N?H-Type Excited-State Proton Transfer in Compounds Possessing a Seven-Membered-Ring Intramolecular Hydrogen Bond

Article abstract of DOI:10.1002/chem.201602425

A series of compounds containing 5-(2-aminobenzylidene)-2,3-dimethyl-3,5-dihydro-4H-imidazol-4-one (o-ABDI) as the core chromophore with a seven-membered-ring N?H-type intramolecular hydrogen bond have been synthesized and characterized. The acidity of the N?H proton and thus the hydrogen-bond strength can be fine-tuned by replacing one of the amino hydrogen atoms by a substituent R, the acidity increasing with increasing electron-withdrawing strength of R, that is, in the order H33. The tosyl and trifluoroacetyl derivatives undergo ultrafast, irreversible excited-state intramolecular proton transfer (ESIPT) that results in proton-transfer emission solely in the red region. Reversible ESIPT, and hence dual emission, involving the normal and proton-transfer tautomers was resolved for the acetyl- and benzyl-substituted counterparts. For o-ABDI, which has the weakest acidity, ESIPT is prohibited due to its highly endergonic reaction. The results clearly demonstrate the harnessing of ESIPT by modifying the proton acidity and hydrogen-bonding strength in a seven-membered-ring intramolecular hydrogen-bonding system. For all the compounds studied, the emission quantum yields are weak (ca. 10^?3) in dichloromethane, but strong in the solid form, ranging from 3.2 to 47.4 %.

Full text of DOI:10.1002/chem.201602425

DOI: 10.1002/chem.201602425
Full Paper
&
Proton Transfer
NꢀH-Type Excited-State Proton Transfer in Compounds
Possessing a Seven-Membered-Ring Intramolecular Hydrogen
Bond
Yi-An Chen+,[a] Fan-Yi Meng+,[a] Yen-Hao Hsu,[a] Cheng-Hsien Hung,[a] Chi-Lin Chen,[a] Kun-
You Chung,[a] Wei-Feng Tang,[b] Wen-Yi Hung,[b] and Pi-Tai Chou*[a]
Abstract: A series of compounds containing 5-(2-aminoben-
zylidene)-2,3-dimethyl-3,5-dihydro-4H-imidazol-4-one (o-
fer emission solely in the red region. Reversible ESIPT, and
hence dual emission, involving the normal and proton-trans-
fer tautomers was resolved for the acetyl- and benzyl-substi-
tuted counterparts. For o-ABDI, which has the weakest acidi-
ty, ESIPT is prohibited due to its highly endergonic reaction.
The results clearly demonstrate the harnessing of ESIPT by
modifying the proton acidity and hydrogen-bonding
strength in a seven-membered-ring intramolecular hydro-
gen-bonding system. For all the compounds studied, the
emission quantum yields are weak (ca. 10ꢀ3) in dichlorome-
thane, but strong in the solid form, ranging from 3.2 to
47.4%.
ABDI) as the core chromophore with a seven-membered-
ring NꢀH-type intramolecular hydrogen bond have been
synthesized and characterized. The acidity of the NꢀH
proton and thus the hydrogen-bond strength can be fine-
tuned by replacing one of the amino hydrogen atoms by
a substituent R, the acidity increasing with increasing elec-
tron-withdrawing strength of R, that is, in the order H<
COCH3 <COPh<Tosyl<COCF3. The tosyl and trifluoroacetyl
derivatives undergo ultrafast, irreversible excited-state intra-
molecular proton transfer (ESIPT) that results in proton-trans-
Introduction
(basicity) ability), reaction kinetics, and even the thermodynam-
ics of ESIPT has been a long-standing quest. Unfortunately,
however, there is no systematic experimental data available for
OH-type ESIPT systems.[5]
Ever since Weller’s first report on excited-state intramolecular
proton transfer (ESIPT) based on salicylic acid and its deriva-
tives,[1] studies have overwhelmingly been focused on the OH-
type of hydrogen-bonding system. Up to now, with the excep-
tion of environmental perturbation, such as external hydrogen
bonding and strong solvent polarity effects[2] or geometry con-
straints,[3] ESIPT is for numerous OH intramolecular hydrogen-
bonded systems a highly exergonic process that takes place
essentially without a barrier and results solely in proton-trans-
fer tautomer emission.[4] This makes the fundamental study of
ESIPT in terms of both reaction dynamics and/or thermody-
namics difficult. For example, the correlation between hydro-
gen-bonding strength (or proton donor (acidity) and acceptor
Recent studies of NꢀH-type intramolecular hydrogen-
bonded systems[6] offer several advantages over the OꢀH hy-
drogen-bonded systems. First, the lower acidity of the NꢀH
proton and hence the associated weaker hydrogen bond may
result in a finite rate of ESIPT suitable for experimental explora-
tion. More importantly, the amino nitrogen allows easy chemi-
cal derivation through the replacement of one of the amino
protons by a substituent R, such that the NRꢀH acidity can be
fine-tuned according to the electronic nature of the R substitu-
ent, thereby resulting in changes in the ESIPT dynamics and/or
thermodynamics, which can be probed systematically.
On the basis of the above analysis, we recently reported an
NꢀH ESIPT system involving PBT-NRH comprising a benzothia-
zole moiety substituted at the 2-position by a phenyl ring con-
taining substituted amino groups (see Scheme 1) in which R
was varied from electron-donating (e.g., CH3) to strongly elec-
tron-withdrawing (e.g., COCF3) groups.[6i] As a result, modifica-
tion of the ESIPT reaction dynamics and thermodynamics as
a function of substitution was demonstrated. The results
showed a correlation between the hydrogen-bonding strength
and the ESIPT dynamics and thermodynamics. Increasing the
electron-withdrawing strength of the R substituent increases
the acidity of the NRꢀH proton and hence enhances the
strength of the hydrogen bond. Interestingly, this is tightly
[a] Y.-A. Chen,+ F.-Y. Meng,+ Y.-H. Hsu, C.-H. Hung, C.-L. Chen, K.-Y. Chung,
Prof. Dr. P.-T. Chou
Department of Chemistry
Center for Emerging Material and Advanced Devices
National Taiwan University, Taipei, 10617 (Taiwan R.O.C.)
E-mail: chop@ntu.edu.tw
[b] W.-F. Tang, Prof. Dr. W.-Y. Hung
Institute of Optoelectronic Sciences
National Taiwan Ocean University
Keelung, 20224 (Taiwan R.O.C.)
[+] These authors contributed equally to this work.
Supporting information and ORCID number from the author for this article
are available on the WWW under http://dx.doi.org/10.1002/
chem.201602425.
Chem. Eur. J. 2016, 22, 1 – 9
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Results and Discussion
Synthesis
Compounds 4[9] and 5a–5d were synthesized starting with imi-
dazolinone derivative 1[10] and 2-aminobenzaldehyde (3),[11]
which were prepared according to methods described in the
literature. Compounds 1 and 3 were subjected to the Knoeve-
nagel condensation reaction,[10] which afforded 4[9] in a yield of
85% (Scheme 2). By using 4 as the precursor, we then strategi-
cally designed and synthesized four amino-substituted deriva-
tives 5a–5d (Scheme 2). Under nitrogen, the reaction of 4 with
acetic anhydride in CH2Cl2 at room temperature for 30 min
gave the corresponding amide 5a in a yield of 76%.[12] Similar-
ly, the treatment of 4 with benzoyl chloride (BzCl) in CH2Cl2 at
Scheme 1. Structures of various hydrogen-bonded ESIPT systems. The
dashed lines denote hydrogen bonds.
linked to the increase in ESIPT reaction rate and exergonicity.
The latter relationship between the kinetics and thermodynam-
ics can be considered in terms of Hammond’s postulate,[7]
which implicitly implies that in a more exothermic reaction the
reactant is closer in energy to the activated complex and
hence there is a lower barrier and faster reaction rate.
We then explored another system based on the N-substitut-
ed 10-aminobenzo[h]quinoline (BQ-NRH) chromophore (see
Scheme 1).[6j] In this system, as a result of the fused aromatic
moiety, the NꢀH forms an intramolecular hydrogen bond with
the pyridinyl nitrogen atom to form a six-membered-ring. The
well-matched angle/distance and higher basicity of the pyridin-
yl nitrogen (cf. the benzothiazole nitrogen in PBT-NRH) lead to
the formation of a strong hydrogen bond. As a result, inde-
pendently of the R group, ESIPT in the BQ-NRH system is
highly exergonic with an ultrafast reaction rate, which results
solely in proton-transfer tautomer emission. Because the
nature of R affects the highest-occupied molecular orbital
(HOMO) levels more than the lowest-unoccupied molecular or-
bital (LUMO) levels (see below), the tautomer emission of the
BQ-NRH system can be greatly altered, spanning the green to
the near-infrared region, by using substituents R with electron-
withdrawing (e.g., COCF3) and -donating (e.g., CH3) properties.
In addition to the above six-membered-ring hydrogen-bond-
ing system, we wished to study in detail another NꢀH-type in-
tramolecular hydrogen-bonding system containing a seven-
membered-ring hydrogen bond. Because of its geometry, the
seven-membered-ring hydrogen bond may have a less suitable
hydrogen-bond angle and length and hence perhaps a weaker
hydrogen-bond strength, which may lead to NꢀH ESIPT prop-
erties distinct from those of the six-membered-ring hydrogen-
bonding systems. In this context, we studied the ortho-Green
Fluoresence Protein chromophore (ortho-GFP) core 5-(2-hy-
droxybenzylidene)-2,3-dimethyl-3,5-dihydro-4H-imidazol-4-one
(o-HBDI; see Scheme 1) in which ESIPT takes places in the OH
seven-membered-ring hydrogen bond and yields a prominent
tautomer emission at 605 nm.[8] Accordingly, the replacement
of OH by NRH may provide invaluable information relating to
Scheme 2. Synthetic route to intermediate 4 and its further derivatization to
5a–5d.
NꢀH-type ESIPT for
a
relatively weak hydrogen-bonding
room temperature for 20 min afforded 5b in a yield of 40%.
Furthermore, p-toluenesulfonyl chloride (TsCl) reacted with 4 in
the presence of pyridine in CH2Cl2 to give 5c in a yield of 74%.
Finally, the desired triflate trifluoroacetamide 5d was obtained
in a yield of 70% by treating 4 with trifluoroacetic anhydride
(TFAA) under similar conditions.
system. Thus, in this study we modified o-HBDI by replacing
the OH group by various NRꢀH groups, thereby forming a new
series of seven-membered-ring NꢀH hydrogen-bonded com-
pounds based on 5-(2-aminobenzylidene)-2,3-dimethyl-3,5-di-
hydro-4H-imidazol-4-one (4[9] and 5a–5d). Details of the syn-
thesis, characterization, and distinct substituent-dependent
NRꢀH ESIPT properties of this new system are elaborated in
the following sections.
Details of the synthetic route and characterization are elabo-
rated in the Experimental Section. To further affirm the struc-
ture of 5c, a crystal of 5c was successfully grown and analyzed
by X-ray diffraction; the crystal structure is shown in Figure 1.
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Figure 1. Single-crystal X-ray diffraction structure of 5c. aN2-C8-C7-
C6=3.2038, aC8-C7-C6-C1=9.4048, N1ꢀN2=2.604 ꢁ. Ellipsoids drawn at
50% probability level. CCDC-1481193.
Figure 2. Normalized steady-state absorption spectra (black dotted-solid
lines) and emission spectra (gray solid lines) of (a) 4, (b) 5a, (c) 5b, (d) 5c,
and (e) 5d in CH2Cl2 at room temperature (lex =380 nm).
The crystallographic data (see the Supporting Information)
reveal that the aniline and imidazol-4-one rings are nearly co-
planar, as evidenced by the rather small dihedral angles for
both aN2-C8-C7-C6 (3.2038) and aC8-C7-C6-C1 (9.4048). This,
in combination with a bond length of 2.604(2) ꢁ for N1ꢀN2
and an angle of 165(3)8 for aN2···H-N1, strongly supports the
existence of an intramolecular hydrogen bond observed in the
form of a seven-membered-ring structure.
showing an intense band with a maximum in the range from
358 to 426 nm with a molar absorptivity (e) ranging from 9162
to 13756mꢀ1 cmꢀ1, which can be assigned to p!p* transi-
tions, as supported by the frontier orbital analyses (shown in
Table S1 in the Supporting Information, see below). The emis-
sion spectrum of 4 (Figure 2a) shows a featureless band cen-
tered at 525 nm in CH2Cl2. The absorption (426 nm) and emis-
sion (525 nm) spectra reveal a mirror image with a regular
Stokes shift of around 4400 cmꢀ1 (peak-to-peak difference in
terms of frequency), which suggests that the emission origi-
nates from the relaxed normal Franck–Condon excited state.
Apparently, ESIPT does not take place in compound 4. The lack
of ESIPT reflects the much weaker NH2 hydrogen bond. In-
stead, the emission of 4 shows charge-transfer character and
hence solvatochromism, that is, a spectral redshift as the sol-
vent polarity is increased.[9]
1
Consistent with this observation, the H NMR spectrum re-
corded for 5c reveals a significantly downfield-shifted signal at
d=13.04 ppm for the amino proton (in CDCl3, see Figure S10
in the Supporting Information). For a similar type of chemical
structure, it has been established that a stronger intramolecu-
lar hydrogen bond is qualitatively correlated with a greater
1
downfield shift of the H chemical shift in the same solvent.[6i]
In the series of molecules studied in this work, compound 4
shows the weakest hydrogen bond, as evidenced by the chem-
ical shift of the NꢀH proton at d=7.46 ppm in CDCl3. The NꢀH
chemical shifts of the series of compounds studied in this work
increase in the order 4 (7.46 ppm)<5a (12.18 ppm)<5b
(12.36 ppm)<5c (13.04 ppm)<5d (13.63 ppm), reflecting the
trend in NꢀH acidity measured in aqueous solution, which in-
creases in the order 4 (pKa >14)ꢁ5a (pKa >14)<5b (pKa =
13.2)<5c (pKa =8.7)ꢁ5d (pKa =8.7). These results reaffirm the
view that an increase in the electron-withdrawing nature of R
increases the NꢀH acidity and hence enhances the hydrogen-
bond strength. We then examined the relationship between
the ESIPT properties of 4 and 5a–5d in terms of dynamics
and/or thermodynamics and the hydrogen-bond strength by
means of emission spectroscopy.
Replacing one of the NꢀH protons in 4 by an electron-with-
drawing group, to form 5a–5d, should facilitate ESIPT due to
the increase in NꢀH acidity and hence the hydrogen-bond
strength, which essentially affects the driving force of ESIPT. As
shown in Figure 2, incorporating an electron-withdrawing
group clearly causes a blueshift of the lowest-lying absorption
band. This can be rationalized by lowering the energy of the
HOMO (see below) of the compounds, which results in an in-
crease in the HOMO–LUMO gap such that a stronger electron-
withdrawing R group gives rise to a larger energy gap, consis-
tent with the trend in the absorption peak wavelength, which
is blueshifted as a function of increasing electron-withdrawing
strength of R.
For strong electron-withdrawing groups such as tosyl and
COCF3 in 5c and 5d, respectively, solely red emission with
a maximum at 631 and 620 nm, respectively, is resolved with
anomalously large Stokes shifts of around 11000 cmꢀ1, which
makes their assignment to proton transfer tautomer emission
unambiguous. Upon replacement by weaker electron-with-
drawing group such as acyl (Ac) and benzoyl (Bz), to form 5a
Photophysical properties
The steady-state absorption and emission spectra of 4 and
5a–5d in dichloromethane at room temperature are shown in
Figure 2, and relevant data are listed in Table 1. All the com-
pounds reveal similar electronic absorption features, each
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Table 1. Photophysical properties of 4 and 5a–5d in CH2Cl2 at room temperature.
Absorption S0!S1
Emission S1!S0
lexp [nm]
lcalcd
[nm] (f)[a]
lexp
[nm][a]
lcalcd
[nm] (f)[a]
Ff
texp [ps]
DEexp
*
DEcalcd
*
(e [mꢀ1 cmꢀ1])
426 (10318)
378 (10639)
[%]
(pre-exp. factor)[b]
[kcalmolꢀ1
]
[kcalmolꢀ1
]
[c]
[c]
4
433 (0.38)
403 (0.39)
N: 525
N: 478
T: 660
N: 455
T: 657
N: 476 (0.43)
N: 460 (0.51)
T: 593 (0.20)
N: 458 (0.47)
T: 580 (0.23)
N: 468 (0.46)
T: 564 (0.27)
N: 456 (0.56)
T: 568 (0.20)
0.74
0.08
82.3[d]
450 nm: 0.9 (0.54), 9.3 (0.46)
650 nm: 0.9 (ꢀ0.43), 9.3 (0.57)
450 nm: 0.6 (0.89), 13.8 (0.11)
650 nm: 0.7 (ꢀ0.73), 13.8 (0.27)
5a
ꢀ0.09
ꢀ1.24
2.72
0.36
5b
5c
5d
362 (13756)
360 (11464)
358 (9162)
410 (0.37)
395 (0.35)
383 (0.48)
0.18
0.27
0.42
T: 631
T: 620
67.2[d]
ꢀ5.84
ꢀ3.72
88.4[d]
[a] N=normal emission; T=tautomer (ESIPT) emission; f=oscillator strength. [b] Experimental emission lifetimes (texp) determined from femtosecond pho-
toluminescence up-conversion. [c] Energy differences (DE*=E*tautomerꢀE*normal) between the normal and tautomer forms in the lowest excited state (S1).
[d] Population decay time constant was applied for the fitting by sub-nanosecond Time Correlated Single Photon Counting (TCSPC).
and 5b, respectively, dual emission was clearly resolved with
peak wavelengths at 478 and 660 nm for 5a and 455 and
657 nm for 5b in CH2Cl2 (see Figure 2b,c). The excitation spec-
tra monitored at short and long wavelengths are identical (see
Figure S1 in the Supporting Information), which are also the
same as the absorption spectrum, thereby confirming that
both emission bands originate from the same ground state. It
is thus reasonable to assign the short and long wavelength
bands to normal and proton-transfer tautomer emissions, re-
spectively. Note that 5a has also been synthesized by Lo
et al.[12] They claimed that only normal emission was resolved
for 5a in acetonitrile. Because no emission spectrum was pro-
vided in their study we reinvestigated 5a in acetonitrile. The
results shown in Figure S2 reveal similar dual emission consist-
ing of normal and tautomer bands with maxima at 450 and
660 nm, respectively. Therefore, ESIPT apparently does operate
in 5a. We believe that the discrepancy arises from weak dual
emission (overall Ff ꢁ5ꢂ10ꢀ4) from 5a in acetonitrile, and that
the red tautomer emission (lmax =660 nm) could not be re-
solved by Lo et al. because their detecting system was not suf-
ficiently sensitive to red emission.
lar system-response-limited rise time and a fast population
decay time of 88.4 ps have been resolved for the tautomer
emission of 5d (see Table 1). Clearly, both 5c and 5d undergo
ultrafast, highly exergonic ESIPT.
For the case of dual emissions, for example, 5a in CH2Cl2, its
early-time relaxation dynamics of normal emission, monitored
at 450 nm (Figure 3b), exhibit a very fast decay component of
0.9ꢂ0.1 ps and a population decay component of 9.3 ps,
whereas the tautomer emission, monitored at 650 nm, consists
of a 0.9ꢂ0.1 ps rise component and a population decay time
of 9.3 ps. For 5b in CH2Cl2, the normal emission, monitored at
450 nm, also decays biexponentially, with decay times of 0.6ꢂ
0.1 and 13.8 ps. On the other hand, the time-resolved tauto-
mer emission monitored at 650 nm consists of a 0.7ꢂ0.1 ps
rise-time component and a decay component with a time con-
stant of 13.8 ps. For both 5a and 5b, the fast decay times of
the normal emission, within experimental error, match well the
rise time of the tautomer emission. This, together with the
identical population decay times for both normal and tauto-
mer emissions, leads us to conclude that both 5a and 5b un-
dergo a reversible type of ESIPT in which there exists a fast
pre-equilibrium between the excited normal (N*) and tautomer
(T*) states (Scheme 3).
To gain further insight into the ESIPT dynamics, time-re-
solved measurements were also carried out and the pertinent
data are listed in Table 1. Despite the lack of ESIPT in 4, its
normal emission, monitored at the wavelength maximum of
525 nm in CH2Cl2, shows fast single exponential decay kinetics
(see Figure S3a in the Supporting Information) with a lifetime
of 82.3 ps. The fast decay corresponds to a low emission quan-
tum yield (Fem) of 0.74%. The fast excited-state quenching
mechanism may invoke rotation around the exo-C6-C7-C8
bond (see Figure 1), which was well established in the study of
derivatives of the o-HBDI core chromophore.[8b]
Compounds 5c and 5d show exclusively proton-transfer
tautomer emission in the steady-state measurement. Therefore,
their rates of ESIPT are expected to be very fast. This is clearly
revealed in the early relaxation dynamics of the tautomer emis-
sion of 5c. As shown in Figure 3a, the rise time of the tauto-
mer emission for 5c in CH2Cl2, monitored at 620 nm, is beyond
the system response time of 150 fs and is followed by a fast
population decay time constant of 67.2 ps (see Table 1). A simi-
In principle, ESIPT of less than a few picoseconds in com-
pounds 5a and 5b may compete with the timescale of internal
conversion. Therefore, we carried out a fluorescence up-con-
version study at two available excitation wavelengths, that is,
at the near-absorption onset (400 nm) and at the absorption
maximum of 380 nm, and, within experimental error, we ob-
tained similar ESIPT kinetics for both 5a and 5b. On the one
hand, this result may imply that the vibrational relaxation is
still faster than the rate of ESIPT. On the other hand, the
energy separation between these two excitation wavelengths
is not large enough to resolve the issue. For 5c and 5d, the
time constant of ESIPT is beyond our system response limit of
150 fs for excitation at either 380 or 400 nm. Therefore, for
these NRꢀH systems, whether the ESIPT dynamics is depen-
dent upon the excitation energy is pending resolution.
The deuterium isotope effect on ESIPT dynamics was also
probed in this study. Using 5a as an example, the time-depen-
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
and angle, the proton transfer (or tunneling) occurs simultane-
ously. Therefore, the deuterium isotope effect, in theory, is not
directly involved, or at least plays a very minor part, in the
ESIPT dynamics.
Under the pre-equilibrium, in which kpt and k are signifi-
ꢀpt
cantly larger than kN* and kT* (Scheme 3) the time-resolved
[N*]t and [T*]t can be expressed by Equations (1) and (2). De-
tailed kinetic derivations have been reported previously (see
also the Supporting Information).[13]
*
½N ꢃ0
1
2
*
ð1Þ
ð2Þ
½N ꢃt ¼
ꢄ ½ðl2ꢀXÞ ꢄ eꢀl tþðXꢀl1Þ ꢄ eꢀl t
ꢃ
l2ꢀl1
*
Kpt ꢄ ½N ꢃ0
1
2
*
½T ꢃt ¼
ꢄ ½eꢀl tꢀeꢀl t
ꢃ
l2ꢀl1
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
2
2
ðXþYÞꢀ ðXꢀYÞ þ4ꢅkpt ꢅk
ðXþYÞþ ðXꢀYÞ þ4ꢅkpt ꢅk
ꢀpt
ꢀpt
Well, l1 ¼
and l2 ¼
*
2
2
With X ¼ kpt þ kN and Y ¼ kꢀpt þ kT
*
When [N*]t =[N*]0 and [T*]t =0 at t=0 and Xꢁkpt, Equa-
tion (3) can be derived.
KN þKT
K
*
1þKeq
eq ; l2 ¼ kptþk
*
l1’
ꢀpt
Accordingly, the pre-exponential factors that contribute to
the fast and slow decay components of [N*]t, denoted as Nfast
and Nslow, respectively, can be expressed by Equations (4) and
(5).
Figure 3. Fluorescence up-conversion decay curves obtained for (a) 5c and
(b) 5a in CH2Cl2. The data points (blue and red) were obtained by monitor-
ing with the emission wavelengths depicted. Solid lines depict the best ex-
ponential fits. Inset: Enlargement of the early rise and decay curve up to
5 ps (lex =380 nm).
kpt
Nfast
’
ð3Þ
ð4Þ
kptþk
ꢀpt
k
ꢀpt
Nslow
’
kptþk
ꢀpt
Nfast and Nslow can be obtained from the biexponential fitting
of [N*]t followed by extrapolation to t=0. The ratio of Nfast to
N
slow, that is, Nfast/Nslow =kpt/kꢀpt, is equivalent to the pre-equilib-
rium constant Keq. The energy differences (DE*) between the
normal (N*) and tautomer (T*) species can thus be deduced
from DE*=ꢀRTlnKeq. Knowing l2 (=kpt +kꢀpt) and Keq (=kpt/
kꢀpt), both the forward (kpt) and backward (kꢀpt) ESIPT rate con-
stants can be deduced. As a result, tpt(=1/kpt), t (=1/kꢀpt),
Scheme 3. Proposed reversible-type ESIPT for 5a and 5b with various relax-
ation rate constants.
ꢀpt
Keq, and DE* were calculated to be 1.67 and 1.96 ps, 1.17, and
ꢀ0.09 kcalmolꢀ1, respectively, for 5a and 0.68 and 5.55 ps,
8.09, and ꢀ1.24 kcalmolꢀ1, respectively, for 5b in CH2Cl2.
Note that the population decay is fast for 5a–5d, which also
correlates well with their low emission yields, which range
from 8.0ꢂ10ꢀ4 for 5a to 4.2ꢂ10ꢀ3 for 5d (see Table 1). Both
the population decay time and emission yield increase in the
order 5a<5b<5c<5d, which indicates that the tautomer
emission may also be quenched by rotation around the exo-
C6-C7-C8 bond (see Figure 1).[8b] The intramolecular hydrogen
bond, in part, hinders this rotation and hence rationalizes the
dent emission of N-deuteriated 5a in CD2Cl2 revealed a similar
result to its NꢀH counterpart in the early relaxation dynamics
(see Figure S4 in the Supporting Information for comparison).
The results confirmed that the ESIPT dynamics are insensitive
to the H/D kinetic isotope effect. This result may not be sur-
prising. It is known that the OH ESIPT system bearing a seven-
membered-ring hydrogen bond undergoes geometry adjust-
ment to assume a configuration favorable for ESIPT reaction.[8b]
In other words, once the hydrogen bond is of the right length
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
increase in tautomer emission yield and lifetime as the
tion of the molecular geometry, for example, a bending
motion, which is associated with changes in the length or
angle of the NꢀH hydrogen bond and induces a non-negligible
barrier. Thus, a finite ESIPT rate is expected for compounds
with weaker NRꢀH acidity, such as 5a and 5b.
strength of the hydrogen bond increases from 5a to 5d.
Computational approaches
To provide complementary support for the experimental re-
sults, a computational analysis was performed by using time-
dependent density functional theory (TDDFT) with the B3LYP
hybrid functional. Also, the polarizable continuum model
(PCM) was incorporated into the calculations to simulate the
experimental conditions in dichloromethane (see the Experi-
mental Section in the Supporting Information for details). For
all the compounds studied, as listed in Table 1, the lowest-
lying electronic transitions (S0!S1) computed in terms of
wavelength agree well with the trend of absorption maxima,
which supports the validity of the current computational ap-
proach. Using 5a as a representative example, the frontier or-
bitals shown in Table S1 in the Supporting Information clearly
indicate that the HOMO commonly possesses higher electron
density on the amino group than the LUMO, which is manifest-
ed as charge transfer in HOMO!LUMO transitions and serves
as the driving force for ESIPT. We then calculated the energies
of the HOMOs and LUMOs and found that HOMO energy is af-
fected more than that of the LUMO by R substitution. As
a result, the increase in electron-withdrawing strength results
in an increase in the energy gap (see Table S2), consistent with
experimental observations (see Figure 2).
From the viewpoint of thermodynamics, a stronger hydro-
gen bond leading to more thermally favorable ESIPT can be
empirically rationalized, at least in part, by the elongation of
the p conjugation induced by the formation of an intramolecu-
lar hydrogen bond. This concept has been applied to explain
the redshift of the absorption of an intramolecular hydrogen-
bonded species compared with one that lacks an intramolecu-
lar hydrogen bond.[4a,8a] Recently, a similar concept was pre-
sented in the study of red fluorescence protein core chromo-
phores subject to hydrogen bonding by adjacent amino acids
or water molecules (i.e., water solvation[14]). However, the
degree to which the hydrogen bond lowers the energy may
be different for normal and tautomer states. Empirically, owing
to the breakdown of the aromatic structure (aromaticity) and
hence the full p delocalization, the excited state of the tauto-
mer is expected to be more stabilized than that of the normal
excited state, which possesses less p delocalization due to its
localized aromatic configuration. This viewpoint is similar to
the interpretation of ESIPT thermodynamics based on the
nodal plane,[15] in which the site of the nodal plane essentially
correlates with the extension of the nonaromatic p conjuga-
tion from the normal to the tautomer species in the excited
state. The stronger hydrogen bond thus lowers the energy of
the tautomer more than that of the normal state, thereby re-
sulting in thermally more favorable ESIPT.
We also calculated the ground-state (S0) and first-excited-
state (S1) energies for both the normal and proton-transfer tau-
tomer for the corresponding optimized geometries. According-
ly, the energy differences between the excited-state normal
and tautomer forms, DEcalcd*=E*tautomerꢀE*normal (the asterisk de-
notes the excited state), for the ESIPT process are listed in
Table 1. The results of the computer and spectroscopic analy-
ses were combined to establish the thermodynamic trends
shown by the compounds, which can be categorized as fol-
lows. 1) For 4, convergence could not be achieved in the tau-
tomer geometry optimization, which implies a much higher
energy, consistent with the prohibition of ESIPT. Hence, only
normal emission could be observed experimentally. 2) The cal-
culated values of DE* are 2.72 and 0.36 kcalmolꢀ1 for 5a and
5b, respectively, which, within calculation uncertainty, matches
well the values of ꢀ0.09 and ꢀ1.24 kcalmolꢀ1 obtained experi-
mentally (see Table 1) and supports the reversible type of
ESIPT and hence dual emission. (3) The calculated values of
DE* of ꢀ5.84 and ꢀ3.72 kcalmolꢀ1 for 5c and 5d, respectively,
reveal highly favorable ESIPT reactions and support the irrever-
sible type of ESIPT, which results solely in tautomer emission.
The above results can be rationalized on the basis of the Nꢀ
H hydrogen-bond strength harnessing ESIPT properties.[3a]
From the kinetics point of view, a strong NꢀH hydrogen bond
in, for example, 5c and 5d, leads to a shorter distance be-
tween the proton donor (NRꢀH) and acceptor. In this case,
ESIPT requires only a small structural variation and hence the
associated barrier is negligible, which results in an ultrafast
rate of reaction. On the other hand, for a weak intramolecular
NꢀH hydrogen-bonding system, ESIPT requires, in part, varia-
Finally, the quenching mechanism invoking rotation around
the exo-C6-C7-C8 bond (see Figure 1 for 5c) may incorporate
large-amplitude motion and should be greatly inhibited in
rigid media. Support for this viewpoint is given by the intense
tautomer emission observed in the solid state for 5a–5d. As
shown in Figure 4 and Table 2, depending on the electron-do-
nating/withdrawing properties of the R substituent, a wide
range of tautomer emission can be achieved from 580 (5d) to
640 nm (5a) with quantum yields as high as 47.4% (5d) in the
solid film. The high emission yields also correlate well with the
rather long population decay times of nanoseconds in the
solid state (see Table 2). Interestingly, as a result of the pre-
equilibrium ESIPT, 5a also shows intense normal emission with
a maximum at 488 nm in the solid film. The equal intensities
of the 488 (normal) and 640 nm (tautomer) emissions lead to
the generation of warm white light with Commission Interna-
tionale de L’Eclairage (CIE) (0.41, 0.42) (see inset of Figure 4),
which demonstrates its potential in lighting applications.
Conclusion
This study of a series of NꢀH seven-membered-ring hydrogen-
bonded molecules, in combination with an analysis of the data
collected previously for NꢀH six-membered-ring hydrogen-
bonded systems,[6i–k] has allowed us to draw the following con-
clusions.
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[R=COCH3 (5a) and COPh (5b)] and then to the highly exer-
gonic cases [R=Tosyl (5c) and COCF3 (5d)], with the corre-
sponding tautomer emission spanning from 580 to 640 nm.
Therefore, this new series of NRꢀH seven-membered-ring hy-
drogen-bonding systems and their corresponding ESIPT prop-
erties broadens the scope of NꢀH-type ESIPT in terms of both
fundamental insight and future applications.
Experimental Section
Synthesis and characterization: All solvents were distilled from
appropriate drying agents prior to use. Commercially available re-
agents were used without further purification unless otherwise
stated. All reactions were monitored by TLC. Column chromatogra-
phy was carried out by using silica gel from Merck (230–400
mesh). 1H and 13C NMR spectra were recorded on a Varian Unity
400 spectrometer at 400 and 100 MHz, respectively. Chemical shifts
(d) are recorded in parts per million (ppm) and coupling constants
(J) are reported in Hertz (Hz). Mass spectra were obtained by using
a Bruker microTOF-QIIwith an ESI source.
Figure 4. Normalized solid-film emission spectra of 4 and 5a–5d
(lex =380 nm). Inset: Emission colors of 4 and 5a–5d viewed by the naked
eye.
Table 2. Photophysical properties of 5a–5d recorded in the solid state.
(Z)-4-(2-Aminobenzylidene)-1,2-dimethyl-1H-imidazol-5(4H)-one
(4): A catalytic amount of piperidine (133 mL, 0.053 mmol) in a sy-
ringe was added to a solution of 1 (33 mg, 0.30 mmol) and 2-ami-
nobenzaldehyde (33 mg, 0.27 mmol) in toluene (35 mL). After stir-
ring at 508C for 12 h, the solvent was removed under reduced
pressure. The crude product and impurities were separated by
flash chromatography (EtOAc/hexanes, 1:2; Rf =0.35 for 4) and re-
crystallized from EtOAc/hexanes to afford 4 (50 mg, 85%) as an
orange solid. 1H NMR (400 MHz, CDCl3): d=2.28 (s, 3H), 3.14 (s,
3H), 6.25 (br., 2H), 6.59–6.67 (m, 2H), 7.10–7.14 (m, 1H), 7.17 (s,
1H), 7.46 ppm (d, J=7.2 Hz, 1H); 13C NMR (100 MHz, CDCl3): d=
15.4, 26.6, 116.7, 117.1, 117.7, 128.7, 132.5, 133.9, 136.0, 148.2,
158.0, 169.8 ppm; HRMS: calcd for C12H14N3O [M+H]+: 216.1137;
found: 216.1149.
lem [nm][a]
Ff [%]
3.2
tobs [ns][b]
1.00
4
562
N*:488
T*: 640
N*: 463[d]
T*: 620
580
5a
5.7
0.86[c]
5b
15.4
3.09[c]
5c
5d
44.6
47.4
7.02
8.40
580
[a] N*=normal emission; T*=tautomer (ESIPT) emission. [b] Lifetime
measured by using the TCSPC system with a pulsed hydrogen-filled lamp
as the excitation source (lex =380 nm). [c] Both normal and tautomer
emissions reveal the same life time. [d] The 463 nm emission is weak but
non-negligible.
(Z)-N-(2-{[1,2-Dimethyl-5-oxo-1H-imidazol-4(5H)-ylidene]methyl}-
phenyl)acetamide (5a): A mixture of 4 (50 mg, 0.23 mmol) and
acetic anhydride (24 mL, 0.25 mmol) in CH2Cl2 (3 mL) was stirred at
room temperature for 30 min. H2O was added and the mixture was
extracted with CH2Cl2 (3ꢂ3 mL). The combined organic extracts
were dried over anhydrous Na2SO4, filtered, and concentrated
under reduced pressure. The residual crude product was purified
on silica gel to give 45 mg (76% yield) of 5a as a yellow solid.
1H NMR (400 MHz, CDCl3): d=2.04 (s, 3H), 2.21 (s, 3H), 3.03 (s, 3H),
6.96–6.90 (m, 2H), 7.30–7.20 (m, 2H), 8.18 (d, J=8.4 Hz, 1H),
12.18 ppm (br., 1H); 13C NMR (100 MHz, CDCl3): d=15.4, 24.5, 26.8,
122.9, 123.6, 123.7, 128.3, 132.1, 135.1, 135.9, 138.0, 160.4, 168.9,
169.1 ppm; HRMS: calcd for C14H16N3O2 [M+H]+) 258.1243; found:
258.1231.
1) From the viewpoint of kinetics, upon Franck–Condon exci-
tation of 5a–5d, structural variation is necessary to assume
a configuration favorable for ESIPT, which most likely involves
a molecular bending motion associated with changes in the
length and angle of the NꢀH hydrogen bond, thereby inducing
a certain barrier. This barrier is dependent upon the acidity of
NꢀH and hence the strength of the hydrogen bond. A finite
rate of ESIPT has been resolved for 5a and 5b with weaker
NRꢀH acidity.
2) A stronger hydrogen bond leads to more thermally favor-
able ESIPT, which can be empirically rationalized, at least in
part, by the elongation of the p conjugation induced by the
formation of the intramolecular hydrogen bond, for which the
tautomer is subject to a greater degree of stabilization (cf. the
normal species) in the S1 state due to more extensive p deloc-
alization. The stronger hydrogen bond thus increases the rela-
tive energy between normal and tautomer states (i.e., more
negative DE, see Table 1).
(Z)-N-(2-{[1,2-Dimethyl-5-oxo-1H-imidazol-4(5H)-ylidene]methyl}-
phenyl)benzamide (5b): A mixture of 4 (115 mg, 0.53 mmol) and
benzoyl chloride (67 mL, 0.58 mmol) in CH2Cl2 (3 mL) was stirred at
room temperature for 20 min. H2O was added and the mixture was
extracted with CH2Cl2 (3ꢂ3 mL). The combined organic extracts
were dried over anhydrous Na2SO4, filtered, and concentrated
under reduced pressure. The residual crude product was purified
on silica gel to give 68 mg (40% yield) of 5b as a yellow solid.
1H NMR (400 MHz, CDCl3): d=2.17 (s, 3H), 3.08 (s, 3H), 7.10–7.15
(m, 2H), 7.38–7.50 (m, 5H), 7.92–7.94 (m, 2H), 8.19 (d, J=8.4 Hz,
1H), 12.36 ppm (br., 1H); 13C NMR (100 MHz, CDCl3): d=15.1, 26.4,
124.3, 124.8, 125.0, 127.7, 128.0, 128.2, 131.35, 131.42, 135.1, 135.5,
136.0, 137.4, 161.4, 167.1, 168.8 ppm; HRMS: calcd for C19H18N3O2
[M+H]+: 320.1399; found: 320.1407.
The combination of (1) and (2) rationalizes the correlation
between hydrogen-bond strength and ESIPT kinetics and ther-
modynamics. By increasing the electron-withdrawing power of
the R group and hence the strength of the hydrogen bond,
the rate of NRꢀH ESIPT increases and the reaction thermody-
namics change from the endergonic [R=H (4)] to equilibrium
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
2013, 42, 1379–1408; d) R. Ghosh, D. K. Palit, Photochem. Photobiol. Sci.
2013, 12, 987–995.
(Z)-N-(2-{[1,2-Dimethyl-5-oxo-1H-imidazol-4(5H)-ylidene]methyl}-
phenyl)-4-methylbenzenesulfonamide (5c): mixture of
A
4
[3] a) M. W. Chung, J. L. Liao, K. C. Tang, C. C. Hsieh, T. Y. Lin, C. Liu, G. H.
Lee, Y. Chi, P. T. Chou, Phys. Chem. Chem. Phys. 2012, 14, 9006–9015;
b) J. Z. Zhao, S. M. Ji, Y. H. Chen, H. M. Guo, P. Yang, Phys. Chem. Chem.
Phys. 2012, 14, 8803–8817; c) G. A. Parada, T. F. Markle, S. D. Glover, L.
Hammarstrom, S. Ott, B. Zietz, Chem. Eur. J. 2015, 21, 6362–6366.
[4] a) P. T. Chou, Y. C. Chen, W. S. Yu, Y. H. Chou, C. Y. Wei, Y. M. Cheng, J.
Phys. Chem. A 2001, 105, 1731–1740; b) S. Ameer-Beg, S. M. Ormson,
R. G. Brown, P. Matousek, M. Towrie, E. T. J. Nibbering, P. Foggi, F. V. R.
Neuwahl, J. Phys. Chem. A 2001, 105, 3709–3718; c) M. Barbatti, A. J. A.
Aquino, H. Lischka, C. Schriever, S. Lochbrunner, E. Riedle, Phys. Chem.
Chem. Phys. 2009, 11, 1406–1415; d) J. Piechowska, K. Virkki, B. Sado-
wski, H. Lemmetyinen, N. V. Tkachenko, D. T. Gryko, J. Phys. Chem. A
2014, 118, 144–151.
(100 mg, 0.46 mmol), toluenesulfonyl chloride (95 mg, 0.5 mmol),
and pyridine (40 mL, 0.5 mmol) in CH2Cl2 (4 mL) was stirred at room
temperature for 1 h. A 10% aqueous solution of HCl was added
and the mixture was extracted with CH2Cl2 (3ꢂ3 mL). The com-
bined organic extracts were dried over anhydrous Na2SO4. After fil-
tration and concentration under reduced pressure, the residual
crude product was purified on silica gel to give 125 mg (74%
1
yield) of 5c as an orange solid. H NMR (400 MHz, CDCl3): d=2.27
(s, 3H), 2.44 (s, 3H), 3.18 (s, 3H), 6.78 (s, 1H), 6.99 (m, 1H), 7.06 (d,
J=8.4 Hz, 2H), 7.25 (m, 2H), 7.48 (d, J=8.4 Hz, 2H), 7.57 (d, J=
8.4 Hz, 1H), 13.04 ppm (br., 1H); 13C NMR (100 MHz, CDCl3): d=
15.5, 21.4, 26.8, 122.5, 124.0, 125.0, 126.6, 126.9, 129.1, 131.9, 134.6,
135.8, 136.76, 136.83, 143.0, 161.0, 168.2 ppm; HRMS: calcd for
C19H20N3O3S [M+H]+: 370.1225; found: 370.1210.
[5] A. J. Stasyuk, P. Bultinck, D. T. Gryko, M. K. Cyranski, J. Photochem. Photo-
biol. A 2016, 314, 198–213.
[6] a) T. P. Smith, K. A. Zaklika, K. Thakur, P. F. Barbara, J. Am. Chem. Soc.
1991, 113, 4035–4036; b) S. Santra, G. Krishnamoorthy, S. K. Dogra,
Chem. Phys. Lett. 1999, 311, 55–61; c) S. Santra, G. Krishnamoorthy, S. K.
Dogra, J. Phys. Chem. A 2000, 104, 476–482; d) C. J. Fahrni, M. M.
Henary, D. G. VanDerveer, J. Phys. Chem. A 2002, 106, 7655–7663;
e) M. M. Henary, Y. G. Wu, C. J. Fahrni, Chem. Eur. J. 2004, 10, 3015–
3025; f) Y. G. Wu, P. V. Lawson, M. M. Henary, K. Schmidt, J. L. Bredas,
C. J. Fahrni, J. Phys. Chem. A 2007, 111, 4584–4595; g) M. K. Nayak, J.
Photochem. Photobiol. A 2012, 241, 26–37; h) A. I. Ciuciu, K. Skonieczny,
D. Koszelewski, D. T. Gryko, L. Flamigni, J. Phys. Chem. A 2013, 117, 791–
803; i) H. W. Tseng, J. Q. Liu, Y. A. Chen, C. M. Chao, K. M. Liu, C. L. Chen,
T. C. Lin, C. H. Hung, Y. L. Chou, T. C. Lin, T. L. Wang, P. T. Chou, J. Phys.
Chem. Lett. 2015, 6, 1477–1486; j) H. W. Tseng, T. C. Lin, C. L. Chen, T. C.
Lin, Y. A. Chen, J. Q. Liu, C. H. Hung, C. M. Chao, K. M. Liu, P. T. Chou,
Chem. Commun. 2015, 51, 16099–16102; k) C. L. Chen, H. W. Tseng, Y. A.
Chen, J. Q. Liu, C. M. Chao, K. M. Liu, T. C. Lin, C. H. Hung, Y. L. Chou,
T. C. Lin, P. T. Chou, J. Phys. Chem. A 2016, 120, 1020–1028.
(Z)-N-(2-[{1,2-Dimethyl-5-oxo-1H-imidazol-4(5H)-ylidene]methyl}-
phenyl)-2,2,2-trifluoroacetamide (5d): A mixture of 4 (70 mg,
0.32 mmol) and trifluoroacetic anhydride (50 mL, 0.35 mmol) in
CH2Cl2 (3 mL) was stirred at room temperature for 20 min. A satu-
rated aqueous solution of NaHCO3 was added and the mixture was
extracted with CH2Cl2 (3ꢂ3 mL). The combined organic extracts
were dried over anhydrous Na2SO4, filtered, and concentrated
under reduced pressure. The residual crude product was purified
on silica gel to give 70 mg (70% yield) of 5d as a yellow solid.
1H NMR (400 MHz, CDCl3): d=2.34 (s, 3H), 3.16 (s, 3H), 7.06 (s, 1H),
7.16–7.20 (m, 1H), 7.38–7.44 (m, 2H), 8.13 (d, J=8.4 Hz, 1H),
13.63 ppm (br., 1H); 13C NMR (100 MHz, CDCl3): d=14.9, 26.7, 116.3
(q, JCꢀF =286 Hz), 124.2, 125.3, 125.7, 126.6, 131.5, 134.6, 135.3,
135.7, 155.6 (q, JCꢀF =38 Hz), 162.2, 168.4 ppm; 19F NMR (376 MHz,
CDCl3): d=ꢀ75.2 ppm; HRMS: calcd for C14H13F3N3O2 [M+H]+:
312.0960; found: 312.0903.
[7] G. S. Hammond, J. Am. Chem. Soc. 1955, 77, 334–338.
[8] a) K. Y. Chen, Y. M. Cheng, C. H. Lai, C. C. Hsu, M. L. Ho, G. H. Lee, P. T.
Chou, J. Am. Chem. Soc. 2007, 129, 4534–4535; b) C. C. Hsieh, P. T.
Chou, C. W. Shih, W. T. Chuang, M. W. Chung, J. Lee, T. Joo, J. Am. Chem.
Soc. 2011, 133, 2932–2943.
For details of the absorption and emission spectroscopic measure-
ments and computational methodology, please see the Supporting
Information.
[9] Y. H. Chen, W. J. Lo, K. S. Sung, J. Org. Chem. 2013, 78, 301–310.
[10] L. X. Wu, K. Burgess, J. Am. Chem. Soc. 2008, 130, 4089–4096.
[11] L. Munoz, M. E. Kavanagh, A. F. Phoa, B. Heng, N. Dzamko, E. J. Chen,
M. R. Doddareddy, G. J. Guillemin, M. Kassiou, Eur. J. Med. Chem. 2015,
95, 29–34.
Acknowledgements
P.T.C. thanks the Ministry of Science and Technology, Taiwan for
financial support.
[12] W. J. Lo, Y. H. Chen, K. S. Sung, J. Org. Chem. 2013, 78, 5925–5931.
[13] K. C. Tang, M. J. Chang, T. Y. Lin, H. A. Pan, T. C. Fang, K. Y. Chen, W. Y.
Hung, Y. H. Hsu, P. T. Chou, J. Am. Chem. Soc. 2011, 133, 17738–17745.
[14] P. Konold, C. K. Regmi, P. P. Chapagain, B. S. Gerstman, R. Jimenez, J.
Phys. Chem. A 2014, 118, 2940–2948.
[15] a) S. Nagaoka, U. Nagashima, N. Ohta, M. Fujita, T. Takemura, J. Phys.
Chem. 1988, 92, 166–171; b) S. Nagaoka, J. Kusunoki, T. Fujibuchi, S. Ha-
takenaka, K. Mukai, U. Nagashima, J. Photochem. Photobiol. A 1999, 122,
151–159.
Keywords: fluorescence · hydrogen bonds · kinetics · proton
transfer · thermodynamics
[1] A. Weller, Z. Elektrochem. 1956, 60, 1144–1147.
[2] a) P. T. Chou, W. S. Yu, Y. M. Cheng, S. C. Pu, Y. C. Yu, Y. C. Lin, C. H.
Huang, C. T. Chen, J. Phys. Chem. A 2004, 108, 6487–6498; b) Y. M.
Cheng, S. C. Pu, C. J. Hsu, C. H. Lai, P. T. Chou, ChemPhysChem 2006, 7,
1372–1381; c) A. P. Demchenko, K. C. Tang, P. T. Chou, Chem. Soc. Rev.
Received: May 21, 2016
Published online on && &&, 0000
&
&
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
8
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
FULL PAPER
&
Proton Transfer
Y.-A. Chen, F.-Y. Meng, Y.-H. Hsu,
C.-H. Hung, C.-L. Chen, K.-Y. Chung,
W.-F. Tang, W.-Y. Hung, P.-T. Chou*
&& – &&
Enhancing proton transfer: Upon in-
creasing the electron-withdrawing
strength of the R substituent in the
amino moiety of a series of amino-imi-
dazole derivatives, the rate of NRꢀH ex-
cited-state intramolecular proton trans-
fer (ESIPT) increases and the reaction
thermodynamics change from the en-
dergonic [R=H (4)] to the equilibrium
[R=COCH3 (5a) and COPh (5b)] and
then to the highly exergonic [R=Tosyl
(5c) and COCF3 (5d)], with the corre-
sponding tautomer emission spanning
from 580 to 640 nm (see figure).
NꢀH-Type Excited-State Proton
Transfer in Compounds Possessing
a Seven-Membered-Ring
Intramolecular Hydrogen Bond
Chem. Eur. J. 2016, 22, 1 – 9
www.chemeurj.org
9
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ