Rh(III)-photosensitized interconversion of norbornadiene and quadricyclane

Article abstract of DOI:10.1021/jp972007h

The utility of two Rh(III) diimine complexes, Rh(phen)₃³⁺ and Rh(phi)₂(phen)³⁺ (phen = 1,10-phenanthroline, phi = 9,10-phenanthrenequinone diimine), as sensitizers for the interconversion of norbornadiene (N) and quadricyclane (Q) has been investigated using steady-state photochemical and laser flash photolysis (LFP) techniques. Irradiation of acetonitrile solutions of Rh(phen)₃³⁺ and N causes slow conversion to Q. The reaction is reversible; irradiation of Rh(phen)₃³⁺ in the presence of Q leads to N. Irradiation of acetonitrile solutions of Rh(phi)₂(phen)³⁺ and Q yields N. However, this reaction is irreversible; irradiation of the Rh-(III) complex in the presence of N fails to afford Q. Irradiation of methanol solutions of either Rh(III) complex in the presence of N or Q affords minor amounts of two methanol-C7 adducts but fails to quench the N-Q interconversion reaction. The results are consistent with N-Q interconversion via an exciplex intermediate. The Rh(III)-sensitized deazatization of two cyclic azoalkane derivatives (Azo-N, Azo-Q) of N and Q was also investigated. Deazatization was achieved by Rh(phen)₃³⁺ but not Rh(phi)₂(phen)³⁺ sensitization. The results are consistent with a mechanism involving triplet energy transfer, but the involvement of exciplex intermediates cannot be ruled out. Bimolecular rate constants for quenching of the Rh(III) excited states by N, Q, Azo-N, and Azo-Q were determined by LFP.

Full text of DOI:10.1021/jp972007h

8834
J. Phys. Chem. A 1997, 101, 8834-8838
Rh(III)-Photosensitized Interconversion of Norbornadiene and Quadricyclane
Gregory W. Sluggett,†,§ Nicholas J. Turro,*,† and Heinz D. Roth*,‡
Departments of Chemistry, Columbia UniVersity, New York, New York 10027, and Rutgers UniVersity,
New Brunswick, New Jersey 08854
ReceiVed: June 19, 1997; In Final Form: September 4, 1997X
The utility of two Rh(III) diimine complexes, Rh(phen)33+ and Rh(phi)2(phen)3+ (phen ) 1,10-phenanthroline,
phi ) 9,10-phenanthrenequinone diimine), as sensitizers for the interconversion of norbornadiene (N) and
quadricyclane (Q) has been investigated using steady-state photochemical and laser flash photolysis (LFP)
3+
techniques. Irradiation of acetonitrile solutions of Rh(phen)3 and N causes slow conversion to Q. The
3+
reaction is reversible; irradiation of Rh(phen)3 in the presence of Q leads to N. Irradiation of acetonitrile
solutions of Rh(phi)2(phen)3+ and Q yields N. However, this reaction is irreversible; irradiation of the Rh-
(III) complex in the presence of N fails to afford Q. Irradiation of methanol solutions of either Rh(III)
complex in the presence of N or Q affords minor amounts of two methanol-C7 adducts but fails to quench
the N-Q interconversion reaction. The results are consistent with N-Q interconversion via an exciplex
intermediate. The Rh(III)-sensitized deazatization of two cyclic azoalkane derivatives (Azo-N, Azo-Q) of N
and Q was also investigated. Deazatization was achieved by Rh(phen)33+ but not Rh(phi)2(phen)3+ sensitization.
The results are consistent with a mechanism involving triplet energy transfer, but the involvement of exciplex
intermediates cannot be ruled out. Bimolecular rate constants for quenching of the Rh(III) excited states by
N, Q, Azo-N, and Azo-Q were determined by LFP.
Introduction
conversion of N to Q. Several Cu(I) compounds form ground-
state charge-transfer complexes with N that can be selectively
irradiated, resulting in efficient conversion to Q (eq 2).7 The
Ir(III)-sensitized conversion of N to Q involves the intermediacy
of excited-state charge-transfer complexes (exciplexes) which
collapse to Q (eq 3),10 while Ru(bpy)32+ (bpy ) 2,2′-bipyridine)
sensitizes the conversion of a substituted N to the Q derivative
by a classical triplet energy transfer mechanism.11
The interconversion of norbornadiene (N) and quadricyclane
(Q) has attracted considerable interest over the last two decades,
largely due to its potential importance in solar energy storage.1
Triplet-sensitized irradiation of N results in clean and efficient
conversion (φ ≈ 0.9) to the higher energy isomer Q.2,3 The
reversion of Q to N, with the liberation of the strain energy as
heat, can be affected by transition metal catalysts. The
conversion of Q to N may also be facilitated by one-electron
oxidation, followed by valence isomerization of Q•+ to N•+ (eq
1).4-14 Chemically induced dynamic nuclear polarization (CI-
Rhodium(III) diimine complexes are strong photooxidants.21-24
3+
For example, Rh(phen)3 and Rh(phi)2(phen)3+ (phen ) 1,-
DNP) studies have established the existence of Q•+ and N•+ as
two distinct species and have provided indirect evidence for
the valence isomerization reaction.15-18 We recently reported
the direct detection of Q•+ by time-resolved ESR and estimated
10-phenanthroline, phi ) 9,10-phenanthrenequinone diimine;
see Chart 1) possess relatively high excited-state reduction
potentials (*E(A
) 2.0 V21 and ca. 1.8 V24 vs SCE,
-/A)
respectively). Given the properties of these complexes, they
are obvious candidates as sensitizers for the N/Q system.
However, there is only a single brief mention of the Rh-
(phen)33+-sensitized conversion of N to Q.6
the rate constant for the valence isomerization to N•+ (kvi
≈
106 s-1).19
One significant problem with the N/Q solar energy storage
system is the relatively high energy of the electronic excited
states of N2,3,20 and the consequent need for high-energy
sensitizers, which typically absorb in the UV region. Since most
solar photons lie in the visible at wavelengths longer than 450
nm, only a few percent of the solar photons may be harvested
by the parent N/Q system. A significant amount of work has
gone into the development of norbornadiene derivatives and
sensitizers that absorb visible light.1 For example, transition
metal complexes containing copper,7 iridium,10 and ruthe-
nium11,12 have been shown to be effective in sensitizing the
To gain a better understanding of the energy surfaces
interconnecting N and Q, we have previously employed two
cyclic azoalkanes, Azo-N and Azo-Q (see Scheme 1), as
alternate precursors to the norbornadienyl and quadricyclanyl
biradicals (BN, BQ)3 and the azoalkane radical cations (Azo-
N•+, Azo-Q•+) generated by one-electron oxidation.25 The
azoalkane radical cations undergo rapid deazatization to afford
radical cations N•+ and Q•+, leading ultimately to the formation
of the neutral hydrocarbons N and Q.25
In this paper, we report results of Rh(III)-photosensitized
interconversion of N and Q and deazatization of Azo-N and
† Columbia University.
3+
Azo-Q. We have employed two Rh(III) sensitizers, Rh(phen)3
‡ Rutgers University.
and Rh(phi)2(phen)3+, with slightly different photophysical and
electrochemical properties. Bimolecular rate constants for the
§ Present address: Merck Research Laboratories, West Point, PA 19486.
X Abstract published in AdVance ACS Abstracts, October 15, 1997.
S1089-5639(97)02007-0 CCC: $14.00 © 1997 American Chemical Society

Interconversion of Norbornadiene and Quadricyclane
J. Phys. Chem. A, Vol. 101, No. 47, 1997 8835
TABLE 1: Oxidation Potentials, Triplet Energies, and
Bimolecular Rate Constants (kq) for Quenching of the
Rh(III) Excited States by Q, N, Azo-Q, and Azo-N in
Deoxygenated Acetonitrile Solution at 23 ( 2 °Ca
CHART 1
kq/109 M-1 s-1
+
E(D/D
)
3+
quencher (V vs SCE) ET (eV) *Rh(phen)3
*Rh(phi)2(phen)3+
Q
N
Azo-Q
Azo-N
0.91b
1.56b
2.7-3.5d
3.1 ( 0.2
1.5 ( 0.1
2.4 ( 0.1
4.1 ( 0.3
10.1 ( 0.4
0.52 ( 0.02
<0.01
2.66e
∼2.1c
2.43f
∼2.3c
2.60f
<0.01
a Errors are reported as twice the standard deviation from the least-
squares analysis of the LFP data according to eq 4. b Data from ref 31.
c Reference 25. d Reference 3. e Reference 20. f Reference 34.
SCHEME 1
in slow decomposition of N without the formation of Q or any
other detectable volatile products. Steady-state photolysis of a
methanol solution of Rh(phi)2(phen)3+ in the presence of Q
afforded N (79%) and trace amounts (ca. 1%) of the two
methanol addition products at ca. 70% conversion.
Rh(III)-Sensitized Irradiations of Azo-N and Azo-Q.
Irradiation (λ ) 300 nm) of an acetonitrile solution of
3+
Rh(phen)3 in the presence of Azo-N afforded Q (82%) and
N (18%) as the sole detected products at ca. 20% conversion.
3+
Photolysis of Rh(phen)3 in the presence of Azo-Q under
similar conditions yielded Q (47%) and N (50%). Photolysis
3+
of a methanol solution containing Rh(phen)3 and Azo-Q
afforded Q (46%), N (47%), and methanol adducts 1 and 2 (7%
combined) at 20% conversion. Similar irradiation of a methanol
solution of Rh(phen)33+ and Azo-N yielded Q (85%), N (14%),
and trace amounts of 1 and 2 (∼1% combined yield) at 15%
conversion. Neither Azo-Q nor Azo-N decomposed upon
prolonged irradiation (λ > 420 nm) of acetonitrile solutions
quenching of the Rh(III) excited states by N and Q and the two
azoalkanes were determined by laser flash photolysis (LFP).
The roles of energy transfer, electron transfer, and exciplex
formation are discussed.
containig Rh(phi)2(phen)3+
.
Laser Flash Photolysis. Laser flash photolysis (LFP) of
dilute deoxygenated acetonitrile solutions of Rh(phen)33+ affords
a readily detectable transient absorption centered at 500 nm
which decays with first-order kinetics (τ ) 170 ( 10 ns). This
transient has been assigned to the 3(π,π*) of the Rh(III)
complex.21-23 Addition of N, Q, Azo-N, and Azo-Q caused
the lifetime of the transient to decrease in each case. LFP of
dilute deoxygenated acetonitrile solutions of Rh(phi)2(phen)3+
affords a transient absorption centered at 460 nm which decays
with first-order kinetics (τ ) 160 ( 10 ns) and has been assigned
to an intraligand charge transfer (ILCT) excited state.24 The
lifetime of the Rh(III) excited state was reduced by additions
of N or Q; however, the lifetime of the transient was unaffected
by the addition of up to 0.15 M Azo-N or Azo-Q. Bimolecular
quenching rate constants (kq) were determined from plots of
the pseudo-first-order rate constant for decay of the Rh(III)
excited states (kdecay) according to eq 4, where ko refers to the
rate constant for decay in the absence of added quencher. The
quenching rate constants are listed in Table 1.
Results
Rh(phen)33+-Sensitized Irradiations of N and Q. Steady-
state irradiation (λ ) 300 nm) of dilute acetonitrile solutions
of Rh(phen)33+ in the presence of Q affords N in 84% chemical
3+
yield at ∼20% conversion. Irradiation of Rh(phen)3 in the
presence of N under the same conditions results in the
production of Q in 77% yield. No products other than N and
1
Q were detected by GC or H NMR analysis in either case.
The quantum yields for Rh(phen)33+-sensitized conversion of
Q to N (φQN) and N to Q (φNQ) are <0.01 in both cases.6
Steady-state photolysis of methanol solutions of Rh(phen)3
3+
in the presence of Q afforded N (72%) and two additional
products (in 23% combined yield). These products had identical
mass spectra (M+ ) 124; C7H8 + CH4O), consistent with
adducts between the C7 hydrocarbon and methanol. Such
adducts have been well documented;26,27 they are assigned to
ethers 1 and 2.27 Irradiation of methanol solutions of Rh-
kdecay ) ko + kq[Q]
(4)
3+
(phen)3 in the presence of N yielded Q (76%) and the same
two methanol adducts (16% combined yield).
Discussion
The two Rh(III) diimine complexes sensitize the conversion
of Q to N, and in the case of Rh(phen)33+ the process is clearly
reversible. However, the various reactions may follow different
pathways depending on the photophysical and electrochemical
properties of the sensitizer and quencher molecules. Our
approach has been to tune the properties of the Rh(III) sensitizers
by changing the ligands. For example, Rh(phen)33+ possesses
a higher excited-state energy and a higher excited-state reduction
Rh(phi)2(phen)3+-Sensitized Irradiations of N and Q.
Steady-state irradiation (λ > 400 nm) of dilute acetonitrile
solutions of Rh(phi)2(phen)3+ in the presence of Q affords N
in 65% yield at ca. 50% conversion. Irradiation of Rh(phi)2-
(phen)3+ in the presence of N under the same conditions resulted
potential (*E(A /A)) than Rh(phi)2(phen)3+ (see Chart 1). The
-
Rh(III) complexes have short-lived singlet excited states21-24

8836 J. Phys. Chem. A, Vol. 101, No. 47, 1997
Sluggett et al.
so that only the triplet excited states are intercepted by
quenchers. The quantum yield for intersystem crossing in *Rh-
(phen)33+ has been reported to be unity at room temperature.23
Potential reactions of the sensitizers with the donors employed
include electron transfer, exciplex formation, or energy transfer.
The free energy for electron transfer (∆G) from a donor to an
electronically excited acceptor may be calculated from the
Rehm-Weller equation28 (eq 5), where E(0,0) is the excited-
toacoustic calorimetry experiments indicated that the triplet
energy of N (2.66 eV)20 may be somewhat lower; thus, energy
transfer may be involved in the generation of Q in this case.
However, classical vertical triplet energy transfer is unlikely to
be important in the Rh(phen)33+-sensitized conversion of N to
Q, given the low efficiency of the reaction (φ < 0.01)6 compared
to that obtained with aryl ketone triplet sensitizers (φ ≈ 0.9).2
The Rh(phen)33+-sensitized conversion of Q to N could
involve full electron transfer, as the valence isomerization of
Q•+ to N•+ (eq 11) has been well documented.15,16,19 The
alternative possibility is reaction via exciplex intermediates (eqs
13 and 14). Classical vertical triplet energy transfer can be ruled
out due to the endothermicity of the process; however, we note
that nonvertical energy transfer mechanisms have been proposed
to account for the inefficient triplet-sensitized conversion of Q
to N.2
∆G ) E(D/D ) - E(A /A) - E(0,0) - e2/ꢀa
(5)
+
-
-/A)
state energy, E(A
is the oxidation potential of the reduced
+
acceptor, E(D/D is the oxidation potential of the donor, and
)
e2/ꢀa is a term to account for ion pairing. The oxidative strength
of an excited state is conveniently expressed by the excited-
state reduction potential (eq 6).29 Given the oxidation potentials
3+
The excited state of Rh(phen)3 is quenched efficiently by
*E(A /A) ) E(A /A) + E(0,0)
(6)
-
-
both N and Q (kq ≈ 109 M-1 s-1) despite the low quantum
yields for their interconversion under these conditions. The rate
constant for quenching by Q is twice as large as the value for
N, consistent with their redox potentials31 (Table 1) and
indicating the involvement of charge-transfer processes. The
quenching rate constants do not correlate with the triplet energies
of N and Q.
of the four donors employed (Table 1), electron transfer from
N and Q to the Rh(III) sensitizers is clearly exergonic, while
electron transfer from Azo-N and Azo-Q is at least slightly
3+
endergonic. Rh(phen)3 possesses a relatively high excited-
state energy (E(0,0) ) 2.75 eV),21-23 leading to the possibility
of triplet energy transfer in competition with electron-transfer
sensitization for the donors employed. On the other hand, Rh-
(phi)2(phen)3+ has an excited-state energy (E(0,0) ) 2.0 eV)24
that is lower than the triplet energies of N, Q, Azo-N, and Azo-Q
(Table 1).
The product studies in neat methanol were key to elucidating
the reaction mechanisms. Under these conditions, any free
radical cations (i.e. N•+, Q•+) should be trapped by methanol
with formation of methyl ethers;26,27 the rate of trapping Q•+
by pure methanol is faster than valence isomerization to N•+.27
The fact that the formation of N from Q (and vice versa) is not
quenched effectively by methanol rules out significant involve-
ment of full electron transfer generating free radical ions.
However, the formation of some methyl ethers (10-20%) in
both reactions requires the involvement of at least partial charge
transfer. The observation of ethers 1 and 2 does not allow one
to differentiate between free radical ions and exciplex intermedi-
ates. The ethers may arise by exo attack of methanol on free
radical ions, N•+ and/or Q•+, or on an endo exciplex.27 Both
reactions have precedent: with 1,4-dicyanobenzene as sensitizer,
N•+ and Q•+ are trapped by exo attack;26,27 likewise, with
1-cyanonaphthalene (CNN) as electron acceptor, an endo
exciplex (or encounter complex) between N and CNN is
captured from the exo face.32
Interconversion of Norbornadiene and Quadricyclane. Q
is inefficiently converted to N by sensitization involving both
Rh(III) complexes. N is converted inefficiently to Q by Rh-
3+
(phen)3 sensitization in acetonitrile solution, consistent with
an earlier preliminary report;6 Rh(phi)2(phen)3+-sensitized ir-
radiation of N fails to yield Q.
Potential reactions of the Rh(III) sensitizers with N and Q
are summarized in Scheme 2. Plausible mechanisms of Rh-
SCHEME 2
Rh(III)
9
hν8 3Rh(III)*
(7)
(8)
3Rh(III)* + D f Rh(III) + 3D*
(D ) N, Q)
We formulate the Rh(phen)33+-sensitized interconversion of
N and Q via excited-state complexes (exciplexes), [Rh-N]*
and [Rh-Q]*, with substantial charge-transfer character (eqs
13 and 14). Such reactions have precedent; for example, the
conversion of N to Q, sensitized by an Ir(III) diimine complex,
has been rationalized via an exciplex mechanism.10 Exciplex
intermediates have been invoked also in the 1-cyanonaphthalene-
sensitized irradiation of N27,32 or of an N derivative.33 However,
the valence isomerization within the complex is inefficient in
both directions; thus, the exciplexes preferentially decay to the
starting materials.
The Rh(phi)2(phen)3+-sensitized conversion of Q to N most
likely proceeds via exciplex intermediates. Vertical triplet
energy transfer is ruled out due to the relatively low excited-
state energy of Rh(phi)2(phen)3+ (Table 1). Electron transfer
from N and Q to the rhodium excited state is predicted to be
exergonic in both cases. Indeed, the rate constant for quenching
of the Rh(phi)2(phen)3+ excited state by Q is near the diffusion-
controlled limit in acetonitrile, whereas the rate constant for
quenching by N is ∼20 times slower, consistent with their
relative oxidation potentials.31 Again, the formation of N from
Q cannot be quenched in neat methanol, consistent with the
3D* f Q and/or N
3Rh(III)* + D f Rh(II) + D•+
Q•+ f N•+
(9)
(10)
(11)
(12)
(13)
Rh(II) + D•+ f Rh(III) + D
3Rh(III)* + D f (δ-Rh(III)- - -Dδ+)*
(
δ-Rh(III)- - -Dδ+)* f Rh(III) + N and/or Q (14)
(phen)33+-sensitized conversion of N to Q involve triplet energy
transfer (eqs 8 and 9) or formation of a (charge-transfer) exciplex
(eq 13) which decays in part to Q (eq 14). Full electron transfer
generating free N•+ (eq 10) followed by valence isomerization
to Q•+ and back electron transfer (eq 12) to yield Q can be
ruled out since N•+ is significantly lower in energy than Q•+.30
Triplet energy transfer is predicted to be an unlikely process
due to the endothermicity of the reaction; the triplet energy of
N (∼3.0 eV)2,3 is significantly higher than the excited-state
3+
energy of Rh(phen)3 (2.75 eV).21-23 However, recent pho-

Interconversion of Norbornadiene and Quadricyclane
J. Phys. Chem. A, Vol. 101, No. 47, 1997 8837
lack of formation of free radical ions to a significant extent.
with the predicted free energies of electron and energy transfer
sensitization.
The key difference appears to be that the [Rh(phi)2(phen)3+
-
Q]* exciplex decomposes in part to N, while the [Rh(phi)2-
(phen)3+-N]* exciplex does not yield Q in detectable amounts.
Deazatization of Azo-N and Azo-Q. Both azoalkanes
undergo efficient deazatization upon Rh(phen)33+ sensitization,
affording mixtures of N and Q. Product studies and LFP
experiments provide key evidence for the role of electron- or
energy-transfer-induced deazatization (Scheme 3). Electron
Summary and Conclusions
The Rh(phen)33+-sensitized irradiation of N or Q results in
inefficient conversion to the respective valence isomer. The
valence isomerization most likely proceeds via an exciplex
(charge-transfer) intermediate rather than via free radical ions
or via vertical triplet energy transfer. This conclusion is based
(i) on the formation of methanol adducts but the inability to
quench N-Q interconversion in methanol; (ii) on low quantum
yields for N-Q interconversion in both directions; (iii) despite
the near diffusion-controlled quenching of the Rh(III) excited
state by both N and Q; (iv) on the correlation of the *Rh-
SCHEME 3
3Rh(III)* + Azo-D f Rh(III) + 3Azo-D*
(D ) N, Q)
(15)
3+
(phen)3 quenching rate constants of N and Q with their
3Azo-D* f Q
(16)
(17)
(18)
(19)
(20)
oxidation potentials; and (v) on the predicted exergonicity of
electron transfer vs the predicted endergonicity of vertical triplet
energy transfer. Rh(phi)2(phen)3+ sensitizes the conversion of
Q to N; however, this reaction is irreversible. Irradiation of
the Rh(III) complex in the presence of N fails to afford Q.
Again, the methanol trapping experiments and the bimolecular
quenching rate constants support a mechanism involving the
intermediacy of an exciplex.
3Rh(III)* + Azo-D f Rh(II) + Azo-D•+
Rh(II) + Azo-D•+ f Rh(III) + Azo-D
Azo-D•+ f D•+
3+
Azo-N and Azo-Q were deazatized by Rh(phen)3 but not
Rh(II) + D•+ f Rh(III) + D
Rh(phi)2(phen)3+ sensitization. The results are consistent with
a mechanism involving classical triplet energy transfer; however,
the involvement of exciplex intermediates cannot be ruled out.
3Rh(III)* + Azo-D f [Rh- - -Azo-D]*
[Rh- - -Azo-D]* f Q and/or N
(21)
(22)
Experimental Section
Materials and Solvents. Norbornadiene (Aldrich) was
passed through activated alumina before use; quadricyclane was
used as received. 3,4-Diazatricyclo[4.2.1.02,5]nona-3,7-diene
transfer from Azo-Q to *Rh(III) is slightly endergonic, whereas
electron transfer from Azo-N is clearly endergonic. On the other
hand, triplet energy transfer from *Rh(phen)3 to both azoal-
3+
(Azo-N),35 3,4-diazaquadricyclo[6.1.0.02,605,9]non-3-ene (Azo-
kanes is exergonic (Table 1).34 Triplet (benzophenone, ac-
etophenone)-sensitized deazatization of both Azo-N and Azo-Q
affords Q in ∼90% yield.3 Thus, the formation of Q upon
sensitized deazatization of the azoalkanes potentially serves as
a probe for the involvement of triplet energy transfer (eqs 15
and 16).
- 24,39
Q),36,37 Rh(phen)33+‚3PF6- 38 and Rh(phi)2(phen)3+‚3PF6
,
were synthesized according to literature procedures. Acetonitrile
(Aldrich spectrophotometric grade) was distilled from calcium
hydride immediately before use. Methanol (Aldrich spectro-
photometric grade) was used as received.
Product Analysis and Characterization. Gas chromato-
graphic analyses were carried out using a Hewlett-Packard 5890
gas chromatograph equipped with a flame ionization detector,
a Hewlett-Packard 3392A integrator, and an HP-1 capillary
column (25 m × 0.2 mm; Hewlett-Packard, Inc.). The GC
injector port was maintained below 175 °C to avoid thermal
decomposition of the azoalkanes. GC/MS analyses were carried
out using a Hewlett-Packard 5890 gas chromatograph equipped
with a HP-5917A mass-selective detector and a HP-1 capillary
column (25 m × 0.2 mm; Hewlett-Packard, Inc.). 1H NMR
spectra were recorded on a Bruker AC80 (80 MHz) NMR
spectrometer in acetonitrile-d3 solution. Ultraviolet-visible
absorption spectra were recorded on a Hewlett-Packard HP8452A
UV spectrometer.
Steady-State Photolyses. The Rh(phen)33+-sensitized steady-
state photolysis experiments were carried out in a Rayonet
photochemical reactor equipped with six or 12 RPR-300 (300
nm) lamps. For the Rh(phi)2(phen)3+-sensitized photolyses, a
150 W Xe lamp was used in conjunction with a glass cutoff
filter (λ > 400 nm). Solutions containing 0.02-0.05 M
substrate, ca. 10-4 M Rh(III) complex, and 0.003 M m-tolunitrile
(internal GC standard) in 1 × 1 cm2 quartz cells were
deoxygenated prior to photolysis with a stream of dry argon
for 15 min and sealed with rubber septa. Alternatively,
deoxygentated acetonitrile-d3 solutions of the substrates, sen-
sitizer, and dichloromethane (internal standard) were irradiated
in 5 mm o.d. NMR tubes. Products were identified by 1H NMR,
On the other hand, Azo-N and Azo-Q show remarkably
different reactivity upon singlet-sensitized one-electron oxida-
tion.25 Azo-Q affords N exclusively, while Azo-N affords Q
as the major product (80%).25 The formation of Q as major
product in the Rh(phen)33+-sensitized irradiation of Azo-N is
consistent with either energy or electron transfer. However,
the comparable yields of N and Q obtained from Azo-Q indicate
that both electron and energy transfer processes may be
involved.
3+
Rh(phen)3 sensitizaton in neat methanol does not quench
the formation of N and Q; this indicates that, consistent with
the suggested free energies (eq 5), the deazatization mechanism
does not involve full electron transfer generating free radical
ions (eqs 17-20). However, the minor additional amounts of
methyl C7 ethers are consistent with the involvement of partial
charge transfer (exciplexes) as a possible minor pathway (eqs
21 and 22).
3+
Both Azo-N and Azo-Q quench *Rh(phen)3 with similar
rate constants (kq ≈ 109 M-1 s-1) despite their different
oxidation potentials; indeed, quenching by Q and N has rate
constants of similar magnitude despite their significantly lower
oxidation potentials (Table 1). This clearly points to the
involvement of energy transfer quenching by the azoalkanes
(eq 15).
Sensitization by Rh(phi)2(phen)3+ failed to cause deazatization
of either azoalkane. Likewise, both Azo-N and Azo-Q failed
to quench *Rh(phi)2(phen)3+ at a measurable rate, consistent

8838 J. Phys. Chem. A, Vol. 101, No. 47, 1997
Sluggett et al.
(8) Borsub, N.; Kutal, C. J. Am. Chem. Soc. 1984, 106, 4826.
GC/MS, and co-injection. Product yields were determined by
GC and H NMR analysis of the crude photolysates and are
based on disappearance of starting material based on the relative
peak areas versus internal standard. Quantum yields were
determined by valerophenone actinometry.40
1
(9) Kelley, C. K.; Kutal, C. Organometallics 1985, 4, 1351.
(10) Grutsch, P. A.; Kutal, C. J. Am. Chem. Soc. 1986, 108, 3108.
(11) Ikezawa, H.; Kutal, C.; Yasufuku, K.; Yamazaki, H. J. Am. Chem.
Soc. 1986, 108, 1589.
(12) Kutal, C.; Kelley, C. K.; Ferraudi, G. Inorg. Chem. 1987, 26, 3258.
(13) Ikezawa, H.; Kutal, C. J. Org. Chem. 1987, 52, 3299.
(14) Baird, N. C.; Draper, A. M.; de Mayo, P. Can. J. Chem. 1988, 66,
1579.
(15) Roth, H. D.; Schilling, M. L. M.; Jones, G., II. J. Am. Chem. Soc.
1981, 103, 1246.
(16) Roth, H. D.; Schilling, M. L. M. J. Am. Chem. Soc. 1981, 103,
7210.
(17) Roth, H. D.; Du, X.; Weng, H.; Lakkaraju, P. S.; Abelt, C. J. J.
Am. Chem. Soc. 1994, 116, 7744.
(18) Liu, Z.; Zhang, M.; Yang, L.; Liu, Y.; Chow, Y. L.; Johansson, C.
I. J. Chem. Soc., Perkin Trans. 2 1994, 585.
(19) Ishiguro, K.; Khudyakov, I. V.; McGarry, P. F.; Turro, N. J.; Roth,
H. D. J. Am. Chem. Soc. 1994, 116, 6933.
(20) Helms, A. M.; Caldwell, R. A. J. Am. Chem. Soc. 1995, 117, 358.
(21) Ballardini, R.; Varani, G.; Balzani, V. J. Am. Chem. Soc. 1980,
102, 1719.
(22) Ohno, T.; Kato, S. Bull. Chem. Soc. Jpn. 1984, 57, 3391.
(23) Indelli, M. T.; Carioli, A.; Scandola, F. J. Phys. Chem. 1984, 88,
2685.
(24) Turro, C.; Evenzahav, A.; Bossmann, S. H.; Barton, J. K.; Turro,
N. J. Inorg. Chim. Acta 1996, 243, 101.
(25) Sluggett, G. W.; Turro, N. J.; Roth, H. D. J. Am. Chem. Soc. 1995,
117, 9982.
(26) Gassman, P. G.; Olson, K. D. Tetrahedron Lett. 1982, 23, 19.
(27) Weng, H.; Roth, H. D. J. Org. Chem. 1995, 60, 4136.
(28) Rehm, D.; Weller, A. Isr. J. Chem. 1970, 8, 259.
(29) Roth, H. D. Top. Curr. Chem. 1992, 163, 133.
(30) Raghavachari, K.; Haddon, R. C.; Roth, H. D. J. Am. Chem. Soc.
1983, 105, 3110.
Laser Flash Photolysis Experiments. Laser flash photolysis
experiments employed the pulses (308 nm, ca. 80 mJ/pulse, 20
ns) from a Lambda Physik Lextra 50 excimer laser or a Lambda
Physik FL 3002 dye laser (Stilbene 420, 420 nm, ca. 2 mJ/
pulse, 20 ns) and a computer-controlled system which has been
described elsewhere.41 The dye laser was pumped by the
excimer laser. Solutions of the Rh(III) complexes were prepared
such that the absorbance at the excitation wavelength was ca.
0.3. Transient absorption spectra were recorded employing a
Suprasil quartz flow cell (1 × 1 cm2) and a flow system to
ensure that a fresh volume of sample was irradiated by each
laser pulse. Quenching rate constants were measured using
argon-saturated static samples contained in 1 × 1 cm2 Suprasil
quartz cells.
Acknowledgment. The authors thank the National Science
Foundation for generous support of this research and C. Turro
for the preparation of the rhodium complexes. G.W.S. thanks
NSERC (Canada) for a postdoctoral fellowship.
Supporting Information Available: Ground-state and tran-
sient UV-vis absorption spectra of the two Rh(III) sensitizers
3+
and a representative kinetic decay trace of the Rh(phen)3
excited state generated by LFP (5 pages). See any current
masthead page for ordering information.
(31) Gassman, P. G.; Yamaguchi, R.; Koser, G. F. J. Org. Chem. 1978,
43, 4392.
(32) Weng, H.; Roth, H. D. Tetrahedron Lett. 1996, 37, 4895.
(33) Jones, G., II; Xuan, P. T.; Schwarz, W. Tetrahedron Lett. 1982,
23, 5505.
(34) Turro, N. J.; Cherry, W. R.; Mirbach, M. J.; Mirbach, M. F.;
Ramamurthy, V. Mol. Photochem. 1978, 9, 111.
(35) Rieber, N.; Alberts, J.; Lipsky, J. A.; Lemal, D. M. J. Am. Chem.
Soc. 1969, 91, 5668.
(36) Moriarty, R. M. J. Org. Chem. 1963, 28, 2385.
(37) Cookson, R. C.; Gilani, S. S. H.; Stevens, I. D. R. J. Chem. Soc.
C 1967, 1905.
(38) Harris, C. M.; McKenzie, E. D. J. Inorg. Nuclear Chem. 1963, 25,
171.
(39) Pyle, A. M.; Chiang, M. Y.; Barton, J. K. Inorg. Chem. 1990, 29,
4487.
(40) (a) Wagner, P. J.; Kemppainen, A. E. J. Am. Chem. Soc. 1968, 90,
5896. (b) Wagner, P. J.; Kemppainen, A. E.; Schott, H. N. J. Am. Chem.
Soc. 1973, 94, 5604.
References and Notes
(1) Bren, V. A.; Dubonosov, A. D.; Minkin, V. I.; Chernoivanov, V.
A. Russ. Chem. ReV. 1991, 60, 451.
(2) (a) Hammond, G. S.; Wyatt, P.; DeBoer, C. D.; Turro, N. J. J. Am.
Chem. Soc. 1964, 86, 2532. (b) Murov, S.; Hammond, G. S. J. Phys. Chem.
1968, 72, 3797.
(3) Turro, N. J.; Cherry, W. R.; Mirbach, M. F.; Mirbach, M. J. J. Am.
Chem. Soc. 1977, 99, 7388.
(4) Jones, G., II; Chiang, S.; Becker, W. G.; Greenberg, D. P. J. Chem.
Soc., Chem. Commun. 1980, 681.
(5) Jones, G., II; Chiang, S.; Becker, W. G.; Welch, J. A. J. Phys.
Chem. 1982, 86, 2805.
(6) Kutal, C. AdV. Chem. Ser. 1978, 168, 158.
(7) (a) Schwendiman, D. P.; Kutal, C. Inorg. Chem. 1977, 16, 719.
(b) Schwendiman, D. P.; Kutal, C. J. Am. Chem. Soc. 1977, 99, 5677. (c)
Grutsch, P. A.; Kutal, C. J. Am. Chem. Soc. 1977, 99, 6460. (d) Sterling,
R. F.; Kutal, C. Inorg. Chem. 1980, 19, 1502. (e) Borsub, N.; Chang, S.;
Kutal, C. Inorg. Chem. 1982, 21, 538.
(41) McGarry, P. F.; Cheh, J.; Ruiz-Silva, B.; Turro, N. J.; Hu, S.; Wang,
J.; Nakanishi, K. J. Phys. Chem. 1996, 100, 646.