Excited-state amine-imine double proton transfer in 7-azaindoline

Article abstract of DOI:10.1021/jp001001g

Ground-state thermodynamics and excited-state amine/imine tautomerism in 7-azaindoline (7AZD) mediated by hydrogen bond formation have been studied by means of absorption and emission spectroscopies. The association constants in cyclohexane (298 K) were determined to be 80, 2.5 × 10², and 7.8 × 10² M^-1, for the formation of 7AZD dimer, 7AZD/azacyclohexanone, and 7AZD/acetic acid dual hydrogen-bonded complexes, respectively. The 7AZD/acetic acid complex undergoes a fast (? 3 × 10⁹ s^-1) excited-state double proton transfer (ESDPT) reaction, resulting in a prominent imine-like tautomer emission. Proton-transfer isomers of 7AZD have been identified through syntheses and spectral characterization of various 7AZD methyl derivatives. In contrast, ESDPT is prohibited in cases of 7AZD dimer and 7AZD/azacyclohexanone hydrogen-bonded complex. The results, in combination with a comparative study on 7-azaindole, generalize the amine/imine tautomerism, which can be fine-tuned by the length of π electron conjugation coupled with types of associated guest molecules, further supporting the proposed catalytic-versus-noncatalytic model for the ESDPT reaction.

Full text of DOI:10.1021/jp001001g

7818
J. Phys. Chem. B 2000, 104, 7818-7829
Excited-State Amine-Imine Double Proton Transfer in 7-Azaindoline
Pi-Tai Chou,*,† Guo-Ray Wu,† Ching-Yen Wei,† Chung-Chih Cheng,‡ Chen-Pin Chang,‡ and
Fa-Tsai Hung§
Department of Chemistry, The National Chung-Cheng UniVersity, Chia Yi, Taiwan R.O.C.,
Department of Chemistry, Fu Jen Catholic UniVersity, Shin Chuang, Taiwan R.O.C., and
The National Hu-Wei Institute of Technology, Yunlin, Taiwan R.O.C.
ReceiVed: March 15, 2000; In Final Form: June 5, 2000
Ground-state thermodynamics and excited-state amine/imine tautomerism in 7-azaindoline (7AZD) mediated
by hydrogen bond formation have been studied by means of absorption and emission spectroscopies. The
association constants in cyclohexane (298 K) were determined to be 80, 2.5 × 102, and 7.8 × 102 M-1, for
the formation of 7AZD dimer, 7AZD/azacyclohexanone, and 7AZD/acetic acid dual hydrogen-bonded
complexes, respectively. The 7AZD/acetic acid complex undergoes a fast (. 3 × 109 s-1) excited-state double
proton transfer (ESDPT) reaction, resulting in a prominent imine-like tautomer emission. Proton-transfer isomers
of 7AZD have been identified through syntheses and spectral characterization of various 7AZD methyl
derivatives. In contrast, ESDPT is prohibited in cases of 7AZD dimer and 7AZD/azacyclohexanone hydrogen-
bonded complex. The results, in combination with a comparative study on 7-azaindole, generalize the amine/
imine tautomerism, which can be fine-tuned by the length of π electron conjugation coupled with types of
associated guest molecules, further supporting the proposed catalytic-versus-noncatalytic model for the ESDPT
reaction.
1. Introduction
noncatalytic process is of importance from a chemistry perspec-
tive since it is plausible that a chemically important isomer of
the guest molecule can be produced via the ESDPT process,
which cannot be otherwise accessed.16c,d
The 7-azaindole (7AI) dimer has long been recognized as a
simplified model for the hydrogen-bonded base pair of DNA,1-3
which upon electronic excitation undergoes the double proton
transfer reaction, resulting in a large Stokes shifted emission
(e.g., λmax ∼ 480 nm in nonpolar solvents). At the molecular
level, such an excited-state double proton transfer (ESDPT)
process provides one possible mechanism for the mutation which
has been proposed to be, in part, due to a “misprint” induced
by the proton-transfer tautomerism of a specific DNA base pair
during replication, recording an error message.3-7 More recently,
studies have been extended to 7AI analogues of biological
importance. Among which, purines possessing a similar elec-
tronic moiety with respect to 7AI have received particular
interest.8,9 Conversely, via the formation of a 7AI(host)/guest
hydrogen-bonded complex, the dynamics of ESDPT incorporat-
ing guest molecules have also received considerable attention.10-17
In alcohols and water, the dynamics of ESDPT in 7AI and its
analogue 7-azatryptophan have been successfully applied to
probe the solvation and/or protein dynamics.12-15 The guest
molecule assisted ESDPT in 7AI also renders an interesting
perspective from the thermodynamic standpoint. The acid-,
alcohol-, and water-assisted ESDPT in 7AI can be specified as
a catalytic processes since the molecular structure of the guest
species (e.g., the acetic acid) remains unchanged. On the other
hand, the double proton transfer in the single-photon excited
7AI dimer results in 7AI(T)*/7AI(T) forms ((T)* denotes the
electronically excited tautomer) in which both 7AI molecules
tautomerize simultaneously. Thus the guest 7AI in the dimeric
form does not act as a catalyst but rather as a reactant. Such a
From a molecular viewpoint, the reaction center of the ESDPT
process in 7AI and its analogues is essentially the amine/imine
proton-transfer tautomerism in which a proton (or hydrogen
atom) in the pyrrolic nitrogen transfers to the pyridinic nitrogen,
forming an imine-like isomer. Such a process is generally highly
endergonic in the ground state. For example, a theoretical
calculation estimates the 7AI dimer f 7AI(T) dimer tautom-
erism to be ∼23.5 kcal/mol.16e It thus becomes crucial that the
S0′ f S1′ transition gap of the tautomer species (hereafter, prime
denotes the imine tautomer state) has to be significantly smaller
than that of the normal species (i.e., the S0-S1 gap) in order to
proceed as a thermally favorable ESDPT reaction. A simple
empirical approach based on the π electron conjugation predicts
that the C2-C3 double bond (see Figure 1) in the imine-like
tautomer of 7AI should be actively involved in the delocalization
of π electrons and hence play a key role to determine the S′0-
S′1 energy gap (i.e., the energy level of the S′1 state). Conversely,
theoretical approaches also indicate that the relative thermody-
namics between amino and imino forms in the ground state
mainly depend on destabilization of the aromaticity of the
pyridine ring (vide infra), and hence are less affected by the
C2-C3 double bond in the pyrrole ring. On this basis, 7-aza-
indoline (7AZD, see Figure 1)18 was synthesized via hydrogena-
tion of the C2-C3 double bond in 7AI. On one hand, 7AZD is
potentially capable of undergoing an amine-imine type of
proton-transfer tautomerism. Thus, studies aiming at spectros-
copy and dynamics of ESDPT in 7AZD are of great importance
to generalize the amine-imine tautomerism being the central
reaction in 7AI analogues. On the other hand, due to the lack
of a C2-C3 double bond, the imine-like tautomer of 7AZD
* To whom correspondence should be addressed.
† The National Chung-Cheng University.
‡ Fu Jen Catholic University.
§ The National Hu-Wei Institute of Technology.
10.1021/jp001001g CCC: $19.00 © 2000 American Chemical Society
Published on Web 07/20/2000

Amine-Imine Double Proton Transfer
J. Phys. Chem. B, Vol. 104, No. 32, 2000 7819
SCHEME 1. Proton transfer tautomerism tuned by
relative thermodynamics. Note that the S0 state has been
normalized to an equivalent energy level. The energy
level is in kcal/mol with its corresponding wavelength (in
parentheses) in nm.
Figure 1. Structures of 7AI, 7AZD and its corresponding proton-
transfer isomers as well as methylated derivatives.
apparently possesses a shorter length of the π electron conjuga-
tion than that of 7AI analogues, and hence a larger S0′-S1′
energy gap is expected. Therefore, 7AZD may serve as an ideal
model to test the catalytic-versus-noncatalytic ESDPT reaction16c,d
fine-tuned by the length of the π electron conjugation.
2. Experimental Section
2.1. Materials Synthesis of 7AZD. 7AI (3 g) in dry toluene
(150 mL) was gradually (2 mL/min) passed through a reaction
column packed with the Pd/δ-Al2O3 catalyst with the reaction
temperature and pressure maintained at 150 °C and 450 psi,
respectively. The product 7AZD mixed with unreacted 7AI was
further separated by column chromatography where ethyl
acetate/hexane (1:3 v/v) was used as an eluent to yield 1.8 g
2.2. Measurements. Steady-state absorption and emission
spectra were recorded by a Cary 3E (Varian) spectrophotometer
and an F4500 (Hitachi) fluorimeter, respectively. Both wave-
length-dependent excitation and emission response of the
fluorimeter have been calibrated according to a previously
reported method.16 Nanosecond lifetime measurements were
performed by an Edinburgh FL 900 photon counting system
with a hydrogen filled flash lamp/or a nitrogen lamp as the
excitation source. The temporal resolution after deconvolution
of the excitation pulse was ∼300 ps. The data were analyzed
using a nonlinear least-squares fitting program with a decon-
volution method reported previously.23
1
(∼60%) of 7AZD. NMR analyses: H NMR(CDCl3, 200 MHz)
7AZD δ3.03(t, J ) 8.42, 8.14 Hz, 2H); 3.59(t, J ) 8.42, 8.44
Hz, 2H); 6.47(t, J ) 6.22, 6.24 Hz, 1H); 7.20(d, J ) 8.06 Hz,
1H), 7.77(d, J ) 4.92 Hz, 1H).
N(1)-methyl-7-azaindoline (1MAZD, see Figure 1) was
synthesized according to a similar procedure to obtain various
methylated purine derivatives.20-22 Briefly, a suspension of
7AZD (0.12 g, 1 mmole) and sodium hydride (57%, 0.17 g, 4
mmole) in N,N-dimethylformamide (∼ 3mL) was stirred under
argon at 0 °C for 20 min. Methyl iodide (1.17 g, 0.0083 mol)
was then added and the resulting mixture was stirred for ∼1 h.
The N,N-dimethylformamide was then removed under reduced
pressure. Column chromatography eluting with 95% ethanol
afforded 0.16 g of 1MAZD. NMR analyses: 1H NMR(CDCl3,
200 MHz) 1MAZD δ1.42(s, 3H); 2.96 (t, J ) 7.2 Hz, 2H);
3.47(t, J ) 8.0 Hz, 2H); 6.41(q, J ) 7 Hz, 1H), 7.15(d, J ) 1.6
Hz, 1H); 7.83(d, J ) 4.6 Hz, 1H).
7-Methyl-2,3-dihydro-7H-pyrrolo[2,3-b]pyridine (7MDPP,
see Figure 1) was prepared by dissolving 7AZD (50 mg) in dry
THF. To the solution was added 0.08 mL CH3I and the mixtures
were refluxed for ∼3 h under N2 atmosphere. After cooling,
the solution containing precipitates was filtered. The precipitate
was then dissolved by addition of excess of potassium carbonate
in the aqueous solution, and the resulting solution was extracted
by ethyl acetate. The ethyl acetate layer was dried, filtered, and
the product was obtained after evaporating the solvent. NMR
analyses: 1H NMR(CDCl3, 400 MHz) 7MDPP δ6.46(m, 2H);
5.50(m, 1H); 3.84(m, 2H); 3.30(s, 3H), 2.88(m, 2H).
2.3. Theoretical Calculations. A detailed account of the
theoretical approach has been previously described.24 Briefly,
the initially optimized structure used for the ab initio calculation
was obtained by the semiempirical AM1 method. Ab initio
molecular orbital calculations were performed by using Gaussian
98 rev. A.7 programs.25 Geometry optimizations for all structures
were carried out with the 6-31G(d,p) basis set at the Hartree-
Fock (HF) level. Vibrational frequencies were also performed
to ensure the global energy minimum for those dimeric and
complex forms. The directly calculated zero-point vibrational
energies (ZPE) were scaled by 0.918126 to account for the
overestimation of vibrational frequencies at the HF level. The
association energy, ∆Hac, was calculated as the change in the
total molecular enthalpy of formation for the conversion of the
optimized monomer individually into the optimized dimer,
which then incorporated with a counterpoise correction
procedure27-29 to correct certain inconsistencies due to the basis-
set superposition (BSSE). In certain cases, a semiempirical
PM3-SM4 solvation model30,31 was applied to obtain the
solvation free energy in cyclohexane, which is then added to
“gas phase” energies obtained from the ab initio method.32
Cyclohexane is of spectrograde quality (Merck Inc.) and used
right after received. Acetic acid (ACID, Merck Inc.) was purified
through a fractional distillation. 2-Azacyclohexanone (Aldrich)
was recrystallized twice from methanol. Since the stable
conformer of 2-azacyclohexanone is in the lactam form we thus
use the abbreviation LACTAM in order to distinguish it from
the LACTIM form (see Scheme 1).
The following sections are organized according to a sequence
of steps where we first performed detailed absorption and
fluorescence titration experiments to determine the thermody-
namic and/or ESDPT properties of 7AZD hydrogen-bonded
complexes.

7820 J. Phys. Chem. B, Vol. 104, No. 32, 2000
Chou et al.
TABLE 1: Thermodynamic and Photophysical Properties of 7AZD, 7AI, and Their Methylated Derivatives in Cyclohexane
absorption
emission
∆G0 (kcal/mol)
λmax (nm)
λ
max (nm)
Φf
τf (ns)
Ka (M-1
)
(298 K)
7AI
7AI dimer
7MPP
7AZD
7AZD dimer
7AZD/ACID
287
292
390
306
NA
318d
320
480
495
353
360
440
0.22a
0.016
0.0073a
0.44
0.15
0.04
1.72a
3.0b
1.67a
4.9
2.1
1.8
2.2 × 103 c
0.8 × 102
7.8 × 102
7.2 × 102 e
2.5 × 102
-2.5
-3.9
7AZD/LACTAM
1MAZD
7MDPP
NA
312
375
368
365
463
0.30
0.48
0.015
3.2
5.2
1.1
-3.2
a Reference 14(b). b Reference 11(b). c Reference 16(d). d Values were obtained from the fluorescence excitation spectrum. e Data were obtained
from the fluorescence titration experiment. NA: not available.
The results in combination with syntheses and spectral
characterization of 7AZD methyl derivatives lead us to deduce
specific hydrogen-bonding sites actively involved in the complex
formation as well as the amine-imine tautomerism. Subse-
quently, complementary supports for the structure and relative
energy levels of various 7AZD/guest hydrogen-bonded com-
plexes were provided via ab initio calculations. Finally, A
general discussion based on amine-imine tautomerism mediated
by the relative excited-state thermodynamics is proposed and
discussed.
drogen-bonding effect. Similar to that proposed in the 7AI self-
association, we first assume a dominant self-dimerization form
possibly possessing a cyclic hydrogen bonding configuration
(see Scheme 1) where the dual hydrogen-bonding sites, i.e., the
N(1)H proton and the pyridinal nitrogen in 7AZD (i.e., N(7))
act as a proton donor and acceptor, respectively. Consequently,
a competitive equilibrium between enol monomer and 7AZD
self-association incorporating two 7AZD molecules can be
depicted as 2(7AZD) h (7AZD)2, and the association constant
Ka for the formation of the 7AZD dimer can be expressed by
Cp
3. Results
Ka )
(1)
3.1. Self-Association. Spectroscopically, the change of UV-
vis absorption spectra associated with the hydrogen-bonding
formation has been used as a tool to obtain thermodynamics of
the 7AI dimer and hydrogen-bonded complexes.2,3,16 We thus
performed a similar method for 7AZD. When the concentration
was prepared to be as low as 1.2 × 10-5 M, 7AZD exhibits an
S0-S1 absorption band maximum at 306 nm (ꢀ306 ∼ 5600
M-1cm-1). The spectral features are similar to that of 1MAZD
which is generally treated as a nonproton-transfer model (i.e.,
the amino form, see Figure 1) due to the lack of an N-H proton.
We thus conclude that at sufficiently low concentrations 7AZD
exists mainly as an amino form in cyclohexane. Despite the
hydrogenation at the C2-C3 double bond, the wavelength at
the S0 f S1 (ππ*) absorption maximum in 7AZD is even red
shifted by ∼20 nm with respect to that of 7AI monomer (see
Table 1). The result can be rationalized by the fact that for 7AI
the lone pair electrons of the pyrrolic nitrogen directly involve
in the π-electron conjugation to gain the aromaticity (i.e., ten
π electrons). Whereas upon hydrogenation of the C2-C3 double
bond, the nitrogen (N(1)) lone pair electrons no longer play
such a role. Instead, upon excitation the occurrence of charge
transfer between electron donor (the alkylamino nitrogen) and
acceptor (the pyridinic nitrogen) is common, resulting in a
bathochromic shift in 7AZD relative to 7AI. More evidence can
be given by a further red shift of the absorption maximum in
1MAZD (λmax ∼ 312 nm, see Figure 1), in which the
dialkylamine possesses a lower ionization energy than that of
the alkylamine in 7AZD. Thus, a stronger excited-state charge-
transfer property is expected in 1MAZD.
(C0 - 2Cp)2
where C0 is the initial concentration of 7AZD and Cp denotes
the concentration of the 7AZD dimeric form. Consequently, the
absorbance of the mixture at a specific wavelength as a function
of initially prepared C0 can be expressed by
C0
1
1
4
)
+
(2)
A - ꢀMC0
C
ꢀD - 2ꢀM
Ka(ꢀD - 2ꢀM)
0
Detailed derivations of eq 2 have been elaborated in ref 9. In
the above equation ꢀM and ꢀD are molar extinction coefficients
of the monomer and dimer monitored at a specific wavelength,
e.g., 340 nm, where an appreciable percentage of absorbance
can be attributed to the self-associated species (see Figure 2).
Under a sufficiently low concentration where only monomer
exists, ꢀ3M40 was measured to be ca. 165 M-1 cm-1. A plot of
C0/(A - ꢀMC0) at 340 nm as a function of 1/C0 shown in the
insert of Figure 2 reveals sufficiently linear behavior, supporting
the assumption of dimeric formation in 7AZD. Accordingly,
by fixing the ꢀM value, the best linear least-squares fit to Figure
2, using eq 2, gives ꢀ3D40 and Ka values to be 1050 M-1 cm-1
and 80 M-1, respectively. In comparison with that of the 7AI
dimer (Ka ∼ 2.2 × 103 M-1 in cyclohexane16d), the relatively
smaller dimerization constant in the case of 7AZD dimer may
be rationalized by a steric effect upon dimerization. The most
stable form of 7AZD monomer is a nonplanar structure in which
the amino hydrogen was calculated to be ∼25.3° with respect
to the pyridine ring (vide infra). Thus, an endothermic hinder
rotation of the amino hydrogen toward the planarity is necessary
prior to the dimerization, which compensates the gain of the
stabilization due to the dual hydrogen-bonding formation. In
contrast, the planar structure in 7AI facilitates the dimerization
with negligible geometry adjustment, resulting in a significantly
larger association constant. Further verification of this viewpoint
will be elaborated in the discussion section.
A concentration-dependent absorption profile at the spectral
region of the S0fS1 transition was observed upon increasing
the concentration (see Figure 2b-e). Although such a change
was not as remarkable as those concentration-dependent absorp-
tion spectra observed in 7AI analogues,3,9,16b,d the results in
comparison with the concentration-independent absorption
profile for 1MAZD unambiguously conclude the formation of
7AZD dimer and/or higher-order aggregates through the hy-

Amine-Imine Double Proton Transfer
J. Phys. Chem. B, Vol. 104, No. 32, 2000 7821
Figure 2. (s) The concentration-dependent absorption spectra of 7AZD in cyclohexane, in which 7AZD (C0) was prepared at (a) 4.5 × 10-4, (b)
9.0 × 10-4, (c) 1.8 × 10-3, (d) 3.6 × 10-3, (e) 7.2 × 10-3 M. The absorption spectra are normalized at 313 nm. (- - -) The absorption spectrum
of 1MAZD (1.0 × 10-5 M) in cyclohexane. Inset: Plot of C0/(A340 - ꢀM340C0) values versus 1/C0 and a best nonlinear least-squares fitting curve
using eq 2. Note that an additional point at C0 ) 2.8 × 10-5 M is added in the plot.
Table 1 lists the steady-state spectral properties of absorption
and emission as well as relaxation dynamics for 7AZD and its
hydrogen-bonded species in cyclohexane. When the concentra-
tion of 7AZD was prepared as low as 1.2 × 10-5 M in
cyclohexane so that the monomer form exists predominantly, a
normal Stokes shifted emission (Φf ∼ 0.44) was observed with
a peak maximized at 353 nm. The emission lifetime was
measured to be 4.9 ( 0.2 ns and was independent of the
excitation as well as the monitored emission wavelength. Both
spectral features and relaxation dynamics of the 7AZD monomer
are similar to those of 1MAZD, exhibiting a normal Stokes
shifted emission maximum at ∼365 nm (τf ∼ 5.2 ns, Φf ∼ 0.48)
in cyclohexane (see Table 1). Therefore, at a sufficiently low
concentration of 7AZD in cyclohexane only one emitting species
exists, which is unambiguously assigned to the amino form.
The steady-state fluorescence spectra as a function of the 7AZD
concentration are shown in Figure 3. In contrast to the distinct
dual emission observed in the concentrated 7AI, in which the
large Stokes shifted band (λmax ∼ 480 nm) is associated with
the tautomer dimeric species resulting from ESDPT, the
fluorescence consists of a spectrally unresolved broad band of
which the fluorescence maximum is gradually red shifted upon
increasing the concentration. For example, when 7AZD was
prepared as high as, e.g., 7 × 10-3 M, where ∼20% exists in
single exponential decay components theoretically expressed as
Ft ) A1e-k t + A2e-k t, where A1 and A2 are the emission
intensity at t ∼ 0 for the decay components 1 and 2, respectively
(see inset of Figure 3). While the ratio for A1 versus A2 is
concentration as well as excitation-wavelength dependent, k1
and k2, within experimental error, were found to be constant,
with a slow and a fast component of 2.1 × 108 s-1 (τ ∼ 4.8 ns)
and 4.8 × 108 s-1 (τ ∼ 2.1 ns). The slower decay component
is essentially identical to that of the 7AZD (monomer) emission
and can thus be unambiguously assigned to be the unassociated
7AZD. Since the entire emission band exhibits a system-
response rise time (i.e., < 300 ps, see the Experimental Section),
it is quite unlikely that one species is the precursor of the other.
Consequently, the results unambiguously conclude that only two
emitting species exist in the entire spectral region (300-500
nm). The longer-lived species can be reasonably ascribed to
the emission from the non-hydrogen-bonded monomer. Con-
versely, the shorter lived, lower frequency emission part with
its excitation spectrum red shifted with respect to the monomer
species unambiguously results from the dimeric species, in
which the red shift of the excitation spectrum is due to the dual
hydrogen-bonding formation.
1
2
3.2. 7AZD/ACID Complexes. Figure 4 shows the absorption
spectra of 7AZD upon adding the acetic acid (ACID) in
cyclohexane. In this experiment, the initial concentration of
7AZD, C0, was prepared to be as low as 1.0 × 10-5 M to avoid
the self-dimerization. The formation of 7AZD/ACID hydrogen-
bonded complexes can be clearly shown by the growth of an
absorption band at ∼318 nm (taken from the excitation
maximum shown in Figure 6) throughout the titration. The
appearance of an isosbestic point located at ∼313 nm indicates
the existence of equilibrium with a common intermediate. The
7AZD hydrogen-bonded complex incorporating stoichiomet-
rically equivalent ACID molecules can be depicted as
the dimeric form deduced from eq 1 with a Ka value of 80 M-1
,
a 335 nm excitation gives rise to a fluorescence maximum of
360 nm, which is ∼7 nm shift with respect to the monomer
emission. The possibility of the spectral distortion due to the
reabsorption in the high sample concentration has been elimi-
nated by the use of square quartz cell with various path length
(0.1-1.0 cm) so that the maximum absorbance can be kept as
low as 0.5. Similarly, the excitation spectrum monitored at the
long-wavelength emission region (e.g., 420 nm) is red shifted
with respect to that monitored at the short-wavelength tail (e.g.,
340 nm), indicating that the emission originates from multiple
ground-state components. Detailed time-resolved studies clearly
indicate that the dynamics of decay were best fitted by two
Ka
7AZD + ACID y\z 7AZD/ACID

7822 J. Phys. Chem. B, Vol. 104, No. 32, 2000
Chou et al.
Figure 3. The concentration-dependent fluorescence spectra of 7AZD in cyclohexane, in which 7AZD was prepared at (a) 1.5 × 10-5, (b) 1.1 ×
10-4, (c) 2.2 × 10-4, (d) 4.4 × 10-4, (e) 1.6 × 10-3 M. The spectra are normalized at an equivalent relative emission intensity. Inset: The
time-dependent fluorescence of 7AZD (1.0 × 10-3 M) monitored at 420 nm (λex: 330 nm) fitted by two single exponential decays (τ ) 4.9 × 10-9
and 2.1 × 10-9 s, ø2 ) 1.005).
Figure 4. The concentration-dependent absorption spectra of 7AZD (1.2 × 10-5 M) in cyclohexane by adding various ACID concentrations (Cg)
of (a) 0, (b) 1.4 × 10-4, (c) 2.8 × 10-4, (d) 4.3 × 10-4, (e) 5.7 × 10-4, (f) 7.1 × 10-4, (g) 1.0 × 10-3, (h) 1.4 × 10-3, (i) 1.8 × 10-3, and (j) 2.7
× 10-3 M. Inset: The plot of A0/(A - A0) at 335 nm as a function of 1/Cg in curves b to j, and a best least-squares fitting curve using eq 3.
Under the condition that the concentration of ACID (or the guest
molecule) is much larger than that of the complex, the
relationship between the measured absorbance as a function of
the initially prepared ACID concentration, Cg, can be expressed
by9
monomer and hydrogen-bonded complex monitored at a specific
wavelength, respectively. The insert of Figure 4 shows the plot
of A0/(A - A0) as a function of 1/Cg at a selected wavelength
of 335 nm. Straight-line behavior supports the validity of the
assumption of 1:1 7AZD/ACID formation, and a best linear
least-squares fit using eq 3 deduces ꢀ3C35 and Ka to be 1400 M-1
cm-1 and 780 M-1, respectively.
A0
ꢀM
1
)
+ 1
(3)
(
)
[
]
A - A0
ꢀC - ꢀM KaCg
Figure 5 shows the fluorescence spectra of 7AZD as a
function of increasing ACID concentration in cyclohexane. Upon
excitation in the region of the isosbestic point (∼313 nm), where
where ꢀM and ꢀC are molar extinction coefficients of the 7AZD

Amine-Imine Double Proton Transfer
J. Phys. Chem. B, Vol. 104, No. 32, 2000 7823
Figure 5. The fluorescence spectra of 7AZD as a function of the acetic acid concentration (Cg) which was prepared to be (a) 0, (b) 1.4 × 10-4
,
(c) 2.8 × 10-4, (d) 4.3 × 10-4, (e) 5.7 × 10-4, (f) 7.1 × 10-4, (g) 1.0 × 10-3, (h) 1.4 × 10-3, (i) 1.8 × 10-3, and (j) 2.7 × 10-3 M. Inset: The
plot of F0/(F - F0) at 430 nm as a function of 1/Cg in curves b to j and a best least-squares fitting curve using eq 4.
Figure 6. The excitation spectra of 7AZD (1.2 × 10-5 M by adding 5.5 × 10-4 M acetic acid in cyclohexane) monitored at (a) 500 nm, (b) 375
nm. Inset: The time-dependent fluorescence monitored at (a) 500 nm (τ ) 1.8 × 10-9 s, ø2 ) 1.005), (b) 350 nm (τ ) 4.9 × 10-9 s, ø2 ) 1.095)
and fitted by one single-exponential decay.
both 7AZD and 7AZD/ACID absorb, distinct dual fluorescence
was observed, consisting of a short-wavelength band maximum
at ∼353 nm (the F1 band) and a large Stokes shifted, vibronically
progressive band maximum at ∼440 nm (the F2 band). The
resulting concentration-dependent spectra reveal an isoemissive
point at ∼403 nm. The excitation spectrum monitored within
the long-wavelength tail of the F2 band (e.g., 500 nm) is red
shifted by ∼12 nm with respect to that monitored within the
region of the F1 band (e.g., 375 nm) (see Figure 6). The decay
dynamics of the F1 (<380 nm) and F2 (> 420 nm) bands are
independent of the added acetic acid concentrations, and both
exhibit single-exponential behavior with a lifetime τf of 4.9 ns
and 1.8 ns, respectively (see inset of Figure 6). The rise time
of both components unfortunately cannot be resolved by our
current photon counting system. Thus, it is quite unlikely that
one species is the precursor of the other. Alternatively, the dual
fluorescence should originate from different ground-state pre-
cursors, most likely the uncomplexed 7AZD monomer and
7AZD/ACID complex. The spectral and dynamic resemblance
leads to the assignment of the F1 band to the monomer emission

7824 J. Phys. Chem. B, Vol. 104, No. 32, 2000
Chou et al.
Figure 7. (s) The absorption (A) and emission (E) spectra of 7AZD (1.2 × 10-5 M) in the acidic aqueous solution (pH ) 1.0). (- - -) The
absorption (A′) and emission (E′) spectra of 7MDPP (2 × 10-5 M) in cyclohexane.
unambiguously. Consequently, the precursor of the F2 band
should be associated with the 7AZD/ACID hydrogen-bonded
complex. Due to the high acidity of the carboxylic proton (pKa
) 4.75) and a drastic increase of the basicity in the excited
pyridinal nitrogen (∆pKa (NH+) ) pKa (NH+) - pKa* (NH+)
) -3.0 in 7AZD),33 the 440 nm band may result from a single
proton transfer of the carboxylic proton to the 7AZD pyridinal
nitrogen, giving rise to a 7AZDH+ (or more strictly the
7AZDH+/ACID- ion pair in cyclohexane) emission. Such a
possibility has been eliminated by the following sequence of
experiments. First, in an HCl gas saturated cyclohexane where
7AZD exists predominantly as a 7AZDH+/Cl- ion pair the
emission maximum was observed to be 392 nm which is similar
to the fluorescence (λmax ∼ 395 nm) of 7AZD in the acidic
aqueous solution (pH ) 1.0), where most of the pyridinal
nitrogen has been protonated (see Figure 7). Furthermore, the
formation of a hydrogen-bonded complex between 1MAZD and
acetic acid has been examined by the ACID concentration-
dependent titration spectra. The Ka value of ∼20 M-1 deduced
from eq 3 for the 1MAZD/ACID complex formation is smaller
than that calculated for 7AZD/ACID by a factor of 40. Such a
difference may not be surprising because theoretically only
single hydrogen-bonded 1:1 1MAZD/ACID complexs can be
formed. Upon increasing the acid concentration, the emission
intensity of 1MAZD increases, accompanied by a slight red shift
of the emission maximum from 363 to 368 nm. The lack of
observing any large Stokes shifted emission in the 1MAZD/
ACID complex discounts the assignment of the 440 nm emission
simply to the excited-state protonation of 7AZD. Thus, the F2
band with an unusually large Stokes shift of the ∼8500 cm-1
(peak-to-peak) relative to the complex S0-S1(ππ*) absorption
(318 nm estimated from the excitation spectrum shown in Figure
6) leads us to propose a possible occurrence of ESDPT, resulting
in an imine-form-like tautomer species (see Figure 1 and Scheme
1). To further verify this viewpoint, a model compound 7MDPP
(see Figure 1) was synthesized, which possesses similar
electronic properties with respect to the imine tautomer except
for the lack of proton-transfer tautomerism. As shown in Figure
7, 7MDPP exhibits a concentration-independent absorption and
fluorescence maxima at 375 and ∼463 nm (τf ∼ 1.1 ns in
cyclohexane), respectively, where the emission spectral features
and relaxation dynamics are similar to the F2 band observed in
exciting the 7AZD/acid complex. In conclusion, spectroscopic
results based on strategically designed 7AZD derivatives related
to 7AZD provide definitive evidence for the assignment of the
F2 band to the imine tautomer emission. In comparison with
the acetic acid-catalyzed tautomer emission of 7AI maximized
at ∼ 500 nm, the F2 band of 7AZD is ∼2700 cm-1 (peak-to-
peak) higher in energy. Such a difference can be rationalized
by the reduction of the length of π electrons configuration in
their corresponding tautomeric species. Details will be elaborated
in the discussion section.
The association constant can also be obtained from the
initially prepared ACID concentration, Cg, and the measured
emission intensity at a selected wavelength expressed in eq 4
F0
ΦMꢀM
1
)
+ 1
(4)
(
)
F - F0
K C
(Φpꢀp - ΦMꢀM)
a
g
where F0 and F denote the measured fluorescence intensity prior
and after adding ACID. Here, ΦM and Φp are fluorescence
quantum yields of the monomer and complex, respectively.9 The
plot of F0/(F - F0) versus 1/Cg shown in the inset of Figure 5
exhibits good linear behavior, supporting the assumption of the
1:1 complex formation. A best linear least-squares fit gives the
slope and intercept to be 9.0 × 10-5 M and 0.065, respectively.
Consequently, a Ka value of 7.2 × 102 M-1 was obtained, which
is consistent with the value of 7.8 × 102 M-1 concluded in the
absorption titration study.
3.3. 7AZD/LACTAM Complex. Appreciable hydrogen-
bonding association was also observed between 7AZD and
LACTAM in the absorption titration study (see Figure 8), and
the value of (A0)/(A0 - A) as a function of 1/Cg also exhibits
a straight-line behavior, indicating the formation of a 1:1 7AZD/
LACTAM hydrogen-bonded complex. The best linear least-
squares fit using eq 3 gives a Ka value of 2.5 × 102 M-1 in
cyclohexane (see Table 1). Since LACTAM also possesses

Amine-Imine Double Proton Transfer
J. Phys. Chem. B, Vol. 104, No. 32, 2000 7825
Figure 8. The concentration-dependent absorption and emission spectra of 7AZD in cyclohexane by adding various LACTAM concentrations (Cg)
of (a) 0, (b) 2.7 × 10-4, (c) 5.5 × 10-4, (d) 7.5 × 10-4, (e) 1.0 × 10-3, and (f) 1.5 × 10-3 M. Inset: The plot of A0/(A0 - A) at 315 nm as a function
of 1/Cg in curves b to f and a best least-squares fitting curve using eq 3. Asterisk (*) denotes the Rayleigh scattering at 335 nm.
bifunctional hydrogen-bonding groups, it is reasonable to expect
the 7AZD/LACTAM complex exhibiting the same photophysi-
cal properties, i.e., the double proton transfer in the excited state,
as those of 7AZD/ACID. In contrast, increasing the LACTAM
concentration results in an increase of the emission intensity
accompanied by a gradual red shift of the emission maximum
from 353 to 368 nm (see Figure 8). Whereas, within the
detection limit, the imine-like tautomer emission expected to
be in the region of 400-600 nm cannot be resolved (see Figure
8). Similar to that observed in the concentrated 7AZD, the
dynamics of decay can be well fitted by two single-exponential
decay components with a lifetime of 5.0 and 3.2 ns which can
be ascribed to the life span of 7AZD and 7AZD/LACTAM
normal emission, respectively. In comparison, the 7AI/LAC-
TAM complex undergoes fast ESDPT dynamics, resulting in a
prominent tautomer emission.16c,d Knowing that both 7AZD
dimer and 7AZD/LACTAM do not undergo a “noncatalytic”
type of ESDPT, the results suggest that the reduction of π
electrons conjugation in 7AZD relative to 7AI should play a
key role for the observation of drastically different relaxation
dynamics between 7AI and 7AZD. Detailed explanations will
appear in the Discussion section.
enthalpy, ∆Hac, was calculated to be approximately -7.0 kcal/
mol. On the other hand, a possible linear, single hydrogen-
bonded dimeric form is also considered, namely the N(1)-
H‚‚‚N(7) single hydrogen-bonded dimer, and ∆Hac was calculated
to be -3.6 kcal/mol, which is apparently much less exothermic
than the cyclic dual hydrogen-bond formation. The results are
in agreement with the absorption and fluorescence titration
studies, concluding that the 7AZD cyclic dual hydrogen-bonded
dimer is the predominant self-associated species.
For the 7AZD/guest hydrogen-bonded complexes, our results
show that, independent of the type of guest molecule studied,
the most stable form of a 1:1 7AZD/guest complex generally
incorporates cyclic dual hydrogen bonds (see Table 2) due to
its much larger formation enthalpy. The enthalpies of association
were estimated to be -11.8 kcal/mol and -9.0 kcal/mol for
7AZD/ACID and 7AZD/LACTAM, respectively. Since ∆Hac
generally correlates well with the hydrogen-bonding strength,
the result may simply indicate that the hydrogen-bonding
strength is on the order of 7AZD/ACID > 7AZD/LACTAM >
7AZD dimer. Evidence can be given by the calculated >N‚‚‚H-
O- hydrogen-bonding distance of 1.88 Å in the case of the
7AZD/ACID, which is shorter by 0.32 Å than the >N‚‚‚H-
N- distance in 7AZD dimer of 2.20 Å (see Figure 9). The trend
in the exothermicity of the association enthalpy also qualitatively
correlates with the experimentally measured order of the
association constants for 7AZD dimer and hydrogen-bonded
complexes (see Table 1), increasing as the association enthalpy
decreases. However, knowing that the association constant is
generally an order of magnitude smaller than that of the 7AI
hydrogen-bonding species (vide supra), other factors such as
the steric effect may play a key role to fine-tune the stability of
various 7AZD hydrogen-bonded species. For the case of 7AZD
dimer the N(1)-H bond was calculated to be ∼14° with respect
to the pyridine plane, while the N(1)-H bond was calculated
to be 25.3° in a geometry optimized 7AZD monomer form. The
result indicates that an endothermic, sterically hindered rotation
of the N(1)-H toward a favorable configuration is necessary
4. Discussion
4.1. The Hydrogen-Bonding Formation. To rationalize the
hydrogen-bonding formation we first performed an ab initio
calculation for the relative energy among various isomers of
7AZD in its monomeric form. The results shown in Table 2
conclude the amino form of 7AZD to be the most stable one in
the gas phase, while the imino tautomer is higher in energy
than the amino form by ∼15.0 kcal/mol. The inclusion of a
solvation free energy based on the PM3-SM4 model obtains
similar results in which the amino form is more stable than the
imino form by 13.6 kcal/mol. Figure 9a-c depicts the full
geometry optimized structures (6-31G(d,p) basis set) of 7AZD
dimer, 7AZD/ACID, and 7AZD/LACTAM dual hydrogen-
bonded complexes. For the case of 7AZD dimer, the association

7826 J. Phys. Chem. B, Vol. 104, No. 32, 2000
Chou et al.
Figure 9. Optimized geometries based on the HF/6-31G(d,p) basis
set (in Å and degree, only critical angles are shown) for the ground
states of (a) 7AZD dimer, (b) 7AZD/ACID, and (c) 7AZD/LACTAM.
prior to the dimerization, which may compensate for the gain
of the stabilization due to the dual hydrogen-bonding formation.
After truncating the 7AZD dimer by removing the dual hydrogen
bonds but holding structures of both 7AZD unchanged, each
7AZD species was calculated to be ∼0.8 kcal/mol higher in
energy than the geometry optimized 7AZD monomer. Note that
a torsional angle of 14° is obtained when the strain energy (an
endothermic process) plus hydrogen-bonding energy (an exo-
thermic process) reach a minimum value upon dimerization.
Thus, the sacrifice of certain hydrogen-bonding energy, e.g.,
the loss of a nonplanar configuration, is necessary to avoid
increasing the strain energy toward planarity. This consequence
explains the calculated ∆Hac of ca. -6.0 kcal/mol for the 7AZD
dimer, which is less exothermic than that of -11 kcal/mol
calculated for the 7AI dimer.16d Similar steric effects were
obtained in the formation of other 7AZD hydrogen-bonded
complexes, giving rise to a lower association constant in
comparison to those of 7AI hydrogen-bonded species where the
planar geometry of 7AI facilitates the dual hydrogen bonded
dimer and/or complex formation.

Amine-Imine Double Proton Transfer
J. Phys. Chem. B, Vol. 104, No. 32, 2000 7827
The calculated free-energy change of the 7AZD hydrogen-
bonded complex generally deviated from the experimental value
by 3-5 kcal/mol at room temperature (see Tables 1 and 2).
Although the use of a lower basis set may introduce certain
deviations, the major discrepancy between the theoretical (in
gas phase) and experimental (in solution) approaches is believed
to be due to solvation energy in the solution phase. Attempts
to resolve the solvation energy based on several different
semiempirical methods have been made. However the results
are scattered. This may not be surprising since the programs
we applied are based on the continuum dielectric model and
cannot accurately describe the 7AZD hydrogen-bonded systems
in solution. In addition, the entropy calculation is based on an
infinitely diluted solute in which the correction term for the
entropy due to the concentration effect has been neglected. Such
an effect is significant for experiments using a finite concentra-
tion, leading to more favorable entropy upon the hydrogen-
bonding association.
in a prominent tautomer emission, whereas ESDPT is thermally
unfavorable in the noncatalytic case of the 4ABI/LACTAM
complex.
Knowing that the normal f tautomer proton transfer process
for most ESIPT and ESDPT molecules is basically an ender-
gonic reaction in the ground state, the S0′ f S1′ transition of
the tautomer has to be significantly smaller than the S0-S1 gap
of the normal species in order for the excited-state proton
transfer reaction to proceed. From a spectroscopic viewpoint,
the ESDPT and/or ESIPT usually result in a conversion of the
aromatic conjugated double bonds in its normal form to a
nonaromatic type of conjugated double bonds. Hence, the πfπ*
transition (i.e., the S0′-S1′ energy gap) of the tautomeric form
is expected to be much more sensitive to the length of π electron
conjugation than that of the normal form. A prominent example
can be shown in a comparative study on 5-hydroxyflavone
(5HF) and 5-hydroxyflavanone (5HFN) where the C2-C3 bond
in the pyrone ring is hydrogenated relative to 5HF, reducing
the length of the conjugated system of the tautomer species.39
Consequently, the 5HFN tautomeric form with an emission
maximized at 550 nm (18 180 cm-1) is ∼3255 cm-1 higher in
energy than that of the S1′-S0′ of the 5HF tautomeric form of
670 nm (14 925 cm-1). Whereas there is only a difference of
4.2. Parameters Tuning ESDPT Reaction. Recently, Na-
gaoka and Nagashima have developed an elegant description
of the occurrence of excited-state proton transfer by considering
the nodal pattern of the wave function and the delocalization
of the lone electrons in the excited-state.34-36 According to the
“nodal plane” model, the selective reaction pathway from
various states can be qualitatively determined. For example, in
the case of an excited-state intramolecular proton transfer
(ESIPT) reaction, the higher reactivity of the La state over the
Lb state in salicylaldehyde and its analogues can be understood
in terms of a nodal plane analysis of their wave function where
the nodal plane across the C1-C2 bond in La state (see Figure
1), leading to a driving force for the proton-transfer reaction. A
similar method has also been applied to the case of the
7-azaindole dimer. Recent femtosecond dynamic studies reveal
a fast proton transfer rate in the La state, while the relaxation
of the Lb state is dominated by the Lb f La internal conversion.
The higher reactivity of the La state over the Lb state in the
dimeric form can also be rationalized in terms of a nodal plane
analysis.37
∼800 cm-1 in the S0-S1 transition between 5HF (28 160 cm-1
)
and 5HFN (28 960 cm-1).
In this study we render another key parameter based on the
length of conjugated double bonds to fine-tune the thermody-
namics of the excited-state proton transfer. Despite the occur-
rence of ESDPT in the 7AZD/ACID complex, the ESDPT is
prohibited in 7AZD dimer and 7AZD/LACTAM complex. The
lifetime of 7AZD dimer and 7AZD/LACTAM complex was
measured to be as long as 2.1 ns (kf ) 4.8 × 108 s-1 ) and 3.2
ns (kf ) 3.1 × 108 s-1, see Table 1), respectively. Thus, it is
quite unlikely that the rate of ESDPT is exceedingly slower
than the lifetime of either excited 7AZD dimer or 7AZD/
LACTAM complex to account for the lack of tautomer emission
unless there exists an unusually high energy barrier. Alterna-
tively, the results may more plausibly be rationalized by a
tautomerization energy-dependent ESDPT reaction. The photo-
physical properties shown in Table 1 reveal an interesting
correlation between the energy gap and its corresponding length
of the π electrons conjugation in which the maximum of S0-
S1 transition of 7MDPP (375 nm; 26,667 cm-1), a model
compound of the 7AZD tautomer, is higher in energy than that
(390 nm; 25,640 cm-1) of 7-methyl-7H-pyrrole[2,3-b]pyridine
(7MPP), a model compound of the 7AI tautomer by ∼1027
cm-1 (see Table 1). On the contrary, a reverse order of the S0-
S1 absorption peak frequency was observed between 7AZD (306
nm; 32 679 cm-1) and 7AI (287 nm, 34 843 cm-1) monomer,
being ∼2160 cm-1 lower in energy than for the 7AZD monomer
(see Table 1). The result indicates that the hydrogenation of
the C2-C3 double bond in the pyrrole system of 7AI alters the
resonance skeleton in the tautomeric form toward a drastic
increase of the πfπ* energy gap, while increasing the charge-
transfer property results in decreasing the S0-S1 transition of
the normal 7AZD (vide supra). Consequently, for various 7AZD
hydrogen-bonded complexes relative energy levels between the
S1 state of the amino form and the S1′ state of the imino tautomer
turn out to be crucial for the observed guest molecular-based
tuning ESDPT, especially the catalytic versus noncatalytic type.
While the theory of the nodal plane model renders a
qualitative, simple method to determine the state-selective
reaction pathway, relative thermodynamics, i.e., the relative
energy levels, between normal (nonproton-transfer) and tauto-
meric (proton-transfer) forms also play a crucial role to further
determine the “fate” of the excited-state proton-transfer reaction.
Such a thermodynamic factor can be fine-tuned by the sub-
stituent effect. In the case of an ESIPT reaction both 1-hydroxy-
2-acetonaphthone (1H2AN) and 1-hydroxy-2-naphthaldehyde
(1H2NA) have been reported to undergo fast proton transfer in
the excited state, resulting in a large Stokes-shifted tautomer
emission.38 In contrast, the ESIPT in methyl 1-hydroxy-2-
naphthoate (1H2MN) with a similar structure as 1H2AN and
1H2NA is prohibited. The results can be rationalized by the
great electron donating property of the methoxy group, which
destabilizes the S1′ state of the tautomer species. As a result,
ESDPT in 1-hydroxy-2-naphthoate is thermally forbidden owing
to its higher tautomer S1′ state with respect to the normal S1
state.38 Conversely, we have recently demonstrated that adding
an electron-rich nitrogen atom in the pyrrole system of 7AI,
forming 4-azabenzimidazole (4ABI), alters the resonance skel-
eton in the tautomeric form toward a drastic increase of the
πfπ* energy gap. Consequently, relative energy levels between
S1 and S1′ states turn out to be crucial for the observed guest
molecular-based tuning ESDPT9 in which a fast ESDPT takes
place in the catalytic case of the 4ABI/ACID complex, resulting
To rationalize the proposed mechanism, we have attempted
to construct relative energy levels for an overall proton-transfer
cycle in 7AZD. Experimentally, due to the thermal prohibition
for the tautomerism, it is not feasible to obtain a relative free

7828 J. Phys. Chem. B, Vol. 104, No. 32, 2000
Chou et al.
energy between the 7AZD dimer (or 7AZD/guest complex) and
its corresponding 7AZD(T) dimer (or 7AZD(T)/guest complex)
in the ground state. Alternatively, using a 6-31G(d,p) basis set,
the change of free energy for 7AZD(T)/ACID, 7AZD(T) dimer
and 7AZD(T)/LACTIM was calculated to be 10.7, 22.5, and
23.4 kcal/mol, respectively, relative to their corresponding 7AZD
(amino form) hydrogen bonding complexes (see Table 2). The
assignment of the 0-0 band for the S0-S1 transition of the
7AZD dimer and/or complexes is relatively difficult due to their
broad absorption profile. For a simplified approach, we take
the frequency where the intensity of the excitation spectrum
(monitored at the tautomer emission of 7AZD/ACID) is ∼1/3
of the maximum to be the 0-0 transition (∼340 nm; 29 412
cm-1) for either the normal 7AZD dimer or the 7AZD/guest
complexes. Conversely, it is rather simple to assign the S0′-
S1′ (0-0) transition of the imine tautomer due to its vibronically
progressive emission observed in the 7AZD/ACID complex. We
then take the first vibronic peak of 412 nm (24 272 cm-1, see
Figure 5) to be the 0-0 transition which is also applied to either
the 7AZD(T) dimer or the 7AZD(T)/LACTIM complex. Given
all of these values, the relative energy levels of a proton-transfer
cycle in the singlet manifold are depicted in Scheme 1.
Apparently, 7AZD/ACID proton-transfer tautomerism in the
excited state is calculated to be -4.1 kcal/mol. On the contrary,
an endergonic value of 7.7 and 8.6 kcal/mol was obtained for
the case of the 7AZD dimer and 7AZD/LACTAM complex,
respectively. Although this approach is qualitative due to the
uncertainty of obtaining accurate excited-state energy levels for
both normal and tautomer species in solution phase, it clearly
shows that for 7AZD dimer and 7AZD/LACTAM complex
ESDPT is thermally unfavorable, consistent with the experi-
mental results. The result can also be rationalized qualitatively
by the noncatalytic type of ESDPT where the guest molecules,
i.e., 7AZD amino form (in the 7AZD dimer) and LACTAM
(in the 7AZD/LACTAM complex) should undergo a corre-
sponding change to the imine and lactim forms, respectively,
during ESDPT. Such a noncatalytic process should raise its S1′
energy level which is a consequence of the simultaneous
tautomerization for both host and guest molecules. The result
of endergonic tautomerism on one hand may simply indicate
that either the 7AZD dimer or the 7AZD/LACTAM complex
may populate predominantly in the excited state through a fast
thermal equilibrium, resulting in a normal Stokes shifted
emission. On the other hand, the more endergonic reaction may
empirically lead to a higher formation free energy of the
activated complex, i.e., the existence of a high energy barrier,
so that ESDPT may be frustrated during the lifetime of the
excited 7AZD dimer or the 7AZD/LACTAM complex. Both
mechanisms unfortunately cannot be distinguished at this stage.
The results open up a study of ESDPT dynamics based on
an generalized amine/imine tautomerism in which the mecha-
nism may be extended to a DNA base such as adenine
possessing similar type of tautomerization proposed in the
mutation process.40 In contrast, despite the dual hydrogen-
bonding association, ESDPT is prohibited in the case of the
7AZD dimer and 7AZD/LACTAM complex, supporting the
proposed noncatalytic type of ESDPT mediated by the length
of the π electron conjugation.
Acknowledgment. This work was supported by the National
Science Council, Taiwan, R.O.C. (grant No. NSC 87 -2119-
M-194-002).
References and Notes
(1) Taylor, C. A.; El-Bayoumi, A. M.; Kasha, M. Proc. Natl. Acad.
Sci. U.S.A. 1969, 65, 253.
(2) Ingham, K. C.; El-Bayoumi, M. A. J. Am. Chem. Soc. 1971, 93,
5023
(3) Ingham. K. C.; El-Bayoumi, M. A. J. Am. Chem. Soc. 1974, 96,
1674.
(4) Watson, J. D.; Crick, F. H. C. Nature (London) 1953, 171, 964.
(5) Watson, D. G.; Sweet, R. M.; Marsh, R. E. Acta Crystallogr. 1965,
19, 573.
(6) Douhal, A.; Kim, S. K.; Zewail, A. H. Nature, 1995, 378.
(7) (a) Chachisvillis, M.; Fiebig, T.; Douhal, A.; Zewail, A. H. J. Phys.
Chem. A 1998, 102, 669. (b) Fiebig, T.; Chachisvillis, M.; Manger, M.;
Zewail, A. H.; Douhal, A.; Garcia-Ochoa, I.; de La Hoz Ayuso, A. J. Phys.
Chem. A 1999, 103, 7419.
(8) Chou, P. T.; Wei, C. Y.; Wu, G. R.; Chen, W. S. J. Am. Chem.
Soc. 1999, 121, 12186.
(9) Chou, P. T.; Wu, G. R.; Wei, C. Y.; Cheng, C. C.; Chang, C. P.;
Hung, F. T. J. Phys. Chem. B 1999, 103, 10042.
(10) McMorrow, D.; Aartsma, T. Chem. Phys. Lett. 1986, 125, 581.
(11) (a) Tokumura, K.; Watanabe, Y.; Itoh, M. J. Phys. Chem. 1986,
90, 2362. (b) Tokumura, K.; Watanabe, Y.; Udagawa, M.; Itoh, M. J. Am.
Chem. Soc. 1987, 109, 1346.
(12) Moog, R. S.; Bovino, S. C.; Simon, J. D. J. Phys. Chem. 1988, 92,
6545.
(13) Koijnenberg, J.; Huizer, A. H.; Varma, C. A. O. J. Chem. Soc.,
Faraday Trans. 2 1988, 84 (8), 1163.
(14) (a) Moog., R. S.; Maroncelli, M. J. Phys. Chem. 1991, 95, 10359.
(b) Chapman, C. F.; Maroncelli, M. J. Phys. Chem. 1992, 96, 8430. (c)
Mentus, S.; Maroncelli, M. J. Phys. Chem. A 1998, 102, 3860.
(15) (a) Negreie, M.; Bellefeuille, S. M.; Whitham, S.; Petrich, J. W.;
Thornburg, R. W. J. Am. Chem. Soc. 1990, 112, 7419. (b) Negrerie, M.;
Gai, F.; Bellefeuille, S. M.; Petrich, J. W. J. Phys. Chem. 1991, 95, 8663.
(c) Negrerie, M.; Gai, F.; Lambry, J.-C.; Martin, J.-L.; Petrich, J. W. J.
Phys. Chem. 1993, 97, 5046. (d) Chen, Y.; Rich, R. L.; Gai, F.; Petrich, J.
W. J. Phys. Chem. 1993, 97, 1770. (e) Chen, Y.; Gai, F.; Petrich, J. W. J.
Am. Chem. Soc. 1993, 115, 10158. (f) Rich, R. L.; Chen, Y.; Neven, D.;
Negrerie, M.; Gai, F.; Petrich, J. W. J. Phys. Chem. 1993, 97, 1781. (g)
Gai, F.; Rich, R. L.; Petrich, J. W. J. Am. Chem. Soc. 1994, 116, 735. (h)
Smirnov, A. V.; English, D. S.; Rich, R. L.; Lane, J. Teyton, L.;
Schwabacher, A. W.; Luo, S.; Thornburg, R. W.; Petrich, J. W. J. Phys.
Chem. 1997, 101, 1B, 2758.
(16) (a) Chou, P. T.; Martinez, M. L.; Cooper, W. C.; McMorrow, D.;
Collin, S. T.; Kasha; M. J. Phys. Chem. 1992, 96, 5203. (b) Chang, C. P.;
Hwang W. C.; Kuo M. S.; Chou P. T.; Clement, J. H. J. Phys. Chem. 1994,
98, 8801. (c) Chou, P. T.; Wei, C. Y.; Chang, C. P.; Chiu, C. H. J. Am.
Chem. Soc. 1995, 117, 7259. (d) Chou, P. T.; Wei, C. Y.; Chang, C. P.;
Kuo, M. S. J. Phys. Chem. 1995, 99, 11994. (e) Chou, P. T.; Yu, W. S.;
Chen, Y. C.; Wei, C. Y.; Martinez, S. S. J. Am. Chem. Soc. 1998, 120,
12927.
(17) Chaban, G. M.; Gordan, M. S. J. Phys. Chem. A 1999, 103, 185.
(18) 7AZD has been used as an intermediate to synthesize various
7-azaindoles.19 However, its photophysical and photochemical properties
have not yet been reported.
(19) Taylor, E. C.; Macor, J. E.; Pont, J. L. Tetrahedron 1987, 43, 5145.
(20) Gonnela, N. C.; Roberts, J. D. J. Am. Chem. Soc. 1982, 104, 3162.
(21) Dryefus, M.; Dodin, G.; Bensaude, O.; Dubois, J. E. J. Am. Chem.
Soc. 1975, 97, 2369.
(22) Robison, M. M.; Robison, B. L. J. Am. Chem. Soc. 1955, 77, 6554.
(23) Demas, J. N.; Crosby, G. A. J. Phys. Chem. 1971, 75, 991-1024.
(24) Chou, P. T.; Wei, C. Y.; Hung, F. T. J. Phys. Chem. B. 1997, 101,
9119.
(25) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A.
5. Conclusion
Summarizing the above results and discussion, we have
studied hydrogen-bonded complexes of 7AZD by means of
absorption, emission, and theoretical calculations.
For the catalytic type of ESDPT such as 7AZD/ACID, where
cyclic dual hydrogen bonds are intrinsically formed, the
photoinduced double proton transfer may require only a small
displacement of the hydrogen atom and/or molecular skeleton,
resulting in a small energy barrier. The rate of such a cooperative
proton-transfer reaction, taking place either stepwise or simul-
taneously, is expected to be much faster than the spontaneous
decay rate of the excited 7AZD/ACID complex, as supported
by the lack of a normal 7AZD/ACID emission as well as the
unresolved imine tautomer fluorescence rise time (τrise < 3 ×
10-10 s).

Amine-Imine Double Proton Transfer
J. Phys. Chem. B, Vol. 104, No. 32, 2000 7829
D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, J.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.;
Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick,
D. K.; Rabuck, a. D.; Raghavachari, K. Foresman, J. B.; Cioslowski, J.;
Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-
Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe,
M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.;
Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian,
Inc.: Pittsburgh, PA, 1998.
(26) Scott, A. P.; Radom, L. J. Phys. Chem. 1996, 100, 16502.
(27) Stewart, J. J. P. J. Comput. Chem. 1989, 10, 221.
(28) Buemi, G. THEOCHEM, 1988, 41, 379.
(29) Bliznyuk, A. A.; Voityuk, A. A. THEOCHEM, 1983, 41, 343.
(30) Cramer, C. J.; Truhlar, D. G. J. Am. Chem. Soc. 1993, 115, 8810.
(31) Cramer, C. J.; Truhlar, D. G. J. Comput. Chem. 1992, 13, 1089.
(32) Wang, J.; Boyd, R. J. Phys. Chem. 1996, 100, 16141.
(33) Through an absorption titration experiment the pKa of 7AZDH+
was determined to be 6.8. The pKa* of 7AZDH+ was calculated from the
Fo¨rster cycle based on the absorption maximum frequencies for 7AZD and
7AZDH+.
(34) Nagaoka, S.; Nagashima, U. Chem. Phys. 1989, 136, 153.
(35) Nagaoka, S.; Nagashima, U. J. Phys. Chem. 1990, 94, 1425.
(36) Nagaoka, S.; Shinde, Y.; Mukai, K.; Nagashima, U. J. Phys. Chem.
A 1997, 101, 61, and references therein.
(37) Takeuchi, S.; Tahara, T. J. Phys. Chem. A 1998, 102, 7740. Note
that a different interpretation regarding the Lb f La internal conversion
has been reported in ref 7b.
(38) Tobita, S.; Yamamoto, M.; Kurahayashi, N.; Tsukagoshi, R.;
Nakamura, Y.; Shizuka, H. J. Phys. Chem. A 1998, 102, 5206.
(39) Martinez, M. L.; Studer, S. L.; Chou, P. T. J. Am. Chem. Soc. 1991,
113, 5881.
(40) Chou, P. T.; Wei, C. Y., in preparation.