Formation and behavior of fluorescent Lewis acid-base exciplexes and triplexes between 3-aminostilbenes and aliphatic amines

Article abstract of DOI:10.1021/jp036723x

The excited singlet states of trans-3-aminostilbene and its N-methyl derivatives are strongly fluorescent in cyclohexane solution and have large singlet state dipole moments. Addition of low concentrations of alkylamines results in a continuous red shift

Full text of DOI:10.1021/jp036723x

J. Phys. Chem. A 2004, 108, 1425-1434
1425
Formation and Behavior of Fluorescent Lewis Acid-Base Exciplexes and Triplexes between
3-Aminostilbenes and Aliphatic Amines
Frederick D. Lewis,* Rajdeep S. Kalgutkar, and Todd L. Kurth
Department of Chemistry, Northwestern UniVersity, EVanston, Illinois 60208-3113
ReceiVed: September 12, 2003; In Final Form: December 2, 2003
The excited singlet states of trans-3-aminostilbene and its N-methyl derivatives are strongly fluorescent in
cyclohexane solution and have large singlet state dipole moments. Addition of low concentrations of alkylamines
results in a continuous red shift of the emission maximum and decreasing fluorescence intensity. Analysis of
the fluorescence behavior using a combination of singular value decomposition with self-modeling and kinetic
analysis provides evidence for the sequential formation of a 1:1 complex (exciplex) and 1:2 complex (triplex)
between the excited stilbene and ground state alkylamine, both of which are strongly fluorescent. Both the
formation and decay of the exciplex and triplex are dependent upon the extent of amine N-alkylation, primary
amines forming the most stable exciplex and triplex. Similarly, N-aminoalkyl derivatives of the aminostilbenes
form intramolecular exciplexes that in turn form 1:1 complexes with added amines. Addition of diaminoalkanes
to the aminostilbenes results in sequential formation of 1:1 and 1:2 complexes rather than the formation of
a triplex with a single molecule of the diaminoalkanes. Excited state complex formation is attributed to a
Lewis acid-base interaction between the excited stilbene (lone pair acceptor) and ground state amine (lone
pair donor). An alternative explanation for the red-shifted emission based on Suppan’s theory of solvent
dielectric enrichment is found to be incompatible with the experimental results.
Introduction
CHART 1
The interactions of excited state aromatic molecules with
ground state tertiary aliphatic amines have been extensively
investigated.1-8 Both inter- and intramolecular interactions
involving tertiary aliphatic amines can result in the formation
of fluorescent, charge-transfer (CT) stabilized exciplexes that
possess large dipole moments.2-6,9 Exciplex stability increases
with decreasing amine ionization potential and is relatively
insensitive to steric effects. Experimental and computational
studies of arene-amine exciplexes indicate that they exist in
shallow conformational minima in which the amine lone pair
overlaps nonspecifically with the delocalized arene π*-orbitals.5
Interactions of singlet arenes with secondary or primary amines
results in the formation of nonfluorescent exciplexes, primary
amines being less reactive as a consequence of their higher
ionization potentials.
Some exciplexes have been observed to interact with a second
ground state molecule to form a termolecular complex.10,11 The
use of a strong donor or acceptor results in the formation of a
CT stabilized triplex or “exterplex” species. Chandross reported
that the intramolecular naphthalene-trialkylamine exciplex can
interact with small polar molecules such as propionitrile or
dimethylformamide in nonpolar solvents to form stoichiometric
fluorescent complexes.9 Fluorescence quenching of molecules
with intramolecular charge-transfer (ICT) excited states and
arene-amine exciplexes by ground state amines correlates with
the nucleophilicity of the amines rather than their ionization
potentials.12-15 The interactions between excited state lone pair
acceptors and ground state donors have received far less
attention than have CT interactions.
complexes.16 Complex formation in nonpolar solvents and the
absence of a correlation of complex stability with amine
ionization potential led us to suggest that these complexes are
stabilized by lone pair donor-acceptor or Lewis acid-base
(LAB) interactions. We report here the results of a detailed
investigation of the formation and behavior of both inter- and
intramolecular LAB exciplex and triplex species formed upon
the reaction of the aminostilbenes 1-3 and their N-aminoalkyl
derivatives 4 and 5 (Chart 1) with primary, secondary, and
tertiary aliphatic amines and with R,ω-diaminoalkanes. The
overlapping fluorescence spectra of the monomer, exciplex, and
triplex have been deconvoluted and the kinetics of their
formation and decay have been resolved using singular value
decomposition with self-modeling (SVD-SM) followed by
kinetic analysis.17,18 The results of this study provide the basis
for an explicit structural model for the exciplex and triplex
species.
Experimental Section
Synthesis of 1-3. m-Nitrobenzaldehyde (Aldrich) was
reacted with triphenylphosphonium chloride (Aldrich) in a CH2-
Cl2-H2O (50% in K2CO3) dual phase system using tetrabutyl-
ammonium iodide (Aldrich) as a phase-transfer catalyst (10 mol
%).19 The reaction mixture was stirred at room-temperature
overnight under a N2 atmosphere. After completion of the
We recently reported that interaction of the singlet state of
3-aminostilbene with ground state primary, secondary, and
tertiary aliphatic amines in nonpolar solvents results in the
sequential formation of fluorescent, stoichiometric 1:1 and 1:2
10.1021/jp036723x CCC: $27.50 © 2004 American Chemical Society
Published on Web 01/29/2004

1426 J. Phys. Chem. A, Vol. 108, No. 8, 2004
Lewis et al.
reaction, the CH2Cl2 layer was washed with water several times
and then dried with potassium carbonate. The trans-m-nitro-
stilbene isomer was enriched by refluxing the resulting cis-
trans mixture in benzene with a catalytic amount of I2 prior to
chromatography. Purification was carried out by column chro-
matography (SiO2/hexanes-ethyl acetate (80:20), 230-400
mesh SiO2) to remove the cis isomer. The trans isomer was
further purified by recrystallization from MeOH, providing a
pale yellow solid in 40% yield. Reduction of the nitro group to
the amino group was carried out by using Zn/HCl-AcOH as
the reducing agent,20 affording 1 in 85% yield. Purification of
1 was carried out by recrystallization from HPLC grade MeOH.
Mp ) 119-120 °C, lit. mp ) 120-121 °C.21 1H NMR (CDCl3,
500 MHz): δ 7.49 (2H, d, J ) 8.0 Hz), 7.34 (2H, t, J ) 8.0
Hz), 7.24 (1H, t, J ) 8.0 Hz,), 7.14 (1H, t, J ) 8.0 Hz), 7.03
(1H, d, J ) 16.0 Hz) 7.01 (1H, d, J ) 16.0 Hz), 6.85 (1H, s),
6.60 (1H, d, J ) 8.0), 6.29 (1H, d, J ) 8.0 Hz), 3.68 (2H, s).
Reaction of 1 with formaldehyde and cyanoborohydride afforded
a mixture of 2 and 3. Separation and purification of 2 and 3
were carried out by column chromatography (SiO2/hexanes-
ethyl acetate (80:20), 230-400 mesh SiO2). For 2: mp ) 50-
and were distilled, under dry nitrogen, from KOH immediately
prior to use. The cyclohexane-amine solutions were prepared
via weight. Refractive Index measurements were made by
utilizing a Bausch and Lomb Abbe-type refractometer.
Dielectric constant measurements were performed using a
GenRad 1658 RLC Digibridge fitted with a Digibridge BNC
adapter connected to a two terminal, shielded, stainless steel
liquid dielectric cell.24 The dielectric constants Cx and dissipa-
tions D of each alkylamine-cyclohexane solvent mixture were
measured directly. Using the capacitances (C) of the empty cell
Ca, corrected with D, and that of cyclohexane Cs, a cell (Cc)
and ground capacitance (Cg) were determined and the dielectric
constants were subsequently obtained; see eqs 1-4.25
Cx
1 + D2
C )
(1)
C0 - Ca
ꢀ0 - ꢀa
Cc )
(2)
(3)
Cg ) C0 - Ca
Cs - Cg
1
52 °C; H NMR (CDCl3, 500 MHz) δ 7.50 (2H, d, J ) 8.0
Hz), 7.34 (2H, t, J ) 8.0 Hz), 7.24 (1H, t, J ) 8.0 Hz), 7.17
(1H, t, J ) 8.0 Hz), 7.13 (1H, d, J ) 16.5 Hz), 7.04 (1H, d, J
) 16.5 Hz), 6.89 (1H, d, J ) 8.0 Hz), 6.75 (1H, s), 6.54 (1H,
d, J ) 8.0 Hz), 3.74 (2H, s), 2.88 (3H, s). For 3: mp ) 74.5-
76.5 °C, lit. mp ) 75.5-76.5 °C;22 1H NMR (CDCl3, 500 MHz)
δ 7.53 (2H, d, J ) 8.0 Hz), 7.36 (2H, t, J ) 8.0 Hz), 7.25 (2H,
t, J ) 8.0 Hz), 7.10 (2H, s), 6.94 (1H, d, J ) 8.0 Hz), 6.8
(1H, s), 6.69 (1H, d, J ) 8.0 Hz), 3.00 (6H, s).
ꢀ )
(4)
Cc
Spectroscopic Measurements. UV-vis spectra were mea-
sured on a Hewlett-Packard 8452A diode array spectrometer
using a 1 cm path length quartz cell. Total emission spectra
were measured on a SPEX Fluoromax spectrometer. All samples
were deaerated for 30 min with dry nitrogen prior to analysis
and had less than 0.15 absorbance at the wavelength of
excitation. Low-temperature spectra were measured in a Suprasil
quartz EPR tube (i.d. ) 3.3 mm) using a quartz liquid nitrogen
coldfinger dewar at 77 K. Total emission quantum yields were
measured by comparing the integrated area under the emission
curve at an equal absorbance and the same excitation wavelength
as an external standard, 9,10-diphenylanthracene (Φf ) 1.0 at
298 K in cyclohexane).26 All emission spectra are uncorrected
and the estimated error for the quantum yields is (10%.
Fluorescence decays were measured on a Photon Technolo-
gies International (PTI) Timemaster stroboscopic detection
instrument with a gated hydrogen or nitrogen lamp using a
scatter solution to profile the instrument response function.27
Nonlinear, least-squares fitting of the decay curves employed
the Levenburg-Marquardt algorithm as described by James et
al. and implemented by the Photon Technologies International
Timemaster (version 1.2) software.28 Goodness of fit was
determined by judging the ø2 (<1.3 in all cases), the residuals,
and the Durbin-Watson parameter (>1.6 in all cases). Solutions
were degassed under vacuum (<0.1 Torr) through five freeze-
pump-thaw cycles.
Synthesis of 4 and 5. Preparation of 4 and 5 was via the
method of Guillard.23 trans-m-Bromostilbene was prepared and
purified, via the method used for the nitro compounds described
above, utilizing m-bromobenzaldehyde as a starting reagent. One
equivalent of trans-m-bromostilbene, 2 equiv of t-BuONa, 2
mol % of 1,1′-bis(diphenylphosphino)ferrocence (dppf), and 1
mol % of (dppf)PdCl2‚CH2Cl2 were combined under dry
nitrogen. Subsequently, 30 mL of anhydrous dioxane was added,
followed by addition via syringe of 3 equiv of either 1,3-
diaminopropane or 1,4-diaminobutane. The solution was re-
fluxed overnight and subsequently reduced in volume, quenched
with minimal water, and then taken up in CH2Cl2. The organic
layer was washed twice with concentrated HCl. The aqueous
layers were made strongly basic and extracted with CH2Cl2.
The resulting oil was then purified by column chromatography
(SiO2/ethyl acetate-isopropylamine (95:5), 230-400 mesh
SiO2). Both 4 and 5 were found to have greater than 98.5%
trans isomer as estimated by GC (typical yields were less than
20%). For 4: mp ) 61-64 °C; 1H NMR (CDCl3, 500 MHz) δ
7.50 (2H, d, J ) 8.0 Hz), 7.34 (2H, t, J ) 8.0 Hz), 7.24 (1H,
t, J ) 8.0 Hz), 7.16 (1H, t, J ) 8.0 Hz), 7.07 (1H, d, J ) 16.5
Hz), 7.03 (1H, d, J ) 16.0 Hz), 6.87 (1H, d, J ) 8.0 Hz), 6.75
(1H, s), 6.53 (1H, d, J ) 8.0 Hz), 3.99 (1H, s), 3.24 (2H, t, J
) 8.0 Hz), 2.87 (2H, t, J ) 7 Hz), 1.79 (2H, p, J ) 7.0 Hz),
Results and Discussion
Intramolecular Complex Formation. The absorption spectra
of the m-aminostilbenes 1 and 3 have been previously de-
scribed.29,30 In contrast to the parent stilbene and its p-amino
derivatives, which display a single long-wavelength absorption
band, 1 and 3 display a long-wavelength band or shoulder near
340 nm and a more intense band near 300 nm. The appearance
of multiple long-wavelength transitions has been attributed to
splitting of the lowest singlet state by configuration interaction
as a consequence of the loss of symmetry in the m-aminostil-
benes.30 A similar effect has been reported for m-(dimethylami-
no)benzonitrile.31 The spectra of the secondary amines 2, 4, and
1
1.54 (2H, s). For 5: mp ) 73-79 °C; H NMR (CDCl3, 500
MHz) δ 7.45 (2H, d, J ) 8.0 Hz), 7.34 (2H, t, J ) 8.0 Hz),
7.24 (1H, t, J ) 8.0 Hz,), 7.16 (1H, t, J ) 8.0 Hz), 7.06 (1H,
d, J ) 16.5 Hz), 7.02 (1H, d, J ) 16.0 Hz), 6.87 (1H, d, J )
8.0 Hz), 6.73 (1H, s), 6.52 (1H, d, J ) 8.0 Hz), 3.17 (2H, t, J
) 7.0 Hz), 2.75 (2H, t, J ) 7.0 Hz), 1.68 (2H, p, J ) 7.0 Hz.),
1.57 (4H, p, J ) 7.0 Hz), 1.55 (2H, s).
Dielectric Constant and Refractive Index Measurements.
HPLC grade cyclohexane was used as the bulk solvent. The
alkylamines were obtained in their highest purity from Aldrich

Lewis Acid-Base Exciplexes and Triplexes
J. Phys. Chem. A, Vol. 108, No. 8, 2004 1427
Figure 1. Absorption spectra of 1-5 in cyclohexane solutions, 298
K: 1 (black), 2 (red), 3 (cyan), 4 (blue), 5 (green).
Figure 2. Fluorescence spectra of 1-5 in cyclohexane solutions, 298
K: 1 (black), 2 (red), 3 (cyan), 4 (blue), 5 (green).
that a change in molecular conformation, presumably the
formation of an intramolecular complex between the amino-
stilbene and the tethered primary amine, is responsible for the
red-shifted solution fluorescence of 4 and 5.
The dipole moments of the ground and excited singlet states
of 1 and 3 have been previously reported (Table 1).29,30 Ground
state dipole moments for 2 and 5 were calculated with PM3/
ZINDO as implemented in CAChe.34 Excited state dipole
moments can be obtained from the dependence of the fluores-
cence maxima on the Lippert-Mataga solvent polarity param-
eter ∆f
5 are more similar in appearance to that of the primary amine
1 than the tertiary amine 3 (Figure 1). N-Methylation is known
to influence the geometry of arylamines, tertiary amines
generally having less planar structures than primary or secondary
arylamines.32 Changes in geometry are reflected in the oscillator
strengths.33 The absorption maxima of 1-5 (Table 1) are
relatively insensitive to solvent polarity, displaying small
frequency shifts (<2 nm) in both polar hydroxylic or nonhy-
droxylic solvents.
The fluorescence spectra of 1 and 3 have also been previously
described and consist of a broad band with a shoulder at high
frequency, which is more pronounced in the case of 3 vs 1.29,30
The spectrum of 2 is intermediate between those of 1 and 3,
both in terms of band shape and emission maximum (Figure
2). The effects of N-methylation on the appearance of the
fluorescence spectra are similar to those for other arylamines.31,32
The fluorescence quantum yield for 2 is also intermediate
between those reported for 1 and 3, and its fluorescence lifetime
and rate constant are similar to those of 1 (Table 1). The smaller
fluorescence rate constant for 3 is consistent with its weaker
long-wavelength absorption band.
The fluorescence spectra of 4 and 5 are red-shifted by ca.
1000 cm-1 with respect to those of 2 (Figure 2). Their
fluorescence quantum yields are similar to that of 2; however,
their singlet lifetimes are somewhat longer, resulting in smaller
fluorescence rate constants. The fluorescence emission and
excitation spectra of 2, 4, and 5 are similar at 77 K in
2-methyltetrahydrofuran (MTHF) glasses (Figure 3), as are their
fluorescence decay times (Table 1). The observation of red-
shifted fluorescence in solution but not in a 77 K glass suggests
ꢀ - 1
2ꢀ + 1
n2 - 1
2n2 + 1
∆f )
-
(5)
where ꢀ and n are the solvent dielectric constant and refractive
index.35 The solvent dependence of the fluorescence maxima
can be described by
1
4πꢀ0
2
νfl ) -
µe(µe - µg)∆f + C
(6)
3
(
[
)
(
)
]
hca
ꢀ0 is the permittivity of free space, h is Planck’s constant, c is
the speed of light, a is the cavity radius, µe is the relaxed excited
state dipole, and µg is the ground state dipole. Lippert-Mataga
plots for 2 and 5 are shown in Figure 4. As expected, the
calculated ground and excited state dipole moments for 2 are
similar to those for 1 and 3. The calculated ground state dipole
moment of 5 is somewhat larger than those of 1-3, plausibly
reflecting the influence of the tethered primary amine. The
TABLE 1: Observed and Calculated Spectral Parametersa
1
2
3
4
5
λabsb, nm
298 (327)
4.38 (3.97)
388
298 (338)
4.38 (3.81)
401
445, 426
3.27
7.5, 12.1
0.74
0.99
276, 298 (343)
4.15,4.15 (3.41)
412
298 (339)
4.37 (3.73)
419
449, 429
3.47
9.6
0.8
298 (338)
4.31 (3.70)
423
454, 427
3.50
12.6, 12.6
0.71
0.56
b
log(ꢀmax
)
λfl(298 K), nm
λfl(298 K, 77 K) (MTHF), nm
Stokes shift (×10-3)c, cm-1
τf(298 K), 77 K, ns
Φf
3.37
7.5
0.78
1.04
1.5
3.21
13.0, 13.0
0.72
0.55
1.7
10-8kf, s-1
0.83
µgd (D)
1.6
12.5
2.9
µee (D)
11.9
12.1
10.9, 13.8f
a 298 K data in cyclohexane and 77 K data in methylcyclohexane unless otherwise noted. b Low energy band in parentheses. c Calculated by
Berlman’s method.26 d Value calculated using PM3/ZINDO. e Calculated using eq 6 and a value of a ) 5 and 6 Å for 5.

1428 J. Phys. Chem. A, Vol. 108, No. 8, 2004
Lewis et al.
Figure 3. Excitation and emission spectra of 2, 4, and 5 in MTHF
solutions at 298 K (solid) and 77 K (dash).
Figure 5. Fluorescence spectra of 1 with PrNH2, ABCO, and DAB at
298 K, each series having an amine concentration range of 0-250 mM.
Added amine results in red-shifted fluorescence.
TABLE 2: Results of SVD-SM and Kinetic Fitting for
Cyclohexane Solutions of 1 and 2a
1S* 1S*:A 1S*:A2
b
λmax λmax
λmax
ke k-e kde
kt k-t kdt kqt
1 PrNH2 461 404
424 10 0.13 0.04 7.8 0.38 0.18 0.58
422 11 0.25 0.09 4.0 0.19 0.24 2
Pr2NH 439 406
Et3N
ABCO
DAB
441 407
409
418
431 11 0.04 0.04 2.4 0.01 0.12 0.26
439 11 0.12 0 2.0 0.61 0.29
8.1 0.1 0.09 0.7 0.05 0.67 0
413
0
2 PrNH2 447 421
430 11 0.01 0.15 0.84 0.07 0.47 0.01
435 11 0.08 0.20 0.32 0.43 0.49 0.10
Pr2NH 437 421
Et3N
438 417
429
7.9 0.17 0.14 0.13 0.62 0
0
Figure 4. Lippert-Mataga plots for 2 (0) and 5 (O).
a Values are 1 × 10-9, kd constrained to equal φfτf -1 ) 0.1 × 109.
b In pure PrNH2, Pr2NH, and Et3N, respectively.
smaller slope of the Lippert-Mataga plot for 5 vs 2 results in
a smaller excited state dipole moment, if the same cavity radius
(a ) 5 Å)30 is assumed. However, the use of a slightly larger
cavity radius (ca. 6 Å) results in a calculated excited state dipole
moment similar to those of 1-3.
There are numerous reports of the formation of fluorescent
ICT exciplexes for tertiary aminoalkylarenes.4-6 However, to
our knowledge there are no previous reports of the formation
of a fluorescent exciplex (either inter- or intramolecular) of a
singlet arene with a primary or secondary alkylamine. In
addition, the ICT exciplexes formed by tertiary aminoalkylarenes
have much larger dipole moments than those observed for 4 or
5. Both monomer and exciplex fluorescence is observed for CT
exciplexes, whereas only a single band is observed for 4 or 5.
Thus the weakly red-shifted emission observed for 4 and 5 in
nonpolar solvents cannot be attributed to the formation of a CT
stabilized exciplex.
solutions of 1-3 in nonpolar solvents results in quenching of
their fluorescence intensity and a continuous red shift in the
fluorescence maxima, as shown for 1 with PrNH2 and ABCO
(azabicyclooctane) in cyclohexane solution in Figure 5. The
extent of the red shift, ca. 2000 cm-1, with low concentrations
of amine (<75 mM) is double that observed for the intramo-
lecular complexes formed by 4 or 5. Diminishing in magnitude,
red shifts continue from 75 to 250 mM amine but remain blue-
shifted with respect to pure amine solvents (Figure 5, Table 2).
Smaller red shifts are observed in moderately polar solvents
such as MTHF upon addition of comparable amine concentra-
tions. Fluorescence quenching but no red-shifted emission is
observed in more polar solvents; e.g., 2 in acetonitrile and PrNH2
displays no red shift with increasing amine concentration and
yields a linear Stern-Volmer plot with kqτ ) KQ ) 3 × 109
M-1
.
Interaction of Excited Aminostilbenes with Monoamines.
Addition of primary, secondary, or tertiary aliphatic amines to
The appearance of red-shifted emission suggests that the
decrease in intensity is not a consequence of straightforward

Lewis Acid-Base Exciplexes and Triplexes
J. Phys. Chem. A, Vol. 108, No. 8, 2004 1429
Figure 7. PrNH2 concentration dependence of the fluorescence
intensity of 1 at 513 (O) and 351 nm (0).
Figure 6. Amine concentration dependence of fluorescence intensity
of 1-3 with PrNH2, Pr2NH, Et3N, ABCO, diaminopropane, diamino-
pentane, diaminobutane (DAB), and diaminoheptane, in cyclohexane.
Stern-Volmer fluorescence quenching of the fluorescent singlet
state to the ground state. Non-Stern-Volmer behavior is also
evident in plots of the integrated normalized fluorescence
intensity (Io/I) for 1-3 vs amine concentration (Figure 6). The
plot for 1 with PrNH2 displays strong upward curvature, whereas
the plots for 1 with Et3N and for 2 with all three amines display
downward curvature. The extent of quenching for a given amine
concentration decreases with N-alkylation of both the amino-
stilbene (1 > 2 > 3) and the alkylamine (PrNH2 > Pr2NH >
Et3N). Only slight quenching is observed for 3 with Et3N (Figure
6). Further evidence for non-Stern-Volmer behavior is provided
by the amine concentration dependence of the fluorescence
intensities at single wavelengths. As shown in Figure 7 for 1
with PrNH2, the intensity at 351 nm decreases continuously with
increasing PrNH2 concentration, whereas the intensity at 513
nm increases to a maximum value near 75 mM PrNH2 and then
decreases slowly with PrNH2 concentration.
Figure 8. PrNH2 concentration dependence of the fluorescence lifetime
of 1 in cyclohexane solutions, 298 K: 351 nm (O) 513 nm (0).
The fluorescence decay times for 1 with PrNH2 also display
non-Stern-Volmer behavior. As may be observed in Figure 8,
the value of τ determined at 351 nm decreases gradually at
amine concentrations below 150 mM, but more precipitously
at higher concentrations. The value of τ determined at 513 nm
initially increases and then decreases, to a lesser degree than 1,
with added amine. At these wavelengths single exponential fits
are satisfactory. At intermediate wavelengths, for each concen-
tration of amine, neither single nor dual exponential fits were
satisfactory. Attempts to apply global analysis to resolve
multiple lifetimes were unsuccessful. Time-resolved fluores-
cence spectra for 1 with 150 mM PrNH2 are shown in Figure
9 along with the steady state spectra of 1 in the absence of amine
and with 150 mM PrNH2. A time-dependent red shift in the
Figure 9. Time-resolved and steady state fluorescence of 1 with PrNH2
in cyclohexane solutions, 298 K. Steady state spectra, 0 and 150 mM
PrNH2 (black) and time-resolved spectra, 150 mM PrNH2 (red). Time-
resolved delay times are 5, 10, 20, 30, and 40 ns after decay maximum.
entire fluorescence band is observed from a value of 402 nm at
short delay times to 412 nm at long delay times. The latter value
is similar to the maximum in the steady state fluorescence of 1
with 150 mM PrNH2.

1430 J. Phys. Chem. A, Vol. 108, No. 8, 2004
Lewis et al.
Figure 11. Relative concentrations of 2 (0), exciplex (O), and triplex
(4), obtained from SVD-SM of the data in Figure 10. The lines are
fits of the data to the model in Scheme 1 corresponding to the
parameters in Table 2.
Figure 10. (Top) fluorescence spectra of 2 in cyclohexane solution in
the presence of 0-310 mM PrNH2. Added amine results in red-shifted
fluorescence. (Bottom) deconvoluted fluorescence spectra of the
monomer (s), exciplex (- - -), and triplex (‚‚‚).
SCHEME 1
Singular Value Decomposition Analysis. A plausible ex-
planation for the effects of added amines on the fluorescence
spectra of the aminostilbenes is the sequential formation of one
or more excited state complexes. Analysis of the spectral
matrixes for the fluorescence with increasing amine concentra-
tion by means of SVD-SM indicates the presence of three
components. The SVD technique treats the fluorescence spectra
as vectors. The spectral vectors form a data matrix (D) that may
be described by a matrix (U) of basis vectors, a diagonal matrix
(S) of singular values, and a matrix (V) consisting of spectral
component evolution vectors (eq 7). The relative magnitude of
D ) USVT
(7)
the singular values determines the number of basis vectors. The
orthogonal basis set of vectors, Un, may be linearly combined
to form either the original spectral vectors or the unresolved
spectral component vectors, Cj (eq 8). These spectral compo-
that the second and third components are formed sequentially.
The growth and decay of the exciplex component is similar to
that of the 513 nm fluorescence intensity shown in Figure 7.
Qualitatively similar results are obtained for 1 and 2 with
each of the amines studied. The growth of the exciplex
component shows similar concentration dependence for all four
amines; however, the growth of the triplex is noticeably retarded
for Et3N when compared to the other amines. Three components
were also obtained by SVD analysis of the fluorescence spectra
of 1 with PrNH2, Pr2NH, and Et3N. The components assigned
to exciplex and triplex display somewhat larger red shifts than
those for 2 (Table 2). The growth of the triplex is noticeably
faster than is the case for 2, especially in the case of 1 with
Et3N. Addition of amines to 3 causes only small red shifts (<10
nm) in the fluorescence maximum. SVD-SM analysis of these
spectra produced only two components which are assigned to
3 and an exciplex.
A kinetic model for reversible sequential formation of the
exciplex and triplex and quenching of the triplex by amine is
shown in Scheme 1, where kd, kde, and kdt are the sum of the
radiative and nonradiative rates for the monomer, exciplex, and
triplex, respectively. Kinetic modeling of the concentration
vectors for 1 and 2 with the aliphatic amines using a multistep
numerical integration technique provides the rate constants
reported in Table 2. Rate constants were optimized using an
iterative BFGS quasi-Newton or Nedler-Mead simplex opti-
mization procedure and represent best fits (e.g., Figure 11).36
No attempt was made to model the kinetics for interaction of 3
with amines.
C ) a U
(8)
∑
j
n,j n,j
n
nents may be grouped to form the product matrix N. To
concurrently determine the set of component spectra (C) and
the set of evolution vectors (K), the coefficients (an,j) of the
linear combination (eq 9) are determined via a self-modeling
K ) N-1D
(9)
(SM) technique.17 In our application, the SM technique is simply
the adherence to spectral nonnegativity of the resulting com-
ponent and evolution vectors. Via bracketing of errant solutions,
uncertainty of the component maxima is similar to that of the
instrument, (3 nm.
The resultant pure component spectra obtained for 2 with
PrNH2 are shown in Figure 10 and are assigned to the monomer,
a 1:1 complex, and a 1:2 complex. Because these complexes
are dissociated in the ground state, we refer to them as exciplex
and triplex. Emission maxima for the exciplex and triplex (Table
2) are red shifted by ca. 20 and 30 nm, respectively, from that
of the monomer (401 nm). The concentration vectors obtained
from the SVD analysis of the PrNH2 spectra are shown in Figure
11. The concentration dependence of the vectors clearly indicates

Lewis Acid-Base Exciplexes and Triplexes
J. Phys. Chem. A, Vol. 108, No. 8, 2004 1431
Rate constants for exciplex formation (ke) for 1 or 2 with
PrNH2 and Pr2NH and for 1 with ABCO are diffusion limited.
The values for Et3N are slightly slower. Formation of the triplex
between 1 and PrNH2 (kt) is somewhat slower than exciplex
formation. Values of kt decrease with amine N-alkylation and
are significantly slower for 2 than for 1, and no triplex can be
resolved for 3. The value of kt for 1 with ABCO is much closer
to that of the secondary amine Pr2NH than that of the tertiary
amine Et3N, plausibly reflecting the decreased steric demand
for the bicyclic tertiary amine ABCO.
exciplex and triplex of 1 with PrNH2. However there is no
indication that a triplex is formed between 1 and a single
molecule of DAB. Similar results were obtained for the other
diaminoalkanes.
Kinetic modeling of the concentration vectors for 1 with DAB
provides rate constants reported in Table 2. The values for ke
and k-e are similar to those for 1 with PrNH2; however the value
of kt is smaller for DAB vs PrNH2 and the value of kqt is larger,
suggesting that the triplex formed with DAB is less stable than
that formed by PrNH2.
Rate constants for dissociation of the exciplexes and triplexes
(k-e and k-t) are much slower than for their formation; however,
assumption of irreversible exciplex formation leads to inferior
results for the kinetic modeling. In the case of 1, unimolecular
decay rate constants for the exciplexes (kde) are smaller than
those for the monomer (kd ) φf/τf ) 0.1 × 109 s-1) whereas
those for the triplexes (kdt) are larger. In the case of 2, decay of
the triplexes is also more rapid than that of the exciplexes. Thus
the rate constants for both the formation and decay of the
triplexes are sensitive to steric effects, methylation of either the
aminostilbene or the aliphatic amine resulting in decreased rate
constants for formation and increased rate constants for dis-
sociation or decay. On the basis of the amine concentration
dependence of the total fluorescence intensity, it is likely that
the exciplex, like the monomer, decays predominantly via
fluorescence, whereas nonradiative decay pathways are more
important for the triplexes.
Model for Stoichiometric Complex Formation. The be-
havior of the excited state complexes formed between the
aminostilbenes and ground state amines differs in several
important respects from that of the CT-stabilized exciplexes
formed between singlet arenes and trialkylamines.2,4-6 First, in
the case of 1 and 2, exciplex formation occurs with diffusion
controlled rates for primary and secondary amines and slightly
slower rates for tertiary amines. Singlet arenes form fluorescent
exciplexes with tertiary amines and nonfluorescent exciplexes
with secondary or primary amines.8 Second, exciplex fluores-
cence from 1 and 2 is observed only in nonpolar solvents,
whereas the stability of arene-amine exciplexes increases with
increasing solvent polarity.1 Third, reaction of the exciplexes
of 1 and 2 with a second equivalent of amine results in the
formation of fluorescent triplexes, whereas reaction of both inter-
and intramolecular arene-amine exciplexes with ground state
amines results exclusively in quenching of the exciplex fluo-
rescence.12,13,15
The decrease in intensity and decay time for the long-
wavelength (513 nm) fluorescence of 1 with Pr2NH with
increasing amine concentration (Figures 7 and 8) indicates that
the triplex is quenched by ground state amine. Weak fluores-
cence is observed for 1-3 in pure alkylamine solvents (Table
2) and is red-shifted with respect to the fluorescence of the
triplexes in cyclohexane solution. This might result from either
the formation of higher order 1:n complexes (n > 2) or from a
bulk solvent polarity effect on the fluorescence of the monomer
or the 1:1 or 1:2 complexes. Because SVD analysis does not
indicate the presence of a fourth component, 1:3 and higher
order complexes are either nonfluorescent or very weakly
fluorescent. Rate constants for quenching of the triplex (kqt) are
generally quite small and cannot be resolved for the triplexes
of 1 or 2 with Et3N.
There are also differences in the apparent steric requirements
for the formation of the excited state complex of the amino-
stilbenes and those of CT-stabilized arene-amine exciplexes.
The intramolecular exciplexes formed by 4 and 5 have
fluorescence maxima similar to that of the intermolecular
exciplex formed by 2 with PrNH2, suggesting that the intra-
and intermolecular exciplexes have similar geometries. In
contrast, the fluorescence maxima for intramolecular arene-
amine exciplexes with tri- or tetramethylene tethers are at higher
energy than their intermolecular analogues, presumably due to
restrictions on the exciplex geometry by their tethers, which
are too short to permit overlap of the nitrogen lone pair orbital
with the arene π* orbital.6
These differences suggest that the excited complexes formed
by the aminostilbenes differ in structure and electronic character
from normal CT-stabilized arene-amine exciplexes. A structural
model for the 1:1 and 1:2 complexes formed from the amino-
stilbenes and amines is shown in Scheme 2. The singlet state
of the aminostilbenes (Scheme 2a) is shown as having aniline-
styrene CT character, in accord with the moderately large dipole
moments of aminostilbenes and other aminoarenes (Table 1).30,31
Exciplex formation is shown to require the specific overlap of
the amine nonbonded orbital with the electron-deficient π-orbital
of the aminostilbene nitrogen (Scheme 2b), and triplex formation
requires a specific interaction with the other lobe of the π-orbital
(Scheme 2c). Thus we refer to the excited complexes as Lewis
acid-base or LAB exciplexes and triplexes, to distinguish them
from the better known CT exciplexes.
Interaction of Diaminostilbenes with Amines and Ami-
nostilbenes with Diaminoalkanes. Reaction of ground state
amines with the excited states of 4 or 5 might be expected to
result in the formation of a 1:1 complex between the intramo-
lecular exciplex and the amine. Addition of PrNH2 to cyclo-
hexane solutions of 5 does in fact result in a decrease in
fluorescence intensity and a red shift in the fluorescence
maximum. SVD-SM analysis of the fluorescence spectra yields
two components with maxima at 423 and 428 nm, which are
assigned to the intramolecular exciplex and bimolecular triplex.
These values are similar to those for the exciplex and triplex of
2 with PrNH2 (421 and 430 nm). The evolution vectors were
not modeled in this case.
The interaction of singlet 1 with several R,ω-diaminoalkanes
results in more efficient quenching of its fluorescence intensity
than is the case for interaction with monoamines (Figure 6).
The efficiency of quenching is dependent upon the alkane chain
length (butyl > propyl > heptyl > pentyl). SVD-SM analysis
of the fluorescence spectra obtained for 1 with 1,4-diaminobu-
tane (DAB, Figure 5) indicates the sequential formation of an
exciplex and triplex with emission maxima at 413 and 439 nm,
respectively. These values are red-shifted in comparison to the
The proposed model for LAB exciplexes and triplexes
(Scheme 2b,c) accounts for both the reactant structure and
solvent dependence for these excited state complexes. Increasing
alkylation of either the aminostilbene or aliphatic amine nitrogen
results in smaller equilibrium constants for LAB complex
formation. When both amines are tertiary, no complex formation
is detected. In this respect, these excited complexes are similar
to the Lewis acid-base complexes formed between ground state

1432 J. Phys. Chem. A, Vol. 108, No. 8, 2004
Lewis et al.
SCHEME 2
Figure 12. Emission energy for 1-3 in alkylamine-cyclohexane
solutions plotted against linear Lippert-Mataga plots for pure solvents
of identical bulk polarity. Without solvent-solute specific interaction
or preferential solvation, the data would be expected to lie on linear
Lippert-Mataga plots.
fluorescence maximum. This shift might result from preferential
solvation rather than exciplex/triplex formation.
In binary solvents, an equilibrium is reached in which the
entropy of solvent demixing (∆S) is balanced with the energy
of stabilization of the solute dipole (∆E, eq 10). For ideal
solutions, i.e., those that obey eq 11, eq 12
amines and Lewis acids such as BF3, the stability constants for
which decrease in the order RNH2 > R2NH . R3N.37 Delo-
calization of the positive charge over two or three nitrogens in
the excited state is not expected to result in an increase in the
CT character of the polar aminostilbene singlet states. In fact,
the polarity of the intramolecular LAB exciplex formed by 5 is
similar to or smaller than that of the aminostilbene 2 (Table 1).
Thus formation of LAB complexes cannot compete with
solvation of the aminostilbene singlet state in polar solvents.
Alternative triplex structural models could have both amines
interacting with the same face of the aminostilbene (monofacial
model, Scheme 2d) or the second amine interacting with the
first amine (“exterplex” model, Scheme 2e). The monofacial
model is analogous to that proposed for the intramolecular triple
complex formed between singlet anthracene and the two
equivalent nitrogens of an attached cryptand.11 The exterplex
model has been invoked to explain the quenching of arene-
amine CT exciplexes by ground state amines and diaminoal-
kanes.15,38 Both of these models would be expected to be favored
for diaminoalkane quenchers and thus are inconsistent with the
observed sequential formation of 1:1 and 1:2 complexes between
singlet 1 and DAB. In addition, formation of an exterplex would
be impossible for ABCO, which is observed to form a 1:2
complex more readily than Et3N (Table 2). The observation of
a large value of kqt for the triplex of 1 with DAB suggests that
an exterplex-type interaction may be responsible for quenching
of the fluorescent triplex. The increased polarity of the local
solvent shell of the 1:1 and 1:2 complex, due to the tethered
primary amine of the diaminoalkanes, can account for the larger
red shift for these complexes vs the complexes formed by
monoamines (Table 2).
d(∆E)
dF
d(∆S)
dF
- T
) 0
(10)
(11)
∆Fbulk ) a∆Fp + b∆Fn
where
ꢀ - 1 n2 - 1
∆Fx )
(
)(
)
n2 + 2
ꢀ + 2
1
2a3
xn
ps
)
1 + e-Z
(12)
µ2∆Fp-n
[
]
xp
∆(∆U)
describes the effect of dielectric enrichment on a polar fluoro-
phores,42 where ∆(∆U) is the emission energy difference
between cyclohexane and the cyclohexane-amine mixture, a
is the aminostilbene solute spherical cavity radius, µ is the solute
dipole moment, ∆Fp-n is the difference between the cyclohexane
and amine polarities, xn/xp is the bulk mole ratio of cyclohexane
to amine, and Zps is the empirical index of preferential solvation.
A plot of 1/∆(∆U) versus xn/xp is predicted to be linear.
However, deviation from linearity indicates a solute-solvent
specific interaction.42 Thus, upon verification of the solvent
mixture ideality, via dielectric constant and refractive index
measurements, eq 12 may be utilized to ascertain the nature of
the solute-solvent interaction.
Alternative Model: Dielectric Enrichment. An alternative
explanation for the observation of continuously red-shifted
emission upon the addition of amines to solutions of the
aminostilbenes is dielectric enrichment of the local solvent
shell.39,40 Suppan and others have shown that numerous cases
of solvent induced emission shifts in binary solvents may be
described by a dynamic increase in the local polarity around a
solute dipole.39,41 In such systems, fluorophores emit at energies
much lower than expected in pure solvents of similar polarity.
This is also the case for aminostilbene-alkylamine-cyclohex-
ane solutions. As seen in Figure 12, addition of small amounts
of amine (<1 mol %) induces a large shift in the aminostilbene
It was found that the measured bulk dielectric constants of
the mixtures varied linearly versus mole fraction of amine and
thus the bulk solvents obey eq 11 (Figure 13). Using the method
of Katrib and Janini,43 the solvent dielectric constant and
refractive index measurements yielded dipole moments of the
amines that agree well with published values, 1.24, 0.98, and
0.77 for PrNH2, Pr2NH, and Et3N, respectively.44
Assuming a dielectric enrichment model for the aminostil-
bene-alkylamine-cyclohexane system, plots of 1/∆(∆U) versus
xn/xp should be linear. Plots for 1-3 with PrNH2, Pr2NH, and
Et3N, are shown in Figure 14. No linearity is observed, from

Lewis Acid-Base Exciplexes and Triplexes
J. Phys. Chem. A, Vol. 108, No. 8, 2004 1433
1:2 complexes with lone pair donor-acceptor character; how-
ever, methods were not readily available at that time for the
deconvolution of the spectra and kinetic analysis.
The quenching of fluorescent arene-trialkylamine exci-
plexes12,13,15 and the ICT states of molecules such as (dimethyl-
amino)benzonitrile7,13 by ground state amines has also previously
been attributed to lone pair donor-acceptor interactions;
however, no shifted fluorescence was reported in theses cases.
In view of the small red shifts that we observe upon addition
of amines to 3, the stabilization of these exciplexes and ICT
states upon LAB triplex formation may be too small to result
in an observable red shift in their fluorescence maxima. It
should, however, be possible to observe the formation of LAB
exciplexes and triplexes upon interaction of the fluorescent
singlet states of other primary and secondary aminoarenes with
ground state amines. We report elsewhere an investigation of
9-aminophenanthrene and several of its derivatives, which
suggests that this is indeed the case.45
Figure 13. Plot of measured solvent polarities ∆F versus alkylamine
concentration.
Acknowledgment. Funding for this project was provided
by NSF grant CHE-0100596.
Supporting Information Available: 500 MHz 1H NMR of
1-5 in CDCl3 a 298 K. This material is available free of charge
via the Internet at http://pubs.acs.org.
References and Notes
(1) Lewis, F. D. Photoaddition Reactions of Amines with Aryl Olefins
and Arenes. In AdVances in Electron-Transfer Chemistry; Mariano, P. S.,
Ed.; JAI Press Inc.: London, 1996; Vol. 5, pp 1; Lewis, F. D. Acc. Chem.
Res. 1979, 12, 152.
(2) Chandross, E. A.; Thomas, H. T. Chem. Phys. Lett. 1971, 9, 393.
Lewis, F. D.; Ho, T.-I. J. Am. Chem. Soc. 1977, 99, 7991. Lewis, F. D.;
Bassani, D. M.; Burch, E. L.; Cohen, B. E.; Engleman, J. A.; Reddy, G.
D.; Schneider, S.; Jaeger, W.; Gedeck, P.; Gahr, M. J. Am. Chem. Soc.
1995, 117, 660.
(3) Kuzmin, M. G.; Guseva, L. N. Chem. Phys. Lett. 1969, 3, 71.
Davidson, R. S.; Trethewey, K. R. J. Chem. Soc. Chem. Commun. 1976,
827. Yang, N. C.; Libman, J. J. Am. Chem. Soc. 1973, 95, 5783. Okada,
T.; Mori, T.; Mataga, N. Bull. Chem. Soc. Jpn. 1976, 49, 3398. Van, S.-P.;
Hammond, G. S. J. Am. Chem. Soc. 1978, 100, 3895. Jacques, P.; Haselbach,
E.; Henseler, A.; Pilloud, D.; Suppan, P. J. Chem. Soc., Faraday Trans.
1991, 87, 3811. Jacques, P.; Burget, D.; Allonas, X. New J. Chem. 1996,
20, 933.
(4) Brimage, D. R. G.; Davidson, R. S. J. Chem. Soc. D, Chem.
Commun. 1971, 1385. Van der Auweraer, M.; Gilbert, A.; De Schryver, F.
C. J. Phys. Chem. 1981, 85, 3198. Van der Auweraer, M.; Grabowski, Z.
R.; Rettig, W. J. Phys. Chem. 1991, 95, 2083.
Figure 14. Inverse energy change verses solvent mole fraction for
1-3 in PrNH2, Pr2NH, and Et3N.
which it is clear that the traditional general solvent models do
not adequately account for the observed interaction of the excited
state aminostilbenes and the ground state amines in nonpolar
solvents.
(5) Van der Auweraer, M.; Swinnen, A. M.; De Schryver, F. C. J.
Chem. Phys. 1982, 77, 4110.
(6) Lewis, F. D.; Reddy, G. D.; Schneider, S.; Gahr, M. J. Am. Chem.
Soc. 1991, 113, 3498. Lewis, F. D.; Cohen, B. E. J. Phys. Chem. 1994, 98,
10591.
(7) Wang, Y. Chem. Phys. Lett. 1985, 116, 286. Wang, Y. J. Chem.
Soc., Faraday Trans. 2 1988, 84, 1809.
Concluding Remarks
(8) Lewis, F. D.; Crompton, E. M. SET Addition of Amines to Alkenes.
In CRC Handbook of Organic Photochemistry and Photobiology, 2nd ed.;
Horspool, W., Lenci, F., Eds.; CRC Press: Boca Raton, 2003.
(9) Chandross, E. A.; Thomas, H. T. Chem. Phys. Lett. 1971, 9, 397.
(10) Beecroft, R. A.; Davidson, R. S.; Goodwin, D. Tetrahedron 1985,
41, 3853. Yang, N. C. C.; Minsek, D. W.; Johnson, D. G.; Larson, J. R.;
Petrich, J. W.; Gerald, R.; Wasielewski, M. R. Tetrahedron 1989, 45, 4669.
Saltiel, J.; Townsend, D. E.; Watson, B. D.; Shannon, P. J. Am. Chem.
Soc. 1975, 97, 5688. Itoh, M.; Takita, N.; Matsumoto, M. J. Am. Chem.
Soc. 1979, 101, 7363. Masuhara, H.; Mataga, N. Chem. Phys. Lett. 1973,
22, 305. Hirata, Y.; Takimoto, M.; Mataga, N.; Sakata, Y.; Misumi, S. Chem.
Phys. Lett. 1982, 92, 76. Beens, H.; Weller, A. Chem. Phys. Lett. 1968, 2,
140. Grellmann, K. H.; Suckow, U. Chem. Phys. Lett. 1975, 32, 250.
Caldwell, R. A.; Creed, D.; DeMarco, D. C.; Melton, L. A.; Ohta, H.; Wine,
P. H. J. Am. Chem. Soc. 1980, 102, 2369.
Interaction of the singlet state of the 3-aminostilbenes with
aliphatic amines results in the sequential formation of fluorescent
stoichiometric 1:1 and 1:2 complexes. Complex formation is
attributed to the interaction of the polar singlet state in which
the electron deficient amine serves as a lone pair acceptor and
the ground state amines serve as lone pair donors, resulting in
the formation of a two-center three-electron bond. These
observations are analogous to those reported by Chandross9,13
for quenching of the intramolecular naphthalene-tertiary amine
exciplex and CT singlet state of 9-(4-(dimethylamino)phenyl)-
anthracene by low concentrations of small polar molecules such
as dimethylformamide in methylcyclohexane solution. The
observation of continuously shifting emission with increasing
DMF concentration was attributed to the formation of 1:1 and
(11) Fages, F.; Desvergne, J. P.; Bouas-Laurent, H. J. Am. Chem. Soc.
1989, 111, 96.
(12) Lewis, F. D.; Bassani, D. M. J. Photochem. Photobiol., A 1992,
66, 43.

1434 J. Phys. Chem. A, Vol. 108, No. 8, 2004
Lewis et al.
(13) Chandross, E. A. Complexes of Dipolar Excited States and Small
Polar Molecules. In The Exciplex; Gordon, M., Ware, W. R., Eds.; Academic
Press: New York, 1975; p 372.
(31) Rettig, W.; Bliss, B.; Dirnberger, K. Chem. Phys. Lett. 1999, 305,
8. Zaitsev, N. K.; Demyashkevich, A. B.; Kuzmin, M. G. Russian J. Appl.
Spectrosc. 1978, 29, 496.
(14) Burget, D.; Jacques, P.; Vauthey, E.; Suppan, P.; Haselbach, E. J.
Chem. Soc., Faraday Trans. 1994, 90, 2481.
(15) Schneider, S.; Geiselhart, P.; Seel, G.; Lewis, F. D.; Dykstra, R.
E.; Nepras, M. J. J. Phys. Chem. 1989, 93, 3112. Hub, W.; Schneider, S.;
Doerr, F.; Oxman, J. D.; Lewis, F. D. J. Phys. Chem. 1983, 87, 4351.
(16) Lewis, F. D.; Li, L.-S.; Kurth, T. L.; Kalgutkar, R. S. J. Am. Chem.
Soc. 2000, 122, 8573.
(17) Volkov, V. V. Appl. Spectrosc. 1996, 50, 320.
(18) Zimanyi, L.; Kulcsar, A.; Lanyi, J. K.; Sears, D. F., Jr.; Saltiel, J.
Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 4408. Zimanyi, L.; Kulcsar, A.;
Lanyi, J. K.; Sears, D. F., Jr.; Saltiel, J. Proc. Natl. Acad. Sci. U.S.A. 1999,
96, 4414. Henry, E. R.; Hofrichter, J. Methods Enzymol. 1992, 210, 129.
(19) Lee, B. H.; Marvel, C. S. J. Polym. Sci., Part A: Polym. Chem.
1982, 20, 393.
(20) Taylor, T. W. J.; Hobson, P. M. J. Chem. Soc. Abstr. 1936, 181.
(21) Boyer, J. H.; Alul, H. J. Am. Chem. Soc. 1959, 81, 2136.
(22) Haddow, A.; Harris, R. J. C.; Kon, G. A. R.; Roe, E. M. F. Trans.
R. Soc. (London) 1948, A 241, 147.
(32) Ru¨ckert, I.; Demeter, A.; Morawski, O.; Kuehnle, W.; Tauer, E.;
Zachariasse, K. A. J. Phys. Chem. A 1999, 103, 1958.
(33) 1H NMR also clearly indicates the small effect of methylation on
geometry; see Supporting Information.
(34) CAChe, v. 5.0; Fujistu Limited: Mihama-Ku, Chiba City, Chiba,
Japan, 2000-2001.
(35) Lippert, E. Z. Naturforsch. 1955, 10a, 541. Liptay, W. Z. Natur-
forsch. 1965, 20a, 1441. Mataga, N.; Ottolenghi, M. In Molecular
Association; Foster, R., Ed.; Academic Press: New York, 1979; Vol. 2; p
1. Mataga, N. In Molecular Interactions; Ratajczak, H., Orville-Thomas,
W. J., Redshaw, M., Eds.; Wiley: New York, 1981; Vol. 2, p 509.
(36) Lagarias, J. C.; Reeds, J. A.; Wright, M. H.; Wright, P. E. SIAM J.
Optim. 1998, 9, 112. Shampine, L. F.; Reichelt, M. W. SIAM J. Sci. Comput.
1997, 18, 1. Gill, P. E.; Murray, W.; Wright, M. H. Practical Optimization;
Academic Press: London; New York, 1981. Rate constants represent best-
fits and errors are estimated to be (10% via bracketing of solutions.
(37) Brown, H. C.; Taylor, M. D. J. Am. Chem. Soc. 1947, 69, 1332.
(23) Beletskaya, I. P.; Bessmertnykh, A. G.; Guilard, R. Tetrahedron
Lett. 1997, 38, 2287. Driver, M. S.; Hartwig, J. F. J. Am. Chem. Soc. 1996,
118, 7217.
(24) Model 350G. Dielectric Products Company, 178 Orchard St.
Watertown, MA 02172; phone 617-924-5688.
(25) Breitung, E. M.; Vaughan, W. E.; McMahon, R. J. ReV. Sci. Instrum.
2000, 71, 224.
(26) Berlman, I. B. Handbook of Fluorescence Spectra of Aromatic
Molecules, 2nd ed.; Academic Press: New York, 1971.
(27) James, D. R.; Siemiarczuk, A.; Ware, W. R. ReV. Sci. Instrum.
1992, 63, 1710.
(28) Photon Technologies Incorporated, 347 Consortium Court, London,
Ontario N6E 2S8, Canada.
(29) Lewis, F. D.; Weigel, W. J. Phys. Chem. A 2000, 104, 8146.
(30) Lewis, F. D.; Kalgutkar, R. S.; Yang, J.-S. J. Am. Chem. Soc. 1999,
121, 12045.
(38) Lewis, F. D.; Reddy, G. D.; Bassani, D.; Schneider, S.; Gahr, M.
J. Photochem. Photobiol., A 1992, 65, 205.
(39) Suppan, P. J. Chem. Soc., Faraday Trans. 1 1987, 83, 495.
(40) Suppan, P.; Ghoneim, N. SolVatochromism; The Royal Society of
Chemistry: Cambridge, U.K., 1997.
(41) Suppan, P. Chem. Phys. Lett. 1983, 94, 272. Petrov, N. K.;
Wiessner, A.; Staerk, H. J. Chem. Phys. 1998, 108, 2326.
(42) Khajehpour, M.; Welch, C. M.; Kleiner, K. A.; Kauffman, J. F. J.
Phys. Chem. A 2001, 105, 5372. Khajehpour, M.; Kauffman, J. F. J. Phys.
Chem. A 2000, 104, 7151.
(43) Janini, G. M.; Katrib, A. H. J. Chem. Educ. 1983, 60, 1087.
(44) Lide, D. R. CRC Handbook of Chemistry and Physics, 82nd ed.;
CRC Press: Boca Raton, FL, 2001.
(45) Lewis, F. D.; Ahrens, A.; Kurth, T. L. Photchem. Photobiol. Sci.,
in press.