The N-arylamino conjugation effect in the photochemistry of fluorescent protein chromophores and aminostilbenes

Article abstract of DOI:10.1002/asia.201000209

To understand the nonradia-tive decay mechanism of fluorescent protein chromophores in solutions, a systematic comparison of a series of (Z)-4-(N-arylamino)benzylidene-2, 3-imidazolinones (ABDIs: 2P, 2PP, 2OM, and 2OMB) and the corresponding trans-4-(N-arylamino)-4′-cya-nostilbenes (ACSs: 1P, 1PP, 1OM, and 1 OMB) was performed. We have previously shown that the parameter φ_f+2 φ_tc, in which φ_f and φ_tc are the quantum yields of fluorescence and trans →cis photoisomerization, respectively, is an effective probe for evaluating the contribution of twisted intramolecular charge transfer (TICT) states in the excited decays of trans-aminostilbenes, including the push-pull ACSs. One of the criteria for postulating the presence of a TICT state is φ_f+2 φ_tc? 1.0, because its formation is decoupled with the C-C bond (φ) torsion pathway and its decay is generally nonradiative. Our results show that the same concept also applies to ABDIs 2 with the parameter φ_f+2 φ_ZE in which φ_ZE is the quantum yield of Z→E photoisomeri-zation. We conclude that the τ torsion rather than the C-C bond (φ) torsion is responsible for the nonradiative decays of ABDIs 2 in aprotic solvents (hexane, THF, acetonitrile). The phenyl-arylamino C-N bond (ω) torsion that leads to a nonradiative TICT state is important only for 2OM in THF and acetonitrile. If the solvent is protic (methanol and 10-20% H₂Oin THF), a new nonradiative decay channel is present for ABDIs 2, but not for ACSs 1. It is attributed to internal conversion (IC) induced by solvent (donor)-solute (acceptor) hydrogen-bonding (HB) interactions. The possible HB modes and the concept of t torsion-coupled proton transfer are also discussed.

Full text of DOI:10.1002/asia.201000209

DOI: 10.1002/asia.201000209
The N-Arylamino Conjugation Effect in the Photochemistry of Fluorescent
Protein Chromophores and Aminostilbenes
Guan-Jhih Huang and Jye-Shane Yang*[a]
Abstract: To understand the nonradia-
tive decay mechanism of fluorescent
protein chromophores in solutions, a
systematic comparison of a series of
(Z)-4-(N-arylamino)benzylidene-2,3-
imidazolinones (ABDIs: 2P, 2PP,
2OM, and 2OMB) and the corre-
sponding trans-4-(N-arylamino)-4’-cya-
nostilbenes (ACSs: 1P, 1PP, 1OM, and
1OMB) was performed. We have pre-
viously shown that the parameter
Ff+2Ftc, in which Ff and Ftc are the
quantum yields of fluorescence and
trans!cis photoisomerization, respec-
tively, is an effective probe for evaluat-
ing the contribution of twisted intramo-
lecular charge transfer (TICT) states in
the excited decays of trans-aminostil-
benes, including the push–pull ACSs.
One of the criteria for postulating the
presence of a TICT state is Ff+2Ftc !
1.0, because its formation is decoupled
with the C=C bond (t) torsion pathway
and its decay is generally nonradiative.
Our results show that the same concept
also applies to ABDIs 2 with the pa-
rameter Ff+2FZE in which FZE is the
quantum yield of Z!E photoisomeri-
zation. We conclude that the t torsion
decays of ABDIs 2 in aprotic solvents
(hexane, THF, acetonitrile). The
ꢁ
phenyl-arylamino C N bond (w) tor-
sion that leads to a nonradiative TICT
state is important only for 2OM in
THF and acetonitrile. If the solvent is
protic (methanol and 10–20% H2O in
THF), a new nonradiative decay chan-
nel is present for ABDIs 2, but not for
ACSs 1. It is attributed to internal con-
version (IC) induced by solvent
(donor)–solute (acceptor) hydrogen-
bonding (HB) interactions. The possi-
ble HB modes and the concept of t tor-
sion-coupled proton transfer are also
discussed.
ꢁ
rather than the C C bond (f) torsion
is responsible for the nonradiative
Keywords: charge transfer · fluo-
rescence · hydrogen bonds · isomer-
ization · photochemistry
Introduction
enhancement. The former was demonstrated by trapping
trans-stilbenes in solvent glass or rigid hosts (e.g., cyclodex-
trins),[7] and the latter is exemplified by meta-amino and N-
arylamino substituted trans-stilbenes, such as m-DCS, 1P,
and 1PP, which were dubbed “the meta-amino effect”[8] and
“the N-arylamino conjugation effect”,[6,9,10] respectively.
In addition to the t torsion, donor–acceptor (D–A) substi-
tuted push–pull alkenes in polar solvents could undergo an-
other type of torsional motion that forms the so-called twist-
ed intramolecular charge transfer (TICT) states.[9–12] One
particular example is provided by aminostilbene 1OM
(Scheme 1) for which the N-(4-methoxyphenyl)amino donor
Photoinduced trans–cis (E–Z) isomerization of alkenes plays
an important role in molecular, biological, and materials
photochemistry.[1–3] For trans-stilbene and many of its deriva-
tives (e.g., p-DCS, Scheme 1),[1,4–6] the nonradiative C=C
bond (t) torsion dominates the excited decays with a quan-
tum yield (Ft) near 1.0, and about 50% of the t torsions
leads to the cis (Z) isomers, corresponding to a trans!cis
isomerization quantum yield (Ftc) of ꢀ0.5 (i.e., Ft =2Ftc).
Nevertheless, this volumne-demanding torsional process can
be supressed mechanically with constrained media or chemi-
cally through substitutions, giving rise to a large fluorescene
is sufficiently strong to initiate a twisting of the stilbenyl-ani-
[10]
ꢁ
lino C N bond (the w torsion). The extremely low quan-
tum yields for both fluorescence and trans!cis isomeriza-
tion for 1OM in acetonitrile (Ff <0.005 and Ftc <0.01)
reveal that the t and w torsions are mutually decoupled and
compete with one another. A restriction of the w torsion by
ring-bridging, as done by 1OMB, restores the high quantum
yield of fluorescence and the t torsion (Ff =0.62 and Ft =
2Ftc =0.36). Consequently, one of the criteria for invoking a
[a] G.-J. Huang, Prof. J.-S. Yang
Department of Chemistry
National Taiwan University
Taipei 10617 Taiwan (Republic of China)
Fax : (+886)2236-36359
E-mail: jsyang@ntu.edu.tw
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/asia.201000209.
Chem. Asian J. 2010, 5, 2075 – 2085
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2075

FULL PAPERS
polar aprotic solvents but be-
comes less than 0.5 in protic
media, such as methanol and
H2O/THF mixed solvents.[16]
These observations led us to
conclude that a) the t, but not
the f torsion, is important to
the excited decay of p-HBDI
and b) there exists an additional
nonradiative decay channel in
protic solvents, attributable to
HB-induced IC. These conclu-
Scheme 1. Structures of DCS and ACSs 1.
sions are based on the assump-
tions of Ft =2FZE and decou-
TICT state for the deactivation of excited trans-stilbenes
should be Ff+2Ftc !1.0. In this context, the previously pro-
pled t and f torsions for p-HBDI in analogy to the cases of
trans-aminostilbenes. A close photochemical relationship be-
tween p-HBDI and trans-aminostilbenes is partly supported
by the facts that a) the amino analog of p-HBDI (i.e., p-
ABDI) possesses the same photochemical behavior as p-
HBDI and b) the meta-amino effect observed for trans-stil-
benes also apply to ABDI, as demonstrated by a large fluo-
rescence enhancement on going from p-ABDI to m-ABDI.
It is also reasonable to consider p-ABDI a neutral isoelec-
tronic structure of the anionic
posed[13] TICT formation arising from the vinyl-anilino C C
ꢁ
bond (f) torsion for p-DCS (Scheme 1) was not supported.
We have concluded that TICT state formation is rather un-
important for aminostilbenes p-DCS, 1P, and 1PP.[10]
The green fluorescent protein (GFP) chromophore (Z)-4-
hydroxybenzylidene-2,3-dimethylimidazolinone
(p-HBDI,
Scheme 2) is a push–pull alkene that displays strong fluores-
form of p-HBDI because of
their
similar
fluorescence
maxima.[21]
To confirm the above conclu-
sions on the nonradiative decay
pathways of p-HBDI in solu-
tions, the features of Ft =2FZE
and decoupled t and f torsions
for fluorescent protein chromo-
phores should be further vali-
dated. To this end, we have in-
Scheme 2. Strucutres of p-HBDI, p-ABDI, m-ABDI, and ABDIs 2.
vestigated the N-arylamino sub-
stituted analogs of p-HBDI
cence in the protein matrix or in solvent glass (Ff ꢀ0.8), but
is essentially nonfluorescent (Ff <10ꢁ3) in fluid solu-
tions.[14,15] Over the past years, the nature of the ultrafast
nonradiative decay channels for p-HBDI in solutions contin-
ues to be controversial.[15–20] The most often discussed mech-
anisms are all associated with torsional motions of either the
(i.e., ABDIs), 2P, 2PP, 2OM, and 2OMB, and compared
with the corresponding trans-4-(N-arylamino)-4’-cyanostil-
benes (ACSs) 1P, 1PP, 1OM, and 1OMB in several solvent
systems. We reported herein that the N-arylamino conjuga-
tion effect on the photochemistry (fluoresence, photoisome-
rization, and TICT formation) of ABDIs 2 parallels that of
ACSs 1 in aprotic solvents, but not in protic solvents. These
results demonstrate the close photochemical relationships
between ABDIs and trans-4-aminostilbenes in terms of tor-
sional relaxations and provide new insights into the HB-in-
duced IC for ABDIs and p-HBDI in aqueous or alcoholic
solutions.
ꢁ
exocyclic C C bond (f) and/or the C=C bond (t). However,
we recently determined the Z!E isomerization quantum
yields (FZE) for p-HBDI and found that the value of
Ff+2FZE is close to 1.0 for p-HBDI in both polar and non-
Abstract in Chinese:
Results
Synthesis
The synthesis of ACSs 1P, 1PP, 1OM, and 1OMB has been
reported.[10] Illustrated in Scheme 3 is the synthesis of ABDI
2P for which the benzylidene-2,3-dimethylimidazolinone
2076
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 2075 – 2085

Photochemistry of Fluorescent Protein Chromophores
Table 1. Maxima of UV/Vis absorption (labs) and fluorescence (lf), quan-
tum yields for fluorescence (Ff) and and trans!cis photoisomerization
(Ftc) for ACSs 1 and Z!E photoisomerization (FZE) for ABDIs 2 in
protic and aprotic solvents at room temperature.
[d]
Solvent[a]
labs [nm]
lf[b,c] [nm]
414
(437)[e]
Ff
Ftc/FZE
1P
Hex
THF
MeCN
10W
20W
MeOH
Hex
THF
MeCN
10W
20W
MeOH
Hex
THF
MeCN
10W
20W
MeOH
Hex
THF
MeCN
10W
20W
MeOH
Hex
THF
MeCN
10W
20W
MeOH
Hex
THF
MeCN
10W
20W
MeOH
Hex
THF
MeCN
10W
20W
MeOH
Hex
THF
MeCN
10W
20W
373[e]
384
379[e]
385
381
384
390[e]
391
388[e]
393
391
390
376[e]
394
384[e]
397
400
407
400[e]
410
408[e]
411
407
406
399
416
413
425
427
430
424
424
423
428
428
430
402
420
416
429
430
432
426
438
440
448
450
454
G
0.11[e]
0.23
0.44[e]
0.46
490
504[e]
514
0.35[e]
0.29
0.33[e]
0.35
Scheme 3. Synthesis of ABDI 2P.
521
0.22
0.31
0.32
529
0.16
backbone was constructed by using our modified Niwaꢁs
two-step protocol[14] starting with 3-bromobenaldehyde (3B)
and via the azalactone intermediate 4B. The bromo group
1PP
430
G
0.79[e]
0.88
501
542[e]
522
526
541
0.14
0.05[e]
0.02
0.92[e]
0.81
in 2B was then replaced by an N-phenylamino group
[22]
0.79
0.04
0.08
ꢁ
through palladium-catalyzed C N coupling reaction with
0.75
aniline. The ABDI 2OM was also prepared from 2B by the
1OM
1OMB
2P
419
N
0.25[e]
0.05[e]
0.27[e]
<0.01[e]
<0.01[e]
0.15
ꢁ
C N coupling reaction with 4-methoxyaniline. The synthesis
530
583[e]
548
<0.005[e]
0.01
of ABDI 2PP and 2OMB is more straightforward, because
the corresponding N-arylamino substituted benzaldehydes
3PP and 3OMB are known[9] and can be directly subjected
to the condensation reactions via intermediates 4PP and
4OMB (Scheme 4). The synthetic details are shown in Ex-
perimental Section.
559
0.01
0.24
<0.01
[f]
<0.001
0.37[e]
0.64[e]
0.62[e]
0.62
447ACHTNUTGRNEUNG
(475)[e]
533
590[e]
560
570
582
457
487
518
497
516
522
467
515
590
556
564
602
466
0.17[e]
0.18[e]
0.16
0.29
0.21
0.46
0.46
0.43
0.32
0.25
0.27
0.49
0.46
0.48
0.33
0.19
0.25
0.43
0.09
0.01
0.25
0.30
0.05
0.48
0.41
0.46
0.23
0.16
0.19
0.59
0.42
0.002
0.002
0.002
0.002
0.002
0.001
0.010
0.035
0.056
0.019
0.031
0.004
0.002
0.003
<0.001
<0.001
<0.001
<0.001
0.005
0.016
0.025
0.015
0.013
0.008
2PP
Scheme 4. Structures of starting materials and intermediates for the syn-
thesis of ABDIs 4PP and 4OMB.
2OM
2OMB
527
[f]
Electronic Spectra
[f]
[f]
[f]
The electronic absorption and emission spectra and the data
of Ff and Ftc for the ACSs 1P, 1PP, 1OM, and 1OMB in
several aprotic solvents, such as hexane, THF, and acetoni-
trile have been reported.[10] For the purpose of comparison
with the ABDIs 2P, 2PP, 2OM, and 2OMB in the same
aprotic (hexane, THF, and acetonitrile) and protic (MeOH,
10% H2O/THF (v/v) (10W), and 20% H2O/THF (v/v)
(20W)) solvents, the previously undetermined data for
ACSs 1 in some of these solvents are provided in this work.
The absorption spectra of ACSs 1 are known to display a
single intense absorption band with a small dependence of
the peak maxima (labs) on the solvent polarity.[10] For exam-
ple, the shifts in peak maxima are generally less than 10 nm
toward the longer wavelength upon changing the solvent
from nonpolar hexane to polar acetonitrile (Table 1). A
switching of the solvent from aprotic acetonitrile to protic
methanol further shifts the peak maxima to the red, except
for the case of 1OMB. Among the four species of ACSs 1,
compound 1PP displays the smallest response to the solvent
polarity and proticity.
484
523
571
561
573
612
MeOH
[a] Hex: hexane; 10W: 10% H2O/THF (v/v); 20W: 20% H2O/THF (v/v).
[b] Fluorescence data are from corrected spectra. [c] Maxima of the
second vibronic bands are given in parentheses. [d] For the purpose of
solubility, Hex, MeCN, and MeOH contain 20% THF for the measure-
ment of FZE and Ftc. [e] Data from Ref. [10]. [f] Fluorescence too weak
to be determined.
Shown in Figure 1 are the absorption spectra of ABDIs
2P, 2PP, 2OM, and 2OMB in hexane and 2P in different
solvents. Like ACSs 1, the ABDIs possess the common fea-
ture of single intense absorption bands. The peak maxima of
ABDIs 2 are at wavelengths 30 nm longer than those of
ACSs 1 of the same N-arylamino substituents (Table 1), and
Chem. Asian J. 2010, 5, 2075 – 2085
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2077

J.-S. Yang and G.-J. Huang
FULL PAPERS
However, this relationship is not always true in the other
solvents (e.g., 2P<2OMB<2PP in acetonitrile). The larger
solvent effect on lf than labs indicates a charge-transfer char-
acter of ABDIs 2 in the excited state. This is consistent with
the AM1-derived[24] frontier orbitals for 2 for which the
HOMOs of all four compounds are delocalized, but their
LUMOs are localized at the imidazolinone moiety
(Figure 3). It is interesting to note that, unlike the larger sol-
vent polarity-induced labs shifts for 2 versus 1, the corre-
sponding shifts of lf are larger for 1 versus 2. Evidently, the
fluorescing excited state of ACSs 1 is more polar than that
of ABDIs 2.
Figure 1. Electronic absorption spectra of (left) a) 2P, b) 2PP, c) 2OM,
and d) 2OMB in hexane and (right) 2P in THF (&), MeCN (*), and
MeOH (~).
the solvent effect on labs is generally larger for ABDIs 2
versus ACSs 1. This indicates that the imidazolinone hetero-
cycle in 2 interacts to a larger extent with the solvent mole-
cules than the benzonitrile group in 1. Along this line, the
relatively small solvent dependence of 1PP and 2PP suggest
that the N-arylamino nitrogen is even more crucial in ac-
counting for the ground-state solvent–solute interactions. It
is known that the basicity of triarylamines is too low to in-
teract with protons or metal ions.[23]
Shown in Figure 2 are the normalized fluorescence spectra
of 2 in hexane, THF, acetonitrile, and methanol. ABDI
2OM is essentially nonfluorescent in acetonitrile and the
three protic solvents. The fluorescence spectra depend not
only on the solvent, but also on the N-arylamino group. In
hexane, the relative fluorescence peak maxima (lf) among
the four compounds follow the same trend as their labs
values in the order of 2P<2OM<2PP<2OMB (Table 1).
Figure 3. The HOMOs and LUMOs of ABDIs 2.
The dipole moment (me) of the fluorescent state of ABDIs
2 can be estimated from the slope (mf) of the plot of the en-
ergies of the fluorescence maxima against the solvent pa-
rameter Df according to Equation (1):[25]
nf ¼ ꢁ½ð1=4 pe0Þð2=hca3Þꢂ½ með meꢁmgÞꢂDfþconstant
Using Equation (2):
ð1Þ
ð2Þ
ð3Þ
Df ¼ ðe0ꢁ1Þ=ð2eþ1Þꢁ0:5ðn2ꢁ1Þ=ð2 n2þ1Þ
and Equation (3)
1=3
a ¼ ð3 M=4 NpdÞ
in which nf is the fluorescence maximum; mg is the ground-
state dipole moment; a is the solvent cavity (Onsager)
radius, which was derived from the Avogadro number (N),
molecular weight (M), and density (d); and e, e0, and n are
the solvent dielectric constant, the vacuum permittivity, and
the solvent refractive index, respectively. To avoid specific
solute–solvent interactions, the solvents employed herein
are aprotic, including hexane, cyclohexane, THF, dichloro-
methane, ethyl acetate, acetone, and acetonitrile. Neverthe-
less, ABDI 2OM is nonfluorescent in solvents more polar
than THF, and thus its me value was not evaluated. The
value of mg was calculated using the AM1 algorithm. As
shown in Table 2, the mg and mf values are lower for ABDIs
2 than ACSs 1 of the same N-arylamino substituent. The me
Figure 2. Normalized fluorescence spectra of ABDIs
b) THF, c) MeCN, d) methanol.
2 in a) hexane,
2078
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 2075 – 2085

Photochemistry of Fluorescent Protein Chromophores
Table 2. Ground- and excited-state dipole moments for ACSs 1 and
ABDIs 2.
Quantum Yields and Lifetimes
[b]
a [ꢃ][a]
mf [cmꢁ1
]
mg [D][c]
me [D]
The absence of thermal E!Z isomerization for p-ABDI in
both protic and aprotic solvents at ambient temperature has
recently been demonstrated.[16] This is also true for ABDIs
2, which allows one to determine the Z!E photoisomeriza-
tion quantum efficiency more straightforwardly.
The values of Ff and Ftc (or FZE) for ACSs 1 and ABDIs
2 in the six solvent systems are shown in Table 1. In general,
the ACSs 1 display larger fluorescence quantum yields than
the corresponding ABDIs 2 in the same solvent, and the op-
posite is true for their relative isomerization quantum yields.
In view of the opposite trends in Ff and Ftc (or FZE), owing
to competitions between fluorescence and the t torsion in
the excited decays of these push–pull alkenes, the following
comparisons of the N-arylamino substituent effect and the
solvent effect will only focus on the changes in Ff unless an
anomaly was found for Ftc (or FZE).
1P[d]
4.90
5.12
5.06
5.19
4.87
5.26
5.16
12560
13454
15832
14620
7515
4.84
5.85
3.93
4.62
2.79
2.46
1.57
14.8ꢃ0.6
16.6ꢃ0.6
16.4ꢃ0.8
16.7ꢃ0.7
11.3ꢃ0.6
16.0ꢃ1.0
13.0ꢃ0.7
1PP[d]
1OM[d]
1OMB[d]
2P
2PP
2OMB
13523
9747
[a] Onsager radius calculated by Equation (3) with d=0.9 gcmꢁ3 for the
ABDIs. [b] Calculated based on Equation (1). [c] Calculated by use of
AM1. [d] Data from Ref. [10].
values are much larger than the mg values for both systems,
consistent with the larger dependence of the fluorescence
versus absorption on the solvent polarity.
Shown in Figure 4 are the fluorescence spectra of 2PP in
methylcyclohexane (MCH) and mixed acetonitrile/THF
(9:1) at different temperatures with an interval of 10 K in
Regarding the N-arylamino substituent effect on Ff, the
two systems 1 and 2 parallel to one another in hexane:
namely, the relative fluorescence quantum efficiency is 1P<
1OM<1OMB<1PP for the ACSs 1 and 2Pꢀ2OM<
2OMB<2PP for the ABDIs 2.
For the solvent effect, the ABDIs 2P, 2PP, and 2OMB
parallel to the corresponding 1P, 1PP, and 1OMB in two as-
pects: 1) The Ff value increases, within the 10% experimen-
tal errors, on going from hexane to THF and to acetonitrile;
2) Their Ff values in the protic solvents lie in between those
of hexane and acetonitrile with a value generally lower in
MeOH than in 10W and 20W. In contrast, highly polar sol-
vents such as acetonitrie and methanol are detrimental to
both the Ff and Ft (=2Ftc or 2FZE) of 1OM and 2OM.
Because of the low fluorescence quantum yields of
ABDIs 2 and the limited resolution of our time-resolved
spectrophotometer (0.1–0.2 ns), fluorescence lifetimes were
selectively determined for 2PP at low temperatures in
MCH and 2-methyltetrahydrofuran (MTHF). Assuming that
the t torsion was the only activated singlet-decay process
and that the fluorescence rate constant kf was temperature
independent, the torsional barrier can be obtained from
nonlinear fitting of the fluorescence lifetimes by using Equa-
tion (4):[27]
Figure 4. Temperature dependence of the fluorescence spectra of 2PP in
methylcyclohexane and acetonitrile/THF (9:1) recorded at intervals of
108C.
the range 160–290 K and 240–340 K, respectively. Addition
of 10% THF to the acetonitrile solutions prevents substrate
aggregation or precipitation at low temperature. A signifi-
cant fluorescence enhancement was observed upon lowering
the temperature for both solutions, which indicates the pres-
ence of activated deactivation processes for the excited
state. Accompanied is a red shift of the peak maxima, which
is larger in MCH (Dlf =11 nm) than in the acetonitrile/THF
mixed solvent (Dlf =4 nm). This can be attributed to the in-
tf ðTÞ ¼ 1=½Sk þ A expðꢁEa=RTÞꢂ
ð4Þ
In which Sk is the sum of all nonactivated processes (fluo-
rescence and intersystem crossing (ISC)) and A and Ea are
the pre-exponential and activation energy for the activated
process, respectively. These results are shown in Figure 5,
and the activation parameters are reported in Table 3. The
value of Ea increases on going from MCH (2.57 kcalmolꢁ1)
to MTHF (5.05 kcalmolꢁ1), but the logA value is also in-
creased. This is reminiscent of enthalpy–entropy compensa-
tion.[28] Thus, the resulting rate constant for the t torsion
(kt) at 296 K is reduced only by half (98 versus 55ꢂ108 sꢁ1).
The temperature-dependent fluorescence lifetimes of
ACS 1PP were also recorded in Figure 5 and the activation
crease of both e and n with decreasing the temperature.[26]
A
similar temperature-dependent fluorescence behavior has
also been observed for 1P in mixed acetonitrile/THF
(9:1).[10]
Chem. Asian J. 2010, 5, 2075 – 2085
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2079

J.-S. Yang and G.-J. Huang
FULL PAPERS
should be taken into account for the excited decays of
ABDIs and in turn the fluorescent protein chromophores.
This concept orignated from our trans-aminostilbene
works,[9,10,12] which has provided valuable information on the
formation and decays of the TICT states of aminostilbenes
(e.g., the ACSs 1). In general, TICT states are weakly or
nonfluorescent and their formation are decoupled with the t
torsions. As the bond that twists corresponds to a single
bond in the ground state (e.g., the w torsion in 1OM), deac-
tivation of TICT states recover the starting ground state.
Consequently, a prerequisite, but not a sufficient require-
ment to argue for the TICT state formation, is the observa-
tion of Ff+2Ftc !1.0. As ACSs 1 and ABDIs 2 have the
push–pull character in common, their analogies and differ-
ences in photochemistry should allow one to address the ap-
plicability of the parameter Ff+2FZE in discussing the non-
radiative decay mechanism of ABDIs. In the following, the
fluorescence and the cis–trans (Z–E) photoisomerization be-
havior of ACSs 1 and ABDIs 2 are comparied in aprotic
and protic solvents. For the purpose of discussion, a summa-
ry of their Ff+2Ftc and Ff+2FZE values in different sol-
vents is depicted in Figure 6.
Figure 5. Temperature-dependent lifetimes and nonlinear fits to Equa-
tion (4) for 1PP and 2PP in a) MCH and b) MTHF.
Table 3. Activation parameters for 1P, 1PP, and 2PP in methylcyclohex-
ane (MCH) and 2-methyltetrahydrofuran (MTHF).
solvent 10ꢁ8
Sk[a,b] [sꢁ1
log A[c]
Ea[c] [kcalmolꢁ1
]
10ꢁ8
[d,e]
]
kt [sꢁ1
]
1P[f] MCH 7.4ꢃ0.1 (5.0) 12.4ꢃ0.1 3.9ꢃ0.5
1PP MCH 5.2ꢃ0.1 (5.3) 12.7ꢃ0.2 6.2ꢃ0.2
MTHF 3.5ꢃ0.1 (4.0) 12.6ꢃ0.3 6.3ꢃ0.4
30.6 (40.5)
1.3 (1.4)
0.91 (0.54)
98.3
2PP MCH 3.6ꢃ0.4
11.9ꢃ0.3 2.6ꢃ0.2
MTHF 2.2ꢃ0.1
13.5ꢃ0.3 5.1ꢃ0.3
55.1
[a] Sum of the nonactivated singlet decay processes. [b] The value given
in parentheses is kf derived from Ff and tf measured at room tempera-
ture. [c] Activation parameters for singlet activated decay from nonlinear
fitting of temperature-dependent lifetimes (Figure 6). [d] Room tempera-
ture double-bond torsional rate calculated from A and Ea. [e] The value
given in parentheses is knr derived from (1ꢁFf) and tf measured at room
temperature. [f] Data from Ref. [10].
parameters are also shown in Table 3. Unlike the solvent de-
pendence of 2PP, the Ea and logA values are essentially the
same for 1PP in MCH and MTHF. The large Ea values and
the medium size of logA for 1PP result in low kt values
(ꢀ1ꢂ108 sꢁ1), which is consistent with its low Ftc values
(Table 1). The values of Sk are comparable to the kf values
derived from Ff and tf for 1PP, indicating that ISC is negli-
gible. It should be noted that the same study has been car-
ried out for 1P, and the results show that the barrier for the
t torsion is smaller (3.9 versus 6.2 kcalmolꢁ1) and the S1!T1
ISC is more important for 1P (FISC ꢀ0.05) versus 1PP
(FISC <0.01) in MCH.[10] The less efficient ISC for 1PP
versus 1P might be attributed to the stronger charge trans-
fer character, as reflected by their me values (Table 2), in the
S1 state.
Figure 6. Plot of Ff+2Ftc or Ff+2 FZE against the solvent polarity and
proticity.
Photochemistry in Aprotic Solvents
The N-arylamino substituent effect on the excited decays of
ACSs 1P, 1PP, 1OM, and 1OMB in aprotic solvents has
been previously elucidated.[10] As compared to p-DCS, the
N-aryl group elongates the p-conjugated backbone of trans-
aminostilbenes, which stabilizes the fluorescing S1 state (1t*)
more than the intermediate of perpendicular geometry (1p*)
and thus raises the t torsional barrier. Consequently, the
fluorescence quantum yield is increased at the expense of
the efficiency of the t torsion and thus reduces the value of
Ftc. The larger fluorescence quantum yields for 1PP versus
1P partly reflect the presence of one additional N-phenyl
group. The current work further suggests that the less effi-
cient ISC for 1PP versus 1P in the excited state also con-
tributes to its higher fluorescence quantum yield. When the
N-aryl group is more electron-donating, such as that in
Discussion
The central theme of this work is to show that the parame-
ter Ff+2FZE is a useful probe for judging whether other
nonradiative decay channels in addition to the t torsion
2080
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 2075 – 2085

Photochemistry of Fluorescent Protein Chromophores
1OM, the TICT state formation becomes kinetically more
favorable[9,10] and thus competes with the t torsion and fluo-
rescence in polar solvents. This accounts for its low quantum
yields in both Ff and Ftc in THF and acetonitrile, giving rise
to Ff+2Ftc <0.25 (Figure 6). The absence of dual fluores-
cence for 1OM suggests that the TICT state is nearly non-
fluorescent. The ring-bridged analog 1OMB was employed
to support the TICT argument and to indentify the bond
that twists in 1OM. On the basis of the distinct fluorescence
spectra and the Ff and Ftc values for 1OM and 1OMB and
of Ff+2Ftc ꢀ1.0 for 1OMB in all three aprotic solvents, the
nonradiative TICT state of 1OM should result from a twist-
tion for ABDIs with the donor–acceptor-substituted stil-
benes ACSs rather than the donor-only stilbenes ASs in the
solvent effect reflects the push–pull electronic nature of
ABDIs.
Photochemistry in Protic Solvents
An apparent feature for the two chromophore systems in al-
coholic solvents is the solvent (HB donor)-solute (HB ac-
ceptor) hydrogen bonding interactions. In the three investi-
gated protic solvents, methanol, 10W, and 20W, the ACSs 1
could form hydrogen bonds with the solvent OH groups
through the N-arylamino and the cyano nitrogens. The me-
thoxy group in 1OM and 1OMB is also an HB acceptor.
For the non-TICT-forming ACSs 1P, 1PP, and 1OMB, the
behavior of Ff+2Ftc ꢀ1.0 is unchanged on going from the
aprotic solvents to all three protic solvents. Evidently, the
possible HB interactions do not induce decay channels
other than fluorescence and the t torsion. For the TICT-
forming 1OM, the low values of Ff and Ftc in acetonitrile
are also retained in methanol, which indicates that the TICT
character (i.e., the w torsion) is also retained in the excited
state.
ꢁ
ing of the stilbenyl-anilino C N bond (i.e., the w torsion,
Scheme 1).
The behavior of Ff+2Ftc ꢀ1.0 observed for 1P, 1PP, and
1OMB, but Ff+2Ftc !1.0 for 1OM are retained when the
stilbene chromophore is changed to benzylideneimidazoli-
none: namely, Ff+2FZE ꢀ1.0 for 2P, 2PP, and 2OMB and
Ff+2FZE !1.0 for 2OM (Figure 6). Evidently, the propensi-
ty of TICT state formation is similar for the two alkene
chromophores 1 and 2. More specifically, the f torsion
could be excluded for all four compounds of ABDIs 2, and
the TICT state for 2OM results from the w torsion
(Scheme 2). On the other hand, the two systems differ large-
ly in the relative efficiency of fluorescnece and the t torsion.
The fluorescence quantum yields are much lower for ABDIs
2 versus ACSs 1, and the opposite is true for the isomeriza-
tion quantum yields. As shown by 1PP and 2PP, this results
from larger rates for the t torsion in S1 and somewhat lower
rates for fluorescence for the ABDIs 2 (Table 3). Neverthe-
less, the fluorescence quantum efficiency for 2P, 2PP, and
2OMB are prominent as compared to the parent p-ABDI,
showing that the N-arylamino conjugation effect on fluores-
cence enhancement applies to both systems. Accordingly,
the t torsional barriers for p-ABDI should be even lower
than that (2.6 kcalmolꢁ1) for 2PP in MCH, because the
former is essentially nonfluorescent (Ff <10ꢁ3) and the t
torsion accounts for the excited decay in all three aprotic
solvents. Assuming that p-ABDI possesses a fluorescence
rate constant and log A value similar to 2PP (kf ꢀSkꢀ3ꢂ
108 sꢁ1 and logA=12 in MCH, Table 3), the rate constant
for the t torsion (kt) should be larger than 3ꢂ1011 sꢁ1 to ac-
count for the nonfluorescent character of p-ABDI. This in
turn suggests that the t torsion for p-ABDI is nearly barrier-
less. The t torsion for p-HBDI in the excited state was also
shown to be nearly barrierless.[29] A close photochemical re-
lationship between p-ABDI and p-HBDI is thus evidenced.
The dependence of Ff on the solvent polarity for the non-
TICT-forming ACSs 1 and ABDIs 2 deserves a comment.
For both alkene systems, the Ff value increases with increas-
ing the solvent polarity. This trend is opposite to that for
trans-4-(N-arylamino)stilbenes without the electron-with-
drawing cyano group (i.e., ASs).[9] The origin of this differ-
ence between ACSs and ASs has previously been discussed
based on a) the interplay between the Ea and logA values
for their t torsions and b) the relative polarity of the fluores-
Although the solvent–solute HB interactions do not
change the decay pathways of ACSs 1, their effect on Ff
and Ftc deserves attention. It appears that the fluorescence
decay is somewhat suppressed in favor of the t torsion, as
shown by the small Ff decrease, but Ftc increases on going
from THF to 10W and to 20W. We would have expected
the opposite trend if we only consider the changes in solvent
polarity. It is interesting to note that this phenomenon is
more significant for the TICT-forming 1OM than the non-
TICT-forming species 1P, 1PP, and 1OM. The phenomenon
of larger isomerization quantum efficiency in polar protic
versus aprotic solvents is also known for other stilbenes.[30]
In addition, it has been shown that the TICT state of N,N-
dimethylaminobenzonitrile (DMABN) can be further stabi-
lized by solvent–solute HB interactions.[31] As 1OM can un-
dergo both the t and w torsions, it would be interesting to
know which torsional motion, t or w, is better promoted by
the solvent–solute HB interactions. On the basis of the
values of Ff+2Ftc in THF (<0.07), 10W (0.31), and 20W
(0.55), we might conclude that the t torsion benefits more
than the w torsion for 1OM upon introducing solvent–
solute HB interactions. Overall, it appears that the HB in-
1
1
teractions stabilize the p* state more than the t* state of
ACSs 1 so that the t torsional barrier is lowered.
The story of the HB interactions on the excited decays of
ABDIs 2 is somewhat different. For the non-TICT-forming
(i.e., without f or w torsion) derivatives 2P, 2PP, and
2OMB, the Ff+2FZE value is in the range 0.33–0.68 in the
three protic solvents, indicating the presence of new decay
channels in addition to fluorescence and the t torsion. This
phenomenon has been observed for p-ABDI and m-ABDI,
and it has been attributed to solute–solvent HB-induced IC
(i.e., the quantum yield FHBIC =1ꢁ
Ultrafast HB-induced IC has been reported for many other
(Ff+2Ftc)=0.32–0.67).[16]
ACHTUNGTRENNUNG
1
1
cent t* state and the p* intermediate.[9,10] A better correla-
Chem. Asian J. 2010, 5, 2075 – 2085
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2081

J.-S. Yang and G.-J. Huang
FULL PAPERS
systems.[32,33] Although there are several potential HB ac-
ceptors in ABDIs 2, the HB mode associated with the N-ar-
ylamino nitrogen is less likely responsible for the excited-
state quenching, as this mode is also present in ACSs 1. As
such, either the imino N or the carbonyl O in the imidazoli-
none group, as depicted for 2P in methanol (Figure 7),
with protic solvent molecules. Also, the longer the excited-
state lifetime of the solute, the more probable the solute–
solvent HB interactions or ESPT that leads to IC. These fea-
tures might account for the much stronger HB-induced fluo-
rescence quenching for m-ABDI versus ABDIs 2 (the deriv-
atives of p-ABDI). In this context, it is interesting to note
that ESPT-induced IC has recently been suggested by Sol-
ntsev et al. for the meta- hydroxy isomer of p-HBDI (i.e.,
m-HBDI), but not for p-HBDI, because the ultrafast excit-
ed-state dynamics of the latter is insensitive to the solvent
nature and the pH of solutions.[36] This is different from our
previous observation of Ff+2FZE !1.0 for p-HBDI (ꢀ0.2–
0.4), p-ABDI (0.34–0.74), and m-ABDI (0.12–0.16) in the
protic solvents,[16] although it is common that the HB-in-
duced IC is considerably more efficient for the meta isomers.
To further account for the fact that HB-induced IC affects
more on the t torsion than the fluorescence of ABDIs 2, we
propose herein that the HB-induced IC is dynamically cou-
pled with the t torsion. This means not only that the HB in-
teractions can promote the t torsion (vide supra), but also
that the t torsion can facilitate the HB-mediated IC, pre-
sumably as a result of charge redistribution along the torsion
coordinate. One of the possible scenarios is the t torsion-as-
sisted proton transfer from the solvent to the imidazolinone
group of the ABDIs, which diverts the reaction from form-
Figure 7. An illustration of the possible HB modes that are likely to be
responsible for the internal conversion of 2P in methanol.
should be responsible. This argument is also supported by
the recent femtosecond-resolved spectroscopic work by Pet-
kova et al. on a series of p-HBDI and p-ABDI deriva-
tives.[20] Nevertheless, the exact HB mode that is responsible
for the nonradiative quenching of the excited state remains
to be established. It should also be noted that the HB
quenching mode could be associated with full proton trans-
fer from the solvent to the solute, owing to the enhanced ba-
sicity of the HB acceptors in the excited state. The behavior
of excited-state proton transfer (ESPT) has been well docu-
mented for many aromatic systems.[34,35] In the case of
TICT-forming 2OM, both the w torsion and the solvent–
solute HB interactions can contribute to the observation of
Ff+2FZE !1.0 (0.10–0.60) in protic solvents. The contribu-
tion from the w torsion is evidenced by the complete fluo-
rescence quenching for 2OM but not for the other ABDIs 2
in all the three protic solvents. Nevertheless, like the case of
1OM, the larger FZE values in 10W and 20W than in THF
once more indicate that the solvent–solute HB interactions
promote the t torsion more than the w torsion.
1
ing the p* state to the conical intersection for IC (aborted
photoisomerization).
Conclusions
A pictorial summary on the nonradiative decay pathways of
ACSs 1 and ABDIs 2 is provided in Figure 8. In aprotic sol-
vents (i.e., absence of the specific solute–solvent HB inter-
actions), both systems, except for 1OM and 2OM, undergo
only t torsion that leads to E–Z isomerization. For 1OM
and 2OM, the w torsion that leads to TICT state formation,
also occurs in THF and acetonitrile. The f torsion can be
excluded for both ACSs 1 and ABDIs 2. In protic solvents,
the solvent–solute HB interactions in both systems facilitate
the t torsion more than the w torsion. However, the HB in-
Compared to the case of m-ABDI,[16] the protic solvent
effect on Ff is different in ABDIs 2P, 2PP, and 2OMB.
Whereas m-ABDI possesses high fluorescence quantum
yields (Ff =0.16–0.34) in the aprotic solvents and becomes
nonfluorescent (Ff <0.001) in all three protic solvents, the
fluorescence quantum yields of ABDIs 2P, 2PP, and 2OMB
are lower by one order of magnitude (Ff =0.002–0.056) in
aprotic solvents but remain fluorescent (Ff =0.001–0.031) in
the protic solvents. Evidently, the HB-induced IC is much
more effective in the meta-amino derivative m-ABDI than
the para-amino derivatives ABDIs 2. The previous studies
on meta- versus para-aminostilbenes have shown that the
meta isomers are of lower fluorescence rates (longer excit-
ed-state lifetimes) and possess a larger extent of charge sep-
aration in S1.[8,12] In principle, the larger the charge separa-
tion in the excited ABDIs, the more the electron density on
the imidazolinone group, and thus the stronger HB interacts
Figure 8. A pictorial summary of the nonradiative decay channels for
ACSs 1 and ABDIs 2. The symbols t, f, and w refer to the torsion path-
ways and HB-IC refers to hydrogen bonding-induced internal conver-
sion.
2082
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 2075 – 2085

Photochemistry of Fluorescent Protein Chromophores
(300 mg) were added. The reaction was heated at 808C with stirring for
12 h. Then ethyl acetate was added and washed with brine directly. The
organic layer was dried over anhydrous MgSO4 and the filtrate was con-
centrated under reduced pressure. The crude product was purified by
silica gel column chromatography with hexane/ethyl acetate/THF (4/2.5/
3.5) to afford 2B (4.3 g, 62%, m.p.: 144–1468C). 1H NMR ([D6]DMSO):
d=2.36 (s, 3H), 3.09 (s, 3H), 6.94 (s, 1H), 7.63 (d, J=8.8 Hz, 2H),
8.14 ppm (d, J=8.8 Hz, 2H); 13C NMR([D6]DMSO): d=15.5, 26.3, 122.9,
123.2, 131.4, 133.1, 133.3, 139.2, 164.9, 169.4 ppm. MS (EI, 70 eV): m/z
(relative intensity) 278 (M+, 100); HRMS calcd for C12H11BrN2O+:
278.0055; found: 278.0045; elemental analysis (%) calcd for
C12H11BrN2O: C 51.63, H 3.97, N 10.04; found: C 51.43, H 4.17, N 9.72.
teractions induce a new IC channel for ABDIs 2, but not for
ACSs 1. These results confirm that a) the assumption of
Ft =2FZE is valid, and thus the value of Ff+2FZE can be a
conclusive probe for evaluating the contribution of decay
channels other than fluorescence and the t torsion for
ABDIs 2 and its parent compound p-ABDI, b) the HB ac-
ceptors responsible for the excited quenching are in the imi-
dazolinone group of ABDIs 2 and p-ABDI, and c) the pho-
tochemistry of p-ABDI and p-HBDI is closely related. Con-
sequently, our previous conclusions on the dual nonradiative
decay channels, the t torsion and the HB-induced IC, for p-
HBDI are further supported.[16] This work led us to further
propose that the HB-induced IC is coupled to a significant
extent with the t torsion for ABDIs 2 and more likely for p-
ABDI and p-HBDI as well. We are currently studying new
fluorescent protein chromophores in order to give a more
detailed picture on the mechanism of the HB-induced IC.
Understanding of the decay mechanism of these GFP-like
chromophores would be benificial for the design of new ma-
terials for light-emitting diodes and solar cells.[37]
Compound 2P
A
mixture of 2B (0.556 g, 2.0 mmol), aniline (0.218 mL, 2.4 mmol),
NaOtBu (0.269 g, 2.8 mmol), (ꢃ)-BINAP (0.019 g, 0.03 mmol), and [Pd2
AHCTUNGRTEG(NNNU dba)3] (0.018 g, 0.02 mmol) in anhydrous toluene (5 mL) under argon
was heated at 1108C for 12 h. The solution was cooled and then the in-
soluble residue was filtered off by CH2Cl2 and ethyl acetate. The filtrate
was concentrated under reduced pressure to afford the crude product.
Further purification was performed by column chromatography (hexane/
ethyl acetate=4/6) to provide the red solid (300 mg, 51%, m.p.: 180–
1838C). Further recrystallization from hexane and CH2Cl2 affords 2P as
a fine crystalline red solid. 1H NMR (CDCl3): d=2.35 (s, 3H), 3.17 (s,
3H), 6.05 (s, 1H), 7.00–7.05 (m, 4H), 7.14 (d, J=8.8 Hz, 2H), 7.30 (t, J=
8.0 Hz, 2H), 8.03 ppm (d, J=8.8 Hz, 2H); 13C NMR
ACHTNUTRGNE(UNG CDCl3): d=16.0,
26.9, 115.4, 119.5, 122.3, 125.8, 127.4, 129.0, 133.6, 135.6, 140.6, 145.0,
159.7, 170.0 ppm; HRMS (ESI+) calcd for C18H18N3O+ [M+H+]:
292.1450; found: 292.1444.
Experimental Section
Methods
Compound 2PP and 4PP
Electronic spectra were recorded at room temperature (23ꢃ18C). UV/
Visible spectra were measured by using a Cary300 double beam spectro-
photometer. Fluorescence spectra were recorded by using a PTI Quanta-
Master C-60 or the Edinburgh FLS920 spectrometers and corrected for
the response of the detector. The optical density (OD) of all solutions
was about 0.1 at the wavelength of excitation. A N2-bubbled (15 min) so-
lution of 9,10-diphenylanthracene (Ff =0.93 in n-hexane)[38] and anthra-
cene (Ff =0.27 in n-hexane)[39] was used as a standard for the fluores-
cence quantum yield determinations of compounds under N2-bubbled
solutions with solvent refractive index correction. An error of 10% is es-
timated for the fluorescence quantum yields. Fluorescence decays were
measured at room temperature by using an Edinburgh FLS920 spectrom-
eter with a gated hydrogen arc lamp using a scatter solution to profile
the instrument response function. The goodness of the nonlinear least-
squares fit was judged by the reduced c2 value (<1.2 in all cases), the
randomness of the residuals, and the autocorrelation function. Quantum
yields of photoisomerization were measured by using optically dense de-
gassed solutions (10ꢁ3–10ꢁ4 m) at l=350 nm by using a 75-W Xe arc lamp
and monochromator. N-phenyl-4-aminostilbenes was used as a reference
standard (Ftc =0.34 in CH2Cl2).[9] The extent of photoisomerization
(<10%) was determined by using HPLC analysis (Waters 600 Controller
and 996 photodiode array detector, Thermo APS-2 Hypersil, hexane and
ethyl acetate mixed solvent) without back-reaction corrections. The re-
producibility error was <10% of the average. AM1 calculations were
performed with the Gaussian03 program.[40]
A mixture of N-acetylglycine (0.45 g, 5.5 mmol), sodium acetate (0.45 g,
5.5 mmol), 3PP (1.1 g, 4 mmol) and acetic anhydride (3 mL) was heated
at 1108C with stirring for 5 h. The solvent was removed under reduced
pressure, and the residue was dissolved in CH2Cl2 and washed with brine.
The organic layer was dried over anhydrous MgSO4 and the filtrate was
concentrated under reduced pressure. Then the crude product 4PP was
added ethanol (0.5 mL) and stirred for 20 min with 40% aqueous methyl-
amine (0.4 mL) solution 20 min at room temperature. Potassium carbon-
ate (10 mg) was added to the solution and the solution was refluxed at
808C for 4 h. The solvent was removed under reduced pressure, and the
residue was dissolved in ethyl acetate and washed with brine. The organic
layer was dried over anhydrous MgSO4 and the filtrate was concentrated
under reduced pressure. The product was purified by silica gel column
chromatography with ethyl acetate/hexane (1/1) to afford 2PP, Further
recrystallization from Ethyl acetate and CH2Cl2 affords 2PP as a fine
crystalline red solid (0.27 g, 18%, m.p.: 207–2128C). 1H NMR
([D6]DMSO): d=2.31 (s, 3H), 3.08 (s, 3H), 6.87 (s, 1H), 6.88 (d, J=
8.4 Hz, 2H), 7.09 (d, J=8.4 Hz, 4H), 7.13 (t, J=8.4 Hz, 2H), 7.35 (t, J=
8.4 Hz, 4H), 8.05 ppm (d, J=8.4 Hz, 2H) ; 13C NMR([D6]DMSO): d=
15.3, 26.2, 120.2, 124.2, 124.4, 125.1, 126.9, 129.5, 133.0, 136.8, 145.9,
148.5, 162.3, 169.4 ppm; MS (EI, 70 eV): m/z (relative intensity) 367
(M+, 100); HRMS calcd for C24H21N3O+ 367.1685; found: 367.1680; ele-
mental analysis (%) calcd for C24H21N3O: C 78.45, H 5.76, N 11.44;
found: C 78.05, H 5.91, N 11.55.
Materials
Compound 2OM
Solvents for spectra and quantum yield measurements all were HPLC
grade and used as received. THF and acetonitrile were dried by sodium
metal and cesium hydride, respectively, and distilled before use. The syn-
thesis of 1P,[10] 1PP,[10] 1OM,[10] 1OMB,[10] 3PP,[41] and 3OMB[10] has
been reported. All the new compounds were identified by using
1H NMR, 13C NMR spectroscopy, and mass spectrometry.
A mixture of 2B (0.72 g, 2.6 mmol), p-anisidine (0.384 g, 3.12 mmol),
NaOtBu (0.35 g, 3.64 mmol), (ꢃ)-BINAP (0.024 g, 0.039 mmol), and [Pd2
AHCTUNGRTEG(NNNU dba)3] (0.024 g, 0.03 mmol) in anhydrous toluene (20 mL) under argon
was heated at 1108C for 12 h. The solution was cooled and then the in-
soluble residue was filtered off by THF. The filtrate was concentrated
under reduced pressure to afford the crude product. Further purification
was performed by column chromatography (hexane/ethyl acetate=1/1)
to provide the red solid (270 mg, 33%, m.p.: 192–1948C). 1H NMR
([D6]DMSO): d=2.31 (s, 3H), 3.08 (s, 3H), 3.73 (s, 3H), 6.84 (s, 1H),
6.90 (d, J=8.0 Hz, 4H), 7.11 (d, J=8.0 Hz, 2H), 8.02 (d, J=8.0 Hz, 2H),
8.51 ppm (s, 1H); 13C NMR([D6]DMSO): d=15.3, 26.2, 55.2, 113.5, 114.4,
Compound 2B
To a solution of 40% aqueous methylamine (2.7 mL, 27 mmol) and THF
(60 mL) was added 4B (6.62 g, 25 mmol). The reaction mixture was
stirred for 20 min, and then water (60 mL) and potassium carbonate
Chem. Asian J. 2010, 5, 2075 – 2085
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2083

J.-S. Yang and G.-J. Huang
FULL PAPERS
121.9, 123.6, 125.7, 133.8, 133.9, 134.9, 147.1, 154.6, 160.7, 169.4 ppm; MS
(EI, 70 eV): m/z (relative intensity) 321 (M+, 100); HRMS calcd. for
C19H19N3O2+: 321.1477; found: 321.1483.
S. Laohhasurayotin, T. S. R. Krishna, Angew. Chem. 2008, 120,
1257–1260; Angew. Chem. Int. Ed. 2008, 47, 1237–1240.
[2] M. Zimmer in Cis-trans Isomerization in Biochemistry (Ed.: C.
Dugave), Wiley-VCH, Weinheim, 2006.
[3] a) H. Meier, Angew. Chem. 1992, 104, 1425–1446; Angew. Chem.
Int. Ed. Engl. 1992, 31, 1399–1420; b) Q.-C. Wang, D.-H. Qu, J.
Ren, K. Chen, H. Tian, Angew. Chem. 2004, 116, 2715–2719;
Angew. Chem. Int. Ed. 2004, 43, 2661–2665; c) B. L. Feringa, J. Org.
Chem. 2007, 72, 6635–6652; d) J.-S. Yang, Y.-T. Huang, J.-H. Ho,
W.-T. Sun, H.-H. Huang, Y.-C. Lin, S.-J. Huang, S.-L. Huang, H.-F.
Lu, I. Chao, Org. Lett. 2008, 10, 2279–2282; e) E. M. Geertsema,
S. J. van der Molen, M. Martens, B. L. Feringa, Proc. Natl. Acad. Sci.
USA 2009, 106, 16919–16924.
[4] a) J. Saltiel, E. D. Megarity, K. G. Kneipp, J. Am. Chem. Soc. 1966,
88, 2336–2338; b) J. Saltiel, J. Am. Chem. Soc. 1967, 89, 1036–1037;
c) J. Saltiel, A. Marinari, D. W. L. Chang, J. C. Mitchener, E. D.
Megarity, J. Am. Chem. Soc. 1979, 101, 2982–2996.
[5] H. Gruen, H. Gçrner, Z. Naturforsch. 1983, 38, 928–936.
[6] J.-S. Yang, S.-Y. Chiou, K.-L. Liau, J. Am. Chem. Soc. 2002, 124,
2518–2527.
[7] a) S. Malkin, E. Fischer, J. Phys. Chem. 1964, 68, 1153–1163;
b) C. A. Stanier, M. J. OꢁConnell, H. L. Anderson, W. Clegg, Chem.
Commun. 2001, 493–494.
[8] a) F. D. Lewis, J.-S. Yang, J. Am. Chem. Soc. 1997, 119, 3834–3835;
b) F. D. Lewis, R. S. Kalgutkar, J.-S. Yang, J. Am. Chem. Soc. 1999,
121, 12045–12053; c) F. D. Lewis, W. Weigel, J. Phys. Chem. A 2000,
104, 8146–8153; d) F. D. Lewis, W. Weigel, X. Zuo, J. Phys. Chem.
A 2001, 105, 4691–4696.
[9] J.-S. Yang, K.-L. Liau, C.-M. Wang, C.-Y. Hwang, J. Am. Chem. Soc.
2004, 126, 12325–12335.
[10] J.-S. Yang, K.-L. Liau, C.-Y. Hwang, C.-M. Wang, J. Phys. Chem. A
2006, 110, 8003–8010.
Compound 2OMB
A mixture of 40% aqueous methylamine solution (0.11 mL, 1.1 mmol),
10 mL methanol, 20 mg potassium carbonate, 4OMB (0.334 g, 1 mmol)
was stirred for 20 min at room temperature, and then heated at 1208C
with stirring for 2 h. The solution was cooled and then concentrated
under reduced pressure. The crude product was purified by silica gel
column chromatography with hexane/ethyl acetate/CH2Cl2 (3/3.5/3.5) to
1
afford 2OMB (220 mg, 63.4%, m.p.: 181–1848C). H NMR ([D6]DMSO):
d=2.32 (s, 3H), 3.08 (s, 3H), 3.12 (t, J=8.8 Hz, 2H), 3.75 (s, 3H), 3.98
(t, J=8.8 Hz, 2H), 6.82 (d, J=8.4 Hz, 1H), 6.84 (s, 1H), 6.96 (d, J=
8.8 Hz, 2H), 7.24 (d, J=8.8 Hz, 2H), 7.75 (d, J=8.4 Hz, 1H), 8.15 ppm
(s, 1H); 13C NMR([D6]DMSO): d=15.3, 26.2, 26.9, 52.5, 55.2, 106.2,
114.4, 120.6, 124.0, 125.9, 127.9, 131.2, 133.5, 134.6, 135.4, 149.3, 154.7,
+
160.4, 169.4 ppm; HRMS (ESI+) calcd for C21H22N3O2 [M+H+]:
348.1712; found: 348.1710.
Compound 4B
A mixture of N-acetylglycine (1.75 g, 15 mmol), sodium acetate (1.2 g,
15 mmol), 4-bromobenzaldehyde (1.85 g, 10 mmol), THF (25 mL), and
acetic anhydride (8 mL) was heated at 808C with stirring for 12 h. The
solvent was removed under reduced pressure, and the residue was dis-
solved in CH2Cl2 and washed with brine. The organic layer was dried
over anhydrous MgSO4 and the filtrate was concentrated under reduced
pressure. The crude product was washed with hexane to afford 4B (2.5 g,
1
95%, m.p.: 126–1298C). H NMR (CDCl3): d=2.39 (s, 3H), 7.04 (s, 1H),
7.55 (d, J=8.4 Hz,2H), 7.92 ppm (d, J=8.4 Hz,2H); 13C NMR
ACTHNUTRGNEN(UG CDCl3):
d=15.8, 125.6, 129.5, 131.8, 131.9, 132.8, 133.2, 166.3, 167.2 ppm; MS (EI,
70 eV): m/z (relative intensity) 265 (M+, 100); HRMS calcd for
C11H8BrNO2+: 264.9738; found: 264.9739.
[11] a) W. Rettig, Angew. Chem. 1986, 98, 969–986; Angew. Chem. Int.
Ed. Engl. 1986, 25, 971–988; b) Z. R. Grabowski, K. Rotkiewicz,
Chem. Rev. 2003, 103, 3899–4031.
Compound 4OMB
[12] a) J.-S. Yang, K.-L. Liau, C.-Y. Li, M.-Y. Chen, J. Am. Chem. Soc.
2007, 129, 13183–13192; b) J.-S. Yang, C.-K. Lin, A. M. Lahoti, C.-
K. Tseng, Y.-H. Liu, G.-H. Lee, S.-M. Peng, J. Phys. Chem. A 2009,
113, 4868–4877.
To
a mixture of N-acetylglycine (1.52 g, 13 mmol), 3OMB (2.53 g,
10 mmol), and acetic anhydride (80 mL) was added 5 drops of 98%
H2SO4, and then the mixture was heated at 808C with stirring for 1.5 h.
The solvent was removed under reduced pressure, and the residue was
dissolved in ethyl acetate, 10% aqueous NaOH solution, and washed
with brine. The organic layer was dried over anhydrous MgSO4 and the
filtrate was concentrated under reduced pressure. The crude product was
purified by silica gel column chromatography with ethyl acetate/hexane/
CH2Cl2 (1/7/2) to afford red solid 4OMB (0.355 g, 10.6%, m.p.: 168–
1718C). 1H NMR (CDCl3): d=2.36 (s, 3H), 3.19 (t, J=8.4 Hz, 2H), 3.81
(s, 3H), 4.01 (t, J=8.4 Hz, 2H), 6.82 (d, J=8.4 Hz, 1H), 6.92 (d, J=
8.8 Hz, 2H), 7.04 (s, 1H), 7.19 (d, J=8.4 Hz, 2H), 7.62 (d, J=8.8 Hz,
[13] a) E. Gilabert, R. Lapouyade, C. Rulliꢄre, Chem. Phys. Lett. 1988,
145, 262–268; b) W. Rettig, W. Majenz, Chem. Phys. Lett. 1989, 154,
335–341; c) R. Lapouyade, K. Czeschka, W. Majenz, W. Rettig, E.
Gilabert, C. Rulliꢄre, J. Phys. Chem. 1992, 96, 9643–9650; d) E.
Abraham, J. Oberlꢅ, G. Jonusauskas, R. Lapouyade, C. Rulliꢄre, J.
Photochem. Photobiol. A 1997, 105, 101–107; e) E. Abraham, J.
Oberlꢅ, G. Jonusauskas, R. Lapouyade, C. Rulliꢄre, Chem. Phys.
1997, 214, 409–423; f) Y. Amatatsu, Theor. Chem. Acc. 2000, 103,
445–450; g) Y. Amatatsu, Chem. Phys. 2001, 274, 87–98; h) D.
Pines, E. Pines, W. Rettig, J. Phys. Chem. A 2003, 107, 236–242;
i) Y. Amatatsu, J. Phys. Chem. A 2006, 110, 8736–8743.
1H), 8.05 ppm (s, 1H); 13C NMR
ACTHNUTRGNEU(GN CDCl3): d=15.7, 27.6, 53.4, 55.6, 106.8,
114.5, 121.5, 123.5, 127.3, 128.3, 131.5, 132.5, 134.7, 135.5, 151.3, 155.7,
162.7, 168.5 ppm; HRMS (ESI+) calcd for C20H19N2O3 [M+H+]:
+
[14] a) H. Niwa, S. Inouye, T. Hirano, T. Matsuno, S. Kojima, M. Kubota,
M. Ohashi, F. I. Tsuji, Proc. Natl. Acad. Sci. USA 1996, 93, 13617–
13622; b) S. Kojima, H. Ohkawa, T. Hirano, S. Maki, H. Niwa, M.
Ohashi, S. Inouye, F. I. Tsuji, Tetrahedron Lett. 1998, 39, 5239–5242.
[15] a) W. Weber, V. Helms, J. A. McCammon, a. P. W. Langhoff, Proc.
Natl. Acad. Sci. USA 1999, 96, 6177–6182; b) S. R. Meech, Chem.
Soc. Rev. 2009, 38, 2922–2934.
334.1396; found: 335.1394.
Acknowledgements
We thank the National Science Council of Taiwan, ROC, for financial
support.
[16] J.-S. Yang, G.-J. Huang, Y.-H. Liu, S.-M. Peng, Chem. Commun.
2008, 1344–1346.
[17] a) M. C. Chen, C. R. Lambert, J. D. Urgitis, M. Zimmer, Chem.
Phys. 2001, 270, 157–164; b) A. D. Kummer, C. Kompa, H. Niwa, T.
Hirano, S. Kojima, M. E. Michel-Beyerle, J. Phys. Chem. B 2002,
106, 7554–7559; c) A. Follenius-Wund, M. Bourotte, M. Schmitt, F.
Iyice, H. Lami, J.-J. Bourguignon, J. Haiech, C. Pigault, Biophys. J.
2003, 85, 1839–1850; d) P. Altoe’, F. Bernardi, M. Garavelli, G. Or-
landi, F. Negri, J. Am. Chem. Soc. 2005, 127, 3952–3963; e) A.
Usman, O. F. Mohammed, E. T. J. Nibbering, J. Dong, K. M. Sol-
ntsev, L. M. Tolbert, J. Am. Chem. Soc. 2005, 127, 11214–11215;
f) K.-Y. Chen, Y.-M. Cheng, C.-H. Lai, C.-C. Hsu, M.-L. Ho, G.-H.
[1] a) J. Saltiel, J. L. Charlton in Rearrangements in Ground and Excited
States, Vol. 3 (Ed.: P. de Mayo), Academic Press, New York, 1980,
pp. 25–89; b) D. H. Waldeck, Chem. Rev. 1991, 91, 415–436; c) H.
Gçrner, H. J. Kuhn, Adv. Photochem. 1995, 19, 1–117; d) A. Simeo-
nov, M. Matsushita, E. A. Juban, E. H. Z. Thompson, T. Z. Hoffman,
A. E. Beuscher, IV, M. J. Taylor, P. Wirsching, W. Rettig, J. K.
McCusker, R. C. Stevens, D. P. Millar, P. G. Schultz, R. A. Lerner,
K. D. Janda, Science 2000, 290, 307–313; e) J. Saltiel, M. A. Bremer,
2084
www.chemasianj.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 2075 – 2085

Photochemistry of Fluorescent Protein Chromophores
Lee, P.-T. Chou, J. Am. Chem. Soc. 2007, 129, 4534–4535; g) V. Vo-
liani, R. Bizzarri, R. Nifosꢆ, S. Abbruzzetti, E. Grandi, C. Viappiani,
F. Beltram, J. Phys. Chem. B 2008, 112, 10714–10722; h) C. Fang,
R. R. Frontiera, R. Tran, R. A. Mathies, Nature 2009, 462, 200–204.
[18] a) M. E. Martin, F. Negri, M. Olivucci, J. Am. Chem. Soc. 2004, 126,
5452–5464; b) S. S. Stavrov, K. M. Solntsev, L. M. Tolbert, D. Hup-
pert, J. Am. Chem. Soc. 2006, 128, 1540–1546; c) L. Wu, K. Burgess,
J. Am. Chem. Soc. 2008, 130, 4089–4096; d) S. Olsen, S. C. Smith, J.
Am. Chem. Soc. 2008, 130, 8677–8689; e) S. Olsen, R. H. McKenzie,
J. Chem. Phys. 2009, 130, 184302; f) S. Olsen, K. Lamothe, T. J. Mar-
tꢆnez, J. Am. Chem. Soc. 2010, 132, 1192–1193.
[32] F. D. Lewis, J.-S. Yang, J. Phys. Chem. 1996, 100, 14560–14568, and
references therein.
[33] a) L. Biczꢇk, T. Bꢅrces, H. Linschitz, J. Am. Chem. Soc. 1997, 119,
11071–11077, and references cited therein; b) H. Shimada, A. Na-
kamura, T. Yoshihara, S. Tobita, Photochem. Photobiol. Sci. 2005, 4,
367–375.
[34] a) J. Waluk, in Conformational Analysis of Molecules in Excited
States (Ed.: J. Waluk), Wiley-VCH, New York, 2000; b) J. Waluk,
Acc. Chem. Res. 2003, 36, 832–838; c) C.-C. Hsieh, K.-Y. Chen, W.-
T. Hsieh, C.-H. Lai, J.-Y. Shen, C.-M. Jiang, H.-S. Duan, P.-T. Chou,
ChemPhysChem 2008, 9, 2221–2229.
[19] a) N. M. Webber, K. L. Litvinenko, S. R. Meech, J. Phys. Chem. B
2001, 105, 8036–8039; b) N. Y. A. Baffour-Awuah, M. Zimmer,
Chem. Phys. 2004, 303, 7–11; c) A. Toniolo, S. Olsen, L. Manohar,
T. J. Martꢆnez, Faraday Discuss. 2004, 127, 149–163; d) D. Mandal,
T. Tahara, S. R. Meech, J. Phys. Chem. B 2004, 108, 1102–1108;
e) C. M. Megley, L. A. Dickson, S. L. Maddalo, G. J. Chandler, M.
Zimmer, J. Phys. Chem. B 2009, 113, 302–308.
[35] H. Shizuka, S. Tobita, in Organic Photochemistry and Photophysics
(Eds.: V. Ramamurthy, K. S. Schanze), CRC Press/Taylor&Francis,
Boca Raton, 2006.
[36] a) J. Dong, K. M. Solntsev, O. Poizat, L. M. Tolbert, J. Am. Chem.
Soc. 2007, 129, 10084–10085; b) K. M. Solntsev, O. Poizat, J. Dong,
J. Rehault, Y. Lou, C. Burda, L. M. Tolbert, J. Phys. Chem. B 2008,
112, 2700–2711.
[20] I. Petkova, G. Dobrikov, N. Banerji, G. Duvanel, R. Perez, V. Dimi-
trov, P. Nikolov, E. Vauthey, J. Phys. Chem. A 2010, 114, 10–20.
[21] L. Wang, J. Xie, A. A. Deniz, P. G. Schultz, J. Org. Chem. 2003, 68,
174–176.
[37] a) Y. You, Y. He, P. E. Burrows, S. R. Forrest, N. A. Petasis, M. E.
Thompson, Adv. Mater. 2000, 12, 1678–1681; b) W.-T. Chuang, B.-S.
Chen, K.-Y. Chen, C.-C. Hsieh, P.-T. Chou, Chem. Commun. 2009,
6982–6984.
[22] a) J. P. Wolfe, S. Wagaw, J.-F. Marcoux, S. L. Buchwald, Acc. Chem.
Res. 1998, 31, 805–818; b) J. F. Hartwig, Angew. Chem. 1998, 110,
2154–2177; Angew. Chem. Int. Ed. 1998, 37, 2046–2067.
[23] J.-S. Yang, Y.-H. Lin, C.-S. Yang, Org. Lett. 2002, 4, 777–780.
[24] M. J. S. Dewar, E. G. Zoebisch, E. F. Healy, J. J. P. Stewart, J. Am.
Chem. Soc. 1985, 107, 3902–3909.
[25] W. Baumann, H. Bischof, J.-C. Frçhling, C. Brittinger, W. Rettig, K.
Rotkiewicz, J. Photochem. Photobiol. A 1992, 64, 49–72.
[26] P. Suppan, C. Tsiamis, J. Chem. Soc. Faraday Trans. 2 1981, 77,
1553–1562.
[27] a) F. D. Lewis, X. Zuo, J. Am. Chem. Soc. 2003, 125, 2046–2047;
b) F. D. Lewis, X. Zuo, J. Am. Chem. Soc. 2003, 125, 8806–8813.
[28] L. Liu, Q-X. Guo, Chem. Rev. 2001, 101, 673–695.
[29] K. L. Litvinenko, N. M. Webber, S. R. Meech, J. Phys. Chem. A
2003, 107, 2616–2623.
[30] a) K. J. Smit, K.-P. Ghiggino, Chem. Phys. Lett. 1985, 122, 369–374;
b) G. Ginocchietti, U. Mazzucato, A. Spalletti, Int. J. Photoenergy
2004, 6, 241–250.
[31] a) W.-M. Kwok, M. W. George, D. C. Grills, C. Ma, P. Matousek,
A. W. Parker, D. Phillips, W. T. Toner, M. Towrie, Angew. Chem.
2003, 115, 1870–1874; Angew. Chem. Int. Ed. 2003, 42, 1826–1830;
b) W. M. Kwok, C. Ma, M. W. George, D. C. Grills, P. Matousek,
A. W. Parker, D. Phillips, W. T. Toner, M. Towrie, Photochem. Pho-
tobiol. Sci. 2007, 6, 987–994.
[38] J. B. Birks, Photophysics of Aromatic Molecules, Wiley, London,
1970.
[39] W. R. Dawson, M. W. Windsor, J. Phys. Chem. 1968, 72, 3251–3260.
[40] Gaussian 03 (version 6.0), M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery,
V., Jr., T., K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J.
Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R.
Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian, J. B. Cross,
C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, P. Y. Ayala, K.
Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg, V. G. Zakr-
zewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K.
Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz,
Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G.
Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox,
T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M. Challa-
combe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, C. Gonza-
lez, J. A. Pople, Gaussian, Inc., Pittsburgh PA, 2003.
[41] S. Sengupta, S. K. Sadhukhan, S. Muhuri, Tetrahedron Lett. 2002, 43,
3521–3524.
Received: March 16, 2010
Published online: July 15, 2010
Chem. Asian J. 2010, 5, 2075 – 2085
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2085