Platinum-catalyzed cycloisomerization reactions of enynes

Article abstract of DOI:10.1021/ja0109343

PtCl₂ constitutes an efficient and practical catalyst for a set of different atom economical rearrangement reactions of enynes. This includes (i) a formal enyne metathesis reaction delivering 1,3-dienes, (ii) the formation of polycyclic vinylcy

Full text of DOI:10.1021/ja0109343

J. Am. Chem. Soc. 2001, 123, 11863-11869
11863
Platinum-Catalyzed Cycloisomerization Reactions of Enynes
Alois Fu1rstner,* Frank Stelzer, and Hauke Szillat
Contribution from the Max-Planck-Institut fu¨r Kohlenforschung, D-45470 Mu¨lheim/Ruhr, Germany
ReceiVed April 11, 2001
Abstract: PtCl2 constitutes an efficient and practical catalyst for a set of different atom economical rearrangement
reactions of enynes. This includes (i) a formal enyne metathesis reaction delivering 1,3-dienes, (ii) the formation
of polycyclic vinylcyclopropane derivatives, and (iii) an unprecedented OfC allyl shift reaction if unsaturated
ethers are employed. Although these transformations produce significantly different structural motifs, they
share a common mechanism comprising a cationic manifold triggered by the π-complexation of Pt(II) onto
the alkyne unit of the substrates. Strong experimental support for the proposed mechanism comes from
deuterium-labeling studies, a careful analysis of the product distribution pattern, and the fact that in some
cases PtCl2 can be replaced by simple Lewis or Brønsted acids as the catalysts.
Introduction
Scheme 1
Chemical reactions proceeding with total mass transfer from
the substrates to the products are inherently attractive.1 To gain
practical importance, however, such “atom economical” trans-
formations must be accompanied by a significant increase in
structural complexity, they should require only catalytic amounts
of reagents, and they must be scalable and user-friendly.
Cycloisomerization reactions of enynes catalyzed by PtCl2 seem
to meet these stringent criteria. Since the pioneering studies of
Murai et al.,2,3 four different reaction modes have been
discovered (Scheme 1) which convert the substrates either to
(i) 1,3-dienes via a formal metathesis pathway (eq 1),2 (ii) 1,4-
dienes if the substrate contains an allylsilane (stannane) entity
(eq 2),4 (iii) unsaturated ethers if the reactions are performed
in alcoholic media (eq 3),5 or (iv) complex polycyclic arrays
if compounds with more than one double bond are employed
(eq 4).6,7
In the following we summarize our investigations in this
field.8,9 During the course of a study aimed at a better
understanding of the platinum-catalyzed enyne metathesis
* Address correspondence to this author: (e-mail) fuerstner@mpi-
muelheim.mpg.de.
(1) Trost, B. M. Angew. Chem. 1995, 107, 285-307; Angew. Chem.,
Int. Ed. Engl. 1995, 34, 259-281.
process, two additional cycloisomerization processes have been
discovered. Although these transformations generate quite
diverse structural motifs, they seem to share a common
mechanism and represent just different productive outlets of a
complex but highly versatile cationic manifold.
(2) (a) Chatani, N.; Furukawa, N.; Sakurai, H.; Murai, S. Organometallics
1996, 15, 901-903. (b) For similar reactions catalyzed by Ru catalysts
see: Chatani, N.; Morimoto, T.; Muto, T.; Murai, S. J. Am. Chem. Soc.
1994, 116, 6049-6050.
(3) In independent work, Trost et al. have pursued similar transformations
using palladol complexes as catalysts, cf.: (a) Trost, B. M.; Trost, M. K.
J. Am. Chem. Soc. 1991, 113, 1850-1852. (b) Trost, B. M.; Yanai, M.;
Hoogsteen, K. J. Am. Chem. Soc. 1993, 115, 5294-5295. (c) Trost, B. M.;
Chang, V. K. Synthesis 1993, 824-832.
Results and Discussion
Enyne Metathesis versus the Formation of Bicyclo[4.1.0]-
heptenes. Driven by a program aimed at the total synthesis of
the immunosuppressive prodigiosin alkaloid streptorubin and
congeners, we investigated if enyne metathesis10 opens an entry
into the azabicyclic pyrrolophane core of this target.11,12 For
(4) Ferna´ndez-Rivas, C.; Me´ndez, M.; Echavarren, A. M. J. Am. Chem.
Soc. 2000, 122, 1221-1222.
(5) Me´ndez, M.; Mun˜oz, M. P.; Echavarren, A. M. J. Am. Chem. Soc.
2000, 122, 11549-11550.
(6) Chatani, N.; Kataoka, K.; Murai, S.; Furukawa, N.; Seki, Y. J. Am.
Chem. Soc. 1998, 120, 9104-9105.
(7) For related reactions of ω-aryl-1-alkynes see: Chatani, N.; Inoue,
H.; Ikeda, T.; Murai, S. J. Org. Chem. 2000, 65, 4913-4918.
(8) For a preliminary communication see: Fu¨rstner, A.; Szillat, H.;
Stelzer, F. J. Am. Chem. Soc. 2000, 122, 6785-6786.
(9) For studies from our laboratory on other Pt(II)-catalyzed C-C-bond
formations see: (a) Fu¨rstner, A.; Voigtla¨nder, D.; Schrader, W.; Giebel,
D.; Reetz, M. T. Org. Lett. 2001, 3, 417-420. (b) Fu¨rstner, A.; Voigtla¨nder,
D. Synthesis 2000, 959-969.
(10) For general discussions of metathesis see the following for timely
reviews: (a) Trnka, T. M.; Grubbs, R. H. Acc. Chem. Res. 2001, 34, 18-
29. (b) Fu¨rstner, A. Angew. Chem. 2000, 112, 3140-3172; Angew. Chem.,
Int. Ed. 2000, 39, 3012-3043. (c) For a review on enyne metathesis see:
Mori, M. Top. Organomet. Chem. 1998, 1, 133-154.
(11) Fu¨rstner, A.; Szillat, H.; Gabor, B.; Mynott, R. J. Am. Chem. Soc.
1998, 120, 8305-8314.
10.1021/ja0109343 CCC: $20.00 © 2001 American Chemical Society
Published on Web 11/07/2001

11864 J. Am. Chem. Soc., Vol. 123, No. 48, 2001
Fu¨rstner et al.
a,b
Table 1. Enyne Metathesis Reactions Catalyzed by PtCl2
Scheme 2
this purpose, the rearrangement of the readily available substrate
1 to product 2 was investigated (Scheme 2). Among the different
catalysts screened, PtCl2 turned out to be by far the most
efficient and convenient one, delivering the desired key building
block in excellent yield.13 In structural terms, this transformation
is quite remarkable as one C-C bond must be broken and two
new ones formed while forging a bicyclic system that incor-
porates a bridgehead alkene moiety. Despite this significant
increase in molecular complexity, the reaction is distinguished
by its “low-tech” character and directly scales-up to multigram
quantities without need for any further optimization.11,14
Subsequently, many other enyne derivatives were found to
undergo equally facile and efficient “metathesis” to the corre-
sponding 1,3-diene products under the same reaction condi-
tions.15 As can be seen from Table 1, substantial structural
variations are accommodated and different functional groups
were found to be compatible, including ethers, esters, sulfones,
ketones, and sulfonamides. The rearrangement process is not
limited to compounds in which the alkene group is part of a
(strained) preexisting ring. Electron-withdrawing groups on the
alkyne facilitate the reaction, and a regular cyclization was
observed even with the polyunsaturated substrate 23.
Surprisingly, however, it was noticed that some closely related
1,6-enynes deviate from the expected course and populate an
entirely different pathway. This striking dichotomy is obvious
by comparing the behavior of sulfone 5 with that of the
corresponding tosylamide derivative 25 (Scheme 3). While the
former undergoes an essentially quantitative conversion to the
expected “metathesis” product 6, the latter delivers only 2% of
the corresponding 1,3-diene 27 together with the previously
unknown cyclopropane derivative 26 as the major product. If
enyne 28 is exposed to PtCl2 in toluene at 80 °C, both reactions
are operative to a similar extent, thus leading to diene 30 and
the tricyclic derivative 29. The unusual structure of the latter
was unequivocally deduced from 2D NMR experiments (see
the Experimental Section in the Supporting Information) and
was confirmed by X-ray analysis (Figure 1).
As can be seen from Table 2, many other substrates
containing heteroelements in the tether between the alkene and
(12) For the syntheses of other prodigiosins and related alkaloids see:
(a) Fu¨rstner, A.; Grabowski, J.; Lehmann, C. W. J. Org. Chem. 1999, 64,
8275-8280. (b) Fu¨rstner, A.; Krause, H. J. Org. Chem. 1999, 64, 8281-
8286. (c) Fu¨rstner, A.; Weintritt, H. J. Am. Chem. Soc. 1998, 120, 2817-
2825. (d) Fu¨rstner, A.; Grabowski, J.; Lehmann, C. W.; Kataoka, T.; Nagai,
K. ChemBioChem 2001, 2, 60-68. (e) Fu¨rstner, A.; Grabowski, E. J.
ChemBioChem 2001, 2, 706-709.
(13) PtCl4 shows similar catalytic activity as PtCl2. In sharp contrast,
the following platinum compounds have been found to be totally inactive:
PtCl2(COD), PtCl2(PhCN)2, PtO2, H2Pt(OH)6, Pt(acac)2, Pt2(dba)3, K2[PtCl6].
Moreover, PdCl2, NiCl2, CoCl2, RhCl3, CrCl2, and MnCl2‚4H2O showed
no catalytic activity in toluene at 80 °C. The palladol complex 56 described
by Trost et al. produced a complex mixture from which the desired product
could be isolated in only 18% yield.
(14) For a short discussion of our strategic goals see: Fu¨rstner, A. Synlett
1999, 1523-1533.
(15) For a related application to a formal total synthesis of roseophilin
see: Trost, B. M.; Doherty, G. A. J. Am. Chem. Soc. 2000, 122, 3801-
3810.
a In toluene at 80 °C with 4-10 mol % of PtCl2. b E ) COOEt.
the alkyne moiety also lead to bicyclo[4.1.0]heptene derivatives
in reasonable to excellent yields.16 This includes even allylsilane
(16) A similar transformation catalyzed by PtCl4 was described for allyl
propargyl ethers; in most cases, however, the yields obtained were rather
low. Conversion of the alkyne into an allene followed by formation of a
metallacycle was proposed as the reaction mechanism, cf.: Blum, J.; Beer-
Kraft, H.; Badrieh, Y. J. Org. Chem. 1995, 60, 5567-5569.

Platinum-Catalyzed Cycloisomerization Reactions of Enynes
J. Am. Chem. Soc., Vol. 123, No. 48, 2001 11865
Table 2. Formation of Cyclopropane Derivatives Catalyzed by
a
PtCl2
Figure 1. ORTEP diagram of the molecular structure of compound
29. Anisotropic displacement parameter ellipsoids are drawn at 50%
probability, hydrogen atoms are omitted for clarity. Selected bond length
(Å) and angles (deg): N1-C2 1.475(2), N1-C12 1.421(2), C2-C3
1.512(2), C12-C11 1.316(3), C11-C10 1.469(3), C3-C10 1.514(3),
C3-C9 1.491(2), C9-C10 1.525(3), C9-C3-C10 60.99(12), C3-
C9-C10 60.24(12), C3-C10-C9 58.77(11).
Scheme 3
a In toluene at 60-80 °C with 4-10 mol % of PtCl2. b Small amounts
(1-15%) of the corresponding 1,3-dienes resulting from formal enyne
metathesis of the substrates were obtained as byproducts in many cases.
c The lower yield is partly due to losses during workup caused by the
lability of the product which is formed as two diastereoisomers, cf.
Supporting Information.
47, although substrates of this type lacking the heteroatom were
previously shown to undergo PtCl2-catalyzed transformations
into 1,4-dienes.4 It is important to note, however, that in most
entries shown in Table 2, minor amounts (1-15%) of the
corresponding 1,3-diene products originating from the formal
metathesis pathway accompany the cyclopropane derivatives
formed as the major reaction products. This may be interpreted
in terms of common reactive intermediates which can stabilize
along two different pathways (vide infra).
Mechanistic Considerations. All attempts to gain insights
into the reaction mechanism by spectroscopic means were
unsuccessful. No intermediates are observed when the reaction
is carried out in an NMR tube even if equimolar amounts of
PtCl2 are employed. However, a set of experimental data has
accumulated that contains substantial indirect evidence for the
elementary steps of these reactions.
complexes, respectively. One of them was launched to explain
the behavior of the electron-deficient palladol complex 56 (R
) COOCH2CF3), which was previously shown to effect similar
enyne metathesis reactions.3,13,17 In this particular case, a truly
organometallic pathway was proposed involving metallacyclo-
pentene derivatives such as 57 as key intermediates of a Pd(II)/
Pd(IV) manifold (Scheme 4). The occasional isolation of
cyclobutenes of type 58 that may originate from 57 by reductive
elimination lends experimental support for this assumption.
Different scenarios can be envisaged for cycloisomerization
reactions of enynes mediated by Pt(II) or electrophilic Pd(II)
(17) See also: Trost, B. M.; Hashmi, A. S. K. Angew. Chem. 1993, 105,
1130-1132; Angew. Chem., Int. Ed. Engl. 1993, 32, 1085-1087.

11866 J. Am. Chem. Soc., Vol. 123, No. 48, 2001
Fu¨rstner et al.
Scheme 4
Scheme 5
Decomplexation followed by an electrocyclic ring opening of
the cyclobutene affords the final product and regenerates the
catalyst.
Alternatively, simple π-complexes formed by coordination
of the transition metal cation onto the alkyne moieties of the
substrates may trigger the rearrangement process. It is well
established that Pt(II)-alkyne complexes are very electrophilic
and are readily attacked by a variety of nucleophiles.18,19 There
is, however, some ambiguity as to whether the π-complex A
itself or the zwitterionic vinylmetal complex B derived thereof
constitutes the actual reactive intermediate.20 No rigorous
distinction between these tautomeric forms has been made so
far. If the nucleophilic reaction partner is a tethered alkene
group, a delocalized cation results which can also be represented
by the forms C-E. As shown in Scheme 5, such species
stabilize along different pathways. The preference for either of
these reaction channels will ultimately depend on which of the
possible resonance forms gains more weight, e.g. by stabilization
via the heteroatom X in the tether and/or the presence of
annellated ring systems.
Scheme 6
6). In contrast, the “cyclopropyl-methyl” cation resonance form
D is “carbenoid” in nature, suffering a rapid 1,2-hydrogen shift
to give bicyclo[4.1.0]heptane derivatives.22
Since this rationale explains the observed dichotomy without
difficulty, further experiments have been carried out to cor-
roborate this mechanistic interpretation. First, the assumed
stabilization of the carbenoid resonance form D by rapid 1,2-
hydrogen shift has been probed by deuterium labeling studies
(Scheme 7).
Specifically, the evolution of the cyclobutyl cation C21 leads
to 1,3-dienes as the products of a formal enyne metathesis. This
may occur either in a concerted or in a stepwise fashion with
formation of a cyclobutene derivative F21 that subsequently
undergoes an electrocyclic ring opening. This facet is important
as it provides an alternative explanation for the possible
intervention of cyclobutenes in these reactions (vide supra).
Upon treatment of enyne 60 with PtCl2, such a compound (61a)
has in fact been isolated; the low yields obtained in this reaction
are mainly due to the poor stability of products 61a,b (Scheme
For this purpose, the gem-dideuterated enyne 62 was prepared,
which rearranges into product 63 bearing deuterium labels at
the Vicinal positions. This finding provides strong evidence for
the assumed pathway. The deuterium incorporation at C-3 is
high, yet incomplete (63:64 ≈ 3:1). The unlabeled product 64
may originate from a competing insertion of the carbene into
traces of water that may be present in the reaction mixture.
Therefore, a control experiment has been carried out in which
the unlabeled analogue 31 was reacted with PtCl2 under the
standard conditions in the presence of D2O (1 eq). As expected,
the product formed was largely unlabeled; interestingly though,
the ratio of isotopomeres is in the same range as above (32:65
≈ 3:1) and the deuterium incorporation occurs selectively at
(18) For reviews on the carbocation-like behavior of Pt(II)-alkyne
complexes see: (a) Chisholm, M. H.; Clark, H. C. Acc. Chem. Res. 1973,
6, 202-209. (b) Belluco, U.; Bertani, R.; Michelin, R. A.; Mozzon, M. J.
Organomet. Chem. 2000, 600, 37-55.
(19) For prototype examples of reactions of alkynes π-complexed to
platinum or Lewis acids with various nucleophiles see the following for
leading references: (a) Kataoka, Y.; Matsumoto, O.; Tani, K. Organome-
tallics 1996, 15, 5246-5249. (b) Baidossi, W.; Lahav, M.; Blum, J. J. Org.
Chem. 1997, 62, 669-672. (c) Murata, T.; Mizobe, Y.; Gao, H.; Ishii, Y.;
Wakabayashi, T.; Nakano, F.; Tanase, T.; Yano, S.; Hidai, M.; Echizen, I.;
Nanikawa, H.; Motomura, S. J. Am. Chem. Soc. 1994, 116, 3389-3398.
(d) Neveux, M.; Seiller, B.; Hagedorn, F.; Bruneau, C.; Dixneuf, P. H. J.
Organomet. Chem. 1993, 451, 133-138. (e) Yeon, S. H.; Han, J. S.; Hong,
E.; Do, Y.; Jung, I. N. J. Organomet. Chem. 1995, 499, 159-165. (f)
Yoshikawa, E.; Gevorgyan, V.; Asao, N.; Yamamoto, Y. J. Am. Chem.
Soc. 1997, 119, 6781-6786. (g) Asao, N.; Shimada, T.; Yamamoto, Y. J.
Am. Chem. Soc. 1999, 121, 3797-3798 and literature cited therein.
(20) A detailed discussion of this issue is found in ref 7.
C-3 as evident from an appropriate set of NMR (1H, H, 13C)
2
experiments. Taken together, these data are fully consistent with
the assumed intervention of carbenoid species that stabilize
preferentially by an intramolecular 1,2-hydrogen (deuterium)
(21) It is emphasized that structures C and F are resonance forms that
represent two extremes of the same intermediate.
(22) For a general treatise of the reactivity of carbenes see: Kirmse, W.
Carbene Chemistry, 2nd ed.; Academic Press: New York, 1971.

Platinum-Catalyzed Cycloisomerization Reactions of Enynes
J. Am. Chem. Soc., Vol. 123, No. 48, 2001 11867
Scheme 7
the concept by intercepting the assumed electrophilic Pt(II)-
alkyne complexes with nucleophiles other than alkenes. For this
purpose, unsaturated ether derivatives of the general structure
shown in Scheme 10 have been designed in which the
coordination of the metal may trigger a cascade comprising a
1,4-addition of the ether oxygen onto the π-complexed alkyne
and simultaneous release of a (presumably metal complexed)
allyl cation. Recapturing of this fragment by the emerging
organoplatinum intermediate leads to a subsequent C-C-bond
formation at the carbon atom adjacent to the newly formed ring.
We were pleased to see that acetylenic esters (Y ) COOR)
as well as acetylenic nitriles (Y ) CN) smoothly undergo this
previously unknown OfC allyl shift reaction on exposure to
catalytic amounts of PtCl2 in toluene at 80 °C. Transition metal
salts other than PtCl2 were either found to be totally inactive or
lead to the rapid decomposition of the starting materials.25
The results compiled in Table 3 show the scope of this
transformation and deserve several comments. Thus, remarkably
high levels of stereoselectivity with regard to the newly formed
tetrasubstituted double bond are obtained in all cases investi-
gated. It is noteworthy, however, that the major products are
(E)-configurated in the case of acetylenic ester substrates,
whereas acetylenic nitriles invariably lead to (Z)-configurated
compounds. The formal trans addition of the ether oxygen and
the allyl part in the latter case is inconsistent with a concerted
reaction mechanism but is well explained by the rationale
depicted in Scheme 10 involving enolate intermediates K and
L which can interconvert via the respective O- or N-metalated
tautomers.26 Since the different stereochemical course of the
reaction in the ester and the nitrile series cannot be explained
by the thermodynamic stability of the products, it must be kinetic
in origin and hence arise from differences in the structure of
these intermediates.27 The fact that ester (as well as ketone)
enolates of nobel metals largely prefer the C-metalated form,28
whereas the nitrile analogues of the same metals are best
described as N-metalated keteniminato complexes,29 provides
a tentative explanation for this distinct behavior. The notion
that organometallic intermediates are involved in the observed
transformation is also supported by the finding that the addition
shift as the dominant pathway, which is faster than a competing
insertion into water (D2O).
The analysis of the byproducts formed in the PtCl2-catalyzed
reaction of substrate 1 also strongly advocates the proposed
cationic scenario. Enyne metathesis of 1 is a clean process
delivering product 2 in 79% yield; however, if the reaction was
carried out on a multigram scale (>7 g), we were able to isolate
minute amounts of compounds 66-68 formed as byproducts.
As has been discussed earlier in more detail,11 the formation of
all of these unusual derivatives is straightforward by assuming
a delocalized carbocation as the key intermediate that originates
from the attack of the alkene onto the π-complexed alkyne
moiety of the substrate. Its possible resonance forms G-J
engender the four different reaction channels featured in Scheme
8 and explain the observed product distribution pattern without
difficulty.
If the reaction mechanism is cationic in nature, it may well
be triggered by catalysts other than PtCl2. In fact, substrate 1
(and related electron deficient enynes) rearranges quite ef-
ficiently to the corresponding “metathesis” product 2 in the
presence of either BF3‚Et2O (5 mol %) or HBF4 (Scheme 9).
This finding is mechanistically relevant since no organometallic
pathway can be operative under these conditions and only a
cationic “Wagner-Meerwein” type process similar to the one
discussed above can account for the observed results. Therefore,
the overall reaction is reminiscent of the Lewis acid-catalyzed
cyclotrimerization of internal alkynes to substituted Dewar-
benzenes, which is initiated by an analogous π-complexation
of the substrates to a metal cation and proceeds via related
metal-cyclobutadiene complexes as reactive intermediates.23
Taken together, these data provide compelling evidence that
the skeletal reorganizations of enynes catalyzed by Pt(II) involve
delocalized cations as the key intermediates.24 This scenario
readily explains the dichotomy of the reaction pathway, is
consistent with the observed product distribution patterns and
the isotope labeling studies, and explains why in some cases
Pt(II) may even be replaced by simple Lewis or Brønsted acids
without changing the course of the reaction.
(25) This includes PtCl2(PPh3)2, PdCl2, Pd(PPh3)4, AgPF6, ZnCl2, BF3‚
Et2O, and a cationic platinum complex formed in situ from PtCl2(PPh3)2/
AgPF6.
(26) The importance of metal enolates as reactive intermediates in these
reactions can also be gleaned from the fact that substrates devoid of the
ester or nitrile group do not undergo this rearrangement process.
(27) Semiempirical calculations have shown that in both series the (E)-
configured products are thermodynamically more stable. For an example
showing slow isomerization of a (Z)-configured product into the more stable
(E)-isomer see: (a) Krueger, S. A.; Bryson, T. A. J. Org. Chem. 1974, 39,
3167-3168. (b) Bryson, T. A. J. Org. Chem. 1973, 38, 3428-3429.
(28) See the following for leading references on structures and reactivity
of ester and ketone enolates of Pt(II) and other nobel metals (Pd, Ni, Re,
etc.): (a) Appleton, T. G.; Chisholm, M. H.; Clark, H. C.; Yasufuku, K. J.
Am. Chem. Soc. 1974, 96, 6600-6605. (b) Chaudhury, N.; Puddephatt, R.
J. Chem. Soc., Dalton Trans. 1976, 915-919. (c) Ito, T.; Yamamoto, A. J.
Organomet. Chem. 1979, 174, 237-245. (d) Bennett, M. A.; Robertson,
G. B.; Whimp, P. O.; Yoshida, T. J. Am. Chem. Soc. 1973, 95, 3028-
3030. (e) Burkhardt, E. R.; Bergman, R. G.; Heathcock, C. H. Organome-
tallics 1990, 9, 30-44. (f) Stack, J. G.; Doney, J. J.; Bergman, R. G.;
Heathcock, C. H. Organometallcis 1990, 9, 453-466. (g) Burkhardt, E.
R.; Doney, J. J.; Stack, J. G.; Heathcock, C. H.; Bergman, R. G. J. Mol.
Catal. 1987, 41, 41-57. (h) Nizova, G. V.; Serdobov, M. V.; Nikitaev, A.
T.; Shul’pin, G. B. J. Organomet. Chem. 1984, 275, 139-144. (i) Yoshida,
T.; Okano, T.; Otsuka, S. J. Chem. Soc., Dalton Trans. 1976, 993-999. (j)
Kurosawa, H.; Majima, T.; Asada, N. J. Am. Chem. Soc. 1980, 102, 6996-
7003.
Platinum-Catalyzed OfC Allyl Shift Reactions. On the
basis of this mechanistic interpretation, we were trying to extend
(23) This transformation is similarly promiscuous with regard to the
Lewis acid that can be used; for a review see: Scha¨fer, W.; Hellmann, H.
Angew. Chem. 1967, 79, 566-573; Angew. Chem., Int. Ed. Engl. 1967, 6,
518 and literature cited therein.
(24) Further support for the proposed cationic mechanism comes from
a recent publication describing elegant labeling studies, cf.: Oi, S.;
Tsukamoto, I.; Miyano, S.; Inoue, Y. Organometallics 2001, 20, 3704-
3709.
(29) See the following for leading references showing that nitrile enolates
of Pt(II) and other nobel metals (Rh, Ir, Pd, etc.) exist as keteniminato
complexes: (a) Lenarda, M.; Baddley, W. H. J. Organomet. Chem. 1972,
39, 217-224. (b) Chaudhury, N.; Kekre, M. G.; Puddephatt, R. J. J.
Organomet. Chem. 1974, 73, C17-C19.

11868 J. Am. Chem. Soc., Vol. 123, No. 48, 2001
Fu¨rstner et al.
Scheme 8
Scheme 9
Scheme 10
of radical scavengers such as p-quinone or dihydroquinone to
the reaction mixture does not inhibit the Pt(II)-catalyzed allyl
shift.
The results shown in entries 2, 3, and 10 provide further
information. In these cases, the substitution patterns of the allyl
ether entities serve as mechanistic probes which indicate if
(metal complexed) allyl cation equivalents form at any stage
along the pathway of the OfC allyl shift reaction. In fact, the
formation of the resulting products is best explained by assuming
an allyl fragment as reactive intermediate that rearranges prior
to recombination at the site where the positive charge is most
stabilized and/or the intermediate is sterically least encumbered.
A crossover experiment has been carried out to gain insight
if the allyl group is transferred inter- or intramolecularly.
Specifically, an equimolar mixture of substrates 69 and 80 has

Platinum-Catalyzed Cycloisomerization Reactions of Enynes
J. Am. Chem. Soc., Vol. 123, No. 48, 2001 11869
Scheme 11
Table 3. Platinum Catalyzed OfC Allyl Shift Reactions (E )
COOMe)
products formed in this reaction. Within the limits of detection
(ca. 1%), no indications for the formation of products derived
from a crossover of the allyl fragments have been found, thus
strongly suggesting an intramolecular delivery of these moieties
as suggested by Scheme 10.
The fact that the PtCl2-catalyzed reaction of compound 92
does not only provide the expected tetrasubstituted alkene 93
but also affords small amounts of the protonated congener 94
lends further credence to the notion that allyl cation equivalents
intervene (Scheme 11). The cycloheptenyl cation initially formed
may not only be recaptured by the organometallic entity but
can also stabilize to some extent via loss of H+, which explains
the competing pathway.
Conclusions. PtCl2 is shown to be an extremely effective,
user-friendly, and versatile catalyst for different structural
reorganizations of enynes. Among the reaction modes that were
already known in the literature, the formal enyne metathesis of
such compounds is particularly attractive and has already served
as the key step in the total synthesis of complex bioactive targets.
Two additional transformations have now been developed: One
of them consists of the formation of bicyclo[4.1.0]heptene
derivatives while the other one represents a previously unknown
OfC allyl shift resulting in the stereoselective formation of
tetrasubstituted alkene moieties. Although these processes lead
to rather diverse structural motifs, strong evidence is provided
that they share a common cationic mechanism. The coordination
of the Pt(II) to the alkyne of the substrate likely triggers all of
these events which belong to the rare category of “low-tech”
yet complexity inducing and atom economical transformations
and should therefore find further applications in advanced
organic synthesis.
Acknowledgment. Generous financial support by the Deut-
sche Forschungsgemeinschaft (Leibniz award to A.F.) as well
as the Fonds der Chemischen Industrie (Kekule´ stipend to F.S.)
is gratefully acknowledged. We thank Dr. Ch. W. Lehmann for
the X-ray analysis of compound 29 and Dr. R. Mynott and co-
workers for their help with the interpretation of the NMR spectra
of some products. We thank the Degussa AG for a generous
gift of platinum salts.
Supporting Information Available: Full experimental
details, including the spectroscopic and analytical data of all
new compounds and information concerning the X-ray structure
of compound 29 (PDF). This material is available free of charge
via the Internet at http://pubs.acs.org.
a Pure (E)-isomer at the exocyclic double bond. b E:Z ) 2.4:1. c E:
Z ) 10.3:1. d With PtCl4 instead of PtCl2. e Pure (Z)-isomer at the
exocyclic double bond. f E:Z ) 1:7.9.
been exposed to catalytic amounts of PtClx (x ) 2, 4) under
standard conditions. Careful inspection of the reaction mixture
by GC/MS showed that compounds 70 and 81 were the only
JA0109343