The di-π-methane reaction of 3,3-dimethyl-1,3-diphenylpropene revisited: Dynamics and barriers for competitive singlet state reactions [7]

Full text of DOI:10.1021/ja001724d

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
J. Am. Chem. Soc. 2000, 122, 8571-8572
8571
Scheme 1
The Di-π-methane Reaction of
3,3-Dimethyl-1,3-Diphenylpropene Revisited:
Dynamics and Barriers for Competitive Singlet State
Reactions
Frederick D. Lewis,*,† Xiaobing Zuo,† Rajdeep S. Kalgutkar,†
Miguel A. Miranda,‡ Enrique Font-Sanchis,‡ and
Julia Perez-Prieto§
Department of Chemistry, Northwestern UniVersity
EVanston, Illinois 60208-3113
Deparatmento de Qu´ımicia-Instituto de Tecnologia Quimica
UniVersidad Polite´cnica de Valencia, Valencia, E-46071-Spain
Deparatmento de Qu´ımicia Organica, Facultad de Farmacia
UniVersidad de Valencia, Burjassot, E-46100-Spain
Table 1. Room Temperature Quantum Yields, Rate Constants (krt)
and Activation Parameters for the Processes in Scheme 1
process
Φa
krt, s-1 b
Eact, kcal/molc
Ac
kf
0.041
0.011e
0.063e
0.81f
6.4 × 107
3.1 × 107
1.7 × 108
1.4 × 109
3.3 × 109
3.3 × 109
d
d
6.6
2.0g
1.9
0.44
d
d
ReceiVed May 22, 2000
kisc
ktwist
kpv
kdiπ
knr
ReVised Manuscript ReceiVed July 13, 2000
1.3 × 1013
4.3 × 1010g
8.0 × 1010
6.9 × 109
The mechanism of the photochemical di-π-methane rearrange-
ment of trans-3,3-dimethyl-1,3-diphenylpropene (DMPP) to trans-
3,3-dimethyl-1,2-diphenylcyclopropane (DMCP) was investigated
both by Hixson1 and by Zimmerman2 during the 1970s.3 Direct
irradiation results in the formation of DMCP in moderately high
quantum yield (Φ ≈ 0.40) along with a small amount of cis-
DMPP (Φ ) 0.065), whereas triplet sensitized irradiation
yields only cis-DMPP.2 A mechanism for decay of 1DMPP* based
on the studies of Hixson1 and Zimmerman2 is outlined in Scheme
1. According to this scheme there are four decay pathways for
1DMPP*. Two of these, fluorescence and intersystem crossing
(kf and kisc), are expected to be unactivated. The other two, phenyl-
vinyl bridging and twisting about the CdC bond (kpv and ktwist),
are expected to be activated. Furthermore, phenyl-vinyl bridging
is proposed to yield an intermediate Ia that can either revert to
ground-state DMPP (knr) or proceed to DMCP via the biradical
intermediate Ib (kdiπ), depending upon which cyclopropane bond
is cleaved (Scheme 1). This scheme is clearly too complex to
permit assignment of rates and barriers by means of simple kinetic
analysis. We report here the measurement of singlet lifetimes (τ)
and quantum yields for fluorescence (Φf) and the formation of
DMCP (Φdiπ) and cis-DMPP (Φc) over a broad temperature
range. Kinetic modeling provides conclusive evidence for the
formation of cis-DMCP via both singlet and triplet pathways and
for DMCP formation and nonradiative decay via a common
intermediate. It further provides activation parameters for both
of the activated primary photoprocesses and for the partitioning
of the intermediate Ia.
0.42
0.39h
a Quantum yields are directly measured except as noted. b Rate
constants obtained from the room-temperature quantum yields and
lifetime. c Activation parameters obtained from kinetic modeling as
described in text. d Unactivated processes. e The measured isomeriza-
tion quantum yield is the sum of these values. f The quantum yield
for phenyl-vinyl bridging is the sum of the quantum yields for di-π-
methane rearrangement and nonradiative decay. g Values of Epv
)
2.1 and Apv ) 3.0 × 1010 are obtained from fitting the Φf data in
Figure 1. h Quantum yield for nonradiative decay calculated using
eq 1.
Values of Φf, Φdiπ, Φc, and τ have been measured over the
temperature range 150-352 K in deoxygenated methylcyclohex-
ane solution. The data for 295.5 K is reported in Table 1 and is
in good agreement with that reported by Hixson1 for hexane
solution and with Zimmerman2 for tert-butyl alcohol solution at
room temperature. Values of τ have also been measure below
the solvent’s glass transition temperature ( ∼140 K) and are single
exponential over the entire temperature range. The temperature
dependence of Φf, Φdiπ, and Φc is shown in Figure 1 and the
Figure 1. Temperature dependence of the quantum yields for trans,cis
photoisomerization (Φc), fluorescence (Φf), di-π-methane rearrangement
(Φdiπ), and nonradiative decay (Φnr). Curve fitting is described in the
text.
temperature dependence of τ in Figure 2.4 The values of Φf
(Figure 1) and τ (Figure 2) increase with decreasing temperature,
to limiting values of 0.66 and 10.1 ns, respectively, below 140
K. Also shown in Figure 1 is the temperature dependence of the
quantum yield for nonradiative decay (Φnr) which is calculated
* E-mail: lewis@chem.nwu.edu.
† Northwestern University.
‡ Universidad Polite´cnica de Valencia.
§ Facultad de Farmacia.
(4) Fluorescence decay times were measured using a PTI Timemaster
apparatus capable of resolving decays g0.1 ns. The shortest decay time
measured is that of DMPP at 352 K (0.29 ns). Low-temperature fluorescence
quantum yield data is obtained by correcting the observed fluorescence
intensities for the temperature dependence of the concentration and refractive
index.
(1) Hixson, S. S. J. Am. Chem. Soc. 1976, 98, 1271-1273.
(2) Zimmerman, H. E.; Steinmetz, M. G.; Kreil, C. L. J. Am. Chem. Soc.
1978, 100, 4146-4162.
(3) For a recent review, see: Zimmerman, H. E.; Armesto, D. Chem. ReV.
1996, 96, 3065-3112.
10.1021/ja001724d CCC: $19.00 © 2000 American Chemical Society
Published on Web 08/17/2000

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
8572 J. Am. Chem. Soc., Vol. 122, No. 35, 2000
Communications to the Editor
mediate formed from the singlet state of DMPP, in accord with
the mechanism proposed by Hixson1 and Zimmerman2 (Scheme
1). These quantum yields are defined in eqs 3, 4 in terms of the
probabilities of forming Ia from the singlet state (P1) and forming
DMCP from Ib (P2). Simultaneous fitting of the values of Φdiπ
and Φnr as shown in Figure 1 gives activation parameters for both
kdiπ and knr (Table 1).
Φnr ) P1 × P2
(3)
(4)
Φ
diπ ) P1 × (1 - P2)
kpv
knr
P1 )
P2 )
kf + kisc + kpv + ktwist
knr + kdiπ
Rate constants and activation parameters for all of the processes
in Scheme 1 are summarized in Table 1. The rate constants for
fluorescence and intersystem crossing are similar to the values
previously reported for trans-1-phenylpropene (kf ) 3.0 × 107
s-1, kisc ) 4.7 × 107 s-1) as are the activation parameters for
singlet state twisting about the CdC bond (Atwist ) 6.3 × 1011
s-1 and Etwist ) 8.8 kcal/mol).5 The appearance of the absorption
and fluorescence spectra of DMPP and 1-phenylpropene are also
similar, in accord with the SCF-CI calculations of Zimmerman2
which indicate that excitation in DMPP is localized on the styryl
moiety. Both the value of A and Eact for pheny-vinyl bridging
are smaller than those for trans,cis isomerization (Table 1). The
smaller value of Apv vs Atwist is consistent with a more ordered
transition state for phenyl-vinyl bridging vs double bond torsion.
However, the difference in activation energies for these processes
is sufficiently large to render phenyl-vinyl bridging the dominant
activated decay process at room temperature. Both at higher
temperatures and very low temperatures trans,cis isomerization
should become the dominant photoprocess due to the large
preexponential for singlet isomerization and the unactivated nature
of triplet isomerization. The two decay pathways for the inter-
mediate Ia also have quite different activation parameters. The
smaller value of Anr vs Adiπ is consistent with a more ordered
transition state for nonradiative decay, which requires overlap of
the benzyl radical with the cyclopropyl bond undergoing frag-
mentation, and the smaller value of Enr vs Ediπ is consistent with
the more highly exergonic nature of the nonradiative decay
process which leads to ground-state DMPP (knr) vs conversion
of Ia to Ib (kdiπ).
The results of this investigation provide a complete description
of the competing reaction pathways for both the singlet state of
DMPP and the first intermediate formed in the di-π-methane
rearrangement. They are consistent with the stepwise mechanism
for 1,2-phenyl migration1,2 and not with a concerted 1,2-migration
process as proposed by Robb et al.6 for the di-π-methane
rearrangement of 1,4-pentadiene. The Robb energy surface may
be applicable to the truncated model but not to molecules with
extended conjugation. The quantum yield for di-π-methane
rearrangement is reported to be significantly lower for 1,3-
diphenylpropene vs DMPP.1 The effects of substituents at C-3
upon both phenyl-vinyl bridging and the behavior of the
intermediate Ia will be the subject of a subsequent report.
Figure 2. Temperature dependence of the singlet lifetime (τ) and the
rate constant for trans,cis photoisomerization (kiso). Curve fitting is
described in the text.
using eq 1
Φnr ) 1 - (Φf + Φdiπ + 2Φc)
(1)
from the sum of the measured quantum yields at each temperature
and the assumption that trans,cis isomerization via either a singlet
or triplet mechanism occurs via a twisted intermediate that decays
to a 1:1 mixture of trans and cis isomers (Φtwist ) 2Φc).5 Unlike
the values of Φdiπ which decrease with decreasing temperature,
the values of Φnr display a maximum at ∼200 K and decrease at
either higher or lower temperature.
Rate constants for each of the primary processes in Scheme 1
can be calculated from the measured quantum yields and lifetimes
at a given temperature. Values obtained at room temperature are
reported in Table 1. The calculated fluorescence rate constant (kf
) Φfτ-1) is independent of temperature. The calculated rate
constant for trans,cis isomerization (kiso ) 2Φcτ-1) is temperature-
dependent (Figure 2). The isomerization data is bimodal, in accord
with the occurrence of both triplet (unactivated) and singlet
(activated) pathways.5 Nonlinear data analysis provides a value
of kisc ) 3.1 × 107 s-1 for intersystem crossing, and values of
Atwist ) 1.3 × 1013 s-1 and Etwist ) 6.6 ( 0.1 kcal/mol for singlet
state torsion. Since the values of kf, kisc, Atwist, and Etwist have been
independently determined, fitting the temperature dependence of
the lifetime (Figure 2) to eq 2 provides values of Apv ) 4.3 ×
1010 s-1 and Epv ) 2.0 ( 0.1 kcal/mol. Fitting the Φf data in
Figure 1 to a modified form of eq 2 provides similar activation
parameters for phenyl-vinyl bridging (Table 1). With kinetic data
for all four of the primary photoprocess in hand it is possible to
compare the experimental and calculated temperature dependence
of Φc, as shown in Figure 1 (solid line). The good agreement
between the experimental and calculated values provides support
for the kinetic model.
twist/RT
τ-1 ) kf + kisc + Apv e-∆E /RT + Atwist e-∆E
(2)
pv
Acknowledgment. Funding for this project was provided by NSF
The temperature dependence of Φdiπ and Φnr is uniquely
consistent with these processes occurring via a common inter-
Grant CHE-9734941 and by Spanish DGES Grant PB97-0339.
JA001724D
(5) Lewis, F. D.; Bassani, D. M.; Caldwell, R. A.; Unett, D. J. J. Am. Chem.
Soc. 1994, 116, 10477-10485.
(6) Ruguero, M.; Bernardi, F.; Jones, H.; Olivucci, M.; Ragazos, I. N.;
Robb, M. A. J. Am. Chem. Soc. 1993, 115, 2073-2074.