Axial Chirality in 1,4-Disubstituted (ZZ)-1,3-Dienes. Surprisingly Low Energies of Activation for the Enantiomerization in Synthetically Useful Fluxional Molecules

Article abstract of DOI:10.1021/ja035136m

Trialkylsilyltrialkylstannes (R₃Si-SnR'₃) add to 1,6-diynes in the presence of Pd(0) and trispentaflurophenylphosphine to give 1,2-dialkylidenecyclopentanes with terminal silicon and tin substituents. The (ZZ)-geometry of these s-cis-1,3-dienes, resulting from the organometallic reaction mechanisms involved, forces the silicon and tin groups to be nonplanar, thus making the molecules axially chiral. There is rapid equilibration between the two helical forms at room-temperature irrespective of the size of the Si and Sn substituents. However, the two forms can be observed by ¹H, ¹³C, and ¹¹⁹Sn NMR spectroscopy at low temperature. The rates of enantiomerization, which depend on the Si and Sn substituents, and the substitution pattern of the cylopentane ring can be studied by dynamic NMR spectroscopy using line shape analysis. The surprisingly low energies of activation (ΔG^? = 52-57 kJ mol^-1) for even the bulky Si and Sn derivatives may be attributed to a widening of the exo-cyclic bond-angles of the diene carbons.

Full text of DOI:10.1021/ja035136m

Published on Web 11/20/2003
Axial Chirality in 1,4-Disubstituted (ZZ)-1,3-Dienes.
Surprisingly Low Energies of Activation for the
Enantiomerization in Synthetically Useful Fluxional Molecules
Sandra Warren, Albert Chow, Gideon Fraenkel, and T. V. RajanBabu*
Contribution from the Department of Chemistry, The Ohio State UniVersity,
100 W. 18th AVenue, Columbus, Ohio 43210
Received March 13, 2003; E-mail: rajanbabu.1@osu.edu
Abstract: Trialkylsilyltrialkylstannes (R3Si-SnR’3) add to 1,6-diynes in the presence of Pd(0) and tris-
pentaflurophenylphosphine to give 1,2-dialkylidenecyclopentanes with terminal silicon and tin substituents.
The (ZZ)-geometry of these s-cis-1,3-dienes, resulting from the organometallic reaction mechanisms
involved, forces the silicon and tin groups to be nonplanar, thus making the molecules axially chiral. There
is rapid equilibration between the two helical forms at room-temperature irrespective of the size of the Si
1
and Sn substituents. However, the two forms can be observed by H, 13C, and 119Sn NMR spectroscopy
at low temperature. The rates of enantiomerization, which depend on the Si and Sn substituents, and the
substitution pattern of the cylopentane ring can be studied by dynamic NMR spectroscopy using line shape
analysis. The surprisingly low energies of activation (∆Gq ) 52-57 kJ mol-1) for even the bulky Si and Sn
derivatives may be attributed to a widening of the exo-cyclic bond-angles of the diene carbons.
Introduction
Chirality incident in molecules with restricted rotation around
a single bond, commonly referred to as axial chirality, plays an
important role in the recognition events that lead to the biological
activity of many natural products.1 In asymmetric catalysis using
metal complexes of biaryl ligands such as 1,1′-binaphthyl deriv-
atives, the origin of enantioselectivity relies on this phenom-
enon.2 Although most synthetic applications of axial chirality
reported to date involve biaryl molecules, studies on other sys-
tems, especially amides with restricted rotation, are beginning
to appear in the literature.3 Other molecules of this class which
received early scrutiny were the highly substituted 1,3-dienes.
In 1967, Boer et al. reported4 that hindered rotation around the
2,3-bond in a 1,3-diene was frozen by intramolecular interaction
between an electrophilic Sn and an appropriately placed Br
(Figure 1, 1). Later, Ko¨brich demonstrated that barrier to rotation
(∆Gq) in hexasubstituted 1,3-butadienes such as 2,3-dihalo-1,4-
tetrabenzyl-1,3-butadiene vary with the size of the halogen (Fig-
ure 1, 2a∼2d).5 Isolation of optically active acyclic6 and cyclic7
Figure 1. Axially Chiral Dienes (Activation energies, ∆Gq, are shown in
brackets, kJ mol-1
.
dissymmetric dienes have also been reported. Free energies of
activation for enantiomerization in dissymmetric 1,2-bis-alky-
lidenecycloalkane derivatives have been determined by NMR
methods, and these results indicate a subtle relationship between
the ring substituents and the activation parameters. For example,
the ∆Gq for the N-chlorosulfonyl imine 37 is above 105 kJ
mol-1, whereas the related compound 48 undergoes facile isom-
erization at room temperature (∆Gq ) 42 kJ mol-1). The pre-
sence of heteroatoms (e.g., 5a, 5b) in the ring8 and size of the
ring9 also have profound effects on the rates of these processes.
(1) For two recent reviews, see: (a) Lloyd-Williams, P.; Giralt, E. Chem. Soc.
ReV. 2001, 30, 145. (b) Bringmann, G.; Breuning, M.; Tasler, S. Synthesis
1999, 525.
(2) Ojima, I., Ed. Catalytic Asymmetric Synthesis, 2nd ed.; Wiley-VCH: New
York, 2000. (b) Jacobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.;
ComprehensiVe Asymmetric Catalysis; Springer-Verlag: Berlin, 1999.
(3) (a) Eliel, E. L.; Wilen, S. H. Stereochemistry of Organic Compounds;
Wiley: New York, 1994. pp 1142-1166. For recent examples see, (b)
Kitagawa, O.; Izawa, H.; Sato, K.; Dobashi, A.; Taguchi, T. J. Org. Chem.
1998, 63, 2634. (c) Clyden, J. Synlett 1998, 810. (d) Curran, D. P.; Hale,
G. R.; Geib, S. J.; Balog, A.; Cass, Q. B.; Degani, A. L. G.; Hermandes,
M. Z.; Freitas, L. C. G. Tetrahedron: Asymmetry 1997, 8, 3955. (e)
Godfrey, C. R. A.; Simpkins, N. S.; Walker, M. D. Synlett. 2000, 388. (f)
Rios, R.; Jimeno, C.; Carroll, P. J.; Walsh, P. J. J. Am. Chem. Soc. 2002,
124, 10 272.
(5) Ko¨brich, G.; Mannschreck, A.; Misra, R. A.; Rissmann, G.; Ro¨sner, M.;
Zu¨ndorf, W. Chem. Ber. 1972, 105, 3794. See also, Ko¨brich, G.; Kolb, B.;
Mannschreck, A.; Misra, R. A. Chem. Ber. 1973, 106, 1601.; Mannschreck,
A.; Jonas, V.; Bo¨decker, H.-O.; Elbe, H.-L.; Ko¨brich, G. Tetrahedron Lett.
1974, 2153.; For another example, see, Bomse, D. S.; Morton, T. H.
Tetrahedron Lett. 1974, 3491.
(6) Ro¨sner, M.; Ko¨brich, G. Angew. Chem., Int. Ed. Engl. 1974, 13, 741.
(7) Pasto, D. J.; Borchardt, J. K. J. Am. Chem. Soc. 1974, 96, 6220.
(8) Jelinski, L. W.; Kiefer, E. F. J. Am. Chem. Soc. 1976, 98, 281.
(4) Boer, F. P.; Doorakian, G. A.; Freedman, H. H.; McKinley, S. V. J. Am.
Chem. Soc. 1970, 92, 1225. For a preliminary report, see, Boer, F. P.; Flynn,
J. J.; Freedman, H. H.; McKinley, S. V.; Sandel, V. R. J. Am. Chem. Soc.
1967, 89, 5068.
9
15402
J. AM. CHEM. SOC. 2003, 125, 15402-15410
10.1021/ja035136m CCC: $25.00 © 2003 American Chemical Society

Axial Chirality in 1,4-Disubstituted (ZZ)-1,3-Dienes
A R T I C L E S
Scheme 1. Formation of a Chiral (Z,Z)-1,3-Diene from a Diyne.
Figure 2. Axially Chiral Dienes Used for Dynamic NMR Study (Ap-
proximate coalescence temperature Tc are shown in brackets).
ization process. In this context, identification of the structural
features of the diene that might produce a higher barrier would
also be useful. With these goals in mind, we have examined
the helix inversion process by temperature-dependent dynamic
NMR spectroscopy. The details are reported in this paper.
Results and Discussion
Although the studies reported in the previous paragraph amply
validated the concept of axial chirality in appropriately substi-
tuted 1,3-dienes, it is difficult to envisage further synthetic
applications of this structural feature in these molecules because
of the recalcitrant nature of the substituents in these molecules.
Recently, we reported a facile synthesis of a new class of 1,4-
disubstituted (ZZ)-1,3-dienes via Pd(0)-catalyzed silyl-stannyl-
ation/cyclization of 1,6-diynes mediated by R3SiSnR′3 (eq 1).10
Synthesis and Characterization of the Dienes. The 1,4-
silyl stannyl dienes 8∼12 (Figure 2) were synthesized using
the procedure reported earlier (eq 1).10 In each case, the structure
of the compound, including the configuration of the double
1
bond, was unambiguously established by H, 13C, 119Sn NMR
1
spectroscopy.11 The data for 8 is typical. The H NMR (500
MHz, CDCl3, 25 °C) is characterized by the following peaks:
δ - 0.70 - + 0.20 (6 H, br, SiCH3), 0.78 (9 H, s, SiCCH3),
3.01 (2 H, br, s, H2), 3.20 (2 H, br, s, H5), 3.77 (6 H, s, OCH3),
5.23 (1 H, s, SiCH), 6.13 (1 H, s, JHSn ) 66.4 Hz, SnCH), 7.39-
7.40 (9 H, m, aromatic), 7.47-7.70 (6 H, m, aromatic).
Irradiation of the olefinic signal at δ 5.23 causes the enhance-
ment of the broad peak centered around 3.01 (4.5%) and
irradiation of the δ 6.13 peak causes the enhancement (5.5%)
of the broad peak at δ 3.20. The broad singlets due to the C-2
and C-5 methylene protons become sharp at 105 °C (toluene-
d8). Upon cooling to -49 °C, these signals evolve into two sets
of AB quartets. Likewise, the extraordinarily broad Si[CH3]2
signals (half-width at room temperature >100 Hz) gives a clear
indication of the possible fluxional nature of this molecule. Upon
warming to 105 °C (toluene-d8), this broad signal transforms
into a sharp singlet, whereas upon cooling to -49 °C two sharp
singlets emerge in the place of the broad peak. The correspond-
ing 13C signal is barely visible above the baseline (half-width
> 400 Hz) at 25 °C between δ -5.5 to -3.5; but is a sharp
singlet at 105 °C (toluene-d8) and a set of two singlets at δ -2.5
The exceptional regio- and stereochemical control arising as a
necessary consequence of the organometallic mechanisms
involved (Scheme 1), results in the placement of the Si and Sn
substituents in an “inside” orientation, thus creating a helical
motif. We expected the enantiomerization process in these
systems to depend on the size of the groups on Si and Sn, and
the substitution pattern around the ring.
As myriad possibilities for further functionalization via the
vinylsilane and vinylstannane moieties can be envisioned, these
fluxional molecules are potentially valuable intermediates for
stereoselective synthesis of carbocylic and heterocyclic com-
pounds. Among the possibilities are dynamic kinetic resolution
of the dienes by reactions with suitable chiral reagents or those
mediated by asymmetric catalysts (e.g., eq 2) and construction
and -5.5 at -60 °C. The other nonaromatic signals in the 13
C
spectra (CDCl3, 125 MHz) are at 17.4 (s, SiCCH3), 26.3 (q,
SiCCH3), 43.8 (t, CH2), 44.4 (t, CH2), 52.9 (q, OCH3), 54.7 (s,
C1), 1224. (d, SnCH), 125.4 (d, SiCH), and 172.0 (s, CO). The
OCH3 and CO signals in the proton-decoupled 13C spectrum
split into two equally sized peaks upon cooling to -60 °C.11
The 119Sn peak appears at δ -154.5 (CDCl3, 185 MHz). The
temperature dependence of the spectrum is completely revers-
ible. Barring a highly unlikely conformational equilibrium
involving the cyclopentane ring, these changes are best explained
by the axial chirality of the molecule. Molecules 9∼12 were
similarly characterized.11 Hydrogen-1, carbon-13, and tin-119
NMR spectra of these compounds recorded (at 500, 125, and
185 MHz respectively) at various temperatures are included in
the Supporting Information.
of extended helical polyolefins (eq 3). Before contemplating
(9) Pasto, D. J.; Scheidt, W. R. J. Org. Chem. 1975, 40, 1444.
(10) Gre´au, S.; Radetich, B.; RajanBabu, T. V. J. Am. Chem. Soc. 2000, 122,
8579.
(11) (a) See the Supporting Information for details and spectra. (b) Graphical
representations of the temperature dependence of chemical shifts of the
relevant protons are included in the Supporting Information.
such applications, it is important to establish the relevant
thermodynamic and kinetic parameters that govern this isomer-
9
J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003 15403

A R T I C L E S
Warren et al.
Figure 3. Exchange of the Diastereotopic Me2Si Groups in 8 upon
Inversion.
To estimate the kinetic parameters for this fast exchanging
system, NMR line shape analysis as described by Kaplan and
Fraenkel,12 was employed. In the present situation, this method
is almost ideal for two reasons: (a) the temperature dependence
of the NMR spectra of the dienes over a wide, readily accessible
range is highly reproducible, and, (b) our synthetic method
makes it possible to systematically change the steric demands
of the “inside” substituents. It is also possible to vary the size
and the substitution pattern of the ring. Because several well-
behaved spin systems undergoing exchange can be readily
identified, the line shape analysis provides the one of the most
accurate means for the determination of kinetic parameters for
this process at equilibrium.
Line Shape Analysis. The first system studied is shown in
Figure 3. Since the diene system is chiral, the two methylsilyl
groups are diastereotopic and appear as two singlets in the NMR
spectrum at temperatures below -5 °C (Figure 4). As the
temperature increases, the shift between the two silyl methyls
progressively averages and the peaks appear as a singlet above
∼28 °C. For line shape analysis purposes, this resonance can
be treated as an uncoupled, equally populated, two pseudo-half
spin exchanging system. The 1H pseudo spins are labeled as A
and B.
1
Figure 4. Experimental (left) and calculated (right) H NMR line shapes
due to methylsilyl groups of 8 at different temperatures with dervied rate
constants.
AB yk1z BA
(4)
k-1
possible using the relation
The absorption is obtained from eq 5
(νA - νB ) -1.6616T + 805.54, R2 ) 0.9987)
Abs(ν) ) - Im(F1 + F2)
(5)
1
Comparisons of the observed with the calculated H NMR
where
spectra (Figure 4) provided the rate constants of the exchange
process at each temperature. The Eyring plot (Figure 5) then
gave the values of ∆Hq and ∆Sq (55.4 ( 2.2 kJ.mol-1 and -
4.3 ( 7.6 J mol-1K-1 respectively).
The second system is depicted in Figure 6. The compound 9
consists of two diastereomers, with each pair of diastereotopic
gem-dimethyls giving rise to two lines. The ratio of the
diastereomers appear to be 1:1 from all spectral data. As
thetemperature increases, they average to 2 lines with some
overlap between them. This system can be treated as the sum
of two independent uncoupled, equally populated, two pseudo-
half spin exchanging systems.
F1 ) R/FA/â , F2 ) R/FB/â
The two elements (F1 and F2) of the density matrix are obtained
by solving the two coupled equations, shown in the matrix form
in eq 6.
i2π(∆Va) - T-1 -k
k
F1
1
) iC
(6)
-1
F
][ ]
2
[
[ ]
1
k
i2π(∆Vb) - T - k
where ∆νi ) ν - νi, νi being the frequency of species i and ν
is the frequency axis of the spectrum, T-1 is the intrinsic line-
width, C is an arbitrary constant, and k is the first-order rate
constant of the exchange process. The line width used in this
instance was T-1 ) 1.5 s-1. In this system, the shift between A
and B varies in a linear fashion with temperature. Shifts over
the temperature range of the study could be estimated by linear
extrapolation11b from the region where direct measurement was
2(AB h BA)
(7)
The absorption is obtained from eq 8
Abs (ν) ) - Im(F1 + F2 + F3 + F4)
(8)
where
F1 ) R/FA/â , F2 ) R/FB/â , F3 ) R/FA′/â , F4 ) R/FB’/â
(12) Kaplan, J. I.; Fraenkel, G. NMR of Chemically Exchanging Systems;
Academic Press: New York, 1980.
9
15404 J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003

Axial Chirality in 1,4-Disubstituted (ZZ)-1,3-Dienes
A R T I C L E S
Figure 5. Eyring plot for inversion of 8 using the lineshapes of the
methylsilyl resonances.
Figure 6. Exchange of the diastereotopic Me2Si groups in 9.
The four elements of the density matrix are obtained by solving
the four coupled equations, shown in matrix form in eq 9. A
line width of 3.4 Hz was used.In this case, there are two options
1
Figure 7. Experimental (left) and calculated (right) H NMR line shapes
due to methylsiyl groups of 9 at different temperatures with derived rate
constants.
for the exchange, the first and third peaks average; the second
and fourth peaks average or the first and fourth peaks average
and the second and third peaks average. Both possibilities were
tested and yielded the same results. Again, shift differences vary
with temperature, but we were able to obtain linear correlations11b
between the shift differences and the temperature as follows
νΑ - νΒ ) -1.8154T + 866.65, R2 ) 0.9994
νA′ - νB′ ) -1.7309T + 835.89, R2 ) 0.9992
νA - νA ) -0.136T + 47.308, R2 ) 0.999
Figure 8. Eyring plot for the inversion in 9 using the line shapes of the
methysilyl resonances in H NMR.
Comparisons of observed and calculated 1H NMR line shapes
are shown in Figure 7. The Eyring plot (Figure 8) then gives
the values of ∆Hq and ∆Sq as 48.8 ( 1.1 kJ mol-1 and -11.6
( 3.8 J mol-1K-1, respectively.
The last two systems studied are shown in Figures 9 and 10.
The mathematical treatments of both systems are identical. Both
1
have diastereotopic methylene groups, giving rise to two AB
systems. As the temperature increases, the shift between A and
B averages and the methylene hydrogens eventually appear as
two singlets. These systems can be treated as the sum of two
9
J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003 15405

A R T I C L E S
Warren et al.
Figure 9. Inversion in 10.
Figure 10. Inversion in 11.
independent coupled, equally populated, two-half-spin exchang-
ing systems.
2(AB h BA)
(10)
m The absorption is obtained from eq 11.
Abs(ν) ) -Im (F1 + F2 + F3 + F4 + F5 + F6 + F7 + F8)
(11)
Figure 11. Experiemental (left) and calculated (right) 1H NMR line shapes
fue to the methylene groups of 10 at different temperatures with derived
rate constants.
where
equations.11b
F1 ) RR/FAB/âR , F2 ) Râ/FAB/ââ , F3 )
RR/FAB/Râ , F4 ) âR/FAB/ââ , F5 ) RR/FA’B’/âR , F6 )
Râ/FA’B’/ââ , F7 ) RR/FA’B’/Râ , F8 ) âR/FA’B’/ââ
νA - νB ) -0.5671T + 368.82, R2 ) 0.9997
νA′ - νB′ ) -0.5671T + 339.42, R2 ) 0.9997
νA - νA′ ) -0.0071T + 31.171, R2 ) 0.8929
The values of these elements of the density matrix are obtained
by solving the eight coupled density matrix equation, shown in
matrix form in eq 12, where J and J′ are the coupling constants
between HA and HB of the two AB systems.
In the case depicted in Figure 9, the AB shift differences
vary linearly with temperature. Shifts in the exchange regime
are estimated by extrapolation as before using the following
The line width used for the left-most system was 2.0 Hz, and
the right-most system was 1.8 Hz. The value of the coupling
constant for former (J) was 14.84 Hz and for the latter (J′) 15.11
1
Hz. Comparisons of observed and calculated H NMR line
shapes are shown in Figure 11. The Eyring plot (Figure 12)
9
15406 J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003

Axial Chirality in 1,4-Disubstituted (ZZ)-1,3-Dienes
A R T I C L E S
Figure 12. Eyring plot for the inversion in 10 using the methylene
resonances in proton line shapes.
then gives the values of ∆Hq and ∆Sq as 54.8 ( 1.3 kJ mol-1
and 0.4 ( 4.8 J mol-1 K-1 respectively for the inversion of
10.
For the case depicted for 11 (Figures 10 and 13), none of the
peaks overlap and there are two options to explore: the two
outer doublets average and the two inner ones average, or the
first and third doublets average and the second and fourth
doublets average. However, only the first of these led to
reasonable results. The second combination led to an enthalpy
of activation of 73.6 kJ mol-1 and an entropy of activation of
60.1 J mol-1K-1, which seemed unreasonable.
The variation of the shifts with temperature was handled by
extrapolation11b as before using the equations
Figure 13. Experimental (left) and calculated (right) 1H NMR line shapes
due to the methylene groups of 11 at different temperatures with derived
rate constants.
νA - νB ) -0.7239T + 415.2, R2 ) 0.9968
νA′ - νB′ ) -0.5684T + 328.73, R2 ) 0.9959
νA - νA′ ) -0.1339T + 61.361, R2 ) 0.9879
The line width used for the inner two doublets was 3.0 Hz and
one for the outer doublets was 3.5 Hz. The values of both of
the geminal coupling constants (J and J′) were 15.00 Hz.
1
Comparisons of the observed and the calculated H NMR line
shapes are shown in Figure 13 and the corresponding Eyring
plot, in Figure 14. For this system the values of ∆Hq and ∆Sq
are 63.2 ( 2.9 kJ mol-1 and 21.0 ( 10.3 J mol-1K-1
respectively.
Line Shape Analysis of molecule 12 was also attempted, but
unfortunately, in this case, the shift of the peaks with temperature
did not appear to be linear and a rigorous analysis could not be
carried out.
Figure 14. Eyring plot for inversion in 11 using methylene resonances in
proton NMR line shapes.
The overall results obtained are summarized in Table 1. For
all molecules studied, the free energies of activation are similar
(52-57 kJ mol-1 at 300 K) and are in the range anticipated
from the NMR behavior. It should be noted that the values of
9
J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003 15407

A R T I C L E S
Warren et al.
positive number) for the triphenylstannane derivatives, but
positive (or near zero) for the tributylstannane derivatives.
Although it is premature to ascribe any chemical significance
to small values of ∆Sq, it is possible that the transition state is
more organized than the ground state for the triphenylstannane
derivatives. In these derivatives, it is possible to have correlated
or sympathetic motion of the aryl units and the silyl substituents
as the crowded transition state for the isomerization is ap-
proached. In a series of classical studies, Mislow has described
stereochemical consequences of correlated rotations in triarylbor-
anes.15 With the silyl substituents protruding into the molecular
propellers described by the Sn-aryl substituents, the present
situation involves more complex motions. However, in this
instance it is not unreasonable to expect higher order when the
Si and Sn substituents are brought together in the transition state,
especially when tin carries phenyl groups. The opposite must
be true for the Sn-n-butyl derivatives, probably because the
n-butyl groups have many more degrees of freedom than the
phenyl groups, and correlated motions are difficult to envision
in this case. Thus, the transition state could be less organized
here, leading to a positive entropy of activation. A very
significant change (∼20 J mol-1K-1) observed in going from
Me3Si to Et3Si derivative (10 and 11) may also support this
hypothesis.16
Even though we have not carried out detailed line shape
analysis on other 1,2-bisalkylidene cyclopentanes prepared
during the synthetic studies,10 the approximate coalescence (Tc)
temperature observed during the variable temperature NMR
studies clearly provide an indication of the fluxional nature of
these molecules. Several examples are included in Table 2.
Perhaps the most striking observation is the low Tc for the
pyrrolidines (-50 to - 70 °C). Apparently, with a very low
barrier for N-inversion,17 an alternate mechanism for the
conformational reorganization of the cycloalkane is now open.
Ongoing studies in our laboratory include attempts to freeze
the helix inversion and explorations of further synthetic ap-
plications of these fascinating molecules.
Figure 15. Bond angles of the diastereomeric chiral dienes derived from
crystal structure of the amine-oxalate salt (ref. 10).
∆Gq and ∆Hq are more reliable and the values ∆Sq, prone to
error. There is significant change in the enthalpies of activation
as the sizes of substituents are changed. For example, in the
tri-n-butylstannyl derivatives 10 and 11, there is 8 kJ mol-1
increase in enthalpy of activation in going from a trimethyl- to
a triethyl-silyl derivative, reflecting the larger size of the latter.
But the entropic contribution brings the overall value of ∆Gq
for the Et3Si derivative to ∼57 kJ mol-1, even though the large
error in the entropy estimates should raise some caution. A
decrease in the bulk at the C-1 carbon of the cyclopentane (in
going from 8 to 9) leads to a significant decrease in the energy
of activation for the helix inversion. This suggests that the
cyclopentane ring is involved in the inversion process, as we
had suspected from the strikingly broad 13C signals in certain
temperature regimes. For example, half-width of >400 Hz was
observed for the SiMe signals of 8 at 25 °C. Similar features
are observed for the ring carbons also (see the Supporting
Information for hardcopies of spectra). In general, the overall
energies of activation for these series of compounds appear to
be lower than would be expected from simple considerations
of size of the silicon and tin moieties. This may be a reflection
of the significant widening of the Si-C1-C2, C1-C2-C3, C2-
C3-C4, and C3-C4-Sn angles (the numbering of the diene-
carbons starting from the Si-Csp2 carbon as C1) of the diene-
carbon atoms. The crystal structure determination10 by X-ray
of the related oxalate salts 13 and 14 provide the angles shown
in Figure 15. The increased length of the silicon-carbon and
tin-carbon bonds, as compared to the C-C bond, could increase
or decrease the effective size of the substituents, depending on
the relative vectorial orientation of the C-Si and C-Sn bonds.
For example, within the diene structure if these bonds are
pointed at each other, longer C-Si or C-Sn bonds (vis-a`-vis
C-C bonds) would increase the effective size of these substit-
uents.13 On the other hand, the progress toward the transition
state results in an orientation where these bonds are pointed
away from each other, the longer bonds would decrease the
steric congestion. It is interesting to note that compounds 8,
10, and 11 have higher energies of activation compared to the
somewhat analogous tetramethyl compound 5c [Z ) C(CO2-
Et)2], described by Kiefer14 (∆Gq ) 46.0 kJ mol-1 at -46 °C).
Another parameter of interest is the bending force constants of
C-Si and C-Sn bonds. Unfortunately, the relevant information
is not available in the literature.
Experimental Section
General. Methylene chloride was distilled from calcium hydride
under nitrogen. Benzene and diethyl ether were distilled under nitrogen
from sodium/benzophenone ketyl. All of the solvents were freshly
(15) (a) Mislow, K. Acc. Chem. Res. 1976, 9, 26. (b) Gust, D.; Mislow, K. J.
Am Chem. Soc. 1973, 95, 1535. (b) Gust, D.; Mislow, K. J. Am Chem.
Soc. 1973, 95, 1535. (c) Finocchiaro, P.; Gust, D.; Mislow, K. J. Am. Chem.
Soc. 1973, 95, 7029. (d) Hummel, J. P.; Gust, D.; Mislow, K. J. Am. Chem.
Soc. 1974, 96, 3679. (e) Blount, J. B.; Finocchiaro, P.; Gust, D.; Mislow,
K. J. Am. Chem. Soc. 1973, 95, 7019.
(16) An intriguing mechanistic possibility for the helix inversion involves a series
of conrotatory electrocyclic ring closures and openings as shown below.
However, this remains highly speculative since we have not seen any
evidence of the cyclobutene or of the (EE)-diene in any of the spectroscopic
or chemical experiments.
Even with the admittedly large errors involved in their
estimation, the entropies of activation display an interesting
behavior, because their values appear to be negative (or a low
(13) The effective sizes of groups are different whether they are in a cis-1,3-
synaxial relationship as in 1,3-cyclohexanes or in the 2,2-positions of a
1,1′-binaphthyl system. In the latter case, the groups are pointed at each
other and the interactions increase as the van der Waals radii or the bond
lengths increase. See Eliel, E. L.; Wilen, S. H.; Mander, L. N. Stereo-
chemistry of Organic Compounds, Wiley: New York, 1994; p. 1144.
(14) Jelinski, L. W.; Kiefer, E. F. J. Am. Chem. Soc. 1976, 98, 282.
(17) Saunders: M.; Yamada, F. J. Am. Chem. Soc. 1963, 85, 1882.
9
15408 J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003

Axial Chirality in 1,4-Disubstituted (ZZ)-1,3-Dienes
A R T I C L E S
temperature was calibrated using methanol.18 The shims and tuning
were maintained as good as possible at all temperatures. At each
temperature, the spectra were recorded 2 to 3 min after the shims had
stabilized. The FIDs of the spectra were submitted to Fourier transform
and phasing only. No other treatments such as line broadening were
performed, as these would affect the shapes of peaks obtained. The
calculations of line shape simulations were performed using Math-
ematica 3.0. Printing of these calculated spectra was done by importing
them into Microsoft Excel 2000. The experimental spectra were printed
on transparencies, allowing for easy comparison.
Table 1. Kinetic Parameters Obtained via Line Shape Analysis
Synthesis of Trialkylsilyl-, Trialkyl-, or Triaryl-Stannanes. Tri-
methylsilyl Tri-n-butylstannane. This and other related compounds
were prepared according to a procedure described earlier.19 A solution
of 0.8 mL (5.71 mmol) of diisopropylamine in 10 mL of THF was
cooled to -78 °C. After 10 min, 2.1 mL (5.00 mmol) of 2.38 M
n-butyllithium was added over 5 min. After 5 min, the cooling bath
was removed and the solution was brought to room temperature. Tri-
n-butyltin hydride (1 mL, 3.72 mmol) was then added over 5 min. The
reaction mixture was stirred at room temperature for 20 min. Tri-
methylsilyl chloride (0.75 mL, 5.91 mmol) was then added to the
solution over minutes. After about 5 min, the solution became cloudy
(formation of lithium chloride). The solution was stirred at room
temperature for 10 more minutes, evaporated and chromatographed on
silica gel (hexanes). The product was obtained in quantitative yield.
Alternatively, the product could be distilled at 85°C under 2 mmHg.
1H NMR (CDCl3, 400 MHz) δ 0.22 (s, 9 H, SiCH3), 0.81-0.89 (m, 15
H, CH2CH3), 1.24-1.51 (m, 12 H, SnCH2 CH2). 13C NMR (CDCl3,
125 MHz) δ 1.49 (q), 7.86 (t), 13.74 (q), 27.61 (t), 30.34 (t).
Table 2. Coalescence Temperatures of Silylstannyl Dienesa
Triethylsilyl Tri-n-butylstannane. A solution of 0.5 mL (3.57
mmol) of di-iso-propylamine in 5 mL of THF was cooled to -70 °C.
After 10 min, 2.5 mL (3.00 mmol) of 1.20 M n-butyllithium was added
over 5 min. After 15 min, the cooling bath was removed and the solution
was brought to room temperature. Tri-n-butyltin hydride (0.6 mL, 2.23
mmol) was then added over 5 min. The reaction mixture was stirred at
room temperature for 30 min. Triethylsilyl chloride (0.5 mL, 2.98 mmol)
was then added to the solution. After about 5 min, the solution became
cloudy (formation of lithium chloride). The solution was stirred at room
temperature for 3 h, evaporated and chromatographed on silica gel
1
(hexanes). The product was obtained in quantitative yield. H NMR
(CDCl3, 500 MHz): δ 0.72 (6 H, q, 7.9 Hz, SiCH2), 0.84-0.89 (15 H,
m, Bu), 0.97 (9 H, t, 7.9 Hz, SiCH2CH3), 1.29 (6 H, sextuplet, 7.4 Hz,
Bu), 1.43-1.49 (6 H, m, Bu). 13C NMR (CDCl3, 125 MHz): δ 5.9 (t,
SiCH2), 8.3 (t, Bu), 8.7 (q, SiCH2C H3), 13.7 (q, Bu), 27.7 (t, Bu),
30.3 (t, Bu). 119Sn NMR (CDCl3 185 MHz): δ -122.8.
tert-Butyldimethylsilyl Triphenylstannane. A solution of 0.5 mL
(3.57 mmol) of diisopropylamine in 5 mL of THF was cooled to -78
°C. After 10 min, 2.5 mL (3.00 mmol) of 1.20 M n-butyllithium was
added over 5 min. After 10 min, the cooling bath was removed and
the solution was brought to room temperature. Triphenyltin hydride
(0.57 mL, 2.23 mmol) was then added over 5 min. The reaction mixture
was stirred at room temperature for 0.5 h. tert-Butyldimethylsilyl
chloride (600 mg, 3.98 mmol) was then added to the solution. The
solution was stirred at room-temperature overnight. The reaction mixture
was then evaporated and chromatographed on silica gel (hexanes). The
product was obtained 99% yield (1.033 g). 1H NMR (CDCl3, 500
MHz): δ 0.40 (6 H, s, SiCH3), 0.96 (9 H, s, CCH3), 7.30-7.31 (9 H,
m, Ph), 7.50-7.51 (6 H, m, Ph). 13C NMR (CDCl3, 125 MHz): δ -2.5
(SiCH3), 19.0 (CCH3), 27.5 (CCH3), 128.1 (Ph), 128.3 (Ph), 137.5 (Ph),
140.6 (Ph). 119Sn NMR (CDCl3) 185 MHz): δ -170.8.
a From 500 MHz 1H NMR VT NMR in CDCl3 unless otherwise noted.
distilled or stored over 3 Å molecular sieves. Analytical TLC was done
on E. Merck precoated (0.25 mm) silica gel 60 F254 plates. Column
chromatography was conducted by using silica gel 40 (Natland). HPLC
analyses were performed on a Chirasil OD column (25 cm length ×
4.6 mm i.d.). Elemental analyses were done by Atlantic Microlab Inc.,
Norcross, GA.
NMR spectra were obtained in the CDCl3 solutions (unless otherwise
stated) using a Bruker DRX-500 NMR and/or Brucker DPX-400
spectrometers. Chemical shifts are reported in parts per million (ppm,
δ) relative to CHCl3 (δ ) 7.240) as an internal standard for proton.
Coupling constants are reported in Hertz (Hz). The NMR spectra
recorded for line shape analysis required special precautions. The probe
Synthesis of Silyl Stannyl Dienes. Di-O-methyl-(3Z,4Z)-3-[(tert-
butyldimethylsilyl)methylene]-4-[(triphenylstannyl)-methylene]cy-
clopentane Carboxylate (8). Di-O-methyldipropargylmalonate (51 mg,
0.245 mmol), (tert-butyldimethylsilyl)-triphenylstannane (115 mg, 0.247
(18) Van Geet, A. L. Anal. Chem. 1970, 42, 679.
(19) Chenard, B. L.; Laganis, E. D.; Davidson, F.; RajanBabu, T. V. J. Org.
Chem. 1985, 50, 3666.
9
J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003 15409

A R T I C L E S
Warren et al.
mmol), Pd2(dba)3 (7 mg, 0.008 mmol) and tris-(pentafluorophenyl)-
phosphine (25 mg, 0.047 mmol) were mixed in 0.5 mL of benzene.
After 16 h of stirring at room temperature, the solvent was evaporated
and the residue was chromatographed on silica gel (hexanes/diethyl
Di-O-methyl-(3Z,4Z)-3-[(tri-n-butylstannyl)methylene]-4-[(trieth-
ylsilyl)methylene]-cyclopentanedicarboxylate (11). Di-O-methyl-
dipropargylmalonate (51 mg, 0.245 mmol), (triethylsilyl)tri-n-butyl-
stannane (99 mg, 0.244 mmol), Pd2(dba)3 (7 mg, 0.008 mmol) and tris-
(pentafluorophenyl) phosphine (22 mg, 0.041 mmol) were mixed in
0.5 mL of benzene. The vial was sealed and placed in an oil bath at 68
°C. After 16 h, the solvent was evaporated and the residue was
chromatographed on silica gel (hexanes/diethyl ether: 95/5; 80/20) to
yield 51 mg (34%) of the desired product and 17 mg of di-O-
1
ether: 95/5; 80/20) to yield 151 mg (91%) of the desired product. H
NMR (CDCl3, 500 MHz): δ - 0.7-0.2 (6 H, brd lump, SiCH3), 0.78
(9 H, s, SiCCH3), 3.01 (2 H, broad s, H2), 3.20 (2 H, broad s, H5), 3.77
(6 H, s, CH3), 5.23 (1 H, s, SiCH), 6.13 (1 H, s, JSn-H ) 66.4 Hz,
SnCH), 7.39-7.40 (9 H, m, Ph), 7.47-7.70 (6 H, m, Ph). 13C NMR
(CDCl3, 125 MHz): δ -5.5-3.5 (lump, SiCH3), 17.4 (s, SiCCH3), 26.3
(q, SiCCH3), 43.8 (t, CH2), 44.4 (t, CH2), 52.9 (q, CH3), 54.7 (s, C1),
122.4 (d, SnCH), 125.4 (d, SiCH), 128.3 (d, Ph), 128.7 (d, Ph), 137.2
(d, Ph), 139.8 (s, Ph), 155.9 (s), 159.2 (s), 172.0 (s, CO). 119Sn NMR
(CDCl3, 185 MHz): δ -154.5. Assignments were confirmed by COSY,
HMQC and DEPT-135. NOE: SnCH f H5: 5.5%; SnCHf SnPh3:
3.8%; SiCH fH2: 4.5%; SiCH ft-Bu: 3.8%.
O-Methyl (3Z,4Z)-3-[(tert-butyldimethylsilyl)methylene]-4-[(tri-
phenylstannyl)-methylene]cyclopentane Carboxylate (9). O-methyl-
2-(2-propynyl)-4-pentynoate (62 mg, 0.413 mmol). (tert-butyldimeth-
ylsilyl)triphenylstannane (154 mg, 0.331 mmol), Pd2(dba)3 (14 mg,
0.015 mmol) and tris-(pentafluorophenyl)phosphine (50 mg, 0.094
mmol) were mixed in 1 mL of benzene. After 15 h at room temperature,
the solvent was evaporated and the residue was chromatographed on
silica gel (hexanes/diethyl ether 95:5 to 90:10) to get 181 mg (89%) of
the desired product. 1H NMR (CDCl3, 500 MHz): δ -0.19 (6 H, broad
s, SiCH3), 0.83 (9 H, s, SiCCH3), 2.68-2.74 (2 H, m, H2), 2.90 (2 H,
d, 7.6 Hz, H5), 3.06 (1 H, quintet, H1), 3.77 (3 H, s, CH3), 5.27 (1 H,
s, SiCH), 6.15 (1 H, s, JHSn ) 69.5 Hz, SnCH), 7.41-7.43 (9 H, m,
Ph), 7.59-7.68 (6 H, m, Ph). 13C NMR (CDCl3, 125 MHz): δ -4.6
(SiCH3), 17.4 (SiCCH3), 26.4 (SiCCH3), 38.3 (C1), 39.5 (CH3), 51.8
(CH3), 121.2 (SnCH), 124.5 (SiCH), 128.3 (Ph), 128.7 (Ph), 137.2 (Ph),
140.0 (Ph), 158.1, 161.2, 175.7 (CO). 119Sn NMR (CDCl3,185 MHz):
δ -154.5. Assignments were confirmed by COSY and HMQC. NOE:
SnCH H5: 5.4%; SnCH f SnPh3: 3.4%; SiCH f H2: 4.4%; SiCH
f t-Bu: 4.0%
1
methyldipropargylmalonate. H NMR (CDCl3, 500 MHz): δ 0.58 (6
H, q, 7.8 Hz, SiCH2), 0.85-0.90 (24 H, m, Bu, SiCH2CH3), 1.27 (6 H,
hex, 7.3 Hz, Bu), 1.37-1.44 (6 H, m, Bu), 2.94 (4 H, broad s, H3, H5),
3.69 (6 H, s, CH3), 5.23 (1 H, s, SiCH), 5.63 (1 H, s, JHSn ) 49.5 Hz,
SnCH). 13C NMR (CDCl3, 125 MHz): δ 4.5 (t, SiCH2), 7.4 (q, SiCH2-
CH3), 10.6 (t, Bu), 13.7 (q, Bu), 27.3 (t, Bu), 29.0 (t, Bu), 44.3 (CH2),
44.4 (CH2), 52.7 (q, CH3), 54.9 (s, C1), 122.3 (d, SiCH), 125.9 (d,
SnCH), 155.6 (s), 156.9 (s), 172.2 (s, CO). 119n NMR (CDCl3, 185
MHz): δ -55.0. Assignments were confirmed by COSY, HETCOR
and DEPT-135. NOE: SnCH f H2: 8.8%; SnCH f SnBu3: 5.5%;
SiCH f H5: 7.7%; SiCH f SiEt3: 5.7%.
O-Methyl-(3Z,4Z)-3-[(tert-butyldimethylsilyl)methylene]-4-[(tri-
n-butylstannyl)-methylene]cyclopentanecarboxylate. A vial was
charged with 200 mg (1.333 mmol) of O-methyl-2-(2-propynyl)-4-
pentynoate, 540 mg (1.333 mmol) of (tert-butyldimethylsilyl)tri-n-
butylstannane, 30 mg (0.033 mmol) of Pd2(dba)3 and 71 mg (0.133
mmol) of tris-(pentafluorophenyl)phosphine in 0.4 mL of benzene. The
vial was sealed and placed in an oil bath at 52 °C for 17.5 h. The
solvent was then evaporated and the residue was chromatographed on
silica gel (hexanes/diethyl ether: 99/1) to yield 32 mg (4%) of the
desired product. 1H NMR (CDCl3, 500 MHz): δ 0.11 (3 H, s, SiCH3),
0.12 (3 H, s, SiCH3), 0.90-0.99 (24 H, m, Bu, t-Bu), 1.30-1.51 (12
H, m, Bu), 2.66 (4 H, broad s, CH2), 2.91 (1 H, quintet, 8.3 Hz, H1),
2.72 (3 H, s, CH3), 5.34 (1 H, s, SiCH), 5.70 (1 H, s, JHSn ) 54.3 Hz,
SnCH). 13C NMR (CDCl3, 125 MHz): δ -4.4 (q, SiCH3), -4.3 (q,
SiCH3), 10.8 (t, Bu), 13.7 (q, Bu), 17.3 s, SiC), 26.5 (q, SiCCH3), 27.3
(t, Bu), 29.1 (t, Bu), 38.4 (d, C1), 39.9 (broad, CH2), 51.8 (q, COOCH3),
121.5 (SiCH), 125.1 (SnCH), 176.0 (COO). 119Sn NMR (CDCl3, 185
MHz): δ -56.5. Assignments were confirmed by COSY, HMQC and
DEPT-135. NOE: SnCH f H5: 4.8%; SnCH f SnBu3: 4.0%; SiCH
f H2: 3.6%; SiCH f t-Bu: 4.0%; SiCH f SiMe2: 1.0%.
(3Z,4Z)-Di-O-methyl-3-[(tri-n-butylstannyl)methylene]-4-[(tri-
methylsilyl)-methylene]cyclopentanedicarboxylate (10). A solution
of 50 mg of di-O-methyldipropargylmalonate (0.240 mmol), 83 µL of
trimethylsilyltri-n-butylstannane (0.239 mmol), 4 mg of Pd2(dba)3 (0.004
mmol), 13 mg of tris-(pentafluorophenyl)phosphine (0.024 mmol) in
0.2 mL of benzene was prepared. The solution was stirred at room
temperature for 22 h. The reaction was then run on silica gel (hexanes/
Acknowledgment. We acknowledge the financial assistance
by the US National Science Foundation (CHE 0079948 and
CHE 0308378) and Petroleum Research Fund of the American
Chemical Society (PRF AC 36617).
1
diethyl ether: 9/1) to yield 97 mg (71%) of the desired product. H
NMR (CDCl3, 500 MHz): δ 0.05 (9 H, s, SiCH3), 0.85-0.88 (15 H,
m, Bu), 1.24-1.44 (12 H, m, Bu), 2.66-3.20 (4 H, brd s, H2, H5),
3.69 (6 H, s, CH3), 5.23 (1 H, s, SiCH), 5.65 (1 H, s, JHSn ) 50.0 Hz,
SnCH). 13C NMR (CDCl3, 125 MHz): δ 0.4 (q, SiCH3), 10.7 (t, Bu),
13.6 (q, Bu), 27.3 (t, Bu), 28.9 (t, Bu), 44.1 (t), 44.10 (t), 52.7 (q,
CH3), 55.0 (s, C1), 125.6 (d, SnCH), 126.2 (d, SiCH), 155.3 (s), 155.9
(s), 172.1 (s, CO). The assignment of the 13C NMR peaks was confirmed
by HMQC. NOE: SnCH f H2: 4.2%; SnCH f SnBu3: 4.0%; SiCH
f H5: 4.2%; SiCH f SiMe3: 2.3%. The retention times on HPLC
were: 34.22 min on Chiralsel OJ (100% hexanes, flow of 0.1 mL/
min) and 51.92 min on Chiralsel OD (100% hexanes, flow of 0.1 mL/
min).
Supporting Information Available: 1H, 13C, and 119Sn NMR
spectra of 8, 9, 10, and 11 including variable temperature
spectra, temperature dependence of relevant chemical shifts
needed for line shape analysis presented in graphical format
(PDF). This material is available free of charge via the Internet
at http://pubs.acs.org.
JA035136M
9
15410 J. AM. CHEM. SOC. VOL. 125, NO. 50, 2003