Stereocontrol in ene-dimerisation and trimerisation of 1-trimethylsilyl-3- phenylcyclopropene

Article abstract of DOI:10.1016/j.tetlet.2005.12.111

1-Trimethylsilyl-3-phenylcyclopropene undergoes a highly stereocontrolled ene-reaction to give a dimer and further reaction leads to one or more trimers derived through two ene-reactions.

Full text of DOI:10.1016/j.tetlet.2005.12.111

Tetrahedron
Letters
Tetrahedron Letters 47 (2006) 2839–2843
Stereocontrol in ene-dimerisation and trimerisation
of 1-trimethylsilyl-3-phenylcyclopropene
a
b,
b
c
*
Andrey E. Sheshenev, Mark S. Baird, Anna K. Croft, Zoya A. Starikova,
d
e
a
Alexandre S. Shashkov, Alexey L. Zhuze and Ivan G. Bolesov
a
Department of Chemistry, Moscow State University, Moscow 119899, Russia
Department of Chemistry, University of Wales, Bangor, Gwynedd LL57 2UW, UK
b
c
A. N. Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, Vavilov Street 28, Moscow 119991, Russia
d
N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, Leninsky Prospect 47, Moscow 117913, Russia
e
V. A. Engelhardt Institute of Molecular Biology, Russian Academy of Sciences, Vavilov Street 32, Moscow 119991, Russia
Received 7 November 2005; revised 12 December 2005; accepted 21 December 2005
Available online 3 March 2006
Abstract—1-Trimethylsilyl-3-phenylcyclopropene undergoes a highly stereocontrolled ene-reaction to give a dimer and further reac-
tion leads to one or more trimers derived through two ene-reactions.
Ó 2006 Elsevier Ltd. All rights reserved.
There has been considerable recent interest in biologi-
cally active compounds containing an array of contigu-
ous cyclopropanes. The synthesis of such molecules in a
introduction of substituents, the stereochemistry (and
particularly the absolute stereochemistry) is more diffi-
cult to control.
1
3–10
stereocontrolled manner has generally been achieved by
2
stepwise introduction of each ring. The ene-dimerisa-
1,3-Diphenylcyclopropene is known to undergo a rapid
ene-reaction to give a single dimer, though the stereo-
tion of cyclopropenes leads initially to a cyclopropyl-
cyclopropene. Formally this reaction could be repeated
with further molecules of cyclopropene leading to the
formation of a single polycyclopropane. In practice, to
achieve such a series of reactions would require an extre-
mely high degree of stereo- (exo- or endo-transition
state) and regio-control in each step and in addition
would require the rate of reaction of the dimer 1
1
1
chemistry of the process is not clear. In the case of
1-chloro-3-trimethylsilylcyclopropene 2, a single dimer
3 is formed apparently, though not explicitly stated, by
the reaction of two enantiomeric molecules through an
exo-transition state, rather than two identical molecules,
which would require an endo-transition state and lead to
1
2a
4
(Scheme 2). However, during the course of this
(
Scheme 1) with cyclopropene (only one possible regio-
work, 1-phenylcyclopropene was reported to undergo
ene-dimerisation and trimerisation, in each case through
an endo-transition state.
and stereochemistry is shown) to be considerably faster
than the dimerisation of this species itself. The known
examples of such reactions for substituted cycloprop-
enes suggest that, although the regiochemistry in dimer-
isations can be largely controlled by appropriate
1
2b
H
H
H
Cl
H
H
H
H
TMS
exo-
H
TMS
TMS
Cl
TMS
H
TMS
Cl
H
H
endo-
TMS
H
Cl
H
TMS
H
H
Cl
Cl
Cl
Cl
TMS
2
3
4
?
H
H
H
Scheme 2.
H
H
1
Scheme 1.
We have recently reported the preparation and studies
1
3
of the chemistry of 3-phenylcyclopropene, and some
of its derivatives including 1-trimethylsilyl-3-phenyl-
cyclopropene 6. We now report the ene-dimerisation
*
Corresponding author. Tel.: +44 1248382374; e-mail: chs028@
bangor.ac.uk
0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2005.12.111

2840
A. E. Sheshenev et al. / Tetrahedron Letters 47 (2006) 2839–2843
and trimerisation of 6. The tribromide 5 was treated
with 2.2 mol equiv of MeLi at À80 °C, allowed to reach
room temperature for 30 min to form the 1-lithio-3-phe-
simply reflect a steric effect slowing the migration of
the trimethylsilyl group in the cis-case and allowing
2
1
the normal cyclopropylidene–allene rearrangement to
compete; clearly, however, the intermediate cannot be
a symmetrical free carbene or both isomers would give
the same product. Moreover, a more subtle effect based
on the stereochemistry of the lithium–bromine exchange
or on the formation of a more complex intermediate is
also possible.
1
4
nylcyclopropene by lithium–halogen exchange, and
then quenched with trimethylsilyl chloride to give cyclo-
1
5
propene 6 that could be trapped in 30% yield (from 5)
1
6
as 7 by reaction with diphenylisobenzofuran (DPIBF).
This reaction also produced a single dimer 8 of the
cyclopropene; this could be isolated in 40% yield in the
1
7
absence of the trap (Scheme 3). The regio- and stereo-
chemistry of 8 is established later (only one enantiomeric
series is shown).
Ph
Ph
Ph
1
.15 mol eq. MeLi
Cpd
Br
Br
TMS
Ph
5, 92%
-
90 oC to -50 oC (20 min)
+
rt, 2 h
Et2O
TMS
TMS
6, 96%
by 1H NMR
TMS
1
1
4
1
0, 4%
1
. MeLi, 2.2 mol eq.,
Ph
Ph
Br
Ph
Br
Br
TMSCl
60 o
to RT
Et2O, -80 oC to RT,
-
C
(20 min)
Scheme 5.
Li
TMS
5
2. RT, 30 min.
6, 30%
by 1H NMR
DPIBF
+
RT
If the cyclopropene was worked up at 20 °C in the absence
of a trap, the dimer 8 was again obtained, in this case
Hc
O
Ph
Ha
RT, 1 h
Hd
2
2
Ph
TMS
accompanied by two trimers, 16 and 17 (Scheme 6).
TMS
Ph
TMS
Ph
Hb 8, 40%
The dimer was efficiently trapped by reaction with cyclo-
Ph
7, 30%
22
pentadiene to give a crystalline adduct 18. Under these
conditions neither of the trimers reacted with the diene.
Scheme 3.
TMS
In order to better understand this process, a route to the
cyclopropene 6 was required which could be carried out
at low temperature to minimize dimer formation. Reac-
Ph
TMS
H
H
Ph
Ph
H
TMS
H
20 oC, 12 h
Et2O
TMS
H
H
+
H
Ph
Ph
TMS
Ph
1
8
H
TMS
tion of 9 with methyllithium was expected to lead to a
,2-silyl shift in an intermediate cyclopropylidene to give
6
(
racemic)
H
H
1
8, 49%
16, 31%
1
9
Cpd
PhH
cyclopropene 6. However, when the reaction was
carried out at À90 °C and quenched with water at À50
0 oC, 24 h
+
2
H
H
TMS
Ph
°
C, an essentially quantitative yield of allene 10 (Scheme
Ph
H
4
) was obtained. At room temperature, the major prod-
TMS
H
H
H
TMS
uct was the acetylene 11, together with allene 10 and a
small amount of the dimer 8.
TMS
H
Ph
Ph
H
H
18, 88%
Ph
TMS
7, 10% by 1H NMR
1
Br Br
Ph
Scheme 6.
MeLi, 1.15 mol eq.
90 oC to -50 o
.
+
8, 1%
-
C
TMS
Ph
Ph
TMS
1
0, 97%
9
Li Br
..
TMS
Ph
TMS
H
Ph
TMS
13
11
12
Scheme 4.
In contrast, the trans-isomer 14 reacted with methyl-
lithium at À90 °C followed by quenching with water at
À60 °C to give the cyclopropene 6 as the major product
(
Scheme 5), together with ca. 4% of the allene 10. The
cyclopropene 6 was again unstable but could be detected
2
0
directly by NMR at À40 °C. The cyclopropene could
be trapped as a single (4+2)-cycloadduct 15 with cyclo-
pentadiene (CPD) in 92% yield based on 6. The forma-
tion of 6 presumably occurs through intermediates 12
and 13 by a lithium–bromine exchange followed by a
formal loss of lithium bromide to give a cyclopropylid-
ene which rearranges by a 1,2-trimethylsilyl shift. The
different outcomes from the cis- and trans-isomers may
Figure 1. X-ray crystal structure of CPD cycloadduct 18.

A. E. Sheshenev et al. / Tetrahedron Letters 47 (2006) 2839–2843
2841
The structure of the adduct 18 (Fig. 1) shows that the di-
TMS
mer 8 was formed from two identical molecules of 6,
rather than two enantiomeric molecules, and that the
reaction occurs through an endo-transition state
H
TMS
H Ph
H
H
H
Ph
H
H
TMS
H
TMS
H
Ph
H
H
Ph
TMS
(
Scheme 7). The formation of one isomer of 8 from 6
TMS
Ph
Ph
creates eight chiral centres in 18 from the single chiral
centre in 6.
Ph
H
H
HPh
Ph
TMS
H
Ph
Ph
TMS
Ph
H
TMS
TMS
H
TMS
H
H
16
H
H
TMS
Ph
H
H
endo-
TMS
H
Ph
H
H
Scheme 9.
TMS
Ph
H
6
TMS
H
8
H
appears to require an exo-transition state (Scheme 10).
The alternative endo-transition state would lead to a
product in which the phenyl and trimethylsilyl groups
on the centre ring were reversed, that is, that involved
C–C bond formation to the phenyl-substituted carbon
in the cyclopropene of the dimer. These two structures
could be distinguished on the basis of the position of
the CHPh group in the proton NMR spectrum (d 1.8),
H
H
H
TMS
Ph
H
Ph
exo-
H
Ph
TMS
H
H
TMS
TMS
Ph
Scheme 7.
the corresponding signal for a CHSiMe group appear-
ing at about d 0.5 in other cases.
To exploit this, the corresponding trimethylsilylcyclo-
propene was obtained from the optically active acid 19
as shown below (Scheme 8). Surprisingly, the dimer
was isolated in only low yield; trimers 16 and 17 were
identified by NMR in a complex mixture with other
products. The reason for the difference between the reac-
tions of the racemic and optically active cyclopropenes is
the subject of further analysis; however, the isolation of
the two trimers from optically active monomer indicates
that they are made up of three identical monomer mole-
cules and not of enantiomeric molecules. Trapping of
3
H
H
TMS
TMS
H
Ph
Ph
H
Ph
H
H
H
TMS
H
TMS
TMS
H
H
TMS
Ph
Ph H H
Ph
H
H
H
(
+)-8 with cyclopentadiene gave a 64% yield of the
H
Ph
Ph
TMS
adduct (+)-18 of the corresponding enantiomerically
enriched cyclopropene.
Ph
H
TMS
TMS
Scheme 10.
Ph
Ph
S
Br
Br
Ph
Br
Br
H
R
TMS
S
S
H
1. HgO/Br2 S
2
. MeLi
CH2Cl2
3. TMSCl
H
H
H
Ph
R
H
CO2H
Br
The trans-isomer 14 reacted with 2.15 mol equiv of
methyllithium at À90 °C followed by quenching with
methyl iodide, trimethylsilyl chloride or carbon dioxide
which led to a lithium–hydrogen exchange in the inter-
mediate cyclopropene 6 to produce 20 which was
trapped by electrophiles to give the corresponding 1,2-
disubstituted 3-phenylcyclopropenes 21–24 mostly in
good yields (Scheme 11).
H
(
-)-(1S, 3S)-19, 91% ee
TMS
(+)-8, 9%
Cpd, benzene
o
Ph
R
20 C, 24 h
R
TMS
R
S
R
H
S
R
H
H
Ph
(
+)-18, 64% H
S
TMS
Scheme 8.
Br
Br
TMS
TMS
The regiochemistry of the major trimer 16 was assigned
based on an endo-transition state as in the ene-dimerisa-
tion, on the basis of the above requirement that the same
enantiomer of the cyclopropene must add to the dimer
MeLi,
E - X
2
.15 mol eq.
Ph
TMS
Ph
Li
Ph
6, 21 - 24
E
14
20
22
(
Scheme 9).
No
E
%
6
H
21
D
-
23
TMS
90
24
Me
97
COOH
It is consistent with the stereochemistry recently
94
57
1
2b
reported for the ene-trimer of 1-phenylcyclopropene.
On the other hand, the structure of the minor isomer
Scheme 11.

2842
A. E. Sheshenev et al. / Tetrahedron Letters 47 (2006) 2839–2843
Attempts to trap the intermediate 20 with deuterated
5. Dowd, P.; Gold, A. Tetrahedron Lett. 1969, 85.
6
. Billups, W. E.; Lee, G.-A.; Arney, B. E.; Whitmire, K. H.
J. Am. Chem. Soc. 1991, 113, 7980.
water, however, did not produce the deuterated cyclopro-
2
3
24
pene 21. Instead, the dimer 25 and trimer 26 (Scheme
2) were observed. Even under the same conditions under
7
. Breslow, R.; Dowd, P. J. Am. Chem. Soc. 1963, 85, 2729.
1
1
8. Lee, G.-A.; Shiau, C.-S.; Chen, J. J. Org. Chem. 1995, 60,
565.
. Garratt, P. J.; Tsonitis, A. J. Org. Chem. 1990, 55, 84.
0. Baird, M. S.; Hussain, H. H.; Clegg, W. J. Chem. Res.
M) 1988, 1101.
which 6 had been observed directly by H NMR, no 21
could be observed, suggesting that the ene-dimerisation
of the 2-D compound proceeds somewhat faster than that
of the 2-H compound. This might result from a secondary
isotope effect in the dimerisation which converts two
pseudo-sp C–H(D) bonds into cyclopropane C–H(D)
bonds. Ab initio calculations have been preformed at
the B3LYP/6-31G* level of theory for a model ene-
dimerisation of 1,2-dideuteriocyclopropene by both the
3
9
1
(
11. Komatsu, K.; Niwa, T.; Akari, H.; Okamoto, K. J. Chem.
Res. (M) 1985, 2847; Durr, H. Chem. Ber. 1970, 103, 369;
Boche, G. Chem. Ber. 1979, 112, 2961.
1
2. (a) Lee, G.-A.; Chen, C.-S. Tetrahedron Lett. 1997, 38,
717; (b) Lee, G.-A.; Chang, C.-Y. J. Org. Chem. 2004, 69,
8
2
5
8949.
exo- and endo-transition states. In both cases, the deu-
terium isotope effect for the dimerisation reaction, k /k ,
1
1
1
3. Sheshenev, A. E.; Baird, M. S.; Bolesov, I. G.; Croft, A.
K. Mendeleev Commun. 2004, 299.
4. Baird, M. S.; Hussain, H. H.; Nethercott, W. J. Chem.
Soc., Perkin Trans. 1 1986, 1845.
H
D
2
6
was calculated at 0.86, in line with the effect expected
for a normal inverse secondary isotope effect. Further
calculations are being carried out to see if this difference
is amplified in more substituted systems. This possible
explanation, and the observation that only one trimer
was observed from 21, whereas two were formed in a
H
5. Cyclopropene 6 was a colourless oil, d (500 MHz)
(À40 °C): 0.28 (9H, s), 2.61 (1H, s), 7.17–7.56 (6H, m);
d
(125 MHz): À1.3+, 20.2+, 116.7, 119.6+, 124.7+,
C
À1
125.0+, 127.7+, 148.4; mmax (film, cm ): 3025 m, 2955 s,
2
897 m, 1693 s, 1602 m, 1492 m, 1446 s, 1248 s, 841 s, 756
3
:1 ratio from 6 are currently being investigated.
s, 698 s.
1
1
6. Baird, M. S. Houben-Weyl, Methods Org. Chem. 1997, E
17d, 182.
H
D
H
7. Dimer 8 was a colourless oil, d
H
(500 MHz): À0.28 (9H, s),
TMS
TMS
Ph
À0.01 (1H, dd, J 7.3, 10.4 Hz), 0.36 (9H, s), 1.45 (1H, ddd,
Ph
Ph
J 4.1, 4.7, 7.3 Hz), 1.96 (1H, d, J 4.1 Hz), 2.20 (1H, dd, J
D O
D
2
Li
D
Ph
4
.7, 10.4 Hz), 7.23–7.62 (10H, m); d
C
(125 MHz): À0.9+,
2
5
À0.6+, 14.4+, 23.5+, 25.9+, 27.6+, 116.8, 125.6+,
Ph
+
TMS
TMS
127.8+, 128.63+, 128.65+, 129.2+, 129.4+, 130.7, 136.9,
D
À1
H TMS
H
141.5; mmax (film, cm ): 2954 s, 2897 m, 1768 m, 1602 m,
2
0
21
Ph
H Ph
D
1488 m, 1446 m, 1248 s, 930 m, 912 m, 840 s, 756 s, 691 s.
TMS
18. Cyclopropanes 9 and 14 were prepared by dibromocyclo-
propanation of the corresponding trans- or cis-1-phenyl-2-
2
6
D
TMS
trimethylsilylethenes using CHBr
cetyltrimethylammonium bromide, and CH
and 90%, respectively). Compound 14, a colourless oil,
(500 MHz): 0.13 (9H, s), 1.48 (1H, d, J 13.2 Hz), 3.24
1H, d, J 13.2 Hz), 7.28–7.46 (5H, m); d (125 MHz):
3
, 50% NaOH–H
2
O,
Scheme 12.
2
Cl (88%
2
d
(
H
Acknowledgements
C
À0.06+, 30.4+, 32.3, 38.9+, 127.3+, 127.9+, 130.2+,
This work was carried out with the support of an IN-
TAS Grant.
À1
1
35.8; m
(film, cm ): 2955 s, 2897 s, 1496 s, 1445 s, 1253
max
s, 1030 s, 1019 s, 965 s, 842 s, 803 s, 766 s, 732 s, 697 s, 654
s. Compound 9, a white powder, mp 20–21 °C, showed d
500 MHz): 0.30 (9H, s), 1.36 (1H, d, J 10.7 Hz), 2.85 (1H,
H
(
References and notes
d, J 10.7 Hz), 7.29–7.42 (5H, m); dC (125 MHz): À1.33+,
2
7.4+, 34.5, 39.1+, 127.5+, 128.2+, 128.7+, 137.1; mmax
À1
1
. Barrett, A. G. M.; Kasdorf, K.; White, A. J. P.; Williams,
D. J. J. Chem. Soc., Chem. Commun. 1995, 649; Barrett, A.
G. M.; Doubleday, W. W.; Kasdorf, K.; Tustin, G. J.;
White, A. J. P.; Williams, D. J. J. Chem. Soc., Chem.
Commun. 1995, 407; Barrett, A. G. M.; Kasdorf, K.;
Tustin, G. J.; Williams, D. J. J. Chem. Soc., Chem.
Commun. 1995, 1143; Barrett, A. G. M.; Kasdorf, K. J.
Chem. Soc., Chem. Commun. 1996, 325; Falck, J. R.;
Mekonnen, B.; Yu, J.; Lai, J.-Y. J. Am. Chem. Soc. 1996,
(film, cm ): 2953 s, 1496 s, 1448 s, 1409 s, 1250 s, 1034 s,
1024 s, 954 s, 846 s, 755 s, 695 s.
19. Baird, M. S.; Dale, C. M.; Al Dulayymi, J. R. J. Chem.
Soc., Perkin Trans. 1 1993, 1373.
20. Compound 6 was relatively stable at À40 °C (an 11%
solution in CDCl
of 50:41:5 monomer–dimer–trimer
3
immediately after reaction rearranged to a 23:56:14
mixture in 27 h) but had reacted completely in 18 h either
neat or as a 7% ethereal solution.
1
18, 6096; Barrett, A. G. M.; Kasdorf, K. J. Am. Chem.
21. Backes, J.; Brinker, U. H. Houben-Weyl, Methoden der
Organischen Chemie. Band 19b. Part 1, 1989; S. 391–
510.
Soc. 1996, 118, 11030; Barrett, A. G. M.; Doubleday, W.
W.; Kasdorf, K.; Tustin, G. J. J. Org. Chem. 1996, 61,
3280; Barrett, A. G. M.; Hamprecht, D.; White, A. J. P.;
22. Compound 16 (30% after chromatography) was a colour-
Williams, D. J. J. Am. Chem. Soc. 1997, 119, 8608.
. See, for example: Verbicky, C. A.; Zercher, C. K.
Tetrahedron Lett. 2000, 41, 8723; McDonald, W. S.;
Verbicky, C. A.; Zercher, C. K. J. Org. Chem. 1997, 62,
less oil, d
H
(500 MHz): À0.37 (9H, s), À0.16 (1H, dd, J 6.9,
2
10.4 Hz), À0.14 (9H, s), 0.14 (9H, s), 0.78 (1H, dd, J 6.6,
10.7 Hz), 1.80 (1H, dd, J 5.1, 6.9 Hz), 2.12 (1H, dd, J 5.1,
10.4 Hz), 2.59 (1H, dd, J 4.7, 6.6 Hz), 2.79 (1H, dd, J 4.7,
1
215.
10.7 Hz), 7.12–7.59 (15H, m); d
C
(125 MHz): À0.96+,
3
4
. Shi, G.-q.; Huang, X.-h. Tetrahedron Lett. 1996, 37, 106.
. Eyman, W.; Hanack, M. Tetrahedron Lett. 1972, 4213.
À0.72+, À0.65+, 13.1+, 17.78+, 17.84+, 23.2, 26.1+,
28.8+, 31.6+, 110.6, 123.3, 125.6+, 126.5+, 126.8+,

A. E. Sheshenev et al. / Tetrahedron Letters 47 (2006) 2839–2843
24. Trimer 26 was a colourless oil, d
2843
1
1
1
27.8+, 128.1+, 128.2+, 128.6+, 129.3+, 131.0, 131.1+,
H
(500 MHz): À0.37 (9H,
À1
39.2, 141.6; mmax (film, cm ): 1840 m, 1601 m, 1496 m,
445 m, 1247 s, 980 m, 909 s, 839 s, 756 s, 735 s, 699 s.
s), À0.14 (9H, s), 0.15 (9H, s), 0.77 (1H, d, J 10.7 Hz), 2.13
(1H, s), 2.79 (1H, d, J 10.7 Hz), 7.13–7.47 (15H, m); dC
(125 MHz): À0.97+, À0.75+, À0.68+, 12.9+, 17.6+,
17.7+, 22.9, 26.0+, 28.6+, 31.5+, 110.5, 123.2, 125.5+,
126.4+, 126.8+, 127.8+, 128.06+, 128.16+, 128.6+,
Compound (17) isolated as a mixture with the trimer 16
showed d
(500 MHz): À0.19 (9H, s), 0.20 (9H, s),
.30 (1H, dd, J 6.9, 10.4 Hz), 0.37 (9H, s), 1.53 (1H,
ddd, J 4.4, 4.7, 6.9 Hz), 1.79 (1H, d, J 8.8 Hz), 2.28 (1 H,
dd, J 4.7, 10.4 Hz), 2.48 (1H, s), 7.13–7.65 (15H, m); d
H
0
À1
129.3+, 131.0, 131.1+, 139.2, 141.5; mmax (film, cm ):
C
3079 m, 3059 m, 3027 s, 2952 s, 2895 m, 1840 s, 1601 s,
1496 s, 1445 s, 1403 m, 1247 s, 1071 m, 1030 m, 984 m, 966
m, 894 m, 837 s, 756 s, 698 s.
(
1
1
125 MHz) including: À0.28+, À0.06+, 0.83+, 18.9+,
9.4+, 20.6, 26.9+, 28.4+, 29.6+, 32.9+, 115.2, 125.62+,
25.75+, 130.9, 136.5, 139.7, 140.4. Complete crystallo-
25. The activation energies calculated for hydrogen-substi-
tuted cyclopropene dimerisations (via both endo- and exo-
transition states) were in good agreement with those
published recently (Deng, Q.; Thomas, B. E.; Houk, K.
N.; Dowd, P. J. Am. Chem. Soc. 1997, 119, 6902).
26. Using unscaled frequencies, the isotope effect was calcu-
lated at 300 K using the Bigeleisen equation (Bigeleisen,
J.; Goeppert-Mayer, M. J. Chem. Phys. 1947, 15, 261) as
presented in; Olsen, L. P.; Li, Y.; Houk, K. N.; Kresge, A.
J.; Schaad, L. J. J. Am. Chem. Soc. 1995, 117, 2992; Bell’s
correction for tunnelling (Bell, R. P. Trans. Faraday Soc.
1959, 55, 1) is not large in this case.
graphic data for compound 18 have been deposited with
the Cambridge Crystallographic Data Centre (registration
no. 283161).
2
3. Dimer 25 was a colourless oil, d
s), À0.01 (1H, d, J 10.4 Hz), 0.37 (9H, s), 2.21 (1H, d, J
0.4 Hz), 7.18–7.64 (10H, m); d
(125 MHz): À0.8+,
25.7 Hz), 25.4+ (t, J
H
(500 MHz): À0.26 (9H,
1
C
À0.6+, 14.3+, 22.9+ (t, J
CD
CD
2
1
2
3.5 Hz), 27.5+, 116.6, 125.6+, 127.8+, 128.63+, 128.65+,
À1
29.2+, 129.4+, 130.7, 136.8, 141.5; mmax (film, cm ):
954 s, 2896 m, 1766 s, 1602 s, 1488 m, 1446 m, 1248 s, 979
m, 916 m, 841 s, 755 s, 698 s.