Elucidation of nonstatistical dynamic effects in the cyclization of enyne allenes by means of kinetic isotope effects

Article abstract of DOI:10.1002/anie.200700709

In no man's land - a reaction without a classic mechanism: A series of enyne allenes follow neither the classical concerted nor stepwise mechanism but rather comply with dynamic nonstatistical laws. The computed potential energy surface (see figure) and e

Full text of DOI:10.1002/anie.200700709

Angewandte
Chemie
DOI: 10.1002/anie.200700709
Reaction Mechanisms
Elucidation of Nonstatistical Dynamic Effects in the Cyclization of
Enyne Allenes by Means of Kinetic Isotope Effects**
Michael Schmittel,* Chandrasekhar Vavilala, and Ralph Jaquet
In the first decades following the birth of physical organic
chemistry, a set of firm paradigms guided our conception of
reactive intermediates, transition states, and reaction coor-
dinates. More recently, the significance of dynamic effects in
organic reactions has become increasingly evident through
the seminal work conducted by Carpenter,[1] Singleton,[2]
Doubleday,[3] Hase,[4] and others.[5] It was suggested quite
often in the context of unimolecular reactions proceeding at
the boundary between concerted versus stepwise mechanisms
that dynamic effects cause a behavior differing from that
predicted by statistical kinetic models, such as the transition-
state theory.[1,4]
Additional reaction mechanisms at the concerted/step-
wise boundary[1,4,6] with strongly contributing dynamic effects
have been recently resolved.[1,2a,5c,7] However, experimental
substantiation of nonstatistical dynamic effects, mostly based
on kinetic isotope effects (KIE) or stereochemical informa-
tion, was usually limited to single case studies and thus
needed to be complemented by computational dynamic
simulations. Herein, we describe how dynamic effects
become manifest in experimental inter- and intramolecular
KIE[8] data at the concerted/stepwise boundary of the thermal
C2–C6 cyclization of enyne allenes E (Scheme 1).[9,10] For
compounds E with a wide range of substituents, the KIE
results defy the primacy of statistical kinetic models. Thus our
data will serve as a copious test case for the validation of
future dynamic trajectory computations.
A few years ago, Engels and co-workers investigated the
ene reaction of enyne allenes at the (U)B3LYP/6-31G(d)
level. They predicted a stepwise process with the radical-
stabilizing group R1 = Ph and a concerted process with R1 =
tBu.[11] A recent study of the intermolecular KIE on the
thermal ene process of enyne allene E4 with R1 = triisopro-
pylsilyl (TIPS), however, suggested a stepwise diradical
process.[10] Lipton and Singleton analyzed the reaction for
E1 by combining insight from a single intermolecular KIE
value, theoretical calculations, and dynamic trajectories.[12]
The intermolecular KIEexp. = 1.43 concurred persuasively
with the theoretically predicted value for a concerted ene
process (KIEcalcd = 1.54). They suggested, based on computa-
tions and a relatively small number of thermalized trajecto-
ries, that both stepwise and concerted ene reactions of enyne
allenes proceed via a single transition state (TS) and that
dynamic effects at a posttransition-state valley–ridge inflec-
tion point determine which pathway is taken. While this was
clearly a fascinating proposal, the Lipton/Singleton proposal
has not found clear experimental support to date.
To sort out the different views arising from the above
experimental work we decided to prepare a set of [D0]-, [D1]-,
and [D2]enyne allenes and to investigate their inter- and
intramolecular KIEs in the thermal ene reaction (Table 1).
Classical statistical kinetic models would predict that inter-
and intramolecular KIEs for a concerted process are close to
each other, differing only in contributions from secondary
effects, while for a stepwise process the isotopic patterns need
not correspond. For the latter case, the intermolecular KIE
should be roughly 1.00–1.05[8c–e,g,j] and the intramolecular KIE
significantly higher than unity.
We chose the benzannulated enyne allenes E2–E5 con-
taining bulky groups such as TIPS, trimethylsilyl (TMS), and
tBu, at the alkyne terminus, as well as TMSand aryl groups at
the allene terminus. Our intent with the aryl group was to test
the effect of a radical-stabilizing group on the diradical to
favor the stepwise over the concerted process.
Scheme 1. The thermal C2–C6 cyclization of enyne allenes: concerted
(path 1) and stepwise (path 2) reaction mechanisms.
[*] Prof. Dr. M. Schmittel, C. Vavilala, Prof. Dr. R. Jaquet
FB 8 (Chemie – Biologie), Universität Siegen
Adolf-Reichwein-Strasse 2, 57068 Siegen (Germany)
Fax: (+49)271-740-3270
Compounds E4 and E5 exhibit intermolecular KIEs of
1.17 and 1.08, respectively, which are close to unity but not
within the range expected from statistical kinetic mod-
els.[8c–e,g,j] To understand the meaning of these values, we
also measured intramolecular KIEs. Neither the values for
[D1]E4 and [D1]E5 (1.003 and 1.001) nor their temperature
independence (see Table 2, [D1]E5), however, can be under-
stood in classical terms. It is clear, though, that both intra- and
intermolecular KIEs rigorously refute a purely concerted
mechanism and point rather to a stepwise process (Scheme 1,
E-mail: schmittel@chemie.uni-siegen.de
[**] We are indebted to the DFG and the Fonds der Chemischen
Industrie for financial support, Prof. Singleton for valuable sug-
gestions, and to the computer centers of NIC-Jülich and HRZ-
Siegen for computer time.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
Angew. Chem. Int. Ed. 2007, 46, 6911 –6914
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
6911

Communications
Table 1: Intermolecular and intramolecular kinetic isotope effects for the ene reaction of enyne allenes E1–E5.
Intermol.
KIE (D0/D2)
Intramol.
KIE (D1)
Cmpd.
Prevailing character of the mechanism
E1
E2
E3
E4
E5
R=OAc, R1 =TMS, R2 =H, R3 =TMS
1.43[a]
1.60[b]
1.24[c]
1.17[d],[e]
1.08[f]
–-
concerted
concerted
boundary
stepwise
stepwise
R=H, R1 =TMS, R2 =nPr, R3 =TMS
R=H, R1 =tBu, R2 =nPr, R3 =TMS
R=H, R1 =TIPS, R2 =nPr, R3 =p-An
R=H, R1 =TMS, R2 =nPr, R3 =p-An
1.352[b]
1.286[c]
1.003[d]
1.001[f]
[a] Ref. [12]. [b] At 908C. [c] At 1258C. [d] At 1008C. [e] Ref. [10]. [f] At 808C.
Table 2: Temperature dependence of intramolecular kinetic isotope
effects for the ene reaction of E2 and E5.
For E3, intra- and intermolecular KIEs lie between those
of E1,E2 and E4,E5. E3 seems to be located in the boundary
zone of the concerted-to-stepwise thermal ene reaction of
enyne allenes as it shows an intermolecular KIE close to that
of E4,E5 but an intramolecular KIE close to that of E2.
Remarkably, the intramolecular KIE is larger than the
intermolecular. Although the full meaning of the KIE data
of E3 has to be inquired further, the intermolecular KIE data
suggests a balanced mixing of concerted and stepwise
trajectories. In order to account for the KIEs in E3 (intra-
molecular > intermolecular), the intramolecular KIE of the
“pristine stepwise” trajectories must be higher than those
measured for E4,E5. This may be a result of a sterically
imposed barrier for the hydrogen transfer from D!P as a
result of the large tert-butyl group at the alkyne terminus.
In order to sharpen the mechanistic picture in light of the
variety of KIEs displayed by E1–E5 we reinvestigated the
potential energy surface of the thermal ene process of E6
(R1 = CH3, R2 = R3 = H, no benzannulation). Owing to the
absence of any radical-stabilizing group, E6 should represent
a prototypical enyne allene following the “concerted” mech-
anism. While a single transition state was located for the
concerted and stepwise process in an earlier computation,[12]
our fully optimized (U)B3LYP/6-31G(d) 2D energy surface of
the C2–C6 cyclization displays a broad transition-state zone
about TS1. At TS1 we find the following results: DE°
T [8C]
Intramol.
Intramol.
KIE [D1]E2
KIE [D1]E5
60
70
80
90
110
1.706
1.561
1.486
1.352
1.075
1.003
1.002
1.001
1.001
1.002
path 2). This interpretation is fully consistent with the
recently observed cyclopropyl ring-opening in the thermolysis
of E4’ (cf. E4 with R2 = 2,2-diphenylcyclopropyl instead of
nPr).[13] Rewardingly, the Lipton/Singleton model[12] provides
a thorough explanation for the experimental KIEs. Accord-
ingly, the intermolecular KIEs of 1.08 and 1.17 propose that
the prevailing stepwise trajectories mix with a small amount
of concerted trajectories (going directly to P), thus raising the
KIE above 1.05. If one assumes that the main fraction
proceeds via the diradical, the near-unity intramolecular KIE
implies that the hydrogen transfer itself is again controlled by
a nonstatistical dynamic process. Apparently the diradical is
formed in a very shallow energy minimum with large excess
energy preventing statistical kinetic behavior.[14]
While E4 and E5 are characterized by aryl substituents at
the allene terminus which provide momentum towards
stepwise diradical formation, E1–E3 are devoid of radical-
stabilizing substituents. As evident in Table 1, an intermolec-
ular KIE of 1.60 was derived from the thermolysis of
[D0,D2]E2, comparable to that of allenyl acetate E1.[12] The
intramolecular KIE for [D1]E2 was measured to 1.352 and
proved to be temperature-dependent (Table 2). Remarkably,
both the inter- and intramolecular KIEs are decisively
different from unity and deviate more from each other than
can possibly be due to secondary isotopic effects.[15] While
these KIEs are more consistent with a concerted mechanism
(Scheme 1, path 1), they are rather low compared to those of
prototypical ene reactions.[8a,b] As in classical terms, inter- and
intramolecular KIEs should match closely for a concerted
mechanism, the experimental discrepancy seems to reflect, as
suggested by the Lipton/Singleton model,[12] the mixing of
concerted and stepwise trajectories, with the former ones
clearly dominating.
(+ZPE) = 31.2 (29.7), DDH°298 = 28.7, DDG°298 = 32.2; kCH
kCD = 1.61, kCH /kCH D = 1.57; Ca–H = 115.5 pm, C2–C6 =
/
3
3
3
2
195.7 pm; barrier heights in kcalmolꢀ1).[3c,16] As shown in
Figure 1—where energies are given relative to diradical D—
there is a smooth increase in energy starting in region E for
changes both in C2–C6 and Ca–H distances. Thermalized
trajectories starting in the E region can cross the TS1
transition-state zone TSZ1(D) at shorter Ca–H distances
(Ca–H = 110 pm,
C2–C6 = 187 pm,
C7–H = 198 pm,
E
ꢁ 31 kcalmolꢀ1) or the TS1 transition-state zone TSZ1(conc)
at longer Ca–H distances (Ca–H = 125 pm, C2–C6 = 207 pm,
C7–H = 157 pm, E ꢁ 32 kcalmolꢀ1); these designated regions
can be interpreted as either stepwise or concerted pathways.
The contours in Figure 1 indicate that within DE = 2 kcal
molꢀ1 trajectories might go directly to the diradical D via
TSZ1(D) or to the product region P via TSZ1(conc). Clearly,
the reaction mechanism can not be addressed without
including dynamic effects.
6912
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 6911 –6914

Angewandte
Chemie
performed far in the product region to explain the intra-
molecular KIEs.
In summary, the present experimental data verify with
hindsight the importance of dynamic effects in the thermal
C2–C6 cyclization of enyne allenes as predicted by Singleton/
Lipton through trajectory calculations on the parent enyne
allene E6. It seems that the influence of substituents in E1–E5
can be explained qualitatively from the dynamic computa-
tions on E6 just by mixing different contributions of
“concerted” and “stepwise” trajectories. Hence, radical-
stabilizing substituents do lead to large percentage of stepwise
trajectories, while other substituents lead to a domination of
concerted ones. Future experimental studies on the C2–C6
cyclization will further extend the set of data, as additional
substitutent changes at C1, C7, and even at C3 open the way for
additional tests. Our experimental KIEs challenge the
theoretical understanding and modeling of nonstatistical
kinetic processes, as now a large set of data must be
reproduced quantitatively in a consistent theoretical treat-
ment.
Figure 1. Energy profile for the C2–C6 cyclization of E6 (R1 =CH3,
R=R2 =R3 =H, no benzannulation). Relative energies are given with
respect to the energy of D. For two fixed distances the energy was
optimized with respect to the other degrees of freedom using the
TURBOMOLE package.[18]
We have performed a few “on the fly” trajectory
calculations on E6 supportive of the analysis in the Single-
ton/Lipton paper.[12] The smooth energy profile behind TS1
(see Figure 2) supports strongly oscillating trajectories
Received: February 15, 2007
Revised: May 25, 2007
Published online: August 6, 2007
Keywords: diradicals · dynamic effects · ene reactions ·
.
enyne allenes · kinetic isotope effect
[1] a) B. K. Carpenter, Acc. Chem. Res. 1992, 25, 520 – 528; b) B. K.
Carpenter, Angew. Chem. 1998, 110, 3532 – 3543; Angew. Chem.
Int. Ed. 1998, 37, 3340 – 3350; c) B. K. Carpenter, Chem. Soc.
Rev. 2006, 35, 736 – 747.
[2] a) B. R. Ussing, D. A. Singleton, J. Am. Chem. Soc. 2005, 127,
2888 – 2899; b) B. R. Ussing, C. Hang, D. A. Singleton, J. Am.
Chem. Soc. 2006, 128, 7594 – 7607.
Figure 2. Mechanistic formulation of the thermal ene reaction. The
above energy profile (before TS1) urged us to consider a variety of
individual thermalized trajectories already starting from the region E.
[3] a) C. Doubleday, M. Nendel, K. N. Houk, D. Thweatt, M. Page, J.
Am. Chem. Soc. 1999, 121, 4720 – 4721; b) C. Doubleday, C. P.
Suhrada, K. N. Houk, J. Am. Chem. Soc. 2006, 128, 90 – 94.
[4] a) C. Doubleday, K. Bolton, W. L. Hase, J. Am. Chem. Soc. 1997,
119, 5251 – 5252; b) C. Doubleday, K. Bolton, W. L. Hase, J.
Phys. Chem. A 1998, 102, 3648 – 3658; c) L. Sun, K. Song, W. L.
Hase, Science 2002, 296, 875 – 878.
[5] a) H. Yamataka, M. Aida, M. Dupuis, Chem. Phys. Lett. 2002,
353, 310 – 316; b) H. Yamataka, M. Aida, M. Dupuis, J. Phys.
Org. Chem. 2003, 16, 475 – 483; c) S. C. Ammal, H. Yamataka,
M. Aida, M. Dupuis, Science 2003, 299, 1555 – 1557; d) A. G.
Leach, E. Goldstein, K. N. Houk, J. Am. Chem. Soc. 2003, 125,
8330 – 8339; e) À. Gonzalez-Lafont, M. Moreno, J. M. Lluch, J.
Am. Chem. Soc. 2004, 126, 13089 – 13094; f) C. P. Suhrada, C.
Selçuki, M. Nendel, C. Cannizzaro, K. N. Houk, P.-J. Rissing, D.
Baumann, D. Hasselmann, Angew. Chem. 2005, 117, 3614 – 3618;
Angew. Chem. Int. Ed. 2005, 44, 3548 – 3552.
[6] a) D. A. Hrovat, S. Fang, W. T. Borden, B. K. Carpenter, J. Am.
Chem. Soc. 1997, 119, 5253 – 5254; b) A. Kless, M. Nendel, S.
Wilsey, K. N. Houk, J. Am. Chem. Soc. 1999, 121, 4524 – 4525.
[7] a) S. L. Debbert, B. K. Carpenter, D. A. Hrovat, W. T. Borden, J.
Am. Chem. Soc. 2002, 124, 7896 – 7897; b) D. A. Singleton, C.
Hang, M. J. Szymanski, M. P. Meyer, A. G. Leach, K. T. Kuwata,
J. S. Chen, A. Greer, C. S. Foote, K. N. Houk,J. Am. Chem. Soc.
2003, 125, 1319 – 1328; c) S. M. Bachrach, J. C. Gilbert, J. Org.
Chem. 2004, 69, 6357 – 6364; d) M. J. McGuire, P. Piecuch, J. Am.
Chem. Soc. 2005, 127, 2608 – 2614.
between diradical and productlike structures. Work is in
progress to analyze the dynamics based on the potential
presented in Figure 1 in more detail using classical and
quantum mechanical wave-packet methods.[17]
We expect that this picture will change appreciably when
we consider quantum mechanical computations of enyne
allenes with radical-stabilizing substituents, as both the
energy of D and TSZ1(D) may be lowered with respect to
TSZ1(conc), and the reaction will proceed in an increasingly
stepwise manner. Clearly, for all thermal enyne allene
thermolyses studied so far experimentally by KIEs, no
single case seems to follow the classical statistical kinetic
model. While our experimental data confirm the Lipton/
Singleton model,[12] one has to assume in light of our
computational results that the model has to be refined by
the assumption of a broad transition-state zone. This suggests
that for future dynamic computations on the C2–C6 cyclization
of enyne allenes, thermalized trajectories starting in region E
must be considered (Figure 2) and that the analysis must be
Angew. Chem. Int. Ed. 2007, 46, 6911 –6914
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
6913

Communications
[8] For intramolecular and intermolecular KIEs in ene reactions
[13] M. Schmittel, A. A. Mahajan, G. Bucher, J. W. Bats, J. Org.
Chem. 2007, 72, 2166 – 2173.
see: a) C. A. Seymour, F. D. Greene, J. Org. Chem. 1982, 47,
5226 – 5227; b) B. B. Snider, E. Ron, J. Am. Chem. Soc. 1985, 107,
8160 – 8164; c) W. Starflinger, G. Kresze, K. Huss, J. Org. Chem.
1986, 51, 37 – 40; d) M. Orfanopoulos, C. S. Foote, J. Am. Chem.
Soc. 1988, 110, 6583 – 6584; e) M. Orfanopoulos, I. Smonou, C. S.
Foote, J. Am. Chem. Soc. 1990, 112, 3607 – 3614; f) T. A.
Blumenkopf, G. C. Look, L. E. Overman, J. Am. Chem. Soc.
1990, 112, 4399 – 4403; g) Z. Song, P. Beak, J. Am. Chem. Soc.
1990, 112, 8126 – 8134; h) Y. Elemes, C. S. Foote, J. Am. Chem.
Soc. 1992, 114, 6044 – 6050; i) M. Stratakis, R. Nencka, C.
Rabalakos, W. Adam, O. Krebs, J. Org. Chem. 2002, 67, 8758 –
8763; j) W. Adam, O. Krebs, M. Orfanopoulos, M. Stratakis,
G. C. Vougioukalakis, J. Org. Chem. 2003, 68, 2420 – 2425.
[9] a) M. Schmittel, A. A. Mahajan, G. Bucher, J. Am. Chem. Soc.
2005, 127, 5324 – 5325, and references therein; b) P. R. Schreiner,
A. Navarro-Vazquez, M. Prall, Acc. Chem. Res. 2005, 38, 29 – 37,
and references therein.
[14] Using the data and method from Figure 1, an intramolecular
kCH /kCD value of 2.46 was computed for the classical stepwise
3
3
process of E6.
[15] a) W. R. Dolbier, Jr., Isotopes in Molecular Rearrangements,
Vol. 1 (Eds.: E. Buncel, C. C. Lee), Elsevier, Amsterdam, 1975,
pp. 51 – 57; b) P. Beak, K. Berger, J. Am. Chem. Soc. 1980, 102,
3848 – 3856; c) C. A. Seymour, F. D. Greene, J. Org. Chem. 1982,
47, 5226 – 5227; d) C. C. Cheng, C. A. Seymour, M. A. Petti, F. D.
Greene, J. F. Blount, J. Org. Chem. 1984, 49, 2910 – 2916; e) W.
Starflinger, G. Kresze, K. Huss, J. Org. Chem. 1986, 51, 37 – 40;
f) M. Orfanopoulos, C. S. Foote, I. Smonou, Tetrahedron Lett.
1987, 28, 15 – 18; g) M. Orfanopoulos, C. S. Foote, J. Am. Chem.
Soc. 1988, 110, 6583 – 6584.
[16] W. v. E. Doering, K. Sachdev, J. Am. Chem. Soc. 1974, 96, 1168 –
1187.
[17] C. Morari, R. Jaquet, J. Phys. Chem. A 2005, 109, 3396 – 3404.
[18] R. Ahlrichs, M. Bꢀr, M. Hꢀser, H. Horn, C. Kꢁlmel, Chem. Phys.
Lett. 1989, 162, 165 – 169.
[10] M. Schmittel, C. Vavilala, J. Org. Chem. 2005, 70, 4865 – 4868.
[11] P. W. Musch, B. Engels, J. Am. Chem. Soc. 2001, 123, 5557 – 5562.
[12] T. Bekele, C. F. Christian, M. A. Lipton, D. A. Singleton, J. Am.
Chem. Soc. 2005, 127, 9216 – 9223.
6914 www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 6911 –6914