G. C. Lloyd-Jones et al.
strate underwent cyclisation, however, the
unable to effect pro-catalyst activation and thus required co-addition
of either the Z isomer or 1 to initiate catalytic turnover.
E
isomer appeared
(38% conversion) rate prior=5.110ꢀ5 msꢀ1
10ꢀ6 msꢀ1; relative rate=8.5.
[42] Stoichiometric quantities of 8 converted (E)-14 to 15a–d at 408C
(3 d being required for 100% conversion)—the major isomer (71%)
was 15a.
[43] The identities of the products 15a-c were confirmed by independent
synthesis (15bc) or by comparison with literature 13C{1H} NMR data
(15b, O. Kitagawa, T. Suzuki, T. Inoue, Y. Watanabe, T. Taguchi,J.
Org. Chem. 1998, 63, 9470–9475).
[44] The procedure for preparation of diene 14 typically generates 3–8%
of 1 as a side product through retro aldol. In the synthesis of
[5,7-13C1]-14, this generates unlabelled 1. In the synthesis of
[1,3-13C1]-14 this produces [1,3-13C1]- [5,7-13C1]-1. The samples of 14
can be freed of much of the contaminating 1 by lengthy column
chromatography on silica gel. In the runs shown in Equations (j)
and (k), the total 1 in each sample amounted to ca. 1.0 and 2.5%,
respectively and thus accounts for 13 and 33% of 3, respectively.
The 13C label distributions in 3 obtained from 1 [cf. Eq. (e)] is dis-
tinct from that obtained from 14 [cf. Eqs. (j) and (k)] and were fully
self-consistent with the initial ratios 14/1.
; rate after=6.0
[36] The geometries of structures 10abe and (Z)-11abe, were optimized
at the MMFF level to obtain the lowest energy conformers as start-
ing structures for subsequent optimization applying density function-
al theory (DFT) at the generalized gradient approximation using
B3LYP hybrid functional Calculations were performed using Spar-
tan’02 (Wavefunction, Inc., Irvine, CA). The standard 6-31G* basis
set was used for C, H, O and F, see J. Kong, C. A. White, A. I.
Krylov, D. Sherril, R. D. Adamson, T. R. Furlani, M. S. Lee, A. M.
Lee, S. R. Gwaltney, T. R. Adams, C. Ochsenfeld, A. T. B. Gilbert,
G. S. Kedziora, V. A. Rassolov, D. R. Maurice, N. Nair, Y. H. Shao,
N. A. Besley, P. E. Maslen, J. P. Dombroski, H. Daschel, W. M.
Zhang, P. P. Korambath, J. Baker, E. F. C. Byrd, T. Van Voorhis, M.
Oumi, S. Hirata, C.-P. Hsu, N. Ishikawa, J. Florian, A. Warshel, B. G.
Johnson, P. M. W. Gill, M. Head-Gordon, J. A. Pople, J. Computa-
tional Chem. 2000, 1532–1548. Calculations indicate that there is
only a small effect from aryl substitution, thus DE 10a/(Z)-11a =
1.0 kcalmolꢀ1; DE 10b/(Z)-11b=0.3 kcalmolꢀ1; DE 10e/(Z)-11e =
0.7 kcalmolꢀ1. It should also be noted that the aryl rings in exo-iso-
mers (Z)-11 are twisted away from co-planarity with the alkylidene
unit p system and thus are not fully conjugated. The twisting arises
from relief of steric strain between the ortho protons and the
C(2)H2 unit on the cyclopentyl ring. DFTsuggests that the dihedral
[45] This can arise from even small amounts of 14 due to the low concen-
tration of the active “ACHTRE[UNG Pd-H]” catalyst and the similar reactivity of
1,6-diene 1 and 1,5-diene 9 towards it.
[46] Such a phenomenon bodes well for trapping of the active intermedi-
ate, allowing detection and identification by spectroscopic methods.
However, despite extensive 1H and 13C{1H} NMR analysis of mix-
tures of [1,3-13C1]-1, [1,3-13C1]-14, [5,7-13C1]-14 and 8 in various sol-
vents, orders of addition and proportions, thus far, we have been
unable to identify or isolate any intermediates from the complex
mixture of products, which we assume to be halide-bridged oligo-
meric forms of 19, that slowly develops.
ꢀ ꢀ
angle C
(ortho) C
N
10a and ꢀ63.5/117.88 in (Z)-10e to ꢀ73.0/107.48 in (Z)-10b.
[37] These substrates reacted at very different rates (times for > 98%
conversion are 6, 20, 5, 1.5 and 22 h for 9a–e, respectively) but when
mixed, all reacted the same (ꢁ 20%) rate (all proceeding to
> 95% conversion in ca. 3 h suggesting that different pre-catalyst
activation abilities, rather than inherent turnover rates, appear to be
predominantly responsible for the differing rates of reaction when
reacted alone.
[47] We do not observe NMR signals attributable to (Z)-15b. The stereo-
chemical assignment for (E)-15b is based on 1H NMR NOE con-
3
tacts, (see ref. [43]) and is supported by the magnitude of JC,C
=
[38] a) E. M. Ban, R. P. Hughes, J. Powell, J. Chem. Soc. Chem.
Commun. 1973, 591–592; b) H. Kurosawa, T. Majoma, N. Asada, J.
Am. Chem. Soc. 1980, 102, 6996–7003.
3.4 Hz between C(3) and the ethylidene methyl group, typical of
such geometry, see: a) J. L. Marshall, L. G. Faehl, R. Kattner, P. E.
Hansen, Org. Magn. Reson. 1979, 12, 169–173; b) P. A. Chaloner, J.
Chem. Soc. Perkin Trans. 2 1980, 1028–1032.
[39] a) The most common mechanism for ligand substitution at d8
square-planar centres, for example PdII, proceeds via an associative
mechanism involving five-coordinate intermediates or transition
states generated via external ligand or solvent attack at the metal
centre. For an overview see: R. J. Cross, Adv. Inorg. Chem. 1989, 34,
219–292; For examples of five-coordinate PdII alkene complexes
see: b) V. G. Albano, C. Castellari, M. E. Cucciolito, A. Panunzi, A.
Vitagliano, Organometallics 1990, 9, 1269–1276; see also c) N. Des-
marais, C. Adamo, B. Panunzi, V. Barone, B. Giovannitti, Inorg.
Chim. Acta 1995, 238, 159–163; for a proposal involving five-coordi-
nate alkene intermediates in PdII catalysis see d) A. Ashimori, B.
Bachand, M. A. Calter, S. P. Govek, L. E. Overman, D. J. Poon, J.
Am. Chem. Soc. 1998, 120, 6488–6489.
[48] Baldwinꢁs “rules for ring closure” (J. E. Baldwin, J. Chem. Soc.
Chem. Commun. 1976, 734–736) state that “As a consequence of
the larger atomic radii and bond distances in atoms of the second
Periodic row the geometric restraints on disfavoured ring closures
may be bypassed.”
[49] The availability of two diastereotopic ethyl protons, both of which
can attain a syn relationship to Pd, potentially allows the generation
of both geometric isomers of 15b [(E/Z)-15b]. Consideration of the
developing strain between the nascent ethylidene group and the
syn-related malonate ester on syn b-H elimination suggests that (E)-
15b will be favoured strongly over (Z)-15b.
[50] The diacetate analogue of 1 undergoes ca. 400-fold faster turnover
than 1 in the presence of the cationic Pd phenanthroline pro-catalyst
system, see ref. [6a].
[40] An alternative effect based on destabilisation of s-alkyl Pd inter-
mediate Div by electron donation from the coordinated vinylarene,
and thus increase of transition state energy relative to displacement
of 11 is also possible. Both effects would work in concert. The effect
of para-anisyl 9b on the regioselectivity of cyclisation of 1, and vice
versa, was found to be negligible (1:1 molar ratio of 1/9b). This may
be understood in terms of the lower steric decompression on libera-
tion of 2 compared to (Z)-11 and the greater reactivity of the uncon-
jugated methylidene in 2 facilitating rapid conversion of Diii ! Div,
irrespective of the identity of the mono-coordinated diene (1 or 9b).
[41] In the absence of exogenous (E)-14, conversion of 1!3 proceeds to
ꢂ98% in about 90 minutes (5 mol% 8, DCE, 408C, [1]0 =0.15m)
with an apparent pseudo zero order profile up to 90% conversion.
The following changes in pseudo zero order gradients were observed
over a 2 h period after addition of (E)-14: i) 2 mol% (E)-14 added
at 20 min (63% conversion) rate prior=6.410ꢀ5 msꢀ1; rate after =
[51] Interestingly, the diol precursor to 1,5-diene 23 did not cycloisomer-
ise. Instead it underwent inefficient isomerisation to what is tenta-
tively assigned as the corresponding 2,5-diene. This stands in stark
contrast to the diol analogue of 1 (4,4-bis(hydroxymethyl)-1,6-hepta-
diene) which underwent cycloisomerisation more rapidly than 1 (see
ref. [14b]). The origins of this difference are unclear but may relate
to the ability of the hydroxyl groups to intramolecularly coordinate
to the Pd intermediate analogous to 19, facilitating alkene displace-
ment and thus b-H elimination. Addition of equimolar PrOH to the
cycloisomerisation of 14 (5 mol% 8, toluene, reflux) resulted in only
2% isomerisation of 14 to its 2,5-diene isomer at 68% conversion
(70 min).
[52] Analysis of mixtures of [1,3-13C1]-1, [1,3-13C1]-14, and [5,7-13C1]-14
with 8 by 13C{1H} NMR, prior to the onset of rapid catalysis, reveals
that there is extensive dynamic line-broadening of the 13C-signals at
terminal double bonds (i.e., at C(1) in [1,3-13C1]-1 and [1,3-13C1]-14).
Based on the effect of the concentration of [(tBuCN)2PdCl2] (8) this
5.210ꢀ6 msꢀ1
; relative rate=12.2; ii) 6 mol% (E)-14 added at
20 min (49% conversion) rate prior=5.810ꢀ5 msꢀ1; rate after=
4.010ꢀ6 msꢀ1; relative rate=14.6; iii) 4.5 mol% 18 added at 20 min
8662
ꢀ 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2006, 12, 8650 – 8663