Understanding difficulties of irregular number-membered ring transition states for intramolecular proton transfer in excited state

Article abstract of DOI:10.1016/j.tet.2016.12.026

This study presents a variety of organic dyes with similar molecular structures that could undergo intramolecular proton transfer in excited states via five-, six- and seven-number-membered ring transition states, respectively. In addition, the dyes without proton transfer segments are also synthesized to use as references. X-ray single crystal diffraction, NMR spectra as well as UV/visible spectra suggests the presence of internal hydrogen bond with different strength in the target dyes. The steady and transient fluorescence measurements demonstrate occurrence of excited state intramolecular proton transfer via a six number-membered ring transition state. In contrast, it cannot be processed through five- and seven-number-membered ring transition states of the studied dyes. The molecular geometry optimization of the studied dyes reveals fundamental factors for the difficulties of intramolecular proton transfer in excited states via five- and seven-number-membered ring transition states.

Full text of DOI:10.1016/j.tet.2016.12.026

Accepted Manuscript
Understanding difficulties of irregular number-membered ring transition states for
intramolecular proton transfer in excited state
Xiaozhuan Qin, Ge Ding, Zhenqiang Wang, Yulong Gong, Fang Gao, Shengtao
Zhang, Ziping Luo, Hongru Li
PII:
S0040-4020(16)31304-7
10.1016/j.tet.2016.12.026
TET 28318
DOI:
Reference:
To appear in: Tetrahedron
Received Date: 9 September 2016
Revised Date: 12 November 2016
Accepted Date: 13 December 2016
Please cite this article as: Qin X, Ding G, Wang Z, Gong Y, Gao F, Zhang S, Luo Z, Li H, Understanding
difficulties of irregular number-membered ring transition states for intramolecular proton transfer in
excited state, Tetrahedron (2017), doi: 10.1016/j.tet.2016.12.026.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphic Abstract
200
175
150
125
100
75
400
(a)
C1
C2
C3
C1
C2
C3
in S
300
1
ESIPT emission
TS*
No, ESIPT
TS*
E*
K*
K*
200
TS* No, ESIPT
K*
Normal emission
E*
E*
100
0
Yes, ESIPT
50
400
450
500
550
600
650
0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
R
O-H
Wavelength (nm)

ACCEPTED MANUSCRIPT
Understanding difficulties of irregular number-membered ring transition states for
intramolecular proton transfer in excited state
Xiaozhuan Qin1, Ge Ding1, Zhenqiang Wang1, Yulong Gong1, Fang Gao1, 2*, Shengtao Zhang1, 2*, Ziping Luo1, 2*, Hongru Li1, 2*
1College of Chemistry and Chemical Engineering, Chongqing University, Chongqing, China, 400044
2National-municipal Joint Engineering Laboratory for Chemical Process Intensification and Reaction, Chongqing University, Chongqing,
China, 400044
Tel/Fax, 86-23-65102531, Emails, fanggao1971@gmail.com, stzhang@cqu.edu.cn, zpluo@126.com, hrli@cqu.edu.cn
ABSTRACT
This study presents a variety of organic dyes with similar molecular structures that could undergo
intramolecular proton transfer in excited states via five-, six- and seven- number-membered ring transition states
respectively. In addition, the dyes without proton transfer segments are also synthesized to use as references. X-ray
single crystal diffraction, NMR spectra as well as UV/visible spectra suggests the presence of internal hydrogen
bond with different strength in the target dyes. The steady and transient fluorescence measurements demonstrate
occurrence of excited state intramolecular proton transfer via a six number-membered ring transition state. In
contrast, it cannot be processed through five- and seven- number-membered ring transition states of the studied
dyes. The molecular geometry optimization of the studied dyes reveals fundamental factors for the difficulties of
intramolecular proton transfer in excited states via five- and seven- number-membered ring transition states
Keywords: Proton transfer; Excited state; Transition sate; Organic dye; Molecular geometry optimization
1. Introduction
.
As one of the significant phenomena in life and nature, excited state intramolecular proton
transfer (ESIPT) receives considerable experimental and theoretical interests.1-6 The four-level
cycle of enol and keto phototautomerization (E→E*→K*→K→E, E, enol, K, keto) is accepted as
1

ACCEPTED MANUSCRIPT
the most representative mechanism to achieve ultrafast (femtosecond or sub-femtosecond time
course) internal proton transfer in excited state of an organic molecule.7-11
Thus to an enol-keto ESIPT organic dye, there are two real chemical reactions in
phototautomerization process, namely E*-K* and K-E. Accordingly, potential energy barrier of
transition state for E*-K* phototautomerization of an organic dye plays crucial roles in occurrence
of ESIPT. This suggests that formation and stability of transition state are significant to achieve an
internal proton transfer in excited state of an organic dye.12 Based on fundamental viewpoint, a
tuned and well-defined barrier is of great importance for occurrence of enol-keto
phototautomerization reaction of an organic chromophore.
Formation and stability of transition state of enol-keto phototatumerization of an organic dye
is association with neighboring hydroxy and carbonyl or imino groups. Hence, transition state of
enol-keto phototatumerization could be determined by intramolecular hydrogen bond strength,
which is related to coplanarity as well as distance between proton donor and acceptor segments.
Owing to good stability, energy barriers of quasi six number-membered ring transition states
yielded by adjacent hydroxy and imino or carbonyl groups in organic dyes are normally small or
even barrierless. Hence, a number of organic dyes can undergo internal proton transfer in excited
states via quasi six number-membered ring transition states.13,14
In view of the chemistry, a five number-membered ring structure possesses a similar stability
as a six number-membered ring one. However, it is confused that only a few organic dyes were
demonstrated to undergo intramolecular proton transfer in excited states via quasi five
number-membered ring transition states.15,16 It is considered that weak internal hydrogen bond can
increase energy barriers of quasi five number-membered ring transition state of enol-keto
2

ACCEPTED MANUSCRIPT
phototautomerization.
It is further observed that fewer organic dyes undergo intramolecular proton transfer in
excited states via quasi seven number-membered ring transition states generated by neighboring
proton donor and acceptor groups in organic dyes.17-19 It is thought that energy barriers of
enol-keto phototautomerization of organic dyes via quasi seven number-membered ring transition
states can be increased by weak internal hydrogen bond.
Hence, enol-keto dye molecules may undergo ESIPT through six-membered ring transition
states. In contrast, ESIPT occurs rarely via five- or seven- number-membered ring transition states
of organic dyes. While so far, no studies have been carried out to reveal the difficulties of five-
and seven- number-membered ring transition states for intramolecular proton transfer in excited
states of organic dyes.
In this work, we present a series of organic dyes carrying proton donor and acceptor
segments with similar molecular skeletons, which could undergo intramolecular proton transfer in
excited states via different number-membered ring transition states. Meanwhile, the dyes lack of
proton donor and acceptor groups are prepared to use as references. To the best of our knowledge,
this is the first attempt to perform comprehensive experimental and theoretical studies to
understand the difficulties of irregular number-membered ring transition states for intramolecular
proton transfer in excited states of organic dyes.
2. Results and discussion
2.1. Internal hydrogen bond strength analyzed by X-single crystal diffraction analysis
Good single crystals of the target dyes were obtained through slow volatilization. Figure 1
shows the presence of intramolecular hydrogen bond between the adjacent phenolic hydroxy and
3

ACCEPTED MANUSCRIPT
imino groups in C1, C2 and C3 respectively. Internal hydrogen bond of these molecules was
compared through analysis of dihedral angles and distances between hydroxy and imino groups
based on X-single crystal diffraction.
Figure 1 suggests that C1 and C3 possess the longer H1-N1 (2.156 Å for C1, 1.844 Å for C2,
1.886 Å for C3) and N1-O1 (2.637 Å for C1, 2.572 Å for C2, 2.652 Å for C3) as well as the
larger dihedral angles of ꢀC-O1-H1-N1 (ꢀC2-O1-H1-N1 for C1, 2.96˚; ꢀC3-O1-H1-N1 for C2,
0.62˚; ꢀC8-O1-H1-N1 for C3, 10.71˚) in phenolic hydroxy and imino groups than C2. This
indicates that there is a stronger internal hydrogen bond in C2 comparing to that in C1 and C3
respectively.
4

ACCEPTED MANUSCRIPT
Figure 1 ORTEP drawing of C1, C2 and C3 with 30% possibility of thermal ellipsoids
Hence, the formation of irregular number-membered ring transition states in C1 and C3
could be harder than that of six number-membered one in C2. Furthermore, the stability of six
number-membered ring transition state in C2 could be greater than that of five- and seven-
number-membered ones in C1 and C3 respectively. Thus, it is deduced that the energy barriers of
five- and seven- number-membered ring transition states in C1 and C3 can be much larger than
six number-membered one in C2 respectively. As a consequence, it is more difficult to undergo
intramolecular proton transfer through five- and seven- number-membered ring of transition states
in C1 and C3 respectively.
X-ray single crystal diffraction analysis also demonstrates that C=N double band in C2
shows the greater coplanarity than C1 and C3 respectively, which means the presence of the
greater steric hindrances between the adjacent hydroxy and imino groups in C1 and C3. It is found
that the dihedral angle of ꢀC8-N1-C1-C2 in C2 is 0.65˚, while the dihedral angle of ꢀ
C1-N1-C7-C8 in C1 is 1.59˚, and the dihedral angle of ꢀC1-N1-C13-C14 in C3 increases to
2.34˚.
2.2. Intramolecular hydrogen bond based on analysis of 1H-NMR spectra
1
It is known that hydrogen bond leads to a downshift of H-NMR peak of a bridge hydrogen
5

ACCEPTED MANUSCRIPT
atom due to deshielding effect.20 1H-NMR chemical shifts of phenolic hydroxy group in the targets
move to the lower magnetic field comparing to those in the references (see Figures S1 in
1
Supplementary materials, such as C1, 8.659 ppm, C7, 8.187 ppm). In addition, H-NMR
chemical shifts of imino groups in the references are located at the higher magnetic field
comparing to those in target molecules (such as C1, 8.489 ppm and C4, 8.257 ppm, Figures S1).
Hence, 1H-NMR spectra further suggest that there is presence of intramolecular hydrogen bond in
the target molecules.
1H-NMR peak of phenolic hydroxy group in C2 shifts to the lower magnetic field than that of
C1 and C3 respectively (such as C1, 8.659 ppm and C3, 10.847 ppm versus C2, 13.634 ppm,
Figure S1). Meanwhile, 1H-NMR peak values of imino group in C1 and C3 are lower than that of
C2 respectively (such as C1, 8.489 ppm and C3, 8.402 ppm versus C2, 8.889 ppm, Figure S1).
1
Hence, H-NMR spectra further suggest that intramolecular hydrogen bond between adjacent
hydroxy and imino groups in C1 and C3 is much weaker than that in C2.
2.3. Intramolecular hydrogen bond effect on ultraviolet/visible spectra
The ultraviolet/visible spectra of the target and reference molecules were studied in various
organic solvents. The typical UV/visible spectra of target molecules C1, C2 and C3 in 1,4-dioxane
and DMF are shown in Figure 2. The representative UV/visible spectral parameters are presented
in Table 1.
It is found that the target dyes C1, C2 and C3 show similar ultraviolet/visible spectral shapes
to the methylated references, which exhibit long-wavelength absorption bands in 350~500 nm
region (also see Figure S2 & Table S1). In contrast, the reduced references are absence of such
long-wavelength absorption bands (Figure S2). The results suggest that there is presence of
6

ACCEPTED MANUSCRIPT
internal charge transfer in ground states of the target dyes and the methylated references, while the
reduced references are deficient in intramolecular charge transfer due to lack of
electron-withdrawing imino group.
0.6
0.4
0.2
0.0
0.6
0.4
0.2
0.0
(a)
(b)
C1
C2
C3
C1
C2
C3
300
350
400
450
500
300
350
400
450
500
Wavelength(nm)
Wavelength (nm)
Figure 2 UV/visible absorption spectra of C1, C2 and C3 in 1,4-dioxane(a) and in DMF (b), the concentration is
2×10-5 mol/L
Table 1 The absorption and emission spectral parameters of C1, C2 and C3 in various solvents, λa,max: the
absorption maximum (nm), λf,max: the emission maximum (nm), εmax: the maximal molar extinction coefficient
(105●cm-1●mol-1●L), Φ: the fluorescence quantum yield
C1
C2
C3
Solvents
λa,max λf,max
Φ
εmax
λa,max
λf,max
Φ
εmax
λa,max λf,max
Φ
εmax
Benzene
1,4-dioxane
THF
379
372
375
373
380
379
374
370
372
433
411
424
413
411
413
422
411
413
0.0671 0.333
0.0321 0.210
0.0772 0.254
0.0191 0.288
0.0339 0.295
0.0328 0.356
0.0668 0.318
0.0797 0.197
0.0647 0.212
388
384
386
379
379
387
384
392
389
433, 527 0.0490 0.286
406, 531 0.0971 0.176
434, 531 0.0324 0.247
435, 526 0.0589 0.106
430, 535 0.0873 0.221
433, 533 0.0739 0.190
436, 542 0.0668 0.240
433, 548 0.0913 0.171
426, 535 0.0847 0.217
372
373
377
373
378
375
378
381
360
426
401
434
434
414
418
434
432
433
0.0835 0.348
0.0560 0.259
0.0528 0.356
0.0484 0.262
0.0457 0.208
0.0367 0.189
0.0189 0.281
0.0366 0.227
0.0704 0.243
EtOAc
CH2Cl2
CHCl3
MeCN
DMSO
DMF
It is further observed that the maximal absorption wavelengths of targets C1, C2 and C3 are
a little red-shifted with respect to those of the methylated references (Figure S2). This means that
intramolecular hydrogen bond in the target molecules enhances extent of intramolecular charge
7

ACCEPTED MANUSCRIPT
transfer in ground states. In addition, the absorption maxima of C1, C2 and C3 are basically in the
order as C2>C3>C1 in various solvents, reflecting the consequence of intramolecular hydrogen
bond strength.
2.4. Steady fluorescence emission spectra of the target dyes
Presence of intramolecular hydrogen bond effect in ground state of the target molecules
means a large possibility to undergo ESIPT.21 In order to get the direct experimental evidence of
internal proton transfer in excited state, fluorescence emission spectra of the target dyes and the
reference molecules were investigated in various organic solvents. The representative fluorescence
spectra of C1, C2 and C3 in 1,4-dioxane and DMF are given in Figure 3. The typical fluorescence
spectral parameters are depicted in Table 1.
500
400
300
200
100
0
400
300
200
100
0
(b)
(a)
C1
C2
C3
C1
C2
C3
ESIPT emission
Normal emission
ESIPT emission
Normal emission
400
450
500
550
600
650
400
450
500
550
600
650
Wavelength (nm)
Wavelength (nm)
Figure 3 Steady emission spectra of C1, C2 and C3 in 1,4-dioxane(a) and in DMF (b), the concentration is
5×10-6 mol/L for the emission spectra, Excitation at 350 nm
It is observed that the targets C1 and C3 show single emission band in various solvents,
which is almost identical to the corresponding reference molecules such as C4 and C6 (Figure
S3). The emission maxima of C1 and C3 and the corresponding methylated references are located
around 430 nm (Table S1), which are red-shifted to those of the reduced references such as C7
and C9 due to more intensive internal charge transfer (Figure S3).
8

ACCEPTED MANUSCRIPT
It is noticed that C2 exhibits dual-emission bands in various solvents. While in contrast, the
reference molecules C5 and C8 display single emission band (Figure S3). It is shown that the
emission maxima of the first emission band of C1 is close to the maximal emission wavelengths
of the reference molecules as well as the other target molecules (Table S1).
It is found that the maximal emission wavelengths of C2 are much red-shifted with respect to
those of the first emission bands (426 nm versus 535 nm in DMF). As a consequence, the
fluorescence emission of the first emission band of C2 possess the normal Stokes-shift (~37 nm in
DMF), while the second fluorescence band of C2 is characterized with the large Stokes-shift
(~146 nm in DMF). Therefore, the results show that the fluorescence emission of C1 and C3 is
yielded by the excited normal enol decay, which is effectively the same as S0→S1 enol absorption
spectral feature. However, the second emission band of C2 is produced by the decay of the excited
keto.22 The results demonstrate that C1 and C3 cannot process internal proton transfer in excited
state through five- and seven- number-membered ring transition states, while C2 undergoes ESIPT
through a six- number-membered ring transition state. This could be ascribed to the stronger
internal hydrogen bond in C2 comparing to that in C1 and C3.
It is further found that the target dyes show similar fluorescence spectral properties in solid
state or in polymer matrix as those in organic solvents respectively in Figure S4. C2 exhibits
strong ESIPT emission, while the solids C1 and C3 show only normal enol emission. The results
further demonstrate that it is more easily to undergo ESIPT through a six- number-membered ring
transition state than irregular number-membered ring ones in various environments.
It is necessary to point out that although the reported benzimidazole-phenol derivatives can
undergo ESIPT via a six number-membered ring transition state, the most of emission maxima of
9

ACCEPTED MANUSCRIPT
keto tautomers are below 500 nm in various states 23-25, and only a few efforts have been made to
achieve above 500 nm keto emission successfully26, 27. It is noticed that phenolic hydroxy group
1
in benzimidazole-phenol derivatives normally shows a higher magnetic field H-NMR chemical
shift (~ 11 ppm), and thus a weaker hydrogen bond could be a detriment to its ESIPT process,
leading to a hyperchromic shift of keto emission maxima due to the influence of competitive
normal (π, π*) decay. While free C=N bond in C2 could offer a greater diversity than restricted
C=N bond in benzimidazole-phenol derivatives, which increases intramolecular hydrogen bond
strength between O-H and C=N groups because it is not limited toward a specific direction. As a
consequence, a six number-ring transition state could be yielded and its potential energy barrier of
E* to K* could be diminished, resulting in increasing ESIPT maximum.
2.5. Time-resolved fluorescence emission spectra
Phototautomers enol and keto forms of C2 can be well distinguished from each other in the
time-resolved fluorescence spectra obtained by the decay kinetics basing on the dependence of
emission wavelength. Figure 4 shows that the maximal transient tautomer enol emissive peak of
C2 in DMF is yielded at 1.5×10-11 s, while the intensity decreases with the decay time course.
Meanwhile, the maximal transient keto peak is found at 3.5×10-11 s decay time range. After then,
the both of transient products show the decrease emission with decay. Thus, time-resolved
fluorescence emission profiles of C2 further demonstrate intramolecular proton transfer in excited
state. In contrast, there are only transient tautomer enol emissive peaks for C1 and C3, suggesting
the absence of intramolecular proton transfer in excited states of C1 and C3 (Figure S5).
10

ACCEPTED MANUSCRIPT
Figure 4 Time-resolved fluorescence spectra of C2 in DMF, the concentration is 5×10-5 mol/L
It is observed that the fluorescence lifetime of the first emission band of C2 is close to that of
the other targets C1 and C3 respectively (such as in 1,4-dioxane, C1, 3.71×10-11 s at 420 nm, C2,
3.62×10-11 s at 416 nm, C3, 3.86×10-11 s at 413 nm). The methylated references also yield the
similar fluorescence lifetimes to those of the first emission bands of C2 (such as in 1,4-dioxane,
C4, 3.53×10-11 s at 413 nm). The results further demonstrate that the excited enol form accounts
for the first emission band of C2 in various solvents. It is further found that the second emission
band of C2 show much shorter fluorescence lifetime in various solvents (such as in 1,4-dioxane,
2.11×10-12 s at 550 nm), indicating that the excited keto form could have a more competitive
radiative transition than the excited enol form.
The radiative and non-radiative transition rates in excited state of a dye can be calculated by
the following equations:
kr=Φf/τf
(1)
(2)
knr=(1-Φf)/τf
11

ACCEPTED MANUSCRIPT
wherein Φf is fluorescence quantum yield of a dye and τf shows lifetime determined at the
maximal emission wavelength of fluorescence band of a dye.
Since the fluorescence emission of C2 includes the enol fluorescence band and the keto
fluorescence band, the fluorescence quantum yield of C2 can be regarded as the sum of the normal
fluorescence quantum yield and ESIPT fluorescence quantum yield, as shown in the following
equation:
Φf (sum) = Φf (enol) + Φf (keto)
(3)
Because the absorption of C2 is assigned to tautomer enol to enol*, the ratio of the integrated
area of the normal emission band or the ESIPT emission band to that of the total emission
represents the percent of the normal emission band or the ESIPT emission band to the total
fluorescence emission respectively. Consequently, the normal fluorescence quantum yield and
ESIPT fluorescence quantum yield of C2 can be calculated respectively. Accordingly, the radiative
and non-radiative transition rates of the tautomers enol decay and keto decay in excited state of C2
can be calculated.
The radiative and non-radiatibe transition rates of C1, C2 and C3 are give in Table 2. The
data suggest that ESIPT emission band of C2 exhibit larger radiative transition rates than the
normal emission band in various organic solvents (such as in DMF, C2, kr(keto), 3.86×1010 s-1, kr(enol)
3.65×109 s-1 ), suggesting that the decay of excited keto form of C2 is strong competitive in
excited state. It is also found that the nonradiative transition rates of C1 and C3 are lower than
those of the normal emission band of C2, which indicates that internal proton transfer decrease the
transition decay of the excited enol form.
,
12

ACCEPTED MANUSCRIPT
Table 2 Radiative transition rates (s-1) of the tautomers enol and keto decay in excited state of C1, C2 and C3
Solvents
Dyes
1,4-dioxane
0.0321
3.71
DMF
0.147
1.65
Φf (enol)
τf (enol) × 10-11 (s)
C1
C2
C3
k
r (enol) ×1010 (s-1)
knr(enol) ×1011 (s-1)
0.0865
0.261
0.891
0.517
0.0500
1.37
Φf (enol)
0.0238
3.62
τf (enol) × 10-11 (s)
k
r (enol) ×1010 (s-1)
knr (enol) ×1011 (s-1)
0.0657
0.270
0.365
0.693
0.0347
0.901
0.386
Φf (keto)
0.0733
2.11
τf(keto) ×10-12 (s)
k
r (keto) × 1011 (s-1)
0.347
knr(keto) ×1012 (s-1)
0.439
1.07
Φf (enol)
0.0560
3.86
0.0704
1.76
τf (enol) × 10-11 (s)
kr (enol) ×1010 (s-1)
knr(enol) × 1011 (s-1)
0. 145
0.245
0.401
0.528
2.6. Molecular geometry optimization
2.6.1. Geometrical parameters
The theoretical calculation was performed to understand the experimental results. The
optimized structures of enol, keto and transition state (TS) forms in excited states of C1, C2 and
C3 are shown in Figure S6, and the corresponding structural parameters are listed in Table 3. It is
shown that TS form of enol-keto phototautomerization in excited state of C1 cannot be obtained,
which means that energy keeps either going up or down all the time during E-K
phototautomerization.
As we expected, the distances of H-O in TS of enol-keto tautomerization of these molecules
are between those of tautomers in S0 or S1 states (such as H-O in S1 of C2, E*, 0.945 Å, TS, 1.214
Å, K, 1.846 Å, Table 3). Furthermore, the distances of O-N in TS forms are smaller than those of
E and K forms in S0 or S1 states (such as O-N in S1 of C2, E*, 2.822Å, TS, 2.395Å, K, 2.668 Å,
13

ACCEPTED MANUSCRIPT
indicating that TS forms of these molecules possess the more intensive internal hydrogen bond
than enol and keto forms.
Table 3 The most important distances association with ESIPT calcualted at HF level for ground state and at CIS
level for excited state to C1, C2 and C3 respectively
Ground state
TS
Excited state
TS*
Structural parameters
Distances (Å)
E
K
E*
K*
RO-H
0.947
2.156
2.693
0.954
1.886
2.711
0.949
3.919
4.264
1.374
1.158
2.342
1.273
1.184
2.385
1.885
1.018
2.539
1.868
1.005
2.656
0.950
2.131
2.666
0.945
2.053
2.822
0.948
2.990
3.848
2.045
0.999
2.576
1.846
1.008
2.668
1.625
1.015
2.597
C1
C2
C3
RN-H
RO-N
RO-H
RN-H
RO-N
RO-H
RN-H
RO-N
1.214
1.240
2.395
1.311
1.323
2.417
As compared with C1 and C3, the smallest N-H in ground and excited states of C2 suggests
that there is the strongest hydrogen bond in C2 as well as the shortest tunnel of intramoleular
proton transfer, which is thus favorable for ESIPT to C2. As shown in Table 4, the important
dihedral angles of ꢀC-H-O-N as well as ꢀC-C-N-C in o-hydroxy-phenyl-imino part of the
target dyes are acquired. As compared with C1 and C3, the greatest coplanar excited TS form of
C2 shows that there is the largest possibility to undergo enol-keto phototautomerization in excited
state of C2 (such asꢀC1-O1-H1-N1 in C2, 0.307 ˚, ꢀC1-O1-H1-N1 in C3, 32.212 ˚). In contrast,
TS form in the excited state of C3 shows the poor coplanarity, which means that it is hard for C3
to process intramolecular proton transfer in excited state.
Table 4 The most important dihedral angle calcuted at HF level for ground state and at CIS level for excited state
to C1, C2 and C3 respectively
14

ACCEPTED MANUSCRIPT
1
H
1
1
1
H
1
1
O
1
H
N
N
N
O
1
O
1
N
3
2
3
1
N
1
4
N
1
4
6
2
4
3
5
2
5
5
7
C1
C2
C3
Ground state
TS
Excited state
TS*
Structural Parameters
Angle (○)
E
K
E*
0.082
K*
ꢀC1-O1-H1-N1
ꢀN1-C3-C4-C5
ꢀC1-C2-N1-C3
ꢀC1-O1-H1-N1
ꢀC3-N1-C4-C5
-5.215
-5.051
29.812
-0.019
41.933
0.012
-0.060
-0.022
0.830
0.001
-0.003
0.002
2.780
35.188
0.006
0.001
0.001
0.103
0.505
C1
C2
-5.547
0.710
30.299
0.361
0.307
0.345
0.046
32.212
6.621
2.899
35.846
ꢀC1-C2-N1-C3
ꢀC1-O1-H1-N1
ꢀC3-C4-N1-C5
ꢀN1-C5-C6-C7
0.800
-44.305
-60.857
0.112
-0.393
-0.101
-7.100
20.325
14.692
5.407
-0.067
49.303
7.584
1.912
C3
2.6.2. Energy barriers of enol-ketotautomerization of in S0 and S1 states
Since enol-keto tautomerization is a typical organic chemical reaction, the potential energy
curves (PEC) of intramolecular proton transfer of the target molecules are drawn considerable
attentions. Hence, the PEC for ESIPT of the studied dyes was calculated by specifying a reaction
coordinate of the distance from O atom to H atom. Figure 5 and Figure S7 show the plots of the
potential energy curves in S0 and S1 states of the molecules C1, C2 and C3 respectively.
Firstly, the instability of the K form in ground state of C3 indicates the inviability of ground
state intramolecular proton transfer. As shown in Figure 5, the energy barrier of E-K in excited
state of C3 is too large (~50 kcal/mol) and the energy of the keto form is much higher than that of
enol form, suggesting the great impossibility of intramolecular proton transfer in excited state of
C3.
It is further found that there is absence of TS form in excited state of C1 and K form is more
unstable in excited state in C1. The energy of E-K in excited state of C1 keeps increasing, which
shows ESIPT (E→E*→K*→K→E) cannot occur to C1. In contrast, the energy of K form in
15

ACCEPTED MANUSCRIPT
excited state of C2 is much lower than that of enol form, suggesting K form is more stable in
excited state of C2. Furthermore, a quite small energy barrier of E-K in excited state of C2 can be
easily overcome (~5 kcal/mol). Hence, intramolecular proton transfer in excited state of C2 is
much more possible. It is interesting that C2 has a smaller reversible energy barrier (K→E) than
C1 and C3 in ground state (see Figure S7), which further suggests that C2 possesses a greater
ability to achieve ESIPT four-level cycle process.
200
C1 ES
C2 ES
C3 ES
175
150
125
100
75
TS*
E*
K*
K*
TS*
E*
E*
K*
50
0.8 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4
R
O-H
Figure 5 Proton transfer reaction potential energy curves relaxed along the O-H distance at the first excited single
state for C1, C2 and C3
2.6.3. Frontier orbitals, Mulliken atomic charge and dipole moment
Since ESIPT occurs in excited state, we particularly concentrate on frontier orbitals of the
excited state enol tautomers. It is noticed that the electron cloud in mainly distributed at the
N,N-dimethyl-phenyl part in HOMO of excited C2 (Figure S8), while it is mainly located at
phenyl-imino part in LUMO. This means that phenyl-imino part could be greatly active in the
excited LUMO of enol form of C2
In contrast, there is presence of major electron cloud density distribution in phenyl-imino part
in HOMOs of excited enol forms of C1 and C3, while it is mainly distributed in
16

ACCEPTED MANUSCRIPT
N,N-dimethyl-phenyl part in LUMOs. The deficiency of electron cloud density distribution in
phenyl-imino part in LUMOs of excited enol forms of C1 and C3 suggests that phenyl-imino part
could be inactive in excited LUMO of enol forms of C1 and C3. Hence, the frontier orbital nature
of the excited enol tautomers shows that C2 has a more possibility to process internal proton
transfer in excited state than C1 and C3.
Table 5 Mulliken net atomic charges of the key atoms of the molecules C1, C2 and C3
E
Mulliken Charge (e)
C
O
H
N
C1
C2
C3
C1
C2
C3
0.381
0.428
0.386
0.342
0.447
0.384
-0.668
-0.683
-0.674
-0.671
-0.653
-0.698
0.369
0.402
0.355
0.373
0.387
0.378
-0.636
-0.682
-0.579
-0.793
-0.782
-0.796
Ground State
Excited State
It is further found that as E of C2 is excited to E*, oxygen atom of phenolic group shows a
decrease of negative charge, and hydrogen atom of phenolic group exhibits a decrease of positive
charge as well (Table 5, -0.683 e to -0.653 e for O atom, 0.402 e to 0.387 e for H atom). While, an
opposite tendency is found during E to E* transition of C1 and C3 (such as to C3, -0.674 e to
-0.698 e for O atom, 0.355 e to 0.378 e for H atom). The results indicate that phenolic group in C2
is more loose and active than that of C1 and C3. As a result, the proton in phenolic group of C2 is
more reactive accordingly.
It is noticed that the tautomers of C2 show a quite lower dipole momenet change from the
ground state to the excited state than those of C1 and C3 (Table 6, such as enol form of C2, 2.925
D to 3.742 D; such as enol form of C3, 3.052 D to 10.596 D), which indicates that it requires a
smaller energy to undergo such diople moment change for C2 as compared with C1 and C3. The
results are consistent with the above energy barrier calculation.
17

ACCEPTED MANUSCRIPT
Table 6 Dipole moment of each tautomer calculated at HF level for ground state and at CIS level for excited state
to C1, C2 and C3 respectively
Ground state
Excited state
Dipole moment (D)
μE
μTS
μK
μE
μTS
μK
C1
C2
C3
4.480 9.551 11.631
8.573
3.742
12.607
5.576
2.925 4.565
3.052
4.971
4.713
10.596 17.429 21.172
3. Conclusions
To be closing, this study aims to deal with a long-staying chemical question why it is hard to
undergo internal proton transfer in excited state through irregular number-membered ring
transition states. For this purpose, a range of organic dyes carrying similar chemical structures that
could undergo intramolecular proton transfer in excited states via different number-membered ring
transition states respectively. The different strength of internal hydrogen bond in the studied dyes
is demonstrated by X-ray single crystal diffraction, NMR spectra as well as UV/visible spectra.
The intramolecular proton transfer in excited state of the studied dyes is strongly competitive
through a six number-membered ring transition state, while no experimental evidence support the
occurrence of ESIPT via five- and seven- number-membered ring transition states of the studied
dyes respectively. The optimized geometric parameters association with internal proton transfer
such as the distance between two atoms and dihedral angles of proton transfer segments suggest
that it is more difficult to undergo ESIPT via irregular five- and seven- number-membered ring
transition states of the studied dyes. The other calculated results including energy barriers of
enol-keto phototautomerization of and frontier orbital properties in S1 states of the studied dyes as
well as Mulliken atomic charge support such conclusions as well. Therefore, in order to achieve
intramolecular proton transfer in excited state through irregular number-membered ring transition
18

ACCEPTED MANUSCRIPT
states, these disadvantages much be minimized. The results presented in this study would be
greatly beneficial for the synthesis of new organic chromophores undergoing ESIPT via different
number-membered ring transition states.
4. Experimental
4.1. Reagents and characterization
The organic solvents were supplied by Aldrich Corporation, which were further processed by
standard laboratory methods.28 The target molecules were synthesized in our own laboratory
(Scheme S1, Supplementary materials), in which C3, C6 and C9 were firstly reported in this
study.
Nuclear magnetic resonance (NMR) spectra of the samples were determined by Bruker 400
MHz, 500 MHz and 600 MHz apparatus in standard NMR tubes at room temperature. The
chemical shifts of hydrogen and carbon atoms in NMR spectra were identified by using
tetramethylsilane (TMS) as an internal reference. Nicolet 550II Fourier transform infrared
spectrometer was employed to measure IR spectra of the samples (KBr pellet method). A CE440
elemental analysis meter from Exeter Analytical Inc was used to conduct elemental analysis of the
samples. High resolution mass spectra (HRMS) electro spray ionization (ESI) was conducted by a
UPLC-Q-ToF MS spectrometer. A Beijing Fukai melting point apparatus was employed to
determine the melting points of the samples.
Slow volatilization of organic solvents in NMR tubes to obtain single crystals of the samples.
A suitable single crystal was put inside a glass fiber capillary to determine molecular structure by
X-ray diffraction, which was analyzed by the direct methods and refined by full-matrix least
squares on F2. XRD data were acquired by a Bruker-AXS CCD area detector equipped with
19

ACCEPTED MANUSCRIPT
diffractometer with Mo Ka (λ = 0.71073 Å) at 298 K. All the hydrogen atoms were identified in
the calculated positions respectively and all the non-hydrogen atoms were analyzed by anisotropic
temperature factors.
4.2. UV/visible absorption and fluorescence emission spectral determination
Spectral grade organic solvents were utilized for spectroscopic measurement of the samples.
The ultraviolet/visible absorption spectra of the samples were recorded by a TU1901
spectrophotometer from Beijing PUXI General Equipment Limited Corporation. The steady
fluorescence emission spectra were determined by Shimadzu RF-531PC spectrofluorophotonmeter.
The fluorescence quantum yields of the samples were measured by using Quinine sulfate in 0.5
mol/L H2SO4 (Φ, 0.546) as the reference29 by the equation (4)30:
n02 A0 If (λf )dλf
∫
Фf = Фf0
(4)
n2 A If0 (λf )dλf
∫
wherein n0 and n represent the refractive indices of the solvents for the reference and sample
respectively, A0 and A are the absorption intensities at excitation wavelength, Φf and Φf0 show the
fluorescence quantum yields of the sample and reference, and the integrals are the areas of the
fluorescence emission bands for the reference and sample respectively. The absorption of the
sample at excited wavelength is guaranteed below 0.1 to minimize experimental errors. The
sample in organic solvents was deaerated by bubbling argon for half of an hour before the
measurement of fluorescence spectra.
The fluorescence lifetimes of the samples were measured by a laser system. The pump laser
pulses were emitted by the amplified frequency doubling directly. Time-dependent laser detection
was obtained through a spectrograph using a connected intensified double photodiode array. The
pulses were focused onto a sample cuvette containing 100 µm thick fused-silica window and a 250
20

ACCEPTED MANUSCRIPT
pm optical path length. The transient spectra were recorded by a ultrafast white-light continuum.
4.3. Geometry optimization computation
Geometry computation of the samples was by 6-31G** basis set in Gaussian 09 program
package. The organic solvents were used as the media in theoretical computation. Geometry
optimization of enol and keto of phototautomers in ground states (S0) was conducted by B3LYP
method based on density functional theory (DFT) and HF (Hartree-Fock) level respectively,31
while geometries in the first singlet excited states (S1) of the molecules was optimized by CIS
method (single-excitation configuration interaction). The energies of optimized geometries of S0
and S1 states were computed by DFT and TD-DFT (time-dependence DFT) method respectively
such as DFT//DFT and DFT//HF (for S0) and TDDFT//CIS (for S1).32
4.4. Synthesis of the target dyes
The synthesis routes of the target and reference dyes are shown in Scheme 1. The preparation
of precursors 2'-hydroxybiphenyl-2-amine and 2'-methoxybiphenyl-2-amine was performed by
condensation of 2-hydroxyphenylboronic acid or 2-methoxybenzeneboronic acid and 2-bromo
aniline with over 80% isolated yields. It is noticed that quite few studies involving free aryl
amines Suzuki coupling reactions have ever been reported.
The synthesis and characterization of the studied molecules are shown in Supplementary
materials.
Acknowledgements
The authors greatly appreciate Chongqing Natural Science Foundation CSTC2012jjB50007
and CSTC2010BB0216.H. Li is grateful to the Postdoctoral Science Foundation of China for
21

ACCEPTED MANUSCRIPT
financial supporting (Grant No. 22012T50762 &2011M501388). This work was also supported by
the Graduate Student Innovation Foundation of Chongqing University (CDJX11131146).
Supplementary materials
The synthesis and characterization of the major products including NMR data, IR data,
high-resolution MS data as well as elementary analysis of C1~C9 are included in Supplementary
materials. The absorption spectra and steady and transient fluorescence emission spectra of the
studied dyes in various media are also provided in Supplementary materials. The optimized
geometries of enol, TS and keto forms in excited state of the target dyes, the spectral parameters
of the dyes in various solutions and the crystallographic data are further contained in the
Supplementary materials. The document is available free of charge from website of this journal.
The supplementary crystallographic data for this study can be found in file numbered by
CCDC 1502865-1502867, which are available http://www.ccdc.cam.ac.uk/conts/retrieving.html or
from the Cambridge crystallographic data center, 12, Union Road, Cambridge CB2 1EZ, UK; fax:
44-1223-336033.
References and notes
1. Suzuki, N.; Fukazawa, A.; Nagura, K.; Saito, S.; Kitoh-Nishioka, H.; Yokogawa, D.; Irle, S.;
Yamaguchi, S. Angew. Chem. Int. Ed. 2014, 53, 8231-8235.
2. Mutai, T.; Sawatani, H.; Shida, T.; Shono, H.; Araki, K. J. Org. Chem. 2013, 78, 2482-2489.
3. Kim, J.; Heo, W.; Joo, T. J. Phys. Chem. B 2014, 119(6), 2620-2627.
4. Zhang, Z. Q.; Yang, C. T.; Liang, L. J.; Xiao, B.; Lu, X.; Liu, J. H.; Sun, Y. Y.; Marder, T. B.; Fu,
Y. Org. Lett. 2014, 16, 6342-6345.
5. (a) Stasyuk, A. J.; Banasiewicz, M.; Cyrański, M. K.; Gryko, D. T. J. Org. Chem. 2012, 77,
22

ACCEPTED MANUSCRIPT
5552-5558. (b) Gargas, D. J.; Gao, H. W.; Wang, H. T.; Yang, P. D. Nano Lett. 2011, 11,
3792-3796.
6. (a) Xie, L. J.; Chen, Y. H.; Wu, W. T.; Guo, H. M.; Zhao, J. Z.; Yu, X. R. Dyes & Pigments 2012,
92, 1361-1369. (b) Cui, G. L.; Lan, Z. G.; Thiel, W. J. Am. Chem. Soc. 2012, 134, 1662-1672.
7. Zhao, J. Z.; Ji, S. M.; Chen, Y. H.; Guo, H. M.; Yang, P. Phys. Chem. Chem. Phys. 2012, 14(25),
8803-8817.
8. Zhang, W.; Yan, Y. L.; Gu, J. M.; Yao, J. N.; Zhao, Y. S. Angew. Chem. Int. Ed. 2015, 54(24),
7125-7129.
9. Gong, Y. L.; Lu, Y.; Wang, Z. Q.; Zhang, S. T.; Luo, Z. P.; Li, H. R.; Gao, F. Tetrahedron Lett.
2015, 56, 236-242.
10. Kim, D.; Kwon, J. E.; Park, S. Y. Adv. Optical Mater. 2016, 4(5), 790-797.
11. Ciuciu, A. I.; Skonieczny, K.; Koszelewski, D.; Gryko, D. T.; Flamigni, L. J. Phys. Chem.
C, 2013, 117(2), 791-803.
12. (a) Gong, Y. L.; Lu, Y.; Ma, H. Y.; Ding G.; Zhang, S. T.; Luo, Z. P.; Li, H. R.; Gao, F. Chem.
Phys. Lett. 2015, 619, 201-207. (b) Ding G.; Lu, Y.; Gong, Y. L.; Ma, L.; Luo, Z. P.; Zhang, S. T.;
Gao, F.; Li, H. R. Tetrahedron 2016, 72, 3040-3056.
13. Jang, D. J.; Kelley, D. F. J. Phys. Chem. 1985, 89(2), 209-211.
14. Bardez, E.; Devol, I.; Larrey, B.; Valeur, B. J. Phys. Chem. B 1997, 101, 7786-7793.
15. Yu, W. S.; Cheng, C. C.; Cheng, Y. M.; Wu, P. C.; Song, Y. H.; Chi, Y.; Chou, P. T. J. Am.
Chem. Soc. 2003, 125, 10800-100801.
16. Singh, R. B.; Mahanta, S.; Guchhait, N. Chem. Phys. 2007, 331(2), 189-199.
17. Guo, Z. Q.; Chen, W. Q.; Duan, X. M. Dyes & Pigments 2012, 92(1), 619-625.
23

ACCEPTED MANUSCRIPT
18. Zhang, Z.; Hsu, Y. H.; Chen, Y. A.; Chen, C. L.; Lin, T. C.; Shen, J. Y.; Chou, P. T. Chem.
Commun. 2014, 50(95), 15026-15029.
19. Chen, K. Y.; Cheng, Y. M.; Lai, C. H.; Hsu, C. C.; Ho, M. L.; Lee, G. H.; Chou, P. T. J. Am.
Chem. Soc. 2007, 129(15), 4534-4535.
20. (a) Casanovas, J.; Namba, A. M.; León, S.; Aquino, G. L.; José da Silva, G. V.; Alemán, C. J.
Org. Chem. 2001, 66, 3775-3782. (b) Wagner, G.; Pardi, A.; Wuethrich, K. J. Am. Chem. Soc. 1983,
105, 5948-5949.
21. (a) Parada, G. A.; Markle, T. F.; Glover, S. D.; Hammarström, L.; Ott, S.; Zietz, B. Chem.-A
Eur. J. 2015, 21, 6362-6366. (b) Tseng, H. W.; Liu, J. Q.; Chen, Y. A.; Chao, C. M.; Liu, K. M.;
Chen, C. L.; Lin, T. C. J. Phys. Chem. Lett. 2015, 6, 1477-1486.
22. Ushuka, B.; Stupaka, A. J. Lumin. 2003, 105, 81-87; (b) Doroshenko, A. O.; Posokhov, E. A.;
Verezubova, A. A.; Ptyagina, L. M. J. Phys. Org. Chem. 2000, 13, 253-255.
23. Park, S.; Kwon, O. H.; Kim, S.; Park, S.; Choi, M. G.; Cha, M.; Park, S. Y.; Jang, D. J. J. Am.
Chem. Soc. 2005, 127(28), 10070-10074.
24. Takagi, K.; Sugihara, K.; Isomura, T. J. Polym. Sci. Part A: Polym. Chem. 2009, 47(18),
4822-4829.
25. Shaikh, M.; Dutta Choudhury, S.; Mohanty, J.; Bhasikuttan, A. C.; Nau, W. M.; Pal, H.
Chem.-A Eur. J. 2009, 15(45), 12362-12370.
26. Padalkar V. S.; Tathe, A.; Gupta, V. D.; Patil, V. S.; Phatangare, K.; Sekar, N. J. Fluoresce.
2012, 22(1), 311-322.
27. Sun, W. H.; Li, S. Y.; Hu, R.; Qian, Y.; Wang, S. Q.; Yang, G. Q. J. Phys. Chem. A 2009,
113(20), 5888-5895.
24

ACCEPTED MANUSCRIPT
28. Perrin, D. D.; Armarego, W. L. F.; Perrin, D. R. Purification of laboratory chemicals;
Pergamon: New York, NY, 1966.
29. (a) Ilies, L.; Sato, Y.; Mitsui, C.; Tsuji. H.; Nakamura, E. Chem. Asian J. 2010, 5, 1376-1381.
(b) Men, G. W.; Chen, C. R.; Zhang, S. T.; Liang, C. S.; Wang, Y.; Deng, M. Y.; Shang, H. X.;
Yang. B.; Jiang, S. M. Dalton T. 2015, 44, 2755-2762.
30. (a) Maus, M.; Reitigg, W.; Bonafoux, D.; Lapouyade, R. J. Phys. Chem. A 1999, 103,
3388-3401. (b) Lukeman, M.; Veale, D.; Wan, P.; Ranjit, V.; Munasinghe, N.; Corrie, J. E. Can. J.
Chem. 2004, 82, 240-253.
31. (a) Zgiershi, M. Z.; Grabowska, A. J. Chem. Phys. 2000, 112, 6329-6337. (b) Yi, P.; Liang, Y.
H.; Cao, C. Z. Chem. Phys. 2005, 315, 297-302.
32. (a) Liang, Y. H.; Yi, P. G. Chem. Phys. Lett. 2007,438, 173-177. (b) Yang, Z. N.; Yang, S. Y.;
Zhang, J. P. J. Phys. Chem. A 2007, 111, 6354-6360.
25