Concentration-dependent isotope effects. The photocyanation of naphthalene

Article abstract of DOI:10.1021/jp9800259

Concentration-dependent isotope effects (IEs) were measured in the competitive photocyanation of naphthalene and perdeuterionaphthalene in acetonitrile in the presence of electron acceptors. The concentration of the reagents naphthalene, cyanide, and oxygen influences the magnitude of the IE. Experiments in which the photocyanation is brought about via selective excitation of an electron acceptor show that the naphthalene-concentration dependence of the IEs is not caused by the naphthalene radical cations, which are intermediates in the photoreaction. This is corroborated by an electrochemical study of (the reactivity of) these radical cations. From these observations and complementary semiempirical PM3 calculations it is concluded that the variation of the IE with naphthalene concentration is due to an excited-state equilibration. A detailed reaction mechanism is proposed to account for the observed IEs.

Full text of DOI:10.1021/jp9800259

5456
J. Phys. Chem. A 1998, 102, 5456-5464
Concentration-Dependent Isotope Effects. The Photocyanation of Naphthalene†
Han Zuilhof,*,‡ Frans A. van Gelderen, Jan Cornelisse, and Gerrit Lodder*
Leiden Institute of Chemistry, Gorlaeus Laboratories, Leiden UniVersity, P.O. Box 9502,
2300 RA Leiden, The Netherlands
ReceiVed: October 30, 1997; In Final Form: January 20, 1998
Concentration-dependent isotope effects (IEs) were measured in the competitive photocyanation of naphthalene
and perdeuterionaphthalene in acetonitrile in the presence of electron acceptors. The concentration of the
reagents naphthalene, cyanide, and oxygen influences the magnitude of the IE. Experiments in which the
photocyanation is brought about via selective excitation of an electron acceptor show that the naphthalene-
concentration dependence of the IEs is not caused by the naphthalene radical cations, which are intermediates
in the photoreaction. This is corroborated by an electrochemical study of (the reactivity of) these radical
cations. From these observations and complementary semiempirical PM3 calculations it is concluded that
the variation of the IE with naphthalene concentration is due to an excited-state equilibration. A detailed
reaction mechanism is proposed to account for the observed IEs.
Introduction
be accounted for within the framework of ground-state concepts,
was observed for the intramolecular triplet-sensitized [2+2]
cyclization of sesquinorbornatrienes.8 On the basis of these data,
which form a limited selection from the literature, it appears to
be difficult to interpret the direction and magnitude of measured
secondary IEs in general in mechanistic terms.
The mechanistic implications of isotope effects (IEs) in
ground-state reactions are often well understood, and interpreta-
tion of these effects is nearly always consistent with the outcome
of other mechanistic probes. Therefore, in many mechanistic
studies use has been made of these effects.1 Much less use of
IEs is made in the study of photochemical reactions,2 mainly
because of interpretative problems with the magnitude and
direction of the effects, which hamper their application in the
elucidation of photochemical reaction mechanisms.
A way to overcome this problem is to study IEs in
photochemical reactions of which the mechanisms have already
been (partially) clarified by other methods. In this way, new
explanations for the IEs can be given a firm basis, as in the
case of the recently discovered concentration-dependent IE.9
This phenomenon was reported from our laboratories for the
meta photocycloaddition of cyclopentene to p-xylene and its
perdeuterated isotopologue.9 The observed IE on the quantum
yield of formation of products varied from 1.01 at low
concentrations of xylene (0.002 M) to 1.52 at high concentra-
tions (1.333 M), which was attributed to energy transfer via
the intermediate formation of xylene excimers. These inter-
mediates allow equilibration between excited xylene-h10 and
xylene-d10 to give excited xylene-d10 and xylene-h10, and vice
versa. This type of IE can occur in any reaction in which
intermediates can exchange excitation energy or charge before
reacting to product(s), and it is not a priori limited to
photochemical reactions. In this paper we report such phe-
nomena for the photocyanation of naphthalene.
The cause of the interpretative problems lies in the multitude
of decay paths of the electronically excited states of organic
molecules, which all can be affected by isotopic substitution.
Deuterium IEs have been reported in detail for photophysical
decay processes of relatively few compounds, mainly aromatic
hydrocarbons.3 For a rather small number of photochemical
reactions, conclusively interpretable IEs have been reported as
well. Clearest is the interpretation of photochemical primary
IEs, in which the same concepts can be used as in ground-state
primary IEs.4,5 With secondary IEs the situation is less
straightforward. The first observation of a (normal) secondary
deuterium IE that could be explained in a way similar to IEs
observed in ground-state reactions was reported for the γ-hy-
drogen abstraction of specifically deuterated bicycloalkyl phenyl
ketones.5 An inverse secondary deuterium IE was observed for
the photorearrangement of 2-deuterio-1-iminopyridinium ylide
and interpreted cautiously within the framework of ground-state
secondary deuterium IE concepts.6 This proved to be impossible
for the photochemical isomerization of dibenzohomobarrelene.7
This reaction shows an extremely large secondary deuterium
IE (kH/kD ) 4.8), which was proposed to originate in different
Franck-Condon factors for different reaction paths. Another
example of a secondary IE (kH/kD ) 1.11-1.16), which cannot
UV irradiation of naphthalene in dry acetonitrile in the
presence of cyanide ion and an electron acceptor such as
p-dicyanobenzene (DCB) results in the clean formation of 1-
and 2-cyanonaphthalene (eq 1). The mechanism of this reaction
has been studied extensively10,11 and is postulated to be as
depicted in eqs 2-5.11
† This paper is dedicated to Prof. Nicholas J. Turro on the occasion of
his 60th birthday.
‡ Present address: Laboratory of Organic Chemistry, Department of
Biomolecular Sciences, Wageningen Agricultural University, Dreijenplein
8, 6703 HB Wageningen, The Netherlands. E-mail: zuilhof@
SG1.OC.WAU.NL.
S1089-5639(98)00025-5 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/17/1998

Photocyanation of Naphthalene
J. Phys. Chem. A, Vol. 102, No. 28, 1998 5457
hν
analysis of every sample was performed at least five times. In
the competition experiments the ratio of naphthalene to per-
deuterionaphthalene was 1:1. Measured IEs on reactions refer
to the ratio of cyano-derivatives arising from naphthalene and
those from perdeuterionaphthalene. The experimental errors are
given as σn-1 standard deviations.
1
C10H8
9
8 (C10H8)*S
(2)
(C10H8)*S1 + DCB
9
ET8 (C10H8)+• + DCB-•
(3)
(4)
•
•
(C10H8)+• + CN- f 1-CN-C10H8 + 2-CN-C10H8
Equipment. GC analyses were carried out with a Packard
model 433 gas chromatograph (50 m CP SIL5-CB, L ) 0.25
mm) and a flame ionization detector. Hydrogen was used as
the carrier gas at an inlet-pressure of 72 kPa. GC analysis was
performed at 90 and 117 °C for analysis of isotopologues of
naphthalene and cyanonaphthalenes respectively, followed by
a rise to 250 °C to clean the column from high boiling trace
products for the next run. Injection and detection temperature
were 250 and 275 °C, respectively.
•
1-CN- or 2-CN-C10H8 + oxidant -H 8
•
1-CN- or 2-CN-C10H7 (5)
The singlet excited state of naphthalene is quenched by DCB
via electron transfer (ET), leading to the radical cation of
naphthalene and the radical anion of DCB (eq 3). The radical
cation reacts with cyanide at position 1 or 2 (eq 4), followed
by loss of hydrogen from the resulting radicals to yield the
products (eq 5). Which reagent functions as the oxidant has
thus far not been clarified.
Low-resolution mass spectra were obtained with a GC-MS
setup consisting of a Packard model 438A gas chromatograph
(25m CP-SIL 19 CB, L ) 0.32 mm) using helium as carrier
gas, coupled with a Finnigan Mat ITD 700 mass spectrometer
using electron impact and chemical ionization. UV spectroscopy
was performed using a Varian DMS 200 spectrophotometer.
Fluorescence emission spectra were recorded on a Spex
Industries, Inc. Fluorolog II spectrofluorimeter.
This mechanism is analogous to that of the anodic cyanation
of naphthalene, which starts with the oxidation of naphthalene
to its radical cation, followed by the reaction of this intermediate
with cyanide. In a solution open to air the resulting arenyl
radicals lose hydrogen to yield the products.12 These, 1- and
2-cyanonaphthalene, are formed in the same ratio as observed
in the photochemical cyanation, demonstrating the analogy
between the photochemical and electrochemical reactions.13
Semipreparative electrochemistry was carried out in a two-
compartment electrochemical cell open to the air with platinum
working and counter electrodes and a saturated calomel refer-
ence electrode (SCE). The potential source used was a PAR
Model 174 Polarographic Analyzer. All reported potentials from
the literature or from our own experiments are with respect to
the SCE.
We report here the IEs observed in the abovementioned
photocyanation of naphthalene and perdeuterionaphthalene at
systematically varied concentrations of all reagents. To un-
derstand more about the causes of the measured IEs the
photocyanation was also performed by selective excitation of
the electron acceptor 9,10-dicyanoanthracene. Furthermore, the
reactivity of the radical cation of naphthalene was studied via
electrochemical means. The study was completed with semiem-
pirical PM3 calculations of the IEs on energy exchange
equilibrations of the lowest singlet and triplet excited states of
naphthalene with the ground state and on the analogous charge-
transfer equilibration of its radical cation.
Reagents and Solvents. Naphthalene (Aldrich) and perdeu-
terionaphthalene (Aldrich) were sublimed twice before use;
DCB, 1-cyanonaphthalene, and 9,10-dicyanoanthracene (all
Aldrich) were sublimed once. 9,9′-bis(N-methylacridinium
nitrate) (BMA), 1,3,5-tri-tert-butylbenzene and 2,4,6-tri-
phenylpyrylium tetrafluoroborate (TPP) (all Aldrich), and
1-pentadecanol (Janssen Chimica) were used as received. 1-Cy-
anoperdeuterionaphthalene was synthesized from perdeuterio-
naphthalene by bromination16 followed by cyanation with
cuprous cyanide in DMF.17 2-Cyanonaphthalene was prepared
analogously from 2-bromonaphthalene.17 N-methylacridinium
iodide (MA) was synthesized according to the literature
procedure.18 Acetonitrile (Janssen, Spectrophotometric grade
or A. C. S. reagent) was dried on activated molecular sieves
(Janssen, 4 Å) and used without further purification.
Experimental and Computational Details
Experiments. The photochemical reactions were performed
at 20 °C under stirring in a 100 mL cylindrical Pyrex reaction
vessel open to the air, in which a tube was immersed, that
contained the lamp (Hanau TNN-15/32 low-pressure mercury
arc (254 nm) or Hanau TQ81 high-pressure mercury arc) and a
cooling liquid. Irradiations were performed at 254 nm (low-
pressure lamp, quartz tube, water as cooling liquid), at >300
nm (high-pressure lamp, Pyrex tube, water) or at >330 nm
(high-pressure lamp, Pyrex tube, acetone). For actinometry (λexc
) 254 nm; 20 °C) the photohydrolysis of 3-nitroanisole (3.5 ×
10-3 M) in a 0.1 M NaOH water/methanol (9:1) solution
(quantum yield ) 0.22)14 was used. The disappearance of
starting materials and appearance of products was studied as a
function of time, using an internal standard (1-pentadecanol)
and the kinetics described earlier.15 This was done by removing
aliquots (1.00 mL) of the reaction mixture at appropriate times,
injecting them in a water/ether mixture (to remove the salts),
and analyzing the ether layer by means of gas chromatography
(GC). An analogous method without internal standard was used
for the determination of the IEs. All irradiations were performed
in triplicate, except those at the two highest naphthalene
concentrations which were performed in duplicate. All elec-
trochemical experiments were performed in duplicate. The GC
Calculations. All calculations were performed using the
programs VAMP (based on AMPAC 1.0 and MOPAC 4.0) and
MOPAC 6.01, using the PM3 parameters.19 Properties of
radicals and radical cations were evaluated using the ROHF
formalism.20 Equilibrium IEs were calculated on the basis of
calculated zero-point energy differences.21
Results
Absorption and Emission. The absorption coefficients of
naphthalene and perdeuterionaphthalene were determined in
acetonitrile at 254 nm, the wavelength of irradiation used in
most of the photochemical experiments. Multiple measurements
of freshly prepared solutions yielded values of ꢀH ) 2632 (
21 and ꢀD ) 2709 ( 32 (M-1 cm-1). This implies an IE of
0.97 ( 0.01 on the absorbance.
Naphthalene and strong electron acceptors form charge-
transfer complexes with two charge-transfer bands.22 In aceto-

5458 J. Phys. Chem. A, Vol. 102, No. 28, 1998
Zuilhof et al.
TABLE 1: Quantum Yields of Formation of 1- and
2-Cyanonaphthalene from Naphthalene at λexc ) 254 nm
with 18-Crown-6 KCN and Various Electron Acceptors
(with Their Reduction Potential) in Acetonitrilea
TABLE 2: Isotope Effects on the Formation of
1-Cyanonaphthalene and Product Ratios at Varying Total
Naphthalene Concentrations (in mM) in Acetonitrile (λexc
254 nm) with DCB as Electron Acceptora
)
quantum
yield of
1-cyano-
quantum
yield of
2-cyano-
entry
[C10L8]
isotope effect
ratio 1-CN/2-CNC10L7
electron acceptor
(concentration;
reduction potential)
1
2
3
4
5
6
7
8
9
0.63
1.01
1.33
3.62
7.50
16.00
32.50
64.50
1.33b
7.50c
0.96 ( 0.02
0.98 ( 0.02
1.04 ( 0.02
1.10 ( 0.02
1.16 ( 0.02
1.20 ( 0.01
1.31 ( 0.01
1.35 ( 0.02
1.07 ( 0.02
0.98 ( 0.02
d
7.27 ( 0.55
8.00 ( 0.68
7.27 ( 0.45
7.29 ( 0.40
7.04 ( 0.59
8.10 ( 0.70
8.05 ( 1.35
6.31 ( 0.43
7.35 ( 0.39
[C10H8] naphthalene naphthalene
TPPb (0.8;-0.29)
8.8
2.4
2.4
8.2
8.0
7.5
1.33
<10-5
<10-5
BMAc (0.8;na.)
<10-5
<10-5
MAd (0.9;-0.46)
<10-5
<10-5
DCAe (0.8;-0.89)
9-CAf (0.8;-1.56)
DCB (0.8;-1.64)
DCB (0.8) + 1,3,5-tri-tert-
butylbenzene (28.2)
DCB (0.8) + argong
3.3 × 10-3
5.2 × 10-3
1.5 × 10-2
2.4 × 10-2
4.6 × 10-4
7.1 × 10-4
2.0 × 10-3
3.9 × 10-3
10
a [DCB] ) 0.8 mM; [18-crown-6-KCN] ) 18 mM. b 28.2 mM 1,3,5-
tri-tert-butylbenzene was added. c IE determined in separate experiments
for naphthalene and perdeuterionaphthalene. d Due to the small amount
of photoproducts formed, no consistent ratio could be obtained.
7.5
8.1 × 10-3
1.2 × 10-3
a All concentrations in mM; reduction potentials in V versus SCE;
[18-crown-6 KCN] ) 18.0 mM. b TPP ) 2,4,6-Tetraphenylpyrylium
tetrafluoroborate. c BMA ) 9,9′-bis(N-methylacridinium nitrate); na )
not available. d MA ) N-methylacridinium iodide. e DCA ) 9,10-
dicyanoanthracene. f 9-CA ) 9-cyanoanthracene. g The reaction was
performed under an argon atmosphere.
product-isomers does not. The data show DCB to be the
preferred electron acceptor of the ones under study, and DCB
was therefore used in most of the other experiments. It is
noteworthy that even if DCB is only present in catalytic
quantities with respect to naphthalene it is not consumed during
the reaction.
nitrile with DCB no such bands are observed, which is due to
the relatively high reduction potential of this acceptor (-1.6V23).24
Therefore the charge-transfer behavior of naphthalene and
perdeuterionaphthalene was studied using tetracyanoethene
(reduction potential: 0.24 V23).25 Maximum charge-transfer
absorptions were found for naphthalene at 558.4 and 427.1 nm
and for perdeuterionaphthalene at 559.0 and 428.1 nm. This
implies isotopic differences of 55 and 156 cal/mol, respectively
(corresponding to IEs of 1.10 and 1.31, respectively, at 20 °C).26
The energy of a charge-transfer transition is proportional to
the ionization potential of the donor molecule.22 The two
charge-transfer transitions observed have energies of 51.2 and
67.0 kcal/mol, respectively, and are ascribed to electron transfer
from the HOMO and HOMO-1 of naphthalene to tetracyano-
ethene. This difference of 15.8 kcal/mol is close to a reported
value in CCl4 of 14.9 kcal/mol22 and to the calculated difference
between HOMO and HOMO-1 energies using the PM3
parametrization: 13.8 kcal/mol.
The fluorescence of naphthalene is quenched by DCB.
Kinetics with this quencher, however, could not be accurately
determined because of the efficient fluorescence of DCB itself
in the spectral region of interest. Stern-Volmer constants were
therefore determined for both isotopologues using 9-cyanoan-
thracene as quencher, and found to be identical within experi-
mental error (kqτ ) 1200 ( 150 M-1). The quenching process
presumably follows the Rehm-Weller kinetics, and its rate
equals the rate of diffusion in acetonitrile.27 On the basis of
these data the fluorescence lifetimes28 of both isotopologues in
acetonitrile solutions open to air were 63 ( 8 ns, in agreement
with the general notion that the effect of perdeuteration on the
fluorescence lifetimes of polycyclic aromatic hydrocarbons is
(almost) nonexistent.3,29 The radiative lifetimes of both naph-
thalene and perdeuterionaphthalene are reported to be 96 ns in
deoxygenated cyclohexane.30 At the naphthalene concentrations
used, no excimer fluorescence was observed.
IEs on the product formation were studied in two ways: (a)
“intraexperimentally”, in which naphthalene and perdeuterio-
naphthalene in a 1:1 ratio were irradiated in one reaction vessel,
and (b) “interexperimentally”, where two experiments were
performed independently in separate reaction vessels, one with
the hydrogen and one with the deuterium compound.
IEs on the formation of both 1- and 2-cyanonaphthalene were
measured by GC analysis.32 The deuterated compounds have
shorter retention times than their hydrogen isotopologues on
all GC columns which were tested for this purpose (OV 101,
Carbowax, CP SIL5, CP SIL19). Base line separation was
achieved between the starting materials naphthalene and per-
deuterionaphthalene. For 1-cyanonaphthalene and 1-cyanoper-
deuterionaphthalene peak-valley ratios > 10 were obtained; for
2-cyanonaphthalene and 2-cyanoperdeuterionaphthalene peak-
valley ratios fell from about 8 at high product concentrations
to about 3 at the lowest product concentrations.
The interpretation of GC peak integrations in terms of IEs
hinges on known response factors of the flame ionization
detector for the different isotopologues. Therefore, 1-cyano-
naphthalene-d7 was synthesized independently, and response
factors were measured for this compound and its h7-isotopo-
logue, and for naphthalene and perdeuterionaphthalene. The
IEs on the response factors were 1.013 ( 0.010 and 1.017 (
0.007, respectively, at the detection temperature of 275 °C, and
independent of the amount of aromatic compound injected.
These values are equal within experimental error, and IEs are
not corrected for possible differences between them.
The IEs measured at various total naphthalene concentrations
([C10L8]; L ) H, D) are given in Table 2, together with the
1-CN-C10L7/2-CN-C10L7 ratios. For entries 2-8 these ratios
average out to 7.57 ( 0.45. Entry 9 presents an IE and product
ratio of a reaction in which part of the naphthalene radical
cations are created as free ions by using 1,3,5-tri-tert-butylben-
zene as cosensitizer.33 All entries refer to intraexperimental IEs,
except entry 10, which presents an interexperimental IE. The
IE results of entries 1-8 are graphically represented in Figure
1.
Photochemistry. When naphthalene is irradiated in a non-
degassed acetonitrile solution, containing cyanide and a suitable
electron acceptor, two products result: 1- and 2-cyanonaph-
thalene, in a ratio of about 7:1. To minimize effects due to ion
pairing of cyanide to its counterion, the complex of potassium
cyanide with 18-crown-6 ether (18C6 KCN) was used as the
source of cyanide.31 The quantum yields of product formation
depend on the electron acceptor (Table 1), but the ratio of
The IEs on the formation of product were also measured by
using 9,10-dicyanoanthracene (DCA) as electron acceptor and

Photocyanation of Naphthalene
J. Phys. Chem. A, Vol. 102, No. 28, 1998 5459
TABLE 4: Intraexperimental IEs on the Formation of
1-Cyanonaphthalene (in Acetonitrile, λexc ) 254 nm, DCB as
Electron Acceptor) with Varying Cyanide and Oxidant
Concentrations (All Concentrations in mM)a
isotope
effect
ratio 1-CN/2-
CNC10L7
entry [C10L8] [cyanide] atmosphere
1
2
3
16.0
16.0
16.0
16.0
7.5
0.52
air
1.04 ( 0.03 6.02 ( 0.23
0.99 ( 0.02 6.50 ( 0.51
1.14 ( 0.01 6.16 ( 1.06
1.20 ( 0.01 7.04 ( 0.59
1.16 ( 0.02 7.29 ( 0.40
0.52b air
2.7
18.0
18.0
18.0
air
air
air
argon
4
5
6c
7.5
2.5 ( 0.4
6.73 ( 0.39
a [DCB] ) 0.8 mM. b 17.48 mM 18-crown-6 KClO4 was added.
c Experiment performed in duplicate.
Figure 1. Isotope effects on the formation of 1-cyanonaphthalene in
the photocyanation of naphthalene in acetonitrile (λexc ) 254 nm), with
DCB as electron acceptor, with varying naphthalene concentration.
TABLE 5: Intraexperimental IEs on the Formation of
1-Cyanonaphthalene in the Anodic Cyanation of
Naphthalene at Two Naphthalene Concentrations (in mM)
and Three Oxidation Potentials (in V versus SCE)a
TABLE 3: Isotope Effects on the Formation of
1-Cyanonaphthalene in Acetonitrile, with DCA as Electron
Acceptor at Various Wavelengths of Excitation
[C10L8]
V
isotope effect
1-CN/2-CNC10L7
3.6
15
15
1.40
1.175
1.25
1.40
1.04 ( 0.02
1.02 ( 0.01
1.01 ( 0.01
1.01 ( 0.01
5.91 ( 0.29
5.17 ( 0.10
5.70 ( 0.13
5.42 ( 0.20
λ
exc (nm)
[C10L8]a
isotope effect
254
254
254
>300
>300
>330
>330
4.7
16.0
29.0
5.0
16.0
12.0
16.0
1.07 ( 0.02
1.16 ( 0.02
1.30 ( 0.01
1.06 ( 0.03
1.15 ( 0.02
1.00 ( 0.02
1.00 ( 0.02
15
a [KCN] ) 2.5 mM.
amounts of several trace products are also formed. It is likely
that, in solutions open to air, oxygen functions as the final
oxidant. This is based on the fact that the photocyanation does
not proceed in the absence of a catalytic amount of DCB, but
that no DCB is consumed. This suggests that DCB is the initial
electron acceptor, and that oxygen picks up the electron from
DCB radical anion, yielding oxygen radical anion.
a All concentrations in mM; [9,10-dicyanoanthracene] ) 0.8 mM;
[18-crown-6 KCN] ) 18.0 mM.
irradiating the solution at λexc ) 254 nm, > 300 nm, and >
330 nm, respectively (Table 3). At the first two wavelengths
of excitation, the same variation in the magnitude of the IEs
with varying naphthalene concentration is observed as with DCB
as electron acceptor. At λexc > 330 nm, there is no IE. In this
case, the acceptor is excited exclusively (eq 6) and the radical
cation of naphthalene is formed via electron transfer to the
excited acceptor (eq 7). Energy transfer from the excited
acceptor to ground-state naphthalene (eq 7a) is energetically
not possible. Thus, under these reaction conditions the singlet
excited state of naphthalene is not involved in the photocya-
nation.
Electrochemistry. Semipreparative electrochemistry was
carried out to investigate the role of naphthalene radical cations
in the determination of the IEs observed in the photochemical
reactions. IEs in the anodic cyanation were measured at
concentrations of naphthalene and perdeuterionaphthalene of 7.5
mM each and 1.8 mM each, and at a cyanide concentration of
2.5 mM, using three different oxidation potentials (1.175-1.40
V versus SCE) (Table 5). The ratio 1-CN-C10L7/2-CN-C10L7
was determined to be 5.55 ( 0.38 (averaged over all averaged
values for each experiment). No systematic variation of this
ratio was observed with varying naphthalene concentrations and
oxidation potential. The ratio is significantly different from that
obtained in the photochemical experiments.
hν
1
DCA
(DCA)*S1 + C10H8
(DCA)*S1 + C10H8 N DCA + (C10H8)*S1
9
8 (DCA)*S
(6)
(7)
ET8 DCA-• + (C10H8)+•
Calculations. PM3 calculations were performed to calculate
the equilibrium IEs which could be expected if the reactions
depicted in eqs 8-10 would occur. The method used has been
discussed in detail before.21 These reactions involve equilibra-
tion of the first singlet and triplet excited states of naphthalene
isotopologues via energy transfer, and the analogous equilibra-
tion between isotopologous naphthalene radical cations via
electron transfer. The first equilibrium may occur at high
concentrations of naphthalene, at which singlet excited naph-
thalene can transfer its energy to another naphthalene molecule
before decaying via bimolecular electron transfer to DCB or
unimolecular decay to the ground state. The second equilibrium
is calculated to show that such an IE is not restricted to singlet
states, although under the reaction conditions used the singlet
state is the reactive one.10 Equation 10 represents the equilibra-
tion that can take place both in the electro- and photochemical
experiments. If a radical cation lives long enough (due to low
cyanide concentration), it can transfer its charge to its isoto-
9
(7a)
The IEs on the formation of product with DCB as ET
sensitizer were also measured at various concentrations of all
reagents other than naphthalene (Table 4). Using constant C10L8
concentrations, the concentration of cyanide was systematically
varied. Furthermore, the effect of the total salt concentration
was tested by combining a low cyanide concentration with a
high total salt concentration (0.52 mM KCN, 17.48 mM KClO4,
entry 2). Variation of the DCB concentration from 0.8 to 8.0
mM, using 3.6 mM as the total naphthalene concentration and
18 mM cyanide, yielded no significant variation of the observed
IE and product ratios (data not in table). In the absence of
oxygen (entry 6), 1- and 2-cyanonaphthalene are still the
dominant products, but dihydrocyanonaphthalenes and small

5460 J. Phys. Chem. A, Vol. 102, No. 28, 1998
Zuilhof et al.
pologue. This would lead to IEs that depend on the cyanide
concentrations used.
C10H8 + (C10D8)*S1 {K
HOMO-LUMO excitation:
∆H ) -165 cal/mol; K ) 1.33
\
} (C10H8)*S1 + C10D8
(8)
CISD-calculation: ∆H ) - (calculation not successful)34
K
C10H8 + (C10D8)*T
{
} (C10H8)*T + C10D8
(9)
1
1
HOMO-LUMO excitation:
∆H ) -157 cal/mol; K ) 1.31
CISD-calculation: ∆H ) -56 cal/mol; K ) 1.10
C10H8 + (C10D8)+•
{
K} C10D8 + (C10H8)+•
(10)
∆H ) -141 cal/mol; K ) 1.27
For the calculation of excited-state properties two methods
were tested: representation of the lowest singlet state by
HOMO-LUMO excitation, or as calculated by the use of
configuration interaction using singly and doubly excited
configurations ()CISD). CISD-calculations were performed
with 10 orbitals: the HOMO-4 up to the LUMO+4, yielding
978 configurations. The 100 configurations with the lowest
energy were then used in the configuration interaction to
describe the molecular state.
As far as reproduction of experimental energies is concerned,
the HOMO-LUMO excitation representation of the lowest
singlet is better than the CISD-calculation: ES1 ) 97 and 76
kcal/mol, respectively, with experimental values of 90-95 kcal/
mol.23,35 Calculated values for ET1 of 51 and 40 kcal/mol
respectively were obtained, where 61 kcal/mol is the experi-
mental value.23 These results compare favorably with ab initio
calculations with large and flexible ANO-type basis sets which
calculated the S1 and T1 energies of naphthalene to be 104 and
78 kcal/mol, respectively (CASSCF),36a or 82 and 69 kcal/mol,
respectively (CASPT2).36b Our results also show that, in the
reproduction of experimental excited-state data, PM3 is no
significant improvement over MNDO, which calculates the S1
and T1 states to have energies of 98 and 57 kcal/mol,
respectively.37
Figure 2. Transition states for hydrogen abstraction from 1-cyano-1-
H-naphthyl radical by (a) the radical anion of oxygen and (b) by oxygen.
oxygen radical anion and by oxygen. For the reaction with
oxygen radical anion, the saddle point on this surface was found
at a geometry with the C-H bond to be broken at 1.352 Å and
the H-O bond length equal to 1.486 Å (Figure 2a). The
activation enthalpy was calculated to be <2 kcal/mol. For the
reaction with oxygen, the activation barrier was calculated to
be 28 kcal/mol, and the transition state was calculated with the
C-H bond to be broken equal to 1.444 Å, as well as the H-O
bond equal to 1.413 Å (Figure 2b). Grid calculations with
systematic variation of both the C-H and O-H distances to
simulate the potential energy surface of the hydrogen abstraction
by the radical anion of oxygen show a slight minimum in the
potential energy surface around r(C-H) ) 1.3 Å and r(O-H)
) 1.7 Å. This suggests that oxygen radical anion possibly forms
a weak hydrogen-bonded complex with 1-cyano-1-H-naphthyl
radical before abstracting the hydrogen atom.
Discussion
Inclusion of explicit configuration interaction using the PM3
parametrization does not lead to a better but, in contrast, to a
worse fit of the calculated energies with the experimental
excited-state energies. It is therefore not to be expected that
the second derivatives of the energy with respect to the nuclear
coordinates, which are calculated to obtain the zero-point
energies, are predicted better by the data obtained via CISD
calculations than by data calculated without inclusion of any
extra electron correlation correction. Therefore, the HOMO-
LUMO excitation data are supposedly more reliable for the
calculation of the equilibrium IEs (eqs 8 and 9). IEs are given
in reaction enthalpies (in cal/mol) and, assuming ∆S ) 0, in
the corresponding equilibrium constant at 293 K, the temperature
at which all experiments were performed.
Photochemical Cyanation. The photocyanation of naph-
thalene in acetonitrile in the presence of electron-accepting
sensitizers smoothly yields two products: 1- and 2-cyanonaph-
thalene, in a ratio of about 7:1. This ratio is in line with the
calculated charge distribution in the naphthalene radical cation.
The calculated charge at position 1 is +0.09, while at position
2 it is -0.05; when the charge of the bound hydrogen atoms is
added to that of the carbon atoms this yields values of +0.25
and +0.09. Both methods for the determination of the
electrostatic attraction of the nucleophile therefore show a
preference for position 1. Attack at this position is also favored
because the resulting radical has a lower energy than the
2-cyano-2-H-naphthyl radical (∆Hf ) 93.8 versus 97.5 kcal/
mol), due to more extensive resonance. The attack must be
relatively fast under the reaction conditions used, since inverse
IEs would have resulted on the product formation if the
combination reaction would have been rate limiting.38 The
n-cyano-n-H-naphthyl radicals lose hydrogen easily to oxygen
Furthermore, calculations were performed to investigate the
potential energy surface for hydrogen atom abstraction from
1-cyano-1-H-naphthyl radical (the intermediate formed after
attack by CN- at position 1 of naphthalene radical cation) by

Photocyanation of Naphthalene
J. Phys. Chem. A, Vol. 102, No. 28, 1998 5461
SCHEME 1: Postulated Mechanism for the
Photocyanation of Naphthalene
the relative concentration of excited C10H8 and excited C10D8
determines the magnitude of the IE.
At higher C10L8 concentrations the IE increases substantially,
approaching a plateau value around 1.35 at 64 mM. This
concentration dependence can be caused by any of the steps in
which naphthalene and perdeuterionaphthalene are competing
to acquire the ability to yield product. Since the reaction is
known to proceed via the lowest excited singlet state of
naphthalene under the reaction conditions used,10 a concentration
dependence involves this state or some intermediate directly
following or preceding this. Three steps are clearly imaginable.
First, excited naphthalene can transfer its excitation energy to
ground-state perdeuterionaphthalene and vice versa (eq 8). At
higher C10L8 concentrations more energy exchange can take
place. Therefore, kinetic IEs will approach the equilibrium IE
for energy exchange. Second, it is well-known that radical
cations of aromatic compounds form dimers with neutral
aromatics.39 For naphthalene the association constant in ben-
zonitrile has been reported to be 520 M-1 40
Via this reversible
.
radical anion. The calculated barrier for hydrogen atom transfer
from 1-cyano-1-H-naphthyl radical to oxygen radical anion is
<2 kcal/mol, while that to oxygen is 28 kcal/mol. This is
tantamount to a relative rate of 1017 at equal concentrations of
oxygen and oxygen radical anion at room temperature. There-
fore, transfer of the hydrogen atom from the radical will almost
exclusively occur to oxygen radical anion, despite the fact that
this is present in much lower concentrations than oxygen. The
reason that oxygen and oxygen radical anion were calculated
as hydrogen atom abstractors and not DCB radical anion is the
large rise in observed IE (from 1.16 to 2.5) under low [O2]
conditions. This primary IE points to a change in the rate-
limiting step, making the hydrogen abstraction step limiting due
to the low [O2-•]. The measured IE is in line with the structure
calculated for the transition state for formation of 1-cyano-
naphthalene, which shows significant C-H bond lengthening
(r(C-H) ) 1.352 Å). Experimental evidence for the hypothesis
that DCB picks up the electron from naphthalene, as well as
that its radical anion transfers this extra electron to oxygen,
comes from the observation that DCB is only needed in catalytic
amounts and that it is not consumed in the reaction. The control
experiment without DCB showed hardly any product formation
on the time-scale studied, while in the experiments with reduced
oxygen concentration DCB (present in 0.7% of the amount of
C10L8) is also not consumed. Under those conditions dihydro-
cyanonaphthalenes, resulting from disproportionation of n-H-
n-CN-naphthyl radicals, are formed besides cyanonaphthalenes,
in quantities slightly less than the cyanonaphthalenes (8.0 and
7.1% respectively, after 16% conversion). The fact that the
amount of cyanonaphthalenes is consistently slightly larger than
the amount of dihydrocyanonaphthalenes shows that some
oxygen is still present, which can reoxidize the small amount
of DCB radical anion formed. The whole reaction scheme is
summarized in Scheme 1.
dimer formation exchange of charge might occur between
isotopologous radical cations (eq 10). On the basis of the
absence of binaphthyl formation in the photochemical cyanation,
as well as on the basics of the IEs observed in the electrochemi-
cal and photochemical experiments (vide infra), this is unlikely.
Third, it has recently been proposed that concentration-depend-
ent photoinduced electron transfer does occur between cyanoan-
thracenes and naphthalene in acetonitrile.41 With increasing
naphthalene concentration, electron transfer results not only in
a 1:1 radical ion complex, but increasingly and more also in
the formation of a 1:2 complex,41,42 in which the positive charge
is distributed over two naphthalene molecules. It is unlikely
that 1:2 complexes are involved in the concentration-dependent
IEs since they display an increased rate of back electron transfer
with respect to analogous 1:1 complexes.41 This would result
in a smaller amount of product after a fixed time of irradiation,
which is incompatible with our experimental results.
We therefore propose the concentration dependence of the
IE on product formation to be caused by the reversible energy
exchange between excited naphthalene and ground-state per-
deuterionaphthalene (eq 8). This is in line with the observation
that the IE ) 1.00 if photocyanation takes place without
excitation of naphthalene (λexc > 330 nm, DCA as sensitizer),
at a concentration of C10L8 at which IE ) 1.15 (λexc > 300 nm;
DCA), 1.16 (λexc ) 254 nm; DCA), or 1.20 (λexc ) 254 nm;
DCB) if naphthalene can be photoexcited (Table 3).
The energy exchange might occur (a) at contact distance (via
excimer formation)9 or (b) via singlet-singlet energy transfer
via either a long-range (Fo¨rster) or a short-range mechanism.43
(a) Two arguments suggest that excimers are not involved in
this process. First, excimer fluorescence could only be observed
at naphthalene concentrations much larger (>0.5 M)44 than those
used in the reaction under study (<0.065 M). Second, the
formation of naphthalene excimers is less likely than that of
naphthalene radical cation dimers: the association constant of
Naphthalene Concentration-Dependent Isotope Effects in
the Photocyanation of C10L8. The photocyanation of naph-
thalene unambiguously shows an IE that is dependent on the
C10L8 concentration, when hydrogen and deuterium isotopo-
logues are competing for photons and material reagents (Table
2, Figure 1). When the C10L8 concentration is low, 0.63 mM,
the IE equals the IE measured interexperimentally, as well as
the difference in absorbance at 254 nm (0.96, 0.98, and 0.97,
respectively). Therefore it seems plausible that the IE at low
total naphthalene concentrations is caused by differences in
absorbance between naphthalene and perdeuterionaphthalene:
naphthalene excimer (0.72 M-1 45
is much lower than that of
)
naphthalene radical cation dimer (520 M-1),40 while the rates
of the quenching reaction of excited naphthalene by DCB and
naphthalene radical cation by cyanide are very similar.46
Binaphthyl products are expected if radical cation dimers are
intermediates, but since those products are not observed, the
intermediacy of excimers is even less likely. (b) On the other
hand, singlet-singlet energy transfer via the Fo¨rster mechanism
is also unlikely. The calculated concentrations at which energy
transfer via this mechanism becomes important for naphthalene

5462 J. Phys. Chem. A, Vol. 102, No. 28, 1998
Zuilhof et al.
and perdeuterionaphthalene are 1.1 and 0.5 M,47 substantially
higher than the concentrations used in this study. This leaves
as a last possibility that energy transfer between C10H8 and
C10D8 occurs via a short-range mechanism.
deuteration in toluene and p-xylene was very close to zero,53
as was the isotopic shift in charge-transfer spectra of benzene
versus perdeuteriobenzene with tetracyanoethene.26 Charge-
transfer spectra of naphthalene and perdeuterionaphthalene,
however, do show isotopic shifts (vide supra). The lowest
energy charge-transfer bands show shifts of 55 and 156 cal/
mol, respectively, comparable to the values for the isotopic shift
observed for toluene-h8 and -d8 and hexamethylbenzene-h18 and
-d18 with tetracyanoethene (117 and 206 cal/mol, respectively).26
The radical cations of these alkylbenzenes show kinetic IEs for
back electron transfer to cyanoanthracene radical anions up to
1.8!53 The kinetic IE has been demonstrated to increase with
the driving force for back electron transfer,53 in line with
theoretical predictions.54 The radical ion pair under study has
a driving force of 3.2 eV (Eox(naphthalene) ) 1.60 V, Ered(DCB)
) -1.60 V), at least 0.4 eV more exergonic than the examples
given in ref 53. It seems therefore reasonable to suppose there
will be a normal kinetic IE of substantial magnitude on the back
electron transfer within the geminate radical ion pair of
naphthalene and DCB (eq 11).
With regard to the calculations of the equilibrium IEs, it
should be noted that they point to the right direction and
magnitude of the observed IE, but are not useful to discriminate
between different IE-effecting steps (eqs 8-10), since the
calculated IEs vary only over a small range (1.27-1.33) for
very different processes.
Isotope Effects in the Anodic Cyanation of Naphthalene.
The above conclusion that, under the reaction conditions chosen,
the equilibration of eq 10 is not important is confirmed by the
IEs obtained in the electrochemical study. The anodic cyanation
of naphthalene shows an IE close to one (1.01-1.04), practically
independent of the total naphthalene concentration used. This
means that the radical cation-nucleophile combination must
be faster than the equilibration of eq 10. If this equilibration
would have been completed before further reaction would take
place, an IE > 1.10 would be expected (on the basis of the
charge-transfer spectra), and the computations suggest this to
be 1.27.48 By performing the electrochemical experiments at
varying potentials around the half-wave potential of naphthalene
(1.26 V versus SCE),49 the concentration of radical cations
produced was varied substantially. As can be seen from the
data in Table 5, electrolyses performed at 1.25 or 1.17 V give
identical results as those performed at 1.40 V. Therefore it can
be concluded that, as long as enough cyanide is available, the
equilibration of eq 10 does not take place, not even at high
oxidation potentials at which the amount of naphthalene radical
cations formed is limited by diffusion of the naphthalene to the
electrode, rather than by oxidation rates of naphthalene residing
near the electrode.50
[C10L8+• ‚‚‚ DCB-•] f C10L8 + DCB
(11)
How does this influence the observed IEs on product
formation with varying cyanide concentration? With decreasing
cyanide concentration the IE decreases. To explain the reduction
in the impact of the singlet energy exchange of eq 8 which
causes the IE, one has to assume that either cyanide acts as an
agent promoting the diffusion of naphthalene radical cation from
the solvent cage which it shares with DCB radical anion or that
cyanide reacts within the solvent cage with the radical cation.
The latter process is more plausible. At lower cyanide
concentrations more back electron transfer will occur, thereby
+•
decreasing the ratio C10H8+•/C10D8
. This will lead to a
Cyanide Concentration-Dependent Isotope Effects in the
Photocyanation of C10L8. Upon decrease of the cyanide
concentration from 18 to 0.52 mM, the IE decreases from 1.20
to 1.04 (Table 4). As added 18-crown-6 KClO4 does not
increase the IE (Table 4, entry 2), the change in solvent polarity
by decreasing salt concentrations is unimportant. Quenching
of excited C10L8 by cyanide is an inefficient process,51 and
therefore will hardly influence the equilibration of eq 8. If any
quenching would occur, however, an increase in the cyanide
concentration hampering the equilibration would reduce the IE
instead of increasing it as is observed. This suggests that the
overall observed IE can be influenced by reaction steps occurring
after the equilibration of eq 8. The first such step, quenching
of singlet excited naphthalene by DCB, occurs with a rate which
is diffusion controlled (vide supra) and will not be subject to
an IE. To explain the variation of the IE with the concentration
of cyanide we postulate that after the quenching a geminate
radical ion pair is formed, which can undergo any of the
following three processes (Scheme 1): (1) back electron transfer,
to give back the reactants naphthalene and DCB; (2) diffusion
of the radical ions out of the solvent cage, followed by capture
of cyanide by free naphthalene radical cation; (3) reaction of
naphthalene radical cation with cyanide in the solvent cage.52
The relative importance of the three processes depends on the
concentration of cyanide and so will the IEs associated with
each step.
reduction of the IE, as observed. Chemical evidence for the
(partial) reaction of cyanide and naphthalene radical cation
within the solvent cage can be found in the fact that the
photochemical reaction of naphthalene with cyanide results in
a different 1- to 2-cyanonaphthalene ratio (7.6:1) than the
electrochemical experiments (5.6:1). The radical cations in the
latter reaction are formed close to a positively charged electrode
from which they diffuse away quickly. Since the reaction with
cyanide is slower than diffusion controlled,11 cyanide can react
with the more or less free ions in that case. The geminate radical
ion pair being somewhat more selective for reaction with
cyanide ion than the free radical ion is consistent with the
observation that the ratio of 1-cyanonaphthalene/2-cyanonaph-
thalene seems to increase slightly with increasing cyanide
concentration (Table 4, entries 1, 3, and 4).
Attempts were made to create free naphthalene radical cations
by photochemical means. Experiments using a positively
charged electron acceptor (TPP or MA) or a doubly charged
one (BMA), to obtain geminate ion pairs with a very short
lifetime due to the lack of Coulombic attraction or the creation
of Coulombic repulsion after electron transfer,55 failed to yield
significant amounts of cyanation product (Table 1). In view
of the high reduction potentials of these three cationic sensitizers,
this is presumably caused by a very efficient energy-wasting
back electron transfer.56 Biphenyl (oxidation potential: 1.96
V),56 which is used in numerous cases to create aromatic radical
cations outside the initial solvent cage,57 has a radical cation
which is a strong enough electron acceptor to oxidize naphtha-
lene. It was, however, cyanated very effectively itself, thereby
hampering the analysis. The use of 1,3,5-tri-tert-butylbenzene
(oxidation potential: 2.01 V33) was more successful (see Table
Deuterium kinetic IEs are known to exist on the back electron
transfer from aromatic radical cations to the radical anion of
cyanoanthracene. Methyl-deuterated methylbenzenes show a
decreased rate of back electron transfer compared to their
hydrogen isotopologues.53 On the other hand, the effect of ring

Photocyanation of Naphthalene
J. Phys. Chem. A, Vol. 102, No. 28, 1998 5463
(15) (a) Van Ginkel, F. I. M.; Cornelisse, J.; Lodder, G. J. Am. Chem.
Soc. 1991, 113, 4261-4272. (b) Van Ginkel, F. I. M.; Cornelisse, J.; Lodder,
G. J. Photochem. Photobiol., A. 1991, 61, 301-315.
(16) (a) Furniss, B. S.; Hannaford, A. J.; Smith, P. W. G.; Tatchell, A.
R. Vogel’s Textbook of Practical Organic Chemistry, 5th ed.; Longman
Scientific and Technical: Harlow, U.K., 1989; p 863. (b) Yonemitsu, T.;
Kubo, K.; Gondo, Y. Memoirs of the Faculty of Science, Kyushu UniVersity,
Ser. C; 1987, 16, 13-18; CA 1988, 108, 221190n.
(17) Friedman, L.; Shechter, H. J. Org. Chem. 1961, 26, 2522-2524.
(18) Gebert, H.; Regenstein, W.; Bendig, J.; Kreysig, D. Z. Phys. Chem.
(Leipzig) 1982, 263, 65-73.
(19) Stewart, J. J. P. J. Comput. Chem. 1989, 10, 209-220. (b) Stewart,
J. J. P. J. Comput. Chem. 1989, 10, 221-245.
(20) (a) Dewar, M. J. S. The Molecular Orbital Theory of Organic
Chemistry; McGraw-Hill: New York, 1969; pp 249-278. Descriptions of
reactions of radical cations with nucleophiles using this method were shown
to be in accordance with experimentally observed reaction patterns; for
example, see: (b) Zuilhof, H.; Vertegaal, L. B. J.; Van der Gen, A.; Lodder,
G. J. Org. Chem. 1993, 58, 2804-2809. (c) Dinnocenzo, J. P.; Zuilhof,
H.; Lieberman, D. R.; Simpson, T. R.; McKechney, M. W. J. Am. Chem.
Soc. 1997, 119, 994-1004.
(21) (a) Zuilhof, H.; Lodder, G. J. Phys. Chem. 1992, 96, 6957-6962.
(b) Zuilhof, H.; Lodder, G. J. Phys. Chem. 1995, 99, 8033-8037.
(22) Foster, R. Organic Charge-Transfer Complexes; Academic Press:
London, U.K., 1969; p 67-72.
(23) Kavarnos, G. J.; Turro, N. J. Chem. ReV. 1986, 86, 401-449 (pp
422-423, Tables 9 and 11).
(24) Also see: (a) Mizuno, K.; Pac, C.; Sakurai, H. J. Chem. Soc., Chem.
Commun. 1975, 553. (b) Yasuda, M.; Pac, C.; Sakurai, H. J. Chem. Soc.,
Perkin Trans. 1 1981, 746-750.
(25) A comprehensive study of complex formation and charge-transfer
bands of (substituted) naphthalene(s) with tetracyanoethene has been
published: Frey, J. E.; Andrews, A. M.; Combs, S. D.; Edens, S. P.; Puckett,
J. J.; Seagle, R. E.; Torreano, L. A. J. Org. Chem. 1992, 57, 6460-6466.
(26) For an earlier example of this technique, see: Martens, F. M.;
Verhoeven, J. W.; De Boer, Th. J. Tetrahedron Lett. 1979, 2919-2920.
(27) (a) For an introduction to Rehm-Weller kinetics, see: Klessinger,
M.; Michl, J. Excited States and Photochemistry of Organic Molecules;
VCH Publishers: New York, 1995; p 285. (b) For the quenching of singlet
excited naphthalene by DCB in acetonitrile, a k-value of 12 × 1010 L/mol
s has been reported,27c supporting our assumption of diffusion-controlled
quenching. (c) Pac, C.; Nakasone, A.; Sakurai, H. J. Am. Chem. Soc. 1977,
99, 5806-5808. (d) Singlet excited naphthalene is quenched by 1,2-
dicyanobenzene, a weaker electron acceptor than DCB, and 9-cyanoan-
thracene in acetonitrile at room temperature with a diffusion-controlled rate
(k ) 1.8 × 1010 L/mol s): Tsuchida, A.; Tsujii, Y.; Ito, S.; Yamamoto,
M.; Wada, Y. J. Phys. Chem. 1989, 93, 1244-1248.
1). The quantum yield increased by a factor of 1.6, and the
ratio of 1-cyano- to 2-cyano-derivative dropped well below the
average value in the absence of 1,3,5-tri-tert-butylbenzene (6.31
versus 7.57) and approached the value obtained in the electro-
chemical experiments (5.55). These data support the hypothesis
that part of the naphthalene radical cations are created via
oxidation by 1,3,5-tri-tert-butylbenzene radical cation, and
therefore as free ions. For reaction with free ions the 1-CN-
to 2-CN-C10L7 ratio is lower. The higher quantum yield (Table
1, entry 7) is due to the reduced importance of energy-wasting
back electron transfer.58
Concluding Remarks
The present study shows: (1) the usefulness of the concept
of concentration-dependent IEs in the study of reaction mech-
anisms, (2) the necessity for adaptation of the proposed
mechanism for the photocyanation of naphthalene on the basis
of the observed concentration-dependent IE with variation of
the cyanide concentration, and (3) the (to the best of our
knowledge) first two examples of concentration-dependent IEs
in a photochemical reaction proceeding via electron transfer:
one involving energy exchange between an electronically excited
molecule and its ground-state isotopologue (eq 8) and the other
involving electron transfer within a ground-state ion pair (eq
11). Concentration-dependent IEs are therefore not limited to
photochemical reactions, but can be found in any reaction in
which intermediates have the possibility to transfer reactivity
to an isotopologue, before leading to product(s).
Acknowledgment. Use of the services and facilities of the
Dutch National SON Expertise Center CAOS/CAMM is grate-
fully acknowledged. The authors thank J. van Brussel for
dedicated technical assistance and Dr. J. L. Timmermans for
computational assistance. They are grateful to Prof. J. P.
Dinnocenzo (University of Rochester, Rochester, NY) for
clarifying discussions.
References and Notes
(28) The calculated lifetimes are based on a value of the diffusion rate
constant in ACN ) 1.9 × 1010 L/mol s. See: Murov, S. L.; Carmichael, I.;
Hug, G. L. Handbook of Photochemistry; Marcel Dekker: New York, 1993;
p 208.
(1) Melander, L.; Saunders: W. H. Reaction Rates of Isotopic
Molecules; Wiley: New York, 1980.
(2) Swenton, J. S. In Isotopes in Organic Chemistry; Buncel, E., Lee,
C. C., Eds.; Elsevier: Amsterdam, 1975; Vol. 1, Chapter 5.
(3) (a) Birks, J. B. Photophysics of Aromatic Molecules;
Wiley-Interscience: London, 1970. (b) Birks, J. B., Ed.; Organic Molecular
Photophysics; Wiley-Interscience: London, 1973 (Vol. 1), 1975 (Vol. 2).
(c) Avouris, P.; Gelbart, W. M.; El-Sayed, M. A. Chem. ReV. 1977, 77,
793-833.
(4) (a) Wagner, P. J.; Truman, R. J.; Puchalski, A. E.; Wake, R. J.
Am. Chem. Soc. 1986, 108, 7727-7738. (b) Engel, P. S.; Kitamura, A.;
Keys, D. E. J. Org. Chem. 1987, 52, 5015-5021. (c) Gritsan, N. P.;
Khmelinski, I. V.; Usov, O. M. J. Am. Chem. Soc. 1991, 113, 9615-9620.
(5) Lewis, F. D.; Johnson, R. W.; Kory, D. R. J. Am. Chem. Soc. 1974,
96, 6100-6107.
(6) Kwart, H.; Benko, D. A.; Streith, J.; Harris, D. J.; Shuppiser, J. L.
J. Am. Chem. Soc. 1978, 100, 6501-6502.
(7) Hemetsberger, H.; Bra¨uer, W.; Tartler, D. Chem. Ber. 1977, 110,
1586-1593.
(8) Paquette, L. A.; Ku¨nzer, H.; Waykole, L. Tetrahedron Lett. 1986,
27, 5803-5806.
(9) De Vaal, P.; Lodder, G.; Cornelisse, J. J. Phys. Org. Chem. 1990,
3, 273-278.
(10) For a summary of mechanistic studies up to 1986, see: Konuk,
R.; Cornelisse, J.; McGlynn, S. P. J. Chem. Phys. 1986, 84, 6808-6815.
(11) Niiranen, J.; Nieminen, K.; Lemmetyinen, H. J. Photochem.
Photobiol., A. 1991, 56, 43-53.
(12) Eberson, L.; Nilsson, S. Discuss. Faraday Soc. 1968, 45, 242-
246.
(13) Nilsson, S. Acta Chem. Scand. 1973, 27, 329-335.
(14) De Jongh, R. O.; Havinga, E. Recl. TraV. Chim. Pays-Bas 1966,
85, 275-283.
(29) (a) Li, R.; Lim, E. C. J. Chem. Phys. 1972, 57, 605-612. (b)
Kanamaru, N.: Bhattacharjee, H. R.; Lim, E. C. Chem. Phys. Lett. 1974,
26, 174-179.
(30) Berlman, I. B. Handbook of Fluorescence Spectra of Aromatic
Molecules; Academic Press: New York, 1971; pp 330-331.
(31) (a) Beugelmans, R.; Le Goff, M.-T.; Pusset, J.; Roussi, G. J. Chem.
Soc., Chem. Comm. 1976, 377-378. (b) Lemmetyinen, H.; Koskikallio, J.;
Lindblad, M.; Kuzmin, M. G. Acta Chem. Scand., Ser. A 1982, 36, 391-
397. (c) Tetraethylammonium cyanide was tested as well, but problems
arose with the necessary removal of the salt before GC analysis.
(32) This method of measuring IEs is both quantitatively superior to
and significantly faster and cheaper than analysis by GC/MS.
(33) The choice of 1,3,5-tri-tert-butylbenzene was based on the condition
that it absorbs >99% of the incoming light, that its radical cation can oxidize
naphthalene (the oxidation potentials of naphthalene and 1,3,5-tri-tert-
butylbenzene are 1.60 and 2.01 V, respectively), and that the bulky
substituent will decrease the rate of back electron transfer. See: Gould, I.
R.; Farid, S. J. Phys. Chem. 1993, 97, 13067-13072.
(34) No data for ∆H of S1 could be obtained using CI or CISD
calculations, since the gradient norm could not be reduced to values smaller
than 20. This was despite the fact that the heat of formation was minimized
to the same value from various starting geometries, using various optimiza-
tion algorithms (BFGS, EF, and NLLSQ), both with and without symmetry
constraints. Force calculations performed on this geometry showed several
imaginary frequencies up to i6000 cm-1
.
(35) (a) Turro, N. J. Modern Molecular Photochemistry, Benjamin/
Cummings Publishing Co.: Menlo Park, CA, 1978; pp 189, and 292. (b)
Huebner, R. H.; Mielczarek, S. R.; Kuyatt, C. E. Chem. Phys. Lett. 1972,
16, 464-469. (c) Dick, B.; Hohlneicher, G. Chem. Phys. Lett. 1981, 84,
471-478.

5464 J. Phys. Chem. A, Vol. 102, No. 28, 1998
Zuilhof et al.
(36) (a) Matos, J. M. O.; Roos, B. O. Theor. Chim. Acta 1988, 74, 363-
379. (b) Suter, H. U.; Ha, T.-K. J. Mol. Struct. (THEOCHEM), 1994, 309,
13-19.
(37) Dewar, M. J. S.; Fox, M. A.; Campbell, K. A.; Chen, C.-C.;
Friedheim, J. E.; Holloway, M. K.; Kim, S. C.; Liescheski, P. B.; Pakiari,
A. M.; Tien, T.-P.; Zoebisch, E. G. J. Comput. Chem. 1984, 5, 480-485.
(38) Reitsto¨en, B.; Parker, V. D. J. Am. Chem. Soc. 1991, 113, 6954-
6958.
(39) (a) Lewis, I. C.; Singer, L. S. J. Chem. Phys. 1965, 43, 2712-
2727. (b) Howarth, O. W.; Fraenkel, G. K. J. Am. Chem. Soc. 1966, 88,
4514-4515. (c) Badger, B.; Brocklehurst, B. Trans. Faraday Soc. 1969,
65, 2588-2594. (d) Tsujii, Y.; Tsuchida, A.; Yamamoto, M.; Momose, T.;
Shida, T. J. Phys. Chem. 1991, 95, 8635-8640.
(40) (a) Kira, A.; Arai, S.; Imamura, M. J. Phys. Chem. 1972, 76, 1119-
1124. (b) Terahara, A.; Ohya-Nishiguchi, H.; Hirota, N.; Oku, A. J. Phys.
Chem. 1986, 90, 1564-1571.
(48) A normal equilibrium isotope effect on the electron transfer between
an organic radical cation and its neutral parent has recently also been
observed for thianthrene: (a) Liu, Z.-L.; Lu, J.-M.; Chen, P.; Wang, X.-L.;
Wen, X.-L.; Yang, L.; Liu, Y.-C. J. Chem. Soc., Chem. Commun. 1992,
76-77. (b) Wen, X.-L.; Liu, Z.-L.; Lu, J.-M.; Liu, Y.-C. J. Chem. Soc.,
Faraday Trans. 1992, 88, 3323-3326.
(49) (a) Eberson, L.; Nyberg, K. Acta Chem. Scand. 1964, 18, 1568-
1570. (b) Bard, A. J.; Faulkner, L. R. Electrochemical Methods; Wiley:
New York, 1980; p 720.
(50) Albery, J. Electrode Kinetics; Clarendon Press: Oxford, 1975.
(51) Bunce, N. J.; Bergsma, J. P.; Schmidt, J. L. J. Chem. Soc., Perkin
Trans. 2 1981, 713-719.
(52) Attack on the radical cation by the nucleophile in the solvent cage
was also proposed by Lemmetyinen and co-workers for the photoamination
of phenanthrene in the presence of DCB, on the basis of flash photolysis
data: Nieminen, K.; Niiranen, J.; Lemmetyinen, H.; Sychtchikova, I. J.
Photochem. Photobiol., A 1991, 61, 235-243.
(53) (a) Gould, I. R.; Farid, S. J. Am. Chem. Soc. 1988, 110, 7883-
7885. (b) Gould, I. R.; Young, R. H.; Moody, R. E.; Farid, S. J. Phys.
Chem. 1991, 95, 2068-2080.
(54) Buhks, E.; Bixon, M.; Jortner, J. J. Phys. Chem. 1981, 85, 3763-
3766.
(55) Todd, W. P.; Dinnocenzo, J. P.; Farid, S.; Goodman, J. L.; Gould,
I. R. J. Am. Chem. Soc. 1991, 113, 3601-3602.
(56) Gould, I. R.; Ege, D.; Moser, J. E.; Farid, S. J. Am. Chem. Soc.
1990, 112, 4290-4301.
(57) Mattes, S. L.; Farid, S. In Organic Photochemistry, Padwa, A.,
Ed.; Marcel Dekker: New York, 1983; Vol. 6, pp 233-326.
(58) This suggests that the diffusing apart of 1,3,5-tri-tert-butylbenzene
radical cation and DCB radical anion is more effective than in the
corresponding case of naphthalene radical cations; cf. ref 33.
(41) Gould, I. R.; Farid, S. J. Am. Chem. Soc. 1993, 115, 4814-4822.
(42) Evidence for the existence of 1:2 complexes of tetracyanoethene
and naphthalene under conditions with a high naphthalene/tetracyanoethene
ratio has recently been presented in ref 25.
(43) See ref 27a, pp 287-294.
(44) (a) Do¨ller, E.; Fo¨rster, Th. Z. Phys. Chem. 1962, 31, 274-277. (b)
Mataga, N.; Tomura, M.; Nishimura, H. Mol. Phys. 1965, 9, 367-375.
(45) Aladekomo, J. B.; Birks, J. B. Proc. R. Soc. A. 1965, 284, 551-
565.
(46) The cyanide concentration is 25 times higher than the DCB
concentration, and quenching of excited naphthalene by DCB is about 20
times faster than the reaction of naphthalene radical cation with cyanide,
under the conditions Lemmetyinen and co-workers used: 1.0 × 1010 and
5.4 × 108 L/mol s (ref 11).
(47) Berlman, I. B. Energy Transfer Parameters of Aromatic Com-
pounds; Academic Press: New York, 1973; pp 308, and 310.