Preparation and fluxionality of a bimetallic platinum-molybdenum complex of 1,2,4,6-cycloheptatetraene

Article abstract of DOI:10.1021/om960340m

Fluxionality in the (Ph₃P)₂Pt l Mo(CO)₃ complex of 1,2,4,6-cycloheptatetraene, 2, is rapid at 20 °C. The process is intramolecular (determined by selective isotopomer irradiation) and requires simultaneous movement of both metals concomitant with inversion of the allene moiety. Fluxionality in the mononuclear complex, 1 is intramolecular at 60 °C but has a superimposed intermolecular component at 80 °C.

Full text of DOI:10.1021/om960340m

3788
Organometallics 1996, 15, 3788-3790
P r ep a r a tion a n d F lu xion a lity of a Bim eta llic
P la tin u m -Molybd en u m Com p lex of
1,2,4,6-Cycloh ep ta tetr a en e
J erzy Klosin,† Xiaoming Zheng, and W. M. J ones*
Department of Chemistry, University of Florida, Gainesville, Florida 32611
Received May 8, 1996X
Summary: Fluxionality in the (Ph3P)2Pt/ Mo(CO)3 com-
plex of 1,2,4,6-cycloheptatetraene, 2, is rapid at 20 °C.
The process is intramolecular (determined by selective
isotopomer irradiation) and requires simultaneous move-
ment of both metals concomitant with inversion of the
allene moiety. Fluxionality in the mononuclear complex
1 is intramolecular at 60 °C but has a superimposed
intermolecular component at 80 °C.
goes a similar fluxional process, but one that both is
faster than for the mononuclear complex and also
requires an achiral intermediate (or transition state).
We also report our discovery that the previously re-
ported intermolecular process for the fluxionality of 1
is in reality superimposed on a lower energy intramo-
lecular process.
The bimetallic complex 2 was prepared by addition
of THF to an equimolar mixture of 1 and (η6-p-xylene)-
Mo(CO)3. The dark brown-red complex was too unstable
for isolation but was stable in solution for few hours at
20 °C (complete decomposition occurred within 2 days),
which permitted complete characterization by multi-
nuclear NMR spectroscopy. As expected for a complex
with C1 symmetry, the 195Pt{1H} and 31P{1H} NMR
spectra of 2 exhibit a doublet of doublets at δ -4476
It is well-established that some transition-metal
complexes of allenes exhibit fluxional behavior by a 1,2-
metal shift between the double bonds of the allene.1
Three different mechanisms have been demonstrated
for this process: an intermolecular mechanism that
occurs by simple dissociation-recombination,2 an in-
tramolecular mechanism (the most common of the three
and sometimes referred to as the Vrieze-Rosenblum
mechanism) that occurs with retention of allene con-
figuration3 and a second intramolecular mechanism
(only one example)3e in which allene chirality is lost and
which has been suggested to proceed via a planar allyl
cation transition state.
Some time ago we investigated the properties of 1 a
Pt(0) complex of cycloheptatetraene, and found that it
undergoes a fluxional process with an activation barrier
of 26.8 kcal/mol.4 This barrier was determined by spin
saturation transfer techniques, which also demonstrated
an intermolecular process as at least one component of
the fluxional mechanism.
1
ppm (1J Pt-P′ ) 3180 Hz, J Pt-P′′ ) 3209 Hz) and two
doublets flanked with 195Pt satellites at δ 22.83 and δ
24.99 ppm (2J P′-P′′ ) 17.5 Hz), respectively. In addition
to PPh3 signals the 1H NMR of 2 shows six different
resonances. The highest field multiplet at δ 2.46 ppm
(2J H1-Pt ) 52.2 Hz) was assigned to H1.5 Since H1 and
H2 are not coupled (no cross-peak in the 2D COSY
spectrum),6 the chemical shift of H2 was assigned to the
peak at δ 3.13 ppm (3J H2-Pt ) 56.1 Hz) on the basis of
an NOE experiment (2% enhancement of H2 was
observed upon irradiation of H1). The remaining proton
assignments are based on a 2D COSY experiment. The
NOE difference spectrum (Figure 1) also showed a
strong negative peak at δ 4.57 ppm (3J H6-Pt ) 61.2 Hz)
which included satellites and a positive NOE at δ 5.85.
These peaks correspond to H6 and H5, respectively, and
were particularly interesting because they indicated
that saturated spin had been transferred from H1 to
H6 (which led to the NOE enhancement at H5); i.e., 3
must be fluxional at 20 °C. Magnetization transfer
experiments were carried out which, upon application
of the approximation of Dahlquist et al.,7 yielded a
fluxional rate of 0.18 ( 0.02 s-1 at 20 °C. Thermal
instability precluded determination of activation pa-
rameters for this process (no transfer at 10 °C; rapid
decomposition above 20 °C).
At this time, we report the preparation of 2, a
bimetallic complex of cycloheptatetraene, which under-
† Present address: The Dow Chemical Company, Central Research
& Development, Midland, MI 48674. E-mail: klosin@dow.com.
X Abstract published in Advance ACS Abstracts, August 1, 1996.
(1) For reviews of transition-metal-allene complexes, see: (a)
Bowden, F. L.; Giles, R. Coord. Chem. Rev. 1976, 20, 81. (b) Shaw, B.
L.; Stringer, A. J . Inorg. Chim. Acta, Rev. 1973, 7, 1.
(2) (a) Volz, H.; Volz-de Lecea, M. J ustus Liebigs Ann. Chm. 1971,
750, 136. (b) Sabastian, J . F.; Grunwell, J . R. Can. J . Chem. 1971, 49,
1779.
(5) (a) Lu, Z.; Abboud, K. A.; J ones, W. M. Organometallics 1993,
12, 1471. (b) Winchester, W. R.; Gawron, M.; Palenik, G. J .; J ones, W.
M. Organometallics 1985, 4, 1894. (c) Otsuka, S.; Nakamura, A.; Tani,
K. J . J . Organomet. Chem. 1968, 14, P30.
(6) See the Supporting Information.
(3) Cf.: (a) Vrieze, K.; Volger, H. C.; Gronert, M.; Pratt, A. P. J .
Organomet. Chem. 1969, 10, 19. (b) Vrieze, K.; Volger, H. C.; Pratt, A.
P. J . Organomet. Chem. 1970, 21, 467. (c) Foxman, B.; Marten, D.;
Rosan, A.; Raghu, S.; Rosenblum, M. J . Am. Chem. Soc. 1977, 99, 2160.
(d) Manganiello, F. J .; Oon, S. M.; Radcliffe, M. C.; J ones, W. M.
Organometallics 1985, 4, 1069. (e) Oon, S. M.; J ones, W. M. Organo-
metallics 1988, 7, 2172.
(7) Dahlquist, F. W.; Longmuir, K. J .; DuVernet, R. B. J . Magn.
Reson. 1975, 17, 406.
(8) In principle, this fluxionality could occur by a process which
involves dissociation of the Mo(CO)3 moiety from the front face of 2
and a 1,2-platinum shift in the resulting cyclohepatetraene complex 1
by either an intermolecular or intramolecular process, followed by
reattachment of the Mo(CO)3 to the open face (front or back) of the
seven-membered ring. This possibility is not viable because fluxionality
in 2 is faster than in 1.
(4) Lu, Z.; J ones, W. M.; Winchester, W. R. Organometallics 1993,
12, 1344.
S0276-7333(96)00340-8 CCC: $12.00 © 1996 American Chemical Society

Communications
Organometallics, Vol. 15, No. 18, 1996 3789
1
F igu r e 1. NOE difference spectrum and part of H NMR spectrum of complex 2.
Sch em e 1
platinum to move from one double bond of the allene to
the other without requiring the molybdenum to move
from the front face of the ring to the back, the config-
uration of the allene must be inverted. This would
probably occur by the allene adopting an achiral car-
bene-tropylium structure as represented by 3 (which
could be either an intermediate or a transition state)
or a dissociated molybenum complex as represented by
5.
To distinguish between the intramolecular process
represented by 2 and 4 and the intermolecular process
as depicted in 5, a magnetization transfer experiment
was carried out in which H1 was irradiated in only the
isotopomers of 2 and 4 with magnetically inactive
platinum nuclei (the central peak).9 The observation
that magnetization was transferred to H6 in only those
isotopomers containing inactive platinum nuclei (no
satellite peaks were observed, indicating no scrambling)
is strong evidence for an intramolecular process (Figure
2). We therefore suggest that fluxionality occurs through
the transition state or intermediate as depicted in 3, a
species that probably has significant tropylium ion
character. We further suggest that Mo(CO)3 accelerates
this process relative to 1 by bonding more strongly to
the tropylium π-system in the transition state (based
on displacement rates10 and X-ray studies11) than to the
highly distorted cycloheptatetraene in the ground state.12
Even more interesting than the fluxionality rate of 2
(about the same at 20 °C as that of the intramolecular
process (vide infra) for 1 at 60 °C) are requisite
characteristics of the mechanism of this fluxionality
process in comparison with other allene complexes.
First, both metals must move. As pictured in Scheme
1, the interconversion between enantiomeric pairs 2 and
4 must involve movement of platinum from one allene
double bond to the other coupled with simultaneous
movement of molybdenum from one heptatriene array
to the other. Second, in both 2 and 4, the platinum is
on the “back face” of the cycloheptatetraene ring and
the molybdenum is on the “front face”. In order for the
(9) Crucial to this technique is ensuring selective irradiation of the
central peak, leaving the satellite peaks unaffected. For a discussion
of control experiments designed for this purpose, see ref 4. In addition,
it may be noted that in this case with identical power settings, no
satellites were observed at 60 °C but were noted at 80 °C.
(10) Cf.: Davis, R.; Kane-Maguire, L. A. P. In Comprehesive Orga-
nometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W., Eds.;
Pergamon: New York, 1982; pp 1232, 1235.
(11) Clarke, G. R.; Palenik, G. J . J . Chem. Soc. D 1969, 667.
(12) Distortion of the cycloheptatriene moiety in 1 can be clearly
seen in its X-ray crystal structure.13 For two stable complexes ((PPh3)2-
Pt(CO)3Mo complexes of 1,2,3,5-cycloheptatetraene and cyclohepta-3,5-
dien-1-yne) that are isomeric with 2 but do not have distorted triene
units, see: Klosin, J .; Abboud, K. A.; J ones, W. M. Organometallics
1996, 15, 596.

3790 Organometallics, Vol. 15, No. 18, 1996
Communications
studied at 60 °C and, in this case, only transfer to the
central peak was observed. In other words, at the lower
temperature the intermolecular process is no longer
observable, which suggests that the latter is superim-
posed on an intramolecular mechanism. At this time
we do not have definitive evidence for the detailed
mechanism of this process, although circumstantial
evidence points to an achiral transition state similar to
3 rather than chiral 6.14
Ack n ow led gm en t. This research was supported by
the National Science Foundation and the Chevron
Research and Technology Co., to whom we are most
grateful.
Su p p or tin g In for m a tion Ava ila ble: Text giving experi-
mental details for the preparation of 2 and details of magne-
tization transfer experiments and figures showing multinu-
clear and 2D COSY spectra of 2 (10 pages). Ordering
information is given on any current masthead page.
F igu r e 2. Magnetization transfer difference spectra: (A)
complex 1 at 80 °C: selective irradiation at δ 3.33 ppm;
(B) complex 1 at 60 °C: selective irradiation at δ 3.33 ppm;
(C) complex 2 at 20 °C: selective irradiation at δ 2.66 ppm.
OM960340M
(14) The intramolecular process could go by either a chiral Vrieze-
Rosenblum mechanism (6) or an achiral carbene mechanism (analogous
to 3). At this time, it is not possible to distinguish between these.
However, we have made two observations that are consistent with the
carbene mechanism although neither demands it. First, intramolecular
fluxionaltiy in 1 is significantly faster (observable as low as 40 °C)
than any process (intramolecular or intermolecular) in the correspond-
ing complex of 1,2-cycloheptadiene (barely detectable at 100 °C). This
is consistent with a carbene-like transition state analogous to 3 which
would be significantly stabilized by the two additional double bonds
(this stabilization exceeds 13.9 kcal/mol in the corresponding Fp+
complexes).3d,15 However, this argument is weakened by Pt-P coupling
constants5a,b that indicate a stronger ground-state metal-allene bond
in the 1,2-cycloheptadiene complex than in 1. Second, when 1 was
prepared from enantiomerically enriched allene (enantiomeric excess
confirmed by trapping with diphenylisobenzofuran),16 within the limit
of experimental error the product was racemic. Unfortunately, this
simply means that 1 is probably optically unstable; it says nothing
about whether the optical instability is associated with the fluxional
process.
Finally, re-examination of magnetization transfer in
1 reproduced our previously reported results at 80 °C
(the only temperature studied); unique irradiation of the
central peak of H1 led to transfer to H6 in all isoto-
pomers, but less to the satellites than to the central peak
(area ratio of satellites to central peak ∼7:1). As
suggested at that time,4 transfer to the satellites
requires some intermolecular reaction but incomplete
transfer also requires a process that is equivalent to an
intramolecular reaction. In our current work, unique
irradiation of the magnetically inactive peak was also
(15) Allison, N. T.; Kawada, Y.; J ones, W. M. J . Am. Chem. Soc.
1978, 100, 5224.
(16) Harris, J . W.; J ones, W. M. J . Am. Chem. Soc. 1983, 104, 7329.
(13) Abboud, K. A.; Lu, Z.; J ones, W. M. Acta Crystallogr. 1992, C48,
909.