [2] which is also acidic.
Apparently the reaction of the organometallic species RM with
the titanium peroxide [8] is much faster than the protonation of
RM by the available proton sources. Similar behaviour of RLi in
other competition reactions is known [9].
t
4
.18. Preparation of BuOOH
The commercial product (Peroxo-Chemie GmbH) is
a 80% solution in water. It is extracted with CH Cl2
and dried over MgSO for 3 days. Removal of the
solvent occurs at 400 mm. At 10 mm, BuOOH distills
2
[
8] Solid state structure and oxidation reactions of a Ti-peroxide: G.
Boche, K. M o¨ bus, K. Harms, M. Marsch, J. Am. Chem. Soc.
118 (1996) 2770.
4
t
at 30°C.
[9] Other reactions of RLi in the presence of acidic protons: (a)
H.W. Whitlock, B.J. Whitlock, R.J. Boatman, J. Am. Chem.
Soc. 99 (1977) 4822. (b) P. Beak, T.J. Musick, C.-W. Chen, J.
Am. Chem. 110 (1988) 3538. (c) W.F. Bailey, J.J. Patricia, T.T.
Nurmi, W. Wang, Tetrahedron Lett. 27 (1986) 1861. (d) N.S.
Narasimhan, N.M. Sunder, R. Ammanamanchi, B.D. Bonde, J.
Am. Chem. Soc. 112 (1990) 4931. (e) A. Schmidt, G. K o¨ brich,
R.W. Hoffmann, Chem. Ber. 124 (1991) 1253.
t
4.19. Preparation of BuOOLi
t
To a solution of BuOOH in THF at −78°C, was
added slowly one molecular equivalent of a 1.6 N
solution of BuLi in hexane. After 20 min at −78°C
the solution is ready for use.
n
t
[
[
[
10] 2- Butylperoxy-1,3,2-dioxaborolan was used for such oxidations
by: R.W. Hoffmann, K. Dittrich, Synthesis (1983) 107. See Ref.
[
11].
11] (a) C.W. Porter, C. Steel, J. Am. Chem. Soc. 42 (1920) 2650. (b)
C. Walling, S.A. Buckler, J. Am. Chem. Soc. 77 (1955) 6032. (c)
S.-O. Lawesson, N.C. Yang, J. Am. Chem. Soc. 81 (1959) 4230.
12] 5% bibenzyl as described in Ref. [6b] are not due to the oxida-
Acknowledgements
This work was supported by the Fonds der Chemis-
chen Industrie and the Deutsche Forschungsgemein-
schaft, Sonderforschungsbereich 260.
tion but to the preparation of PhCH MgBr from benzyl bromide
2
and magnesium.
[13] Oxidation of Zn compounds: (a) P. Knochel, I. Klement, H.
L u¨ tjens, Tetrahedron Lett. 36 (1995) 3161. (b) I. Klement, P.
Knochel, Synlett (1995) 1113.
[
14] (a) G.M. Whitesides, J. San Filippo, J.P. Casey, E.J. Panek, J.
Am. Chem. Soc. 89 (1967) 5302. (b) B.H. Lipshutz, K. Sieg-
mann, E. Garcia, F. Kayser, J. Am. Chem. Soc. 115 (1993) 9276.
References
[
1] (a) Shell Oil, Eng. Patent 1, 249, 079, 1971; Chem. Abstr. 74
1971) 12981m. (b) R.A. Sheldon, J. Mol. Catal. 7 (1980) 107. (c)
P.A. Jacobs, L. DeBruycker, D.R.C. Huybrechts, Nature 345
1990) 240. (d) T. Tatsumi, M. Nakamura, S. Negishi, H.
[15] G. Boche, K. M o¨ bus, K. Harms, J.C.W. Lohrenz, M. Marsch,
Chem. Eur. J. 2 (1996) 604.
[16] This was also observed by: Bettina H o¨ lzer, Ph.D. Thesis, in
progress, R.W. Hoffmann, personal communication, Marburg,
2000.
[17] (a) D. Schneider, Dissertation, Universit a¨ t M u¨ nchen, 1977. (b)
D. Seyferth, H. Yamazaki, D.L. Alleston, J. Org. Chem. 28
(1963) 703. (c) C.L. Osborn, T.C. Shields, B.A. Shoulders, J.F.
Krause, H.V. Cortez, P.D. Gardner, J. Am. Chem. Soc. 87
(1965) 3158. (d) J.W. Hauser, M.J. Grubber, J. Org. Chem. 37
(1972) 2648. (e) D. Seebach, R. H a¨ ssig, J. Gabriel, Helv. Chem.
Acta 66 (1983) 308. (f) T. Hirao, T. Hasegawa, Y. Ohshiro, I.
Ikeda, J. Org. Chem. 58 (1993) 6529.
(
(
Tominaga, J. Chem. Soc. Chem. Commun. (1990) 476. (e) W.
Adam, A. Corma, T. Indrasena Reddy, M. Renz, J. Org. Chem.
6
2 (1997) 3631. (f) I.W.C.W. Arends, R.A. Sheldon, M. Wallau,
U. Schuchardt, Angew. Chem. 109 (1997) 1190; Angew. Chem.
Int. Ed. Engl. 36 (1997) 1144.
[
[
2] (a) T. Katsuki, K.B. Sharpless, J. Am. Chem. Soc. 102 (1980)
5
974. (b) M.G. Finn, K.B. Sharpless, in: J.D. Morrison (Ed.),
Asymmetric Synthesis, Vol. 5, Aademic Press, New York, 1986,
p. 247. (c) M.G. Finn, K.B. Sharpless, J. Am. Chem. Soc. 113
(
1991) 113.
3] (a) P. Pitchen, H.B. Kagan, Tetrahedron Lett. 25 (1984) 1049.
b) C. Girard, H.B. Kagan, Angew. Chem. 110 (1998) 3088;
[18] E.A. Dehmlow, Angew. Chem. 89 (1977) 521; Angew. Chem.
Int. Ed. Engl. 16 (1977) 493.
(
[19] Stereochemistry: N.I. Yakushkina, G.A. Zakharova, L.S. Sur-
mina, I.G. Bolesov, J. Org. Chem. USSR 19 (1980) 1553.
[20] The stereoselective preparation, configurational stability and
stereochemistry of the reaction of cyclopropyllithium and cyclo-
propyl Grignard compounds with electrophiles is discussed in
detail in: G. Boche, H.M. Walborsky, in: S. Patai, Z. Rappoport
(Eds.), Cyclopropane Derived Reactive Intermediates, Wiley,
New York, 1990.
[21] The stereoselective formation, configurational stability and reac-
tion with electrophiles with retention of configuration of cyclo-
propyl cuprates has been studied by: (a) K. Kitatani, T. Hiyama,
H. Nozaki, Bull. Chem. Soc. Jpn 1977, 50, 1600. (b) H. Ya-
mamoto, K. Kitatami, T. Hiyama, H. Nozaki, J. Am. Chem.
Soc. 1977, 99, 5816. (c) J.P. Marino, L.J. Brown, J. Org. Chem.
41 (1976) 3629.
Angew. Chem. Int. Ed. Engl. 37 (1998) 2922; this review article
summarizes the publications of H.B. Kagan on this topic. For
further reactions of this type: (c) M. Bouchio, S. Calloni, F. Di
Furia, G. Licini, G. Modena, S. Moro, W.A. Nugent, J. Am.
Chem. Soc. 119 (1997) 6935 cit. lit. (d) N. Komatsu, M.
Hashizume, T. Sugia, S. Uemura, J. Org. Chem. 58 (1993) 4529.
4] M. Schulz, R. Kluge, M. Sch u¨ ler, G. Hoffmann, Tetrahedron 51
[
[
[
(
1995) 3175.
5] (a) W. Adam, K. Peters, M. Renz, J. Org. Chem. 62 (1997) 3183.
b) W. Adam, M.J. Richter, Acc. Chem. Res. 27 (1994) 57.
(
6] (a) M. Julia, V. Pfeuty-Saint-Jalmes, K. Ple, J.-N. Verpeaux, in
part G. Hollingworth, Bull. Soc. Chim. Fr. 133 (1996) 15, earlier
lit. cit. (b) G. Boche, F. Bosold, J.C.W. Lohrenz, Angew. Chem.
1
06 (1994) 1228; Angew. Chem. Int. Ed. Engl. 33 (1994) 1161. (c)
E. M u¨ ller, T. T o¨ pel, Chem. Ber. 72 (1939) 273. (d) P. Warner,
S.-L. Lu, J. Org. Chem. 41 (1976) 1459. (e) G.M. Whitesides, J.
San Filippo, J.P. Casey, E.J. Panek, J. Am. Chem. Soc. 89 (1967)
[22] For GC analytical studies the reaction mixture was diluted with
−
1
Et O (c=0.025 g ml ) after the extraction workup procedure.
2
The yields were determined by calculation from calibration plots.