Excited-state intramolecular proton transfer molecules bearing o -hydroxy analogues of green fluorescent protein chromophore

Article abstract of DOI:10.1021/jo2012384

o-Hydroxy analogues, 1a-g, of the green fluorescent protein chromophore have been synthesized. Their structures and electronic properties were investigated by X-ray single-crystal analyses, electrochemistry, and luminescence properties. In solid and nonpo

Full text of DOI:10.1021/jo2012384

Article
pubs.acs.org/joc
Excited-State Intramolecular Proton Transfer Molecules Bearing o-
Hydroxy Analogues of Green Fluorescent Protein Chromophore
Wei-Ti Chuang,† Cheng-Chih Hsieh,† Chin-Hung Lai,† Cheng-Hsuan Lai,† Chun-Wei Shih,†
Kew-Yu Chen,*,‡ Wen-Yi Hung,*,§ Yu-Hsiang Hsu,§ and Pi-Tai Chou*,†
†Department of Chemistry, National Taiwan University, Taipei 106, Taiwan, Republic of China
‡Department of Chemical Engineering, Feng Chia University, Taichung 40724, Taiwan, Republic of China
§Institute of Optoelectronic Sciences, National Taiwan Ocean University, Keelung 202, Taiwan, Republic of China
S
* Supporting Information
ABSTRACT: o-Hydroxy analogues, 1a−g, of the green fluorescent protein chromophore have been synthesized. Their
structures and electronic properties were investigated by X-ray single-crystal analyses, electrochemistry, and luminescence
properties. In solid and nonpolar solvents 1a−g exist mainly as Z conformers that possess a seven-membered-ring hydrogen
bond and undergo excited-state intramolecular proton transfer (ESIPT) reactions, resulting in a proton-transfer tautomer
emission. Fluorescence upconversion dynamics have revealed a coherent type of ESIPT, followed by a fast vibrational/solvent
relaxation (<1 ps) to a twisted (regarding exo-C(5)−C(4)−C(3) bonds) conformation, from which a fast population decay of a
few to several tens of picoseconds was resolved in cyclohexane. Accordingly, the proton-transfer tautomer emission intensity is
moderate (0.08 in 1e) to weak (∼10−4 in 1a) in cyclohexane. The stronger intramolecular hydrogen bonding in 1g suppresses
the rotation of the aryl−alkene bond, resulting in a high yield of tautomer emission (Φ f ≈ 0.2). In the solid state, due to the
inhibition of exo-C(5)−C(4)−C(3) rotation, intense tautomer emission with a quantum yield of 0.1−0.9 was obtained for 1a−g.
Depending on the electronic donor or acceptor strength of the substituent in either the HOMO or LUMO site, a broad tuning
range of the emission from 560 (1g) to 670 nm (1a) has been achieved.
residue such as E222,3 resulting in a very intense anion
1. INTRODUCTION
Green fluorescent protein (GFP)1 has received intense
fluorescence. From a chemistry point of view, most of the
research has focused on the chemical modification of p-HBDI4
attention because of its ubiquitous applications in molecular
biology and biochemistry. GFP takes advantage of the presence
of a chromophore, (Z)-4-(4-hydroxybenzylidene)-1,2-dimethyl-
1H-imidazol-5(4H)-one (p-HBDI; see Scheme 1), which
analogues at the C(1) position (see Scheme 1), such that the
emission color can be tuned via the substituent effect.5 Studies
nevertheless reveal a cutoff in properties between wild type
GFP (or certain GFP mutants) and the synthetic analogue
chromophores of p-HBDI. The GFP-free p-HBDI gives
Scheme 1. Structural Isomers of p-HBDI and o-HBDI
virtually no emission in fluid solution at room temperature.
The results suggest an effective radiationless transition
operating in p-HBDI, plausibly induced by conformational
relaxation along torsional deformation6 of the two exocyclic C−
C bonds from the initially prepared Franck−Condon state to a
nonfluorescent twisted intermediate.7 The shallow potential
energy surface of the intermediates may conically intersect with
undergoes excited-state proton transfer (ESPT)2 via the proton
Received: June 27, 2011
relay of the water network and/or some residues to a remote
Published: September 26, 2011
© 2011 American Chemical Society
8189
dx.doi.org/10.1021/jo2012384|J. Org. Chem. 2011, 76, 8189−8202

The Journal of Organic Chemistry
Article
Scheme 2. (A) Synthetic Route of o-Hydroxy Analogues (1a−g) of the Green Fluorescent Protein Chromophore and (B)
Synthesis of Z and E Forms of 4-((1H-Pyrrol-2-yl)methylene)-1-methyl-2-phenyl-1H-imidazol-5(4H)-one (1h)
that of the ground state, inducing the dominant radiationless
deactivation.8 Such a conformational relaxation is greatly
suppressed in wild GFP by its proton relay network, forming
a rigid environment to restrain the conformational relaxation.
Recently, with an aim to probe the associated excited-state
proton transfer phenomena, two research groups have made
different approaches via switching the position of the hydroxyl
group in p-HBDI. On the one hand, Tolbert and co-workers
have developed a promising strategy toward the m-GFP
chromophore,9 m-HBDI, to probe a stepwise proton transfer
mechanism in protic solvents. The results serve as a paradigm
for the elementary proton transfer steps in relevant systems. On
the other hand, we have strategically designed and synthesized
an o-GFP chromophore, o-HBDI (see Scheme 1).10 o-HBDI
possesses a seven-membered-ring hydrogen bond, from which
the excited-state intramolecular proton transfer (ESIPT) takes
place, resulting in a remarkable proton transfer tautomer
emission of ∼605 nm in organic solvents such as cyclohexane.
ESIPT also takes place in the solid film, giving rise to a ∼595
nm tautomer emission with a quantum yield as high as 0.4.
While applications of o-HBDI are pending exploration, it is
believed that the chemical derivation as well as the associated
chemical and photophysical properties can be further explored.
In view of the photophysical properties, we have recently
performed comprehensive studies of o-HBDI during an overall
proton transfer cycle. The results conclude that ESIPT in o-
8190
dx.doi.org/10.1021/jo2012384|J. Org. Chem. 2011, 76, 8189−8202

The Journal of Organic Chemistry
Article
Scheme 3. Proposed Mechanism
HBDI is essentially triggered by low-frequency motions
associated with hydrogen bonding and may be barrierless
along the reaction coordinate. Femtosecond UV−vis transient
absorption and IR spectra also provide supplementary evidence
for the structural evolution during the reaction.11
characterization, and the associated chemical, electrochemical,
and ESIPT properties are elaborated in the following sections.
2.2. Synthesis and Characterization. Similar to the
preparation of p-HBDI and its analogues,8 in our earlier
attempt, we expected that o-HBDI (1d) and the corresponding
analogues might be synthesized via the reaction of o-
hydroxybenzaldehyde (see Scheme 3) with N-acetylglycine in
the presence of acetic anhydride and sodium acetate. However,
the intermediate (4Z)-4-(2-hydroxybenzylidene)-2-methyloxa-
zol-5(4H)-one (o-HBMO; see Scheme 3) obtained from the
reaction between N-acetylglycine and o-hydroxybenzaldehyde
was obtained in a poor yield of <20%. Instead, as shown in
Scheme 3, a substantial amount of N-(2-oxo-2H-chromen-3-
yl)acetamide was isolated and identified by X-ray single-crystal
analyses (see Figure S47 in the Supporting Information).
In view of the mechanistic approach, we propose that the
presence of the o-hydroxyl group plays a key role in the
observed reaction pattern. In basic solution, the phenolate acts
as a nucleophile to attack the carbonyl oxygen, resulting in a
cyclization reaction. A similarly low yield (<20%) of
intermediates, i.e. derivatives of o-HBMO, was obtained when
various o-hydroxybenzaldehyde derivatives were used as starting
reactants. As a result, the subsequent workup procedure by
treating methylamine in the presence of K2CO3-basified
ethanol, followed by neutralization, gave the corresponding o-
HBDI derivatives 1a−g in a poor yield of <10%.
From the viewpoint of chemical derivation, in this study, we
have made intense efforts to carry out the chemical
derivatization of o-HBDI. Our aim is twofold. First, we plan
to systematically study the synthetic routes and the associated
reaction mechanism. Second, on the basis of this series of
ESIPT molecules we will be able to gain more insight into the
relationship for the photophysical behavior versus structural
properties. We also expect that, through the synthetic approach,
it is feasible to fine-tune the proton transfer tautomer emission
covering a broad spectral range from visible to near-IR,
establishing a new class of o-GFP dyes. Finally, the successful
fabrication of OLEDs using o-HBDI provides a new perspective
of utilizing proton transfer dyes in electroluminescence devices.
2. RESULTS AND DISCUSSION
2.1. Design Strategy. A preliminary result of the
computational approach for o-HBDI10 indicated that the
frontier orbitals contributing to S0 → S1 as well as S0′ → S1′
(the prime denotes the proton-transfer tautomer) transitions in
the normal and proton-transfer tautomer species, respectively,
can be mainly ascribed to o-hydroxyl phenyl ring (HOMO) and
imidazolone (LUMO) moieties. On this basis, a series of
derivatives of o-HBDI (hereafter, o-HBDI is denoted as 1d in
this study), 1a−g, can be strategically designed by functionaliz-
ing different substituents at the R1 or R2 position (see Scheme
2A).
Via alteration of the electron donating/withdrawing strength
or extension of π-conjugation of the substituent, the correlation
among molecular structure, hydrogen-bonding strength, and
the corresponding ESIPT properties can be systematically
probed. The interplay between HOMO and LUMO energy
levels should fine-tune the energy gap of proton-transfer
tautomer emission covering a broad spectral range. To probe
the hydrogen bond−geometry relationship, we also took one
more step via replacing the phenol ring in 1a by the pyrrolic
moiety, forming 1h (see Scheme 2B). Details of synthesis,
Alternatively, a more effective synthetic route to achieve the
substituted o-HBDI (1a−g) began with o-methoxybenzalde-
hyde or its derivatives (see Scheme 2A). The lack of an o-
hydroxyl group and hence the intramolecular lactonation leads
to the formation of 3a−g in a good yield of 71−76%.
Subsequent reaction of 3a−g with methylamine, followed by
deprotection of the methyl group of 2a−g by BBr3, afforded
1a−g in an overall product yield of >40%. Detailed synthetic
1
procedures as well as H NMR, 13C NMR, MS, and HRMS
data are provided in the Experimental Section. It is also worth
noting that treatment of 2b with BBr3 (4 equiv, 0 °C, 4 h)
resulted in cleavage of the methyl ether group only at the C(6)
position (see Scheme 1 for numbering), giving 1b in 89% yield.
The high regioselectivity of 2b with BBr3 can be rationalized by
the combination of two key factors. Chelation of the imine
nitrogen and the ether oxygen to the electron-deficient BBr3 is
8191
dx.doi.org/10.1021/jo2012384|J. Org. Chem. 2011, 76, 8189−8202

The Journal of Organic Chemistry
Article
expected to be thermally favorable, forming the intermediate
ImA shown in Scheme 4. Moreover, in comparison to the
withdrawing substituent at the para position is evident. The
results clearly follow a trend in that an increase of the acidity of
the hydroxyl proton renders an increase of the intramolecular
hydrogen bonding strength. Note that the substitution of CH3
in 1d by a phenyl ring at the C(1) position (Scheme 1, also see
R2 in Scheme 2), forming 1a, seems to increase the basicity of
the proton acceptor, i.e. the N(2) atom, through the resonance
effect. As a result, 1a exhibits a larger downfield shift of the O−
H proton and hence a stronger hydrogen bond relative to 1d.
Further support of the above assignments is given by the X-
ray single-crystal analyses. Except for 1f, growth of single
crystals was successful for all compounds 1a−g and their
associated structures have been resolved by X-ray analyses.
Using 1a as a prototype, Figure 1A depicts the X-ray-resolved
Scheme 4. Proposed Mechanism
phenolate moiety at the C(9) position, o-phenolate is
considered to be a better leaving group, due to the resonance
effect that weakens the O−CH3 bond. Accordingly, the
chelation exerts a regiodirective bias and results in the
subsequent cleavage of the methyl ether at the ortho position
(intermediate ImB; see Scheme 4).
The dominance of a Z isomer for 1a−g, namely intra-
molecular hydrogen-bond formation between O(2)−H and
N(2), is firmly supported by a combination of 1H NMR and X-
ray single-crystal analyses. In the 1H NMR studies, the
existence of a strong hydrogen bond between O(2)−H and
N(2) is evidenced by the observation of a large downfield shift
of the proton peak at δ >13 ppm for all compounds 1a−g, the
values of which are in the order 1g (15.11 ppm) > 1e (14.70
ppm) > 1f (14.18 ppm) > 1a (13.94 ppm) > 1c (13.82 ppm) >
1d (13.68 ppm) > 1b (13.26 ppm) in dry CD2Cl2. For the
C(9) position substituents (R1; see Scheme 2) that are in para
positions with respect to the hydroxyl group, the trend of the
O−H proton peak of 1g > 1e > 1f > 1c > 1d > 1b is in good
correlation with the associated electron-withdrawing strength of
1g (−NO2) > 1e (−CN) > 1f (−CF3) > 1c (−Br) > 1d (−H)
> 1b (−OCH3), in which OCH3 in 1b is an electron-donating
group. If we assume that the chemical shift of the hydroxyl
proton in 1a−g belongs to the class of aromatic alcohols, the
hydrogen bonding energy, ΔE in kcal/mol, can be empirically
estimated by introducing Shaefer’s correlation,6 expressed as
Figure 1. Molecular structures of (A) 1a and its intermolecular
relationship and (B) 1h-E. Thermal ellipsoids are drawn at the 50%
probability level.
structure to address its salient featured, while the structures of
1b−e,g are provided in the Supporting Information (see
Figures S49−S53). First of all, according to the single-crystal
structure, 1a−g all reveal a Z configuration, in which the O−H
proton is hydrogen-bonded with the N(2) nitrogen (see
Scheme 1). Apparently, the results of X-ray analyses all indicate
a short distance of <2.65 Å between O(2) and N(2) for 1a−e,g,
supporting the formation of an O(2)−H- - -N(2) intra-
molecular hydrogen bond. Moreover, with the exception of
1b (2.588 Å), where resonance effects pertaining to the
methoxy group may dominate the hydrogen-bonding character
rather than inductive effects, the distance between O(2) and
N(2) along the O(2)-H- - -N(2) hydrogen bond is in the order
of 1d (2.630 Å) > 1a (2.610 Å) ≈ 1c (2.608 Å) > 1e (2.591 Å)
> 1g (2.579 Å), consistent with the hydrogen-bonding strength
(1)
where Δδ is given in parts per million for the difference
between chemical shift of the O−H peak of 1a−g and that of
phenol (δ 4.29).6 Accordingly, the hydrogen-bonding energy is
estimated to be 1g (11.22 ± 0.2 kcal/mol) > 1e (10.81 ± 0.2
kcal/mol) > 1f (10.28 ± 0.2 kcal/mol) > 1a (10.04 ± 0.2 kcal/
mol) > 1c (9.92 ± 0.2 kcal/mol) > 1d (9.78 ± 0.2 kcal/mol) >
1b (9.36 ± 0.2 kcal/mol). Although these values may be
overestimated due to the use of an intermolecular hydrogen-
bonding model (i.e., eq 1) for aromatic alcohols instead of a
intramolecular model (1a−g), the trend of an increase in the
hydrogen-bonding strength upon an increase of the electron-
1
estimated from H NMR measurements (vide supra). We will
elaborate in the following and later sections that the hydrogen-
bonding strength plays a role in accounting for the structure
versus luminescence relationship.
8192
dx.doi.org/10.1021/jo2012384|J. Org. Chem. 2011, 76, 8189−8202

The Journal of Organic Chemistry
Article
As for 1a, evidenced by the dihedral angle ∠N(2)−C(3)−
C(5)−C(6) = −2.39°, the X-ray structure reveals a planar
configuration between the phenol and imidazolidinone rings.
This, together with the distance O(2)−N(2) = 2.61 Å and
∠N(2)−H−O(2) = 175°, strongly supports a planar, seven-
membered-ring intramolecular hydrogen-bond formation.
Similar planar structures between phenol and imidazolidinone
rings were obtained for compounds 1b−e,g (see the
Supporting Information). Although we did not obtain an X-
ray structure for 1f, it is reasonable to assume a similar
molecular planarity. For 1a, however, the phenyl substituent at
the C(1) position was tilted by an angle of ∼37.9° with respect
to imidazolidinone (see Figure 1).
Despite a planar molecular geometry, 1a−g all revealed no
notable intermolecular π···π contacts in the solid crystal. For
example, 1a appears to possess an alternating slab in the
molecular packing and the distance between two adjacent
phenol rings, estimated by the C(6)−C(26) distance, is 5.07 Å,
while the distance between phenol (e.g., C(26)) and the
imidazolidinone (C(2)) is estimated to be 3.3 Å. The lack of
strong intermolecular π interactions is also supported by similar
spectral features of the proton-transfer emission for 1a−g in
both solution and the solid phase (vide infra). The weak π−π
interaction could be due to the alternating slabs and lack of full
π electron delocalization exerted between the phenol and
imidazolidinone moieties.
Å and angle ∠N(1)−H(1)−O(1) = 147°, strongly supports the
strong seven-membered-ring intramolecular hydrogen-bonding
formation.
In a computational approach based on the B3LYP/6-
31g(d,p) method incorporating solvation (PCM model in
cyclohexane; see the Experimental Section), the 1h-E form is
calculated to be more stable than 1h-Z by 1.08 kcal/mol,
consistent with the experimental observation. With the same
theoretical level and basis sets, we also performed the
calculation of relative energy between the Z (seven-
membered-ring) and E (eight-membered-ring) forms for o-
HBDI (1d). The results show that the 1d-Z form is more stable
than the 1d-E form by as much as 7.05 kcal/mol. Moreover, as
for the 1d-E form, the dihedral angles for C(4)−C(5)−C(6)−
O(2) and C(3)−C(4)−C(5)−C(6) are calculated to be 42.3
and −24.8°, respectively, indicating a nonplanar structure.
Evidently, for 1a−g the eight-membered-ring hydrogen bond
imposes a steric hindrance. The combination of these results
leads us to conclude that steric effects play an important role
regarding seven- versus eight-membered-ring hydrogen-bond
formation for the 1a−g.
2.3. Photophysical Properties. Figure 2A shows the
absorption and emission spectra of 1a−g in cyclohexane, while
The lack of an E isomer for 1a−g, i.e. the formation of a
C(2)O(1)- - -H−O(2) eight-membered-ring hydrogen-
bonding configuration, is of fundamental interest, which may
be attributed to two possible factors. (1) The C(2)O(1)
carbonyl group is less basic than the N(2) nitrogen and hence
forms weaker hydrogen bonds with respect to the −O(2)H
proton. (2) The CO(1)- - -H−O(2) eight-membered-ring
hydrogen-bonded configuration (E form) is sterically hindered
and hence is thermally unfavorable (see Scheme 1 for the Z and
E forms). To verify these viewpoints, we then intentionally
replaced the phenol by a pyrrolic moiety in 1a and successfully
synthesized the compounds 4-((1H-pyrrol-2-yl)methylene)-1-
methyl-2-phenyl-1H-imidazol-5(4H)-one as both E (1h-E) and
Z (1h-Z) isomers, which could be further separated via column
chromatography (see Scheme 2B and Figure S54 in the
Supporting Information). The assignment of each isomer can
be preliminarily supported by 1H NMR of the olefinic
hydrogen. It has been reported that the olefinic hydrogen
atom on the cis-NC−C6H5 type of moiety is further
downfield than that of the trans olefinic hydrogen.12
Accordingly, 1h with olefinic hydrogen peaks at 7.37 and
7.18 ppm (see Figures S43 and S45 in the Supporting
Information) is assigned to be the E form (1h-E) and Z form
(1h-Z), respectively. As a result, the ratio 1h-E:1h-Z is
calculated to be 4:5. As indicated by two downfield-shifted
proton peaks of 13.21 ppm (1h-E) and 11.12 ppm (1h-Z) in
CD2Cl2, the formation of seven- and six-membered hydrogen
bonds is evident in 1h-E and 1h-Z, respectively. Interestingly,
due to the geometry fit, the E form having a seven-membered-
ring N−H- - -O< hydrogen bond seems to be stronger than
the Z form possessing a six-membered-ring N−H- - -N
hydrogen bond. The structure of 1h-E was further confirmed
by single-crystal X-ray diffraction analysis. As depicted in Figure
1B, the nearly planar configuration between pyrrole and
imidazolidinone rings was established by ∠N(1)−H(1)−
O(1)−C(7) = −0.43° and ∠N(1)−C(4)−C(6)−C(7) =
−0.39°. This, together with the distance O(1)−H(1) = 1.71
Figure 2. Absorption and emission spectra in terms of absorption
extinction coefficient and the normalized emission spectra in
cyclohexane: (A) 1a (red □), 1b (orange ○), 1c (green △), 1d
(black ▽), 1e (blue ◇), 1f (brown ☆), and 1g (magenta ■) in
cyclohexane; (B) 2a (red □), 2b (orange ○), 2c (green △), 2d
(black ▽), 2e (blue ◇), and 2f (brown ☆).
the pertinent photophysical data are given in Table 1. The
absorption spectra are characterized by a low-lying band
maximized at 380 nm (1g)−415 nm (1a), the ε values for
which above 1.5 × 104 M−1 cm−1 make their assignments to the
π → π* transition unambiguous. In comparison to the
absorption spectra of the corresponding methoxyl derivatives,
i.e. 2a−g (see Figure 2B), the absorption peak wavelengths in
1a−g are obviously red-shifted, indicating that the formation of
a O(2)−H- - -N(2) intramolecular hydrogen bond induces
further π-electron delocalization and hence a smaller S0−S1
energy gap.
As for the steady-state emission, 1a−g all exhibit solely an
anomalously long wavelength emission (>550 nm) in cyclo-
hexane. Figure 2A clearly shows a large separation of the energy
gap between the 0−0 onset of the absorption and emission.
The Stokes shift of the emission, defined by peak (absorption)-
to-peak (emission) gap in terms of frequency, is calculated to
be >8000 cm−1 for 1a−g. To further verify the origin of the
8193
dx.doi.org/10.1021/jo2012384|J. Org. Chem. 2011, 76, 8189−8202

The Journal of Organic Chemistry
Article
Table 1. Photophysical Properties of 1a−g and Their Methoxy Derivatives 2a−g
λabs
λem
relaxation dynamics
Φ
λabs
λem
relaxation dynamics
(ps)
Φ
a
solvent
C6H12
(nm)
(nm)
(ps)
(×10−3
)
solvent
C6H12
(nm)
(nm)
(×10−3
)
b
1a
415
413
410
390
425
420
410
395
400
397
390
380
670
660
645
465
655
660
645
455
610
615
625
440
τ1 = 0.07 (0.41)
0.08
0.11
0.17
7.3
1d
385
386
383
372
385
382
380
373
383
380
376
370
605
603
602
425
578
578
570
423
575
573
570
422
τ1 = 0.11 (0.12)
τ2 = 31.9 (0.88)
τ1 = 0.26 (0.23)
τ2 = 16.6 (0.77)
τ1 = 0.39 (0.27)
τ2 = 7.8 (0.73)
τ1 = 0.63 (0.55)
τ2 = 4.1 (0.45)
τ1 = 0.71 (0.03)
τ2 = 835 (0.97)
τ1 = 0.47 (0.13)
τ2 = 177 (0.87)
τ1 = 0.14 (0.31)
τ = 20.9 (0.69)
τ1 = 0.40 (0.34)
τ2 = 6.7 (0.66)
τ1 = 0.85 (0.07)
τ2 = 247 (0.93)
τ1 = 0.32 (0.09)
τ2 = 66.8 (0.91)
τ1 = 0.43 (0.25)
τ2 = 17.0 (0.75)
τ1 = 0.21 (0.10)
τ2 = 30.4 (0.90)
τ = 2000
3.1
τ2 = 3.7 (0.59)
τ1 = 0.10 (0.47)
τ2 = 3.0 (0.53)
τ1 = 0.49 (0.52)
τ2 = 2.4 (0.48)
τ1 = 0.14 (0.13)
τ2 = 30.7 (0.87)
τ1 = 0.18 (0.29)
τ2 = 8.8 (0.71)
CH2Cl2
CH3CN
C6H12
CH2Cl2
CH3CN
C6H12
2.2
1.5
0.5
80
20
3.1
24
20
5
2a
1b
2d
1e
C6H12
0.35
0.28
1.6
C6H12
CH2Cl2
CH3CN
C6H12
CH2Cl2
CH3CN
C6H12
2b
1c
τ1 = 1.3 (0.02)
τ2 = 339 (0.98)
τ1 = 0.72 (0.06)
τ2 = 81.4 (0.94)
τ1 = 0.22 (0.23)
τ2 = 35.4 (0.77)
τ1 = 0.19 (0.26)
τ2 = 14.3 (0.74)
τ1 = 0.24 (0.27)
τ2 = 28.6 (0.73)
64
2e
1f
C6H12
4.6
C6H12
CH2Cl2
CH3CN
C6H12
2.1
CH2Cl2
CH3CN
C6H12
1.1
1.2
10
2c
9.2
2f
1g
C6H12
380
382
567
570
123
57
CH2Cl2
τ = 620
a
b
The relaxation dynamics pumped at 400 nm and monitored at the emission peak maximum. Data in parentheses indicate the fitted pre-exponential
factor, which was normalized to 1. For 1b in polar solvents, please see the text for explanation.
emission, the methoxy derivatives of 1a−g, i.e. 2a−g (see
Scheme 2), were also investigated. Owing to their lack of a
hydroxyl proton, 2a−g serve as models to represent the
prohibition of the proton transfer reaction. As depicted in
Figure 2B, for 2a−g, the S0 → S1 absorption versus the
corresponding emission features reveals a good mirror image
with a normal peak-to-peak Stokes shift of <4000 cm−1.
Accordingly, the assignment of ∼560−670 nm emission for
1a−g in cyclohexane to a proton-transfer tautomer emission is
unambiguous, and ESIPT takes place from the phenolic proton
(O(2)−H) to the N(2) nitrogen, forming the zwitterionic
species depicted in Scheme 5. The lack of any normal emission
Scheme 5
Figure 3. Absorption and emission spectra of 1h-E (black) and 1h-Z
(red).
for 1a−g in a steady-state manner indicates that the rate of
ESIPT must be ultrafast, as evidenced by the recent study of
1d11 and the reaction dynamics of the rest of the title
compounds elaborated in the next section.
definitive explanation, we tentatively propose that the
prohibition of ESIPT in 1h might be due to the weakening
of acidity at the pyrrolic hydrogen. We will have more
discussion on this in section 2.5.
Also, interestingly, despite the formation of an intramolecular
hydrogen bond, ESIPT is prohibited in both 1h-E and 1h-Z
forms, as indicated by the normal Stokes-shifted emission λmax
470−480 nm with respect to the absorption peak: 410 nm for
1h-Z and 415 nm for 1h-E (see Figure 3), while no proton
transfer emission can be resolved. Though it is pending a
2.4. ESIPT and Relaxation Dynamics. The relaxation
dynamics of 1a−g have been measured with the femtosecond
fluorescence upconversion in combination with time-correlated
single photon counting techniques in various aprotic solvents.
For the representative data, the best fitted time constants of the
relaxation dynamics monitored at near the peak of the emission
8194
dx.doi.org/10.1021/jo2012384|J. Org. Chem. 2011, 76, 8189−8202

The Journal of Organic Chemistry
Article
are given in Table 1. A fast time-resolved profile (<10 ps) using
1f as a prototype in cyclohexane is depicted in Figure 4.
indicates that the excited state charge transfer character is rather
small or even negligible. Thus, dynamic Stokes shifts should not
play a role in the observed relaxation dynamics of the title
compounds. Accordingly, the common ∼1 ps decay component
is attributed to vibrational relaxation coupled with solvent
collisional deactivation due to an exergonic ESIPT reaction.
The short viscosity-sensitive population decay is intriguing.
Since the population decay and the steady-state emission yield
are independent of the O−D substitution, the quenching
process associated with high-frequency O−H vibration motions
can be discarded. Instead, the dominant quenching process has
been concluded to originate from one-bond (the exo C(5)−
C(4)−C(3) bond) flip cis−trans isomerization diabatically
from the excited-state cis tautomer to the ground-state trans
tautomer. A comprehensive discussion of the viscosity-depend-
ent lifetimes has been elaborated in our recent report.11 This
viewpoint is also supported qualitatively by the increase in the
population lifetimes upon increasing the solvent viscosity for
other title complexes (see Table 1). It is also noteworthy that
the decay dynamics of 1b in polar solvents is very complicated,
involving multiple decay components. Due to the −OCH3
electron donating group, the hydrogen-bonding strength of 1b
is substantially weaker than those of the other derivatives (vide
supra). The weak hydrogen bonding strength, on the one hand,
indicates the possible E/Z equilibrium in the ground state for
1b in solution. On the other hand, for such a weak hydrogen
bonding system, excited-state deprotonation to the surrounding
solvent rather than intramolecular proton transfer may take
place in a polar medium, forming the anionic species.
Overlapping among E, Z, and anionic forms seriously
complicates the results and analyses. The relevant discussion
is complicated and deviates from the core issue of this study.
Thus, only the lifetime of 1b in cyclohexane is reported in
Table 1.
Figure 4. Fluorescence decay (black) and theoretical fitting (red) of 1f
at (a) 490 nm and (b) 580 nm, cross-correlation traces between the
excitation and the gate pulse (blue) at (c) 620 nm and (d) 660 nm,
and (e) deuterium substitution decay dynamics at 580 nm recorded in
cyclohexane. The excitation wavelength was set to be 400 nm.
Upon monitoring at the region of e.g. 490 nm, presumably
ascribed to the normal emission that could not be resolved by
steady-state means, the upconversion signal consists of very fast
rise and decay dynamics. For comparison, an experimental trace
of the pump (400 nm) and gate (800 nm) cross-correlation
function was also shown in Figure 4, which gives an fwhm of a
Gaussian shape-like response profile of ∼180 fs and within
experimental error is indistinguishable with that obtained from
the 490 nm upconversion signal. Figure 4 also reveals the time-
dependent upconversion signal of the tautomer emission for 1f
as a function of the monitored emission wavelengths of >550
nm.
In general, independent of the monitored wavelength, the
temporal resolution of the tautomer emission consists of both
ultrafast rise and decay components, i.e. a spike, the shape of
which is indistinguishable from that obtained from e.g. 385
(pump) and 770 nm (gate) cross-correlation functions (not
shown here), followed by a small decay component of ∼0.9 ps
and a population decay time of 10 ps. The population decay
time was further resolved to be 247 ps in cyclohexane by time-
correlated single photon counting techniques. The population
decay is strongly solvent dependent: 67 ps in CH2Cl2 and 18 ps
in CH3CN (see Table 1). Similar relaxation patterns, i.e.
consisting of a spike, an ∼1 ps decay component, and a
population decay of several tens of picoseconds, are resolved
for other title compounds upon monitoring at the peak of the
tautomer emission. Using an ultrashort pulse (25 fs) excitation,
our recent time-resolved studies of o-HBDI (1d)11 have
concluded that the spike can be further resolved, which
shows a coherent vibration motion of low-frequency bending
vibrations associated with hydrogen bonds. The ultrafast (25 fs)
fluorescence upconversion experiments of 1a−g are not
relevant to the goals of this study, and the results will be
published elsewhere. Also, as indicated by the only slight
solvent polarity dependent shift of the emission spectra, the
charge transfer character of tautomer emission for these o-
HBDI derivatives is slim. This has been observed and discussed
for compound 1d in our recent report.11 In another example,
the emission peak wavelength of 1f is measured to be 575 and
570 nm in cyclohexane and acetonitrile, respectively. Such a
small and even slightly blue shift of the emission maximum
The population decay time in cyclohexane correlates well
with the weak to moderate steady-state tautomer emission, for
which the quantum yield ranges from (8.1 ± 2) × 10−5 for 1a
to (2.0 ± 0.2) × 10−1 for 1g. As observed in several derivatives
of the core chromophore of GFP, p-HBDI, the population
decay is even faster, which has been correlated with an efficient
quenching process invoking similar cis−trans isomerization
mechanisms.7 However, in comparison to the Φf value of
<10−4 for p-HBDI, the substantially higher tautomer emission
yield for the title compounds, except for 1a, may be attributed
to intramolecular hydrogen bond formation, which in part
hinders the exocyclic torsional deformation such that the
radiationless deactivation is reduced. The rather low yield of
(8.1 ± 2) × 10−5 for 1a in cyclohexane may be rationalized by
its longest wavelength tautomer emission at 670 nm, such that
any additional quenching is possible due to the rather low
energy gap; for example, the energy gap law, i.e. the quenching
of emission by high-frequency vibration overtones and/or low-
frequency motions associated with hydrogen bonds, may be
operative.13
The quenching mechanism invoking rotation around the
exo-C(5)−C(4)−C(3) bond incorporates large-amplitude
motion and should be greatly inhibited in rigid media. Support
of this viewpoint is given by the intense tautomer emission
observed in the solid state for 1a−g. As shown in Figure 5 and
Table 2, depending on the electronic properties and position of
the substituent, a wide range of tautomer emission can be tuned
from 670 nm (1a) to 560 nm (1g) with a quantum yield as
high as 0.1−0.9 in the solid film. The naked-eye view of the
8195
dx.doi.org/10.1021/jo2012384|J. Org. Chem. 2011, 76, 8189−8202

The Journal of Organic Chemistry
Article
Figure 5. Solid-state emission spectra of 1a (red □), 1b (orange ○),
1c (green △), 1d (black ▽), 1e (blue ◇), and 1f (purple ☆).
Table 2. Photophysical Properties of 1a−g Recorded in the
a
Solid State
compd
substituent
λem (nm)
ΦF
τobsd (ns)
τr (ns)
1a
1b
1c
1d
1e
1f
−Ph
−OMe
−Br
649
643
605
595
570
560
560
0.09
0.19
0.59
0.4
0.5
0.7
3.5
1.7
6
5.9
3.7
5.9
−H
4.3
−CN
−CF3
−NO2
0.84
0.85
0.9
7.1
6.3
9.2
7.4
1g
10.2
a
τr stands for the radiative decay time and is calculated by τr = τobsd
/
ΦF, in which τobsd is the observed decay time and ΦF denotes the
emission yield.
emission color for 1b,d,f in the solid state is also demonstrated
in the abstract and table of contents artwork. The high emission
yield also correlates well with the rather long population decay
time of subnanoseconds to nanoseconds for 1a−g in the solid
state (see Table 2).
2.5. Computation and Electrochemistry. Supplementa-
ry support of the above structural and photophysical properties
and their relationship are further provided by a computational
approach. Theoretical confirmation of the underlying basis for
the photophysical properties of 1a−g and their tautomerized
isomers are provided by Hartree−Fock calculations (see the
Experimental Section). The optimized geometries of 1a−g all
demonstrated the existence of a seven-membered-ring intra-
molecular hydrogen bond between −OH and the N(2) atom
(vide supra). On the basis of the TDB3LYP/6-31++G(d′,p′)
method, the S0 → S1 transition of 1a−g can be mainly
attributed to a HOMO → LUMO transition, for which
HOMOs and LUMOs for the representative 1b, 1d, 1e, and 1f
are drawn to exemplify the electron donor OCH3, −H, and
electron acceptors CN and CF3 as the substituents, respectively,
at the R1 position (see Scheme 2A). Furthermore, the Franck−
Condon type of emission can also be calculated on the basis of
the TDB3LYP/HF method. Accordingly, HOMOs and
LUMOs contributing to the S0′ → S1′ (the prime denotes the
proton-transfer tautomer species) transition of the tautomer
also can be calculated for 1a−g, among which the
representative 1d is depicted in the last row of Figure 6.
In agreement with the aforementioned absorption spectros-
copy, both S0 → S1 and S0′ → S1′ types of transitions can be
clearly ascribed to a π → π* transition. For both transitions, the
electron density was mainly located in the phenol fragment in
the HOMO, while it is largely located at the imidazolone
Figure 6. HOMOs and LUMOs involved in the S0 → S1 transition for
1b,d−f and the tautomer of 1d.
moiety for the LUMO. Intuitively, adding an electron-donating
group at the benzene (imidazolone) moiety causes an increase
of energy in the HOMO (LUMO), hence the decrease
(increase) of the lowest lying transition. Vice versa, adding an
electron-withdrawing substituent should cause the inverse
effect.
The above viewpoint is also supported experimentally by
electrochemical measurements, in which the oxidation potential
(HOMO) of 1a−g was measured by cyclic voltammetry (CV)
with ferrocene as the reference. The LUMO energy level of 1f
was then calculated by adding the lowest energy UV−vis
absorption gap onto the HOMO level. As given in Table 3, it is
evident that substituents with electron-donating nature at the
C(9) position, such as −OCH3 in 1b, give a HOMO value of
−5.55 eV, which is higher than that (−5.75 eV) of 1d. For 1g
possessing a strong −NO2 electron-withdrawing nature at the
C(9) position, the HOMO value of −6.15 eV is the lowest
among 1a−g. Conversely, in comparison to the LUMO of
−3.00 eV in nonsubstituted 1d, adding a phenyl group at the
C(1) position in 1a decreases the LUMO energy to −3.32 eV
simply due to the elongation of the resonance effect in LUMO.
The results firmly support the original synthetic strategy aimed
8196
dx.doi.org/10.1021/jo2012384|J. Org. Chem. 2011, 76, 8189−8202

The Journal of Organic Chemistry
Article
a
7), which corresponds well to the thin film PL spectrum shown
in Figure 2. In our preliminary test, the device revealed a
maximum brightness of 5790 cd m−2 at 17.5 V (1610 mA
cm−2) with the CIE coordinates of (0.55, 0.43). As Figure 7
Table 3. Electrochemical Properties of 1a−g
Eox (V vs Ag/
HOMO
c
LUMO
c
energy gap
b
d
compd
AgCl)
(eV)
(eV)
(eV)
1a
1b
1c
1d
1e
1f
1.47
1.17
1.3S
1.37
1.52
1.53
1.77
−5.85
−5.55
−5.76
−5.75
−5.90
−5.91
−6.15
−3.32
−3.11
−3.15
−3.00
−3.05
−3.01
−3.00
2.53
2.44
2.61
2.75
2.85
2.90
3.15
1g
a
Values obtained with 1.0 × 10 M solutions of 1a−g in acetonitrile
using 0.10 M tetrabutylanmonium pcrchlorate (TBAP) as supporting
b
electrolyte. The Ag/AgNO3 reference electrode was calibrated using
c
Fc/Fc+. The energy levels of the LUMO were calculated using the
+
equations EHOMO (eV) = −4.88 − (Eox − EFc/Fc ) and ELUMO (eV) =
EHOMO + Eg. The energy gap (Eg) is calculated from the absorption
d
onset of each compound.
Figure 7. External quantum (ηext) and power efficiencies (η p) as a
function of brightness. Inset: EL spectrum of the device.
at fine-tuning the HOMO and LUMO energetics and hence the
luminescence energy gap.
shows, the maximum external quantum efficiency (η ext) and
power efficiency (ηp) were 0.40% (0.82 cd A−1) and 0.24 lm
W−1, respectively (also see Table S1 in the Supporting
Information). We believe that the efficiency of this o-
hydroxy-GFP device can be further improved by optimizing
device configurations. Work regarding full optimization of
devices fabricated with 1a−g is in progress.
Figure 6 also depicts the HOMO and LUMO for both 1h-E
and 1h-Z. Upon careful examination, one can note that upon
replacing the phenol by the pyrrolic moiety, forming either 1h-
E or 1h-Z, the electron density of HOMO in 1h is mainly
located at the π-conjugated system spreading from the pyrrolic
to the immidazole moieties, whereas the pyrrolic N atom is not
involved. This may result in a negligible increase of acidity of
the pyrrolic proton upon excitation, rationalizing the
prohibition of ESIPT for both 1h-E and 1h-Z forms (vide
supra).
3. CONCLUSION
In summary, we have reported the synthesis, characterization,
and corresponding electrochemical and photophysical proper-
ties of a series of o-hydroxy analogues, compounds 1a−g, of the
green fluorescent protein chromophore. All title compounds
possess a Z conformation and undergo a remarkable ESIPT
reaction via a pre-existing seven-membered-ring hydrogen
bond. In comparison to the natural p-hydroxy-GFP core
chromophore, i.e. p-HBDI, the intramolecular hydrogen bond
in 1a−g plays a key role in suppressing the nonradiative
deactivation pathway associated with cis (Z form)−trans (E
form) isomerization.11 Depending on the electron donating/
withdrawing or resonance properties of the substituents, the
proton transfer emission can be fine-tuned from 560 nm (1g)
to 670 nm (1a). Although the tautomer emission yield is weak
to moderate in solution, due to the lack of rotation at the exo
C−C bond, the quantum yield is as high as 0.1−0.9 for the title
compounds in the solid state. Fabrication of OLEDs using 1d
as a prototype has been successfully carried out, demonstrating
its latent potential in recently popular proton-transfer types of
lighting sources.16 Synthetic approaches with other relevant
analogues are feasible, as supported by another series of pyrrolic
analogues 1h (see Scheme 2). Also, one can envisage that
further substitution at the R2 site (see Scheme 2A) by e.g.
imidazole may lead to the derivation of a series of o-hydroxy red
fluorescent protein (RFP) core chromophores (the Kaede
dye).1a−d We also anticipate that the radiationless quenching
process may be further reduced by anchoring bulky groups at
the C(4) position (see Scheme 1), generating a series of
isomers of p-HBDI with remarkable ESIPT properties and
intense proton-transfer tautomer emission suitable for lighting
applications.
We then simply utilize the (lowest) absorption and emission
peak frequencies as the first excitation energy gap for normal
and proton-transfer tautomer, respectively. These values are
added to the calculated relative energy between normal and
tautomer species in the ground state. The results indicate that
ESIPT for 1a−g is a thermodynamically favorable process. For
example, as depicted in Figure S55 (see the Supporting
Information), ESIPT is estimated to be 8.82 kcal/mol thermally
favorable for the case of 1d. Moreover, upon Franck−Condon
excitation and execution of the geometry relaxation (see the
Experimental Section), the TDDFT method could not locate
the energy minimum of the excited normal species, the result of
which is consistent with a barrierless, perhaps coherent type of
ESIPT process concluded experimentally using 1d.11
2.6. Fabrication of OLEDs. To test the applicability of the
titled o-hydroxy GFP chromophores as emitters in electro-
luminescence (EL),14 we then selected 1d as the dopant with
the structures of ITO/PEDOT:PSS (30 nm)/NPB (20 nm)/
TCTA (5 nm)/CBP:1d (5 wt %, 25 nm)/TPBI (50 nm)/LiF
(0.5 nm)/Al (100 nm). We used 4,4′-N,N-dicarbazolyl-1,1′-
biphenyl (CBP) as a host that possesses suitable energy levels
(HOMO/LUMO = −5.9/−2.6 eV) to confine excitons within
the guest emitter.15 Here, the conducting polymer poly-
(ethylene dioxythiophene)/poly(styrene sulfonate) (PE-
DOT:PSS) was used as the hole-injection layer, 4,4′-bis[N-(1-
naphthyl)-N-phenylamino]biphenyl (NPB) and 4,4′,4″-tri(N-
carbazolyl)triphenylamine (TCTA) were used as hole-transport
layers, 1,3,5-tris(N-phenylbenzimidizol-2-yl)benzene (TPBI)
was used as an electron-transport and hole-blocking layer, LiF
was used as an electron-injection layer, and Al was used as a
cathode, respectively. The device configuration is depicted in
Figure S56 of the Supporting Information. Consequently, an
orange EL spectrum from 1d was obtained (the inset of Figure
8197
dx.doi.org/10.1021/jo2012384|J. Org. Chem. 2011, 76, 8189−8202

The Journal of Organic Chemistry
4. EXPERIMENTAL SECTION
Article
1
to afford 3b: yellow solid; yield 3.2 g (76%); mp 104−105 °C; H
NMR (400 MHz, CDCl3, ppm) 8.26 (d, J = 2.4 Hz, 1H), 7.67 (s, 1H),
6.93 (dd, J1 = 8.8 Hz, J2 = 2.4 Hz, 1H), 6.81 (d, J = 8.8 Hz, 1H), 3.83
(s, 3H), 3.80 (s, 3H), 2.36 (s, 3H); 13C NMR (100 MHz, CDCl3,
ppm) 167.9, 165.4, 153.9, 153.4, 131.9, 125.3, 122.6, 119.2, 116.5,
111.9, 56.1, 55.8, 15.6; MS (EI, 70 eV) m/z (relative intensity) 247
(M+, 100); HRMS calcd for C13H13O4N 247.0845, found 247.0841.
Anal. Calcd for C13H13O4N: C, 63.15; H, 5.30; N, 5.67. Found: C,
63.01; H, 5.36; N, 5.81.
4.1. Synthesis. The synthetic route of 1d (o-HBDI) has been
described in our previous report and is thus omitted from the
following synthetic elaboration.10
4.2. Physical and Spectroscopic Characterization Data for
Compounds 1a−g, 1h-Z, and 1h-E. 4.2.1. (Z)-4-(2-Methox-
ybenzylidene)-2-phenyloxazol-5(4H)-one (3a). Hippuric acid (5.0 g,
27.9 mmol), sodium acetate (2.3 g, 28.0 mmol), o-anisaldehyde (3.8 g,
27.9 mmol), and acetic anhydride (50 mL) were heated at 80 °C with
stirring for 4 h. After the solvent was removed, the crude product was
purified by silica gel column chromatography with ethyl acetate/n-
hexane (1/2) as eluent to afford 3a: yellow solid; yield 5.5 g (71%);
mp 110−111 °C; 1H NMR (400 MHz, CD2Cl2, ppm) 8.89 (d, J = 8.8
Hz, 1H), 8.20−8.17 (m, 2H), 7.82 (s, 1H), 7.66 (t, J = 7.4 Hz, 1H),
7.58 (m, 2H), 7.48 (t, J = 8.6 Hz, 1H), 7.13 (t, J = 7.6 Hz, 1H), 6.99
(d, J = 8.4 Hz, 1H), 3.95 (s, 3H); 13C NMR (100 MHz, CD2Cl2, ppm)
167.5, 163.1, 159.3, 133.1, 132.9, 132.7, 132.5, 128.9, 128.1, 125.8,
125.3, 122.5, 120.8, 110.8, 55.7; MS (EI, 70 eV) m/z (relative
intensity) 279 (M+, 100); HRMS calcd for C17H13O3N 279.0895,
found 279.0891. Anal. Calcd for C17H13O3N: C, 73.11; H, 4.69; N,
5.02. Found: C, 73.37; H, 4.75; N, 5.22.
4.2.2. (Z)-4-(2-Methoxybenzylidene)-1-methyl-2-phenyl-1H-imi-
dazol-5(4H)-one (2a). (Z)-4-(2-Methoxybenzylidene)-2-phenyloxa-
zol-5(4H)-one (3a; 1.0 g, 3.6 mmol) was added to potassium
carbonate (0.5 g, 3.6 mmol) in a 50 mL round-bottom flask, in which
20 mL of ethanol (95%) and 1.0 mL of methylamine (40% aqueous)
were then added. The reaction mixture was refluxed for 4 h. After
cooling, the mixture was neutralized with 10% HCl and extracted with
CH2Cl2 (3 × 25 mL). After solvent was removed, the crude product
was purified by silica gel column chromatography with ethyl acetate/n-
hexane (1/2) as eluent to afford 2a: yellow solid; yield 0.7 g (67%);
mp 126−127 °C; 1H NMR (400 MHz, CDCl3, ppm) 8.92 (d, J = 7.6
Hz, 1H), 7.85−7.83 (m, 3H), 7.56−7.52 (m, 3H), 7.37 (t, J = 8.4 Hz,
1H), 7.04 (t, J = 7.6 Hz, 1H), 6.91 (d, J = 8.3 Hz, 1H), 3.90 (s, 3H),
3.37 (s, 3H); 13C NMR (100 MHz, CDCl3, ppm) 171.7, 161.8, 159.3,
138.4, 133.3, 131.9, 131.3, 129.5, 128.7, 128.6, 123.4, 122.9, 120.9,
110.5, 55.5, 29.0; MS (EI, 70 eV) m/z (relative intensity) 292 (M+,
100); HRMS calcd for C18H16O2N2 292.1212, found 292.1219. Anal.
Calcd for C18H16O2N2: C, 73.95; H, 5.52; N, 9.58. Found: C, 73.69;
H, 5.56; N, 9.80.
4.2.3. (Z)-4-(2-Hydroxybenzylidene)-1-methyl-2-phenyl-1H-imi-
dazol-5(4H)-one (1a). (Z)-4-(2-Methoxybenzylidene)-1-methyl-2-
phenyl-1H-imidazol-5(4H)-one (2a; 300 mg, 1.0 mmol) was dissolved
in 10 mL of dichloromethane in a 50 mL round-bottom flask, and the
flask was placed in an ice bath at 0 °C. A solution of boron tribromide
(4.0 mL, 1.0 M solution in dichloromethane) was added carefully to
the stirred solution under a nitrogen atmosphere. After 4 h, the
reaction mixture was cooled and then hydrolyzed by careful shaking
with 10 mL of water and extracted twice with 10 mL of
dichloromethane. The combined organic phases were then dried
over magnesium sulfate, filtered, and evaporated in vacuo, and the
crude product was purified by silica gel column chromatography with
ethyl acetate/n-hexane (1/2) as eluent to afford 1a: yellow solid; yield
265 mg (94%); mp 150−151 °C; 1H NMR (400 MHz, CD2Cl2, ppm)
13.94 (s, 1H), 7.84−7.81 (m, 2H), 7.66−7.57 (m, 3H), 7.38−7.35 (m,
2H), 7.28 (s, 1H), 6.94 (d, J = 8.5 Hz, 1H), 6.88 (t, J = 7.4 Hz, 1H),
3.39 (s, 3H); 13C NMR (100 MHz, CD2Cl2, ppm) 168.5, 158.7, 157.6,
136.4, 134.2, 133.3, 132.0, 130.7, 129.0, 128.3, 127.8, 119.9, 119.3,
119.0, 29.1; MS (EI, 70 eV) m/z (relative intensity) 278 (M+, 100);
HRMS calcd for C17H14O2N2 278.1055, found 278.1049. Anal. Calcd
for C17H14O2N2: C, 73.37; H, 5.07; N, 10.07. Found: C, 73.21; H,
5.11; N, 10.29. Yellow needle-shaped crystals suitable for the
crystallographic studies reported here were isolated over a period of
6 weeks by slow evaporation from the chloroform solution.
4.2.5. (Z)-4-(2,5-Dimethoxybenzylidene)-1,2-dimethyl-1H-imida-
zol-5(4H)-one (2b). (Z)-4-(2,5-Dimethoxybenzylidene)-2-methyloxa-
zol-5(4H)-one (3b; 1.2 g, 4.8 mmol) was added to potassium
carbonate (0.7 g, 5.1 mmol) in a 50 mL round-bottom flask, in which
15 mL of ethanol (95%) and 1.0 mL of methylamine (40% aqueous)
were then added. The reaction mixture was refluxed for 4 h. After
cooling, the mixture was neutralized with 10% HCl and extracted with
CH2Cl2 (3 × 25 mL). After solvent was removed, the crude product
was purified by silica gel column chromatography with ethyl acetate/n-
hexane (1/1) as eluent to afford 2b: yellow solid; yield 0.81 g (65%);
mp 118−119 °C; 1H NMR (400 MHz, CDCl3, ppm) 8.39 (d, J = 3.0
Hz, 1H), 7.58 (s, 1H), 6.87 (dd, J1 = 9.0 Hz, J2 = 3.0 Hz, 1H), 6.78 (d,
J = 9.1 Hz, 1H), 3.79 (s, 3H), 3.77 (s, 3H), 3.12 (s, 3H), 2.29 (s, 3H);
13C NMR (100 MHz, CDCl3, ppm) 170.6, 161.9, 153.8, 153.4, 138.3,
123.6, 120.8, 117.8, 117.0, 111.7, 56.1, 55.7, 26.4, 15.6; MS (EI, 70 eV)
m/z (relative intensity) 260 (M+, 54), 229 (100); HRMS calcd for
C14H16O3N2 260.1161, found 260.1156. Anal. Calcd for C14H16O3N2:
C, 64.60; H, 6.20; N, 10.76. Found: C, 64.42; H, 6.26; N, 10.92.
4.2.6. (Z)-4-(2-Hydroxy-5-methoxybenzylidene)-1,2-dimethyl-1H-
imidazol-5(4H)-one (1b). (Z)-4-(2,5-Dimethoxybenzylidene)-1,2-di-
methyl-1H-imidazol-5(4H)-one (2b; 286 mg, 1.1 mmol) was dissolved
in 10 mL of dichloromethane in a 50 mL round-bottom flask, and the
flask was placed in an ice bath at 0 °C. A solution of boron tribromide
(4.4 mL, 1.0 M solution in dichloromethane) was added carefully to
the stirred solution under a nitrogen atmosphere. After 4 h, the
reaction mixture was cooled and then hydrolyzed by careful shaking
with 10 mL of water and extracted twice with 10 mL of
dichloromethane. The combined organic phases were then dried
over magnesium sulfate, filtered, and evaporated in vacuo, and the
crude product was purified by silica gel column chromatography with
ethyl acetate/n-hexane (1/1) as eluent to afford 1b: yellow solid; yield
241 mg (89%); mp 146−147 °C; 1H NMR (400 MHz, CD2Cl2, ppm)
13.26 (s, 1H), 7.09 (s, 1H), 6.97 (dd, J1 = 8.9 Hz, J2 = 3.0 Hz, 1H),
6.88 (d, J = 8.9 Hz, 1H), 6.76 (d, J = 3.0 Hz, 1H), 3.76 (s, 3H), 3.20
(s, 3H), 2.36 (s, 3H); 13C NMR (100 MHz, CD2Cl2, ppm) 167.9,
157.7, 152.8, 152.2, 133.2, 129.8, 121.7, 120.0, 119.4, 118.3, 55.7, 26.7,
15.1; MS (EI, 70 eV) m/z (relative intensity) 246 (M+, 100); HRMS
calcd for C13H14O3N2 246.1004, found 246.1008. Anal. Calcd for
C13H14O3N2: C, 63.40; H, 5.73; N, 11.38. Found: C, 63.26; H, 5.77;
N, 11.56. Yellow needle-shaped crystals suitable for the crystallo-
graphic studies reported here were isolated over a period of 5 weeks by
slow evaporation from the chloroform solution.
4.2.7. (Z)-4-(5-Bromo-2-methoxybenzylidene)-2-methyloxazol-
5(4H)-one (3c). N-Acetylglycine (2.0 g, 17.1 mmol), sodium acetate
(1.4 g, 17.1 mmol), 5-bromo-2-methoxybenzaldehyde (3.6 g, 16.7
mmol), and acetic anhydride (40 mL) were heated at 80 °C with
stirring for 4 h. After solvent was removed, the crude product was
purified by silica gel column chromatography with ethyl acetate/n-
hexane (1/2) as eluent to afford 3c: yellow solid; yield 3.6 g (71%);
mp 145−146 °C; 1H NMR (400 MHz, CDCl3, ppm) 8.75 (d, J = 1.7
Hz, 1H), 7.57 (s, 1H), 7.46 (dd, J1 = 8.4 Hz, J2 = 1.7 Hz, 1H), 6.78 (d,
J = 8.8 Hz, 1H), 3.86 (s, 3H), 2.39 (s, 3H); 13C NMR (100 MHz,
CDCl3, ppm) 167.6, 166.3, 158.0, 135.0, 134.7, 132.7, 124.0, 123.5,
113.3, 112.4, 55.9, 15.6; MS (EI, 70 eV) m/z (relative intensity) 295
(M+, 70), 225 (100); HRMS calcd for C12H10BrO3N 294.9844, found
294.9848. Anal. Calcd for C12H10O3NBr: C, 48.67; H, 3.40; N, 4.73.
Found: C, 48.49; H, 3.42; N, 4.87.
4.2.4. (Z)-4-(2,5-Dimethoxybenzylidene)-2-methyloxazol-5(4H)-
one (3b). N-Acetylglycine (2.0 g, 17.1 mmol), sodium acetate (1.4
g, 17.1 mmol), 2,5-dimethoxybenzaldehyde (2.8 g, 16.8 mmol), and
acetic anhydride (40 mL) were heated at 80 °C with stirring for 4 h.
After solvent was removed, the crude product was purified by silica gel
column chromatography with ethyl acetate/n-hexane (1/2) as eluent
4.2.8. (Z)-4-(5-Bromo-2-methoxybenzylidene)-1,2-dimethyl-1H-
imidazol-5(4H)-one (2c). (Z)-4-(5-Bromo-2-methoxybenzylidene)-2-
methyloxazol-5(4H)-one (3c; 1.3 g, 4.4 mmol) was added to
potassium carbonate (0.7 g, 5.1 mmol) in a 50 mL round-bottom
8198
dx.doi.org/10.1021/jo2012384|J. Org. Chem. 2011, 76, 8189−8202

The Journal of Organic Chemistry
Article
flask, in which 15 mL of ethanol (95%) and 1.0 mL of methylamine
(40% aqueous) were then added. The reaction mixture was refluxed
for 4 h. After cooling, the mixture was neutralized with 10% HCl and
extracted with CH2Cl2 (3 × 25 mL). After solvent was removed, the
crude product was purified by silica gel column chromatography with
ethyl acetate/n-hexane (1/1) as eluent to afford 2c: yellow solid; yield
C14H13O2N3 255.1008, found 255.1010. Anal. Calcd for C14H13O2N3:
C, 65.87; H, 5.13; N, 16.46. Found: C, 65.65; H, 5.19; N, 16.66.
4.2.12. (Z)-3-((1,2-Dimethyl-5-oxo-1H-imidazol-4(5H)-ylidene)-
methyl)-4-hydroxybenzonitrile (1e). (Z)-3-((1,2-Dimethyl-5-oxo-
1H-imidazol-4(5H)-ylidene)methyl)-4-methoxybenzonitrile (2e; 280
mg, 1.1 mmol) was dissolved in 10 mL of dichloromethane in a 50 mL
round-bottom flask, and the flask was placed in an ice bath at 0 °C. A
solution of boron tribromide (4.4 mL, 1.0 M solution in dichloro-
methane) was added carefully to the stirred solution under a nitrogen
atmosphere. After 4 h, the reaction mixture was cooled and then
hydrolyzed by careful shaking with 10 mL of water and extracted twice
with 10 mL of dichloromethane. The combined organic phases were
then dried over magnesium sulfate, filtered, and evaporated in vacuo,
and the crude product was purified by silica gel column
chromatography with ethyl acetate/n-hexane (1/1) as eluent to afford
1e: yellow solid; yield 250 mg (95%); mp 186−187 °C; 1H NMR (400
MHz, CD2Cl2, ppm) 14.70 (s, 1H), 7.62 (s, 1H), 7.55 (d, J = 8.4 Hz,
1H), 6.99 (s, 1H), 6.95 (d, J = 8.5 Hz, 1H), 3.20 (s, 3H), 2.40 (s, 3H);
13C NMR (100 MHz, CD2Cl2, ppm) 167.2, 162.0, 160.1, 140.3, 136.1,
1
0.88 g (68%); mp 157−158 °C; H NMR (400 MHz, CDCl3, ppm)
8.88 (d, J = 2.7 Hz, 1H), 7.52 (s, 1H), 7.42 (dd, J1 = 8.8 Hz, J2 = 2.4
Hz, 1H), 6.76 (d, J = 8.8 Hz, 1H), 3.86 (s, 3H), 3.17 (s, 3H), 2.37 (s,
3H); 13C NMR (100 MHz, CDCl3, ppm) 170.5, 162.8, 157.9, 138.9,
135.0, 133.8, 125.0, 119.49, 113.2, 112.2, 55.8, 26.5, 15.6; MS (EI, 70
eV) m/z (relative intensity) 308 (M+, 92), 279 (100); HRMS calcd for
C13H13BrO2N2 308.0160, found 308.0168. Anal. Calcd for
C13H13O2N2Br: C, 50.50; H, 4.24; N, 9.06. Found: C, 50.28; H,
4.40; N, 9.28.
4.2.9. (Z)-4-(5-Bromo-2-hydroxybenzylidene)-1,2-dimethyl-1H-
imidazol-5(4H)-one (1c). (Z)-4-(5-Bromo-2-methoxybenzylidene)-
1,2-dimethyl-1H-imidazol-5(4H)-one (2c; 340 mg, 1.1 mmol) was
dissolved in 10 mL of dichloromethane in a 50 mL round-bottom
flask, and the flask was placed in an ice bath at 0 °C. A solution of
boron tribromide (4.4 mL, 1.0 M solution in dichloromethane) was
added carefully to the stirred solution under a nitrogen atmosphere.
After 4 h, the reaction mixture was cooled and then hydrolyzed by
careful shaking with 10 mL of water and extracted twice with 10 mL of
dichloromethane. The combined organic phases were then dried over
magnesium sulfate, filtered, and evaporated in vacuo, and the crude
product was purified by silica gel column chromatography with ethyl
acetate/n-hexane (1/1) as eluent to afford 1c: yellow solid; yield 300
134.2, 126.3, 120.6, 120.4, 118.6, 102.6, 26.7, 15.1; MS (EI, 70 eV) m/
z (relative intensity) 241 (M+, 100); HRMS calcd for C13H11O2N3
241.0851, found 241.0847. Anal. Calcd for C13H11O2N3: C, 64.72; H,
4.60; N, 17.42. Found: C, 64.60; H, 4.64; N, 17.60. Yellow needle-
shaped crystals suitable for the crystallographic studies reported here
were isolated over a period of 5 weeks by slow evaporation from the
chloroform solution.
4.2.13. (Z)-4-(2-Methoxy-5-(trifluoromethyl)benzylidene)-2-meth-
yl-4H-oxazol-5-one (3f). N-Acetylglycine (2.0 g, 17.1 mmol), sodium
acetate (1.4 g, 17.1 mmol), 2-methoxy-5-(trifluoromethyl)-
benzaldehyde (3.5 g, 17.1 mmol), and acetic anhydride (40 mL)
were heated at 80 °C with stirring for 4 h. After solvent was removed,
the crude product was purified by silica gel column chromatography
with ethyl acetate/n-hexane (1/2) as eluent to afford 3f: yellow solid;
1
mg (92%); mp 172−173 °C; H NMR (400 MHz, CD2Cl2, ppm)
13.82 (s, 1H), 7.39 (d, J = 9.3 Hz, 1H), 7.36 (s, 1H), 7.03 (s, 1H), 6.84
(d, J = 9.3 Hz, 1H), 3.23 (s, 3H), 2.40 (s, 3H); 13C NMR (100 MHz,
CD2Cl2, ppm) 167.6, 158.4, 157.6, 137.6, 136.3, 133.5, 128.3, 121.4,
121.1, 110.8, 26.8, 15.2; MS (EI, 70 eV) m/z (relative intensity) 294
(M+, 100); HRMS calcd for C12H11BrO2N2 294.0004, found 294.0003.
Anal. Calcd for C12H11O2N2Br: C, 48.84; H, 3.76; N, 9.49. Found: C,
48.66; H, 3.80; N, 9.61. Yellow needle-shaped crystals suitable for the
crystallographic studies reported here were isolated over a period of 4
weeks by slow evaporation from the chloroform solution.
1
yield 4.0 g (82%); mp 161−162 °C; H NMR (400 MHz, CDCl3,
ppm) 8.97 (s, 1H), 7.66 (s, 1H), 7.63 (d, J = 8.8 Hz, 1H), 6.98 (d, J =
8.8 Hz, 1H), 3.96 (s, 3H), 2.42 (s, 3H); 13C NMR (100 MHz, CDCl3,
ppm) 167.2, 162.1, 136.1, 135.7, 133.8, 123.7, 122.5, 120.7, 116.7,
112.4, 111.4, 56.2, 15.6; MS (EI, 70 eV) m/z (relative intensity) 285
(M+, 100); HRMS calcd for C13H10F3NO3 285.0613, found 285.0616.
Anal. Calcd for C13H10O3NF3: C, 54.74; H, 3.53; N, 4.91. Found: C,
54.52; H, 3.61; N, 5.05.
4.2.10. (Z)-4-Methoxy-3-((2-methyl-5-oxooxazol-4(5H)-ylidene)-
methyl)benzonitrile (3e). N-Acetylglycine (2.0 g, 17.1 mmol),
sodium acetate (1.4 g, 17.1 mmol), 3-formyl-4-methoxybenzonitrile
(2.7 g, 17.4 mmol), and acetic anhydride (40 mL) were heated at 80
°C with stirring for 4 h. After solvent was removed, the crude product
was purified by silica gel column chromatography with ethyl acetate/n-
hexane (1/2) as eluent to afford 3e: yellow solid; yield 3.1 g (77%);
4.2.14. (Z)-5-(2-Methoxy-5-(trifluoromethyl)benzylidene)-2,3-di-
methyl-3,5-dihydroimidazol-4-one (2f). (Z)-4-(2-Methoxy-5-
(trifluoromethyl)benzylidene)-2-methyl-4H-oxazol-5-one (3f; 1.0 g,
3.5 mmol) was added to potassium carbonate (0.7 g, 5.1 mmol) in a
50 mL round-bottom flask, in which 15 mL of ethanol (95%) and 1.0
mL of methylamine (40% aqueous) were then added. The reaction
mixture was refluxed for 4 h. After cooling, the mixture was neutralized
with 10% HCl and extracted with CH2Cl2 (3 × 25 mL). After solvent
was removed, the crude product was purified by silica gel column
chromatography with ethyl acetate/n-hexane (1/1) as eluent to afford
1
mp 151−152 °C; H NMR (400 MHz, CDCl3, ppm) 8.98 (s, 1H),
7.65 (d, J = 8.5 Hz, 1H), 7.55 (s, 1H), 6.97 (d, J = 8.8 Hz, 1H), 3.95
(s, 3H), 2.41 (s, 3H); 13C NMR (100 MHz, CDCl3, ppm) 167.2,
161.4, 136.6, 135.8, 133.7, 123.2, 122.2, 118.7, 111.4, 104.6, 56.1, 15.7.
(one carbon nuclei could not be observed or resolved); MS (EI, 70
eV) m/z (relative intensity) 242 (M+, 100); HRMS calcd for
C13H10O3N2 242.0691, found 242.0693. Anal. Calcd for C13H10O3N2:
C, 64.46; H, 4.16; N, 11.56. Found: C, 64.34; H, 4.22; N, 11.68.
4.2.11. (Z)-3-((1,2-Dimethyl-5-oxo-1H-imidazol-4(5H)-ylidene)-
methyl)-4-methoxybenzonitrile (2e). (Z)-4-Methoxy-3-((2-methyl-
5-oxooxazol-4(5H)-ylidene)methyl)benzonitrile (3e; 1.1 g, 4.5
mmol) was added to potassium carbonate (0.7 g, 5.1 mmol) in a 50
mL round-bottom flask, in which 15 mL of ethanol (95%) and 1.0 mL
of methylamine (40% aqueous) were then added. The reaction
mixture was refluxed for 4 h. After cooling, the mixture was neutralized
with 10% HCl and extracted with CH2Cl2 (3 × 25 mL). After solvent
was removed, the crude product was purified by silica gel column
chromatography with ethyl acetate/n-hexane (1/1) as eluent to afford
2e: yellow solid; yield 0.69 g (60%); mp 162−163 °C; 1H NMR (400
MHz, CDCl3, ppm) 9.11 (s, 1H), 7.61 (d, J = 8.8 Hz, 1H), 7.48 (s,
1H), 6.95 (d, J = 8.8 Hz, 1H), 3.95 (s, 3H), 3.19 (s, 3H), 2.39 (s, 3H);
13C NMR (100 MHz, CDCl3, ppm) 170.3, 163.8, 161.4, 139.8, 136.9,
1
2f: yellow solid; yield 0.64 g (62%); mp 172−173 °C; H NMR (400
MHz, CDCl3, ppm) 9.08 (s, 1H), 7.58 (s, 1H), 7.56 (d, J = 8.8 Hz,
1H), 6.96 (d, J = 8.8 Hz, 1H), 3.93 (s, 3H), 3.19 (s, 3H), 2.38 (s, 3H);
13C NMR (100 MHz, CDCl3, ppm) 170.3, 164.3, 138.8, 137.4, 134.7,
124.6, 121.4, 119.9, 116.2, 112.1, 111.3, 56.0, 26.9, 15.6; MS (EI, 70
eV) m/z (relative intensity) 298 (M+, 100); HRMS calcd for
C14H13F3N2O2 298.0929, found 298.0934. Anal. Calcd for
C14H13O2N2F3: C, 56.38; H, 4.39; N, 9.39. Found: C, 56.52; H,
4.45; N, 9.17.
4.2.15. (Z)-5-(2-Hydroxy-5-(trifluoromethyl)benzylidene)-2,3-di-
methyl-3,5-dihydroimidazol-4-one (1f). (Z)-5-(2-Methoxy-5-
(trifluoromethyl)benzylidene)-2,3-dimethyl-3,5-dihydro-imidazol-4-
one (2f; 200 mg, 0.7 mmol) was dissolved in 10 mL of
dichloromethane in a 50 mL round-bottom flask, and the flask was
placed in an ice bath at 0 °C. A solution of boron tribromide (2.8 mL,
1.0 M solution in dichloromethane) was added carefully to the stirred
solution under a nitrogen atmosphere. After 4 h, the reaction mixture
was cooled and then hydrolyzed by careful shaking with 10 mL of
water and extracted twice with 10 mL of dichloromethane. The
134.8, 124.2, 119.1, 118.0, 111.1, 104.4, 55.9, 26.6, 15.6; MS (EI, 70
eV) m/z (relative intensity) 255 (M+, 100); HRMS calcd for
8199
dx.doi.org/10.1021/jo2012384|J. Org. Chem. 2011, 76, 8189−8202

The Journal of Organic Chemistry
Article
combined organic phases were then dried over magnesium sulfate,
filtered, and evaporated in vacuo, and the crude product was purified
by silica gel column chromatography with ethyl acetate/n-hexane (1/
1) as eluent to afford 1f: yellow solid; yield 160 mg (84%); mp 190−
4.2.19. (Z)-4-((1H-Pyrrol-2-yl)methylene)-1-methyl-2-phenyl-1H-
imidazol-5(4H)-one (1h-Z). (Z)-4-((1H-Pyrrol-2-yl)methylene)-2-
phenyloxazol-5(4H)-one12 (2h-Z; 1.0 g, 4.2 mmol) was added to
potassium carbonate (0.6 g, 4.3 mmol) in a 50 mL round-bottom flask,
in which 20 mL of ethanol (95%) and 1.0 mL of methylamine (40%
aqueous) were then added. The reaction mixture was refluxed for 3 h.
After cooling, the mixture was neutralized with 10% HCl and extracted
with CH2Cl2 (3 × 25 mL). After solvent was removed, the crude
product was purified by silica gel column chromatography with ethyl
acetate/n-hexane (1/3) as eluent to afford 1h-Z: yellow solid; yield
1
191 °C; H NMR (400 MHz, CD2Cl2, ppm) 14.18 (s, 1H), 7.68 (s,
1H), 7.56 (d, J = 8.4 Hz, 1H), 7.13 (s, 1H), 7.03 (d, J = 8.4 Hz, 1H),
3.25 (s, 3H), 2.51 (s, 3H); 13C NMR (100 MHz, CD2Cl2, ppm) 167.1,
162.3, 138.4, 137.8, 135.9, 124.6, 121.3, 119.7, 116.0, 112.0, 111.1,
26.9, 15.6; MS (EI, 70 eV) m/z (relative intensity) 284 (M+, 100);
HRMS calcd for C13H11F3N2O2 284.0773, found 284.0776. Anal.
Calcd for C13H11O2N2F3: C, 54.93; H, 3.90; N, 9.86. Found: C, 54.75;
H, 3.96; N, 9.98.
1
0.69 g (65%); mp 183−184 °C; H NMR (400 MHz, CD2Cl2, ppm)
11.12 (br, 1H), 7.79 (m, 2H), 7.57 (m, 3H), 7.18 (s, 1H), 7.11(s, 1H),
6.69 (s, 1H), 6.31 (dd, J1 = 6.0 Hz, J2 = 2.8 Hz, 1H), 3.39 (s, 3H); 13
C
4.2.16. (4Z)-4-(2-Methoxy-5-nitrobenzylidene)-2-methyloxazol-
5(4H)-one (3g). N-Acetylglycine (2.0 g, 17.1 mmol), sodium acetate
(1.4 g, 17.1 mmol), 2-methoxy-5-nitrobenzaldehyde (3.1 g, 17.1
mmol), and acetic anhydride (40 mL) were heated at 80 °C with
stirring for 4 h. After solvent was removed, the crude product was
purified by silica gel column chromatography with ethyl acetate/n-
hexane (1/2) as eluent to afford 3g: yellow solid; yield 3.2 g (71%);
mp 185−186 °C; 1H NMR (400 MHz, CDCl3, ppm) 9.52 (d, J = 2.6
Hz, 1H), 8.28 (dd, J1 = 9.2 Hz, J2 = 2.6 Hz, 1H), 7.56 (s, 1H), 6.99 (d,
J = 9.2 Hz, 1H), 4.02 (s, 3H), 2.45 (s, 3H); 13C NMR (100 MHz,
CDCl3, ppm) 167.5, 167.2, 162.9, 114.6, 134.1, 128.2, 127.6, 122.6,
122.3, 110.6, 56.6, 15.8; MS (EI, 70 eV) m/z (relative intensity) 262
(M+, 100); HRMS calcd for C12H10O5N2 262.0590, found 262.0596.
Anal. Calcd for C12H10O5N2: C, 54.97; H, 3.84; N, 10.68. Found: C,
54.75; H, 3.92; N, 10.84.
NMR (100 MHz, CD2Cl2, ppm) 170.2, 158.5, 133.5, 131.0, 129.7,
129.4, 128.8, 128.4, 125.7, 119.3, 118.8, 111.1, 28.8; MS (EI, 70 eV)
m/z (relative intensity) 251 (M+, 100); HRMS calcd for C15H13ON3
251.1059, found 251.1058. Anal. Calcd for C15H13ON3: C, 71.70; H,
5.21; N, 16.72. Found: C, 71.42; H, 5.27; N, 16.98.
4.2.20. (E)-4-((1H-Pyrrol-2-yl)methylene)-1-methyl-2-phenyl-1H-
imidazol-5(4H)-one (1h-E). (E)-4-((1H-Pyrrol-2-yl)methylene)-2-
phenyloxazol-5(4H)-one12 (2h-E; 1.0 g, 4.2 mmol) was added to
potassium carbonate (0.6 g, 4.3 mmol) in a 50 mL round-bottom flask,
in which 20 mL of ethanol (95%) and 1.0 mL of methylamine (40%
aqueous) were then added. The reaction mixture was refluxed for 3 h.
After cooling, the mixture was neutralized with 10% HCl and extracted
with CH2Cl2 (3 × 25 mL). After solvent was removed, the crude
product was purified by silica gel column chromatography with ethyl
acetate/n-hexane (1/3) as eluent to afford 1h-E: yellow solid; yield
4.2.17. (4Z)-4-(2-Methoxy-5-nitrobenzylidene)-1,2-dimethyl-1H-
imidazol-5(4H)-one (2g). (4Z)-4-(2-Methoxy-5-nitrobenzylidene)-2-
methyloxazol-5(4H)-one (3g; 1.2 g, 4.6 mmol) was added to
potassium carbonate (0.7 g, 5.1 mmol) in a 50 mL round-bottom
flask, in which 15 mL of ethanol (95%) and 1.0 mL of methylamine
(40% aqueous) were then added. The reaction mixture was refluxed
for 4 h. After cooling, the mixture was neutralized with 10% HCl and
extracted with CH2Cl2 (3 × 25 mL). After solvent was removed, the
crude product was purified by silica gel column chromatography with
ethyl acetate/n-hexane (1/1) as eluent to afford 2g: yellow solid; yield
1
0.64 g (61%); mp 196−197 °C; H NMR (400 MHz, CD2Cl2, ppm)
13.21 (br, 1H), 7.76 (m, 2H), 7.55 (m, 3H), 7.37 (s, 1H), 7.19 (s,
1H), 6.79 (s, 1H), 6.41 (dd, J1 = 5.2 Hz, J2 = 2.0 Hz, 1H), 3.41 (s,
3H); 13C NMR (100 MHz, CD2Cl2, ppm) 169.2, 155.4, 132.7, 130.6,
129.5, 129.4, 128.8, 128.2, 128.0, 125.8, 121.3, 112.2, 29.1; MS (EI, 70
eV) m/z (relative intensity) 251 (M+, 100); HRMS calcd for
C15H13ON3 251.1059, found 251.1054. Anal. Calcd for C15H13ON3: C,
71.70; H, 5.21; N, 16.72. Found: C, 71.52; H, 5.25; N, 16.90. Yellow
needle-shaped crystals suitable for the crystallographic studies reported
here were isolated over a period of 6 weeks by slow evaporation from
the chloroform solution.
1
0.79 g (62%); mp 197−198 °C; H NMR (400 MHz, CDCl3, ppm)
9.65 (d, J = 3.1 Hz, 1H), 8.22 (dd, J1 = 8.9 Hz, J2 = 3.1 Hz, 1H), 7.49
(s, 1H), 6.94 (d, J = 8.9 Hz, 1H), 3.98 (s, 3H), 3.18 (s, 3H), 2.40 (s,
3H); 13C NMR (100 MHz, CDCl3, ppm) 170.3, 164.1, 162.9, 141.5,
140.1, 128.5, 126.5, 123.7, 118.0, 110.3, 56.3, 26.6, 15.7; MS (EI, 70
eV) m/z (relative intensity) 275 (M+, 100); HRMS calcd for
C13H13O4N3 275.0906, found 275.0902. Anal. Calcd for C13H13O4N3:
C, 56.72; H, 4.76; N, 15.27. Found: C, 56.50; H, 4.80; N, 15.51.
4.2.18. (4Z)-4-(2-Hydroxy-5-nitrobenzylidene)-1,2-dimethyl-1H-
imidazol-5(4H)-one (1g). (4Z)-4-(2-Methoxy-5-nitrobenzylidene)-
1,2-dimethyl-1H-imidazol-5(4H)-one (2g; 300 mg, 1.1 mmol) was
dissolved in 10 mL of dichloromethane in a 50 mL round-bottom
flask, and the flask was placed in an ice bath at 0 °C. A solution of
boron tribromide (4.4 mL, 1.0 M solution in dichloromethane) was
added carefully to the stirred solution under a nitrogen atmosphere.
After 4 h, the reaction mixture was cooled and then hydrolyzed by
careful shaking with 10 mL of water and extracted twice with 10 mL of
dichloromethane. The combined organic phases were then dried over
magnesium sulfate, filtered, and evaporated in vacuo, and the crude
product was purified by silica gel column chromatography with ethyl
acetate/n-hexane (1/1) as eluent to afford 1g: yellow solid; yield 260
4.3. Measurements. Steady-state absorption and emission spectra
were recorded with an UV−vis spectrophotometer and a fluorometer,
respectively. Quinine sulfate with an emission yield of Φ ≈ 0.54 ± 0.02
in 1.0 N sulfuric acid solution served as the standard to calculate the
emission quantum yield. An integrating sphere was applied to measure
the quantum yield in the solid state, for which the solid thin film was
prepared via direct vacuum deposition and was excited by an argon ion
laser at ∼363 nm. The resulting luminescence was acquired with an
intensified charge-coupled detector for subsequent quantum yield
analyses according to a reported method.17
A detailed setup of femtosecond dynamical measurements has been
elaborated in our previous report.18 Briefly, the fundamental of a
Ti:sapphire laser at 750−840 nm (80 MHz, 120 fs) was used to
produce second harmonics (SH) at 375−420 nm. The resulting
fluorescence and the optical delayed remaining fundamental pulses
were collected and focused on a BBO type I crystal (0.5 mm) for the
sum−frequency generation. The upconversion signal was then
separated and detected via a photon counting PMT. The cross
correlation between SH and the fundamental had a full width at half-
maximum (fwhm) of 220 fs, which was chosen as a response function
of the system. A half-wave plate was placed in the pump beam path to
ensure that the polarization of the pump laser was set at the magic
angle (54.78°) with respect to that of the probe laser to eliminate the
fluorescence anisotropy. For picosecond lifetime measurements, a
time-correlated single photon counter (TCSPC) was used as a
detecting system. The excitation source was similar to that for the
femtosecond dynamic measurements by using SH of the Ti:sapphire
laser. The fundamental pulse was used as the trigger signal to
synchronize TCSPC. The resolution of the time-correlated photon
counting system is limited by the detector response of ∼30 ps. The
fluorescence decays were analyzed by the sum of exponential functions
1
mg (91%); mp 224−225 °C; H NMR (400 MHz, CD2Cl2, ppm)
15.11 (s, 1H), 8.29 (d, J = 2.2 Hz, 1H), 8.18 (dd, J1 = 9.1 Hz, J2 = 2.2
Hz, 1H), 7.12 (s, 1H), 7.01 (d, J = 9.1 Hz, 1H), 3.24 (s, 3H), 2.44 (s,
3H); 13C NMR (100 MHz, CD2Cl2, ppm) 167.2, 164.0, 159.5, 140.2,
134.2, 132.0, 128.6, 127.4, 120.1, 119.6, 26.9, 15.3; MS (EI, 70 eV) m/
z (relative intensity) 261 (M+, 100); HRMS calcd for C12H11O4N3
261.0750, found 261.0748. Anal. Calcd for C12H11O4N3: C, 55.17; H,
4.24; N, 16.09. Found: C, 55.01; H, 4.22; N, 16.35. Yellow needle-
shaped crystals suitable for the crystallographic studies reported here
were isolated over a period of 6 weeks by slow evaporation from the
chloroform solution.
8200
dx.doi.org/10.1021/jo2012384|J. Org. Chem. 2011, 76, 8189−8202

The Journal of Organic Chemistry
Article
with an iterative convolution method.18 The solid-state emission was
collected >45° from the pump beam to avoid the scattering line
reflecting from the sample surface. The solid film was prepared using
vacuum deposition on quartz for uniform and homogeneous surface
properties.
Academic Press: San Diego, CA, 1999. (i) Chalfie, M. Green
Fluorescent Proteins, Properties, Applications and Protocols; Wiley-Liss:
New York, 1998.
(2) (a) Agmon, N. Biophys. J. 2005, 88, 2452. (b) Stoner-Ma, D.;
Jaye, A. A.; Matousek, P.; Towrie, M.; Meech, S. R.; Tonge, P. J. J. Am.
Chem. Soc. 2005, 127, 2864. (c) Hosoi, H.; Mizuno, H.; Miyawaki, A.;
Tahara, T. J. Phys. Chem. B 2006, 110, 22853.
(3) (a) Stoner-Ma, D.; Melief, E. H.; Nappa, J.; Ronayne, K. L.;
Tonge, P. J.; Meech, S. R. J. Phys. Chem. B 2006, 110, 22009.
(b) Abbyad, P.; Childs, W.; Shi, X.; Boxer, S. G. Proc. Natl. Acad. Sci.
U.S.A. 2007, 104, 20189. (c) Stoner-Ma, D.; Jaye, A. A.; Ronayne, K.
L.; Nappa, J.; Meech, S. R.; Tonge, P. J. J. Am. Chem. Soc. 2008, 130,
1227. (d) Fang, C.; Frontiera, R. R.; Tran, R.; Mathies, R. A. Nature
2009, 462, 200.
(4) (a) Brejc, K.; Sixma, T. K.; Kitts, P. A.; Kain, S. R.; Tsien, R. Y.;
Ormo, M.; Remington, S. J. Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 2306.
(b) Litvinenko, K. L.; Webber, N. M.; Meech, S. R. J. Phys. Chem. A
2003, 107, 2616. (c) Mandal, D.; Tahara, T.; Meech, S. R. J. Phys.
Chem. B 2004, 108, 1102. (d) Usman, A.; Mohammed, O. F.;
Nibbering, E. T. J.; Dong, J.; Solntsev, K. M.; Tolbert, L. M. J. Am.
Chem. Soc. 2005, 127, 11214. (e) Dong, J.; Solntsev, K. M.; Tolbert, L.
M. J. Am. Chem. Soc. 2006, 128, 12038.
4.4. Theoretical Approach. To gain further insight into the
experimental evidence, the basis set HF/6-31++G(d′,p′) was used to
obtain the ground-state geometries of each species and their
tautomerized forms.19 Each minimum was checked by frequency
analysis to confirm that the number of imaginary frequencies is 0. On
the basis of the HF-optimized geometries, TDB3LYP/6-31++G(d′,p′)
was used to access the vertical excitation energies of each species and
their tautomerized forms.20 All calculations were done with the
Gaussian 03 program.21 In addition, on the basis of the time-
dependent DFT method (TDDFT/B3LYP/cc-pVDZ and aug-cc-
pVDZ) implemented in the TURBOMOLE 5.8 software package,22
we also carried out the geometry relaxation after Franck−Condon
excitation to examine the feasibility of ESIPT in a qualitative manner.
4.5. Fabrication of OLEDs. OLEDs were fabricated by vacuum
deposition of the materials at 10−6 Torr onto ITO-coated glass
substrates having a sheet resistance of 15 Ω square−1. The ITO surface
was cleaned through ultrasonication sequentially with acetone,
methanol, and deionized water, followed by treatment with UV−
ozone. A hole-injection layer of PEDOT:PSS was spin-coated onto the
substrates and dried at 130 °C for 30 min to remove residual water.
Organic layers were then vacuum-deposited at a deposition rate of ca.
1−2 Å s−1. Subsequently, LiF was deposited at 0.1 Å s−1 and then
capped with Al (ca. 5 Å s−1) by shadow masking without breaking the
vacuum. The current−voltage−brightness (I−V−L) characteristics of
the devices were measured simultaneously using a Keithley 6430
source meter and a Keithley 6487 picoammeter equipped with a
calibration Si-photodiode. EL spectra were measured using an Ocean
Optics spectrometer.
(5) He, X.; Bell, A. F.; Tonge, P. J. Org. Lett. 2002, 4, 1523.
(6) Schaefer, T. J. Phys. Chem. 1975, 79, 1888.
(7) (a) Stavrov, S. S.; Solntsev, K. M.; Tolbert, L. M.; Huppert, D. J.
Am. Chem. Soc. 2006, 128, 1540. (b) Gepshtein, R.; Huppert, D.;
Agmon, N. J. Phys. Chem. B 2006, 110, 4434. (c) Usman, A.;
Mohammed, O. F.; Nibbering, E. T. J.; Dong, J.; Solntsev, K. M.;
Tolbert, L. M. J. Am. Chem. Soc. 2005, 127, 11214.
(8) (a) Webber, N. M.; Litvinenko, K. L.; Meech, S. R. J. Phys. Chem.
B 2001, 105, 8036. (b) Polyakov, I. V.; Grigorenko, B. L.; Epifanovsky,
E. M.; Krylov, A. I.; Nemukhin, A. V. J. Chem. Theory Comput. 2010, 6,
2377.
(9) (a) Dong, J.; Solntsev, K. M.; Poizat, O.; Tolbert, L. M. J. Am.
Chem. Soc. 2007, 129, 10084. (b) Solntsev, K. M.; Poizat, O.; Dong, J.;
Rehault, J.; Lou, Y.; Burda, C.; Tolbert, L. M. J. Phys. Chem. B 2008,
112, 2700. (c) Baldridge, A.; Feng, S.; Chang, Y.-T.; Tolbert, L. M.
ACS Comb. Sci. 2011, 13, 214.
(10) 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, 4534.
(11) Hsieh, C. C.; Chou, P. T.; Shih, C. W.; Chuang, W. T.; Chung,
M. W.; Lee, J.; Joo, T. J. Am. Chem. Soc. 2011, 133, 2932.
(12) (a) Rao, Y. S.; Filler, R. Synthesis 1975, 12, 749. (b) Sidhu, G. S.;
Venkataratnam, R. V.; Prasad, K. K.; Iyengar, D. S. Indian J. Chem.
1972, 10, 448.
(13) (a) Siebrand, W. J. Chem. Phys. 1967, 47, 2411. (b) Chen, K. Y.;
Hsieh, C. C.; Cheng, Y. M.; Lai, C. H.; Chou, P. T. Chem. Commun.
2006, 4395.
(14) You, Y.; He, Y.; Burrows, P. E.; Forrest, S. R.; Petasis, N. A.;
Thompson, M. E. Adv. Mater. 2000, 12, 1678.
ASSOCIATED CONTENT
■
S
* Supporting Information
Figures, tables, text, and CIF files giving H and 13C NMR
1
spectra of the synthesized compounds, X-ray crystallography
data for 1a−e,g and 1h-E, EL performances of the device (1a),
and theoretical studies. This material is available free of charge
via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*K.-Y.C.: e-mail, kyuchen@fcu.edu.tw; tel, +886-4-2451-7250;
fax, +886-4-2451-0890. W.-Y.H.: e-mail, wenhung@mail.ntou.
edu.tw; tel, +886-2-2462-2192 ext. 6718; fax, + 886-2-2463-
4360. P.-T.C.: email, chop@ntu.edu.tw; tel, +886-2-3366-3894;
fax, +886-2-2369-5208.
■
(15) Shin, M. G.; Thangaraju, K.; Kim, S. O.; Park, J. W.; Kim, Y. H.;
Kwon, S. K. Org. Electron. 2011, 12, 785.
ACKNOWLEDGMENTS
■
(16) (a) Ma, D.; Liang, F.; Wang, L.; Lee, S. T.; Hung, L. S. Chem.
Phys. Lett. 2002, 358, 24. (b) Kim, S.; Seo, J.; Jung, H. K.; Kim, J. J.;
Park, S. Y. Adv. Mater. 2005, 17, 2077. (c) Gaenko, A. V.; Devarajan,
A.; Tselinskii, I. V.; Ryde, U. J. Phys. Chem. A 2006, 110, 7935.
(d) Park, S.; Kwon, J. E.; Kim, S. H.; Seo, J.; Chung, K.; Park, S. Y.;
Jang, D. J.; Medina, B. M.; Gierschner, J.; Park, S. Y. J. Am. Chem. Soc.
2009, 131, 14043. (e) Kim, S. H.; Park, S.; Kwon, J. E.; Park, S. Y. Adv.
Funct. Mater. 2011, 21, 644. (f) Yao, D.; Zhao, S.; Guo, J.; Zhang, Z.;
Zhang, H.; Liu, Y.; Wang, Y. J. Mater. Chem. 2011, 21, 3568.
(17) de Mello, J. C.; Wittmann, H. F.; Friend, R. H. Adv. Mater.
1997, 9, 230.
(18) Chou, P. T.; Chen, Y. C.; Yu, W. S.; Chou, Y. H.; Wei, C. Y.;
Cheng, Y. M. J. Phys. Chem. A 2001, 105, 1731.
(19) Petersson, G. A.; Al-Laham, M. A. J. Chem. Phys. 1991, 94, 6081.
(20) (a) Bauemschmitt, R.; Ahlrichs, R. Chem. Phys. Lett. 1996, 256,
454. (b) Casida, M. E.; Jamorski, C.; Casida, K. C.; Salahub, D. R. J.
Chem. Phys. 1998, 108, 4439.
We thank the National Science Council (Grant No. 99-1989-
2004) for financial support. We are also grateful to the National
Center for High-Performance Computing of Taiwan for
allowing us generous amounts of computing time.
REFERENCES
■
(1) (a) Chalfie, M.; Tu, Y.; Euskirchen, G.; Ward, W. W.; Prasher, D.
C. Science 1994, 263, 802. (b) Lippincott-Schwartz, J.; Patterson, G. H.
Science 2003, 300, 87. (c) Ormo, M.; Cubitt, A. B.; Kallio, K.; Gross, L.
A.; Tsien, R. Y.; Remington, S. J. Science 1996, 273, 1392. (d) Zimmer,
M. Chem. Rev. 2002, 102, 759. (e) Tsien, R. Y. Annu. Rev. Biochem.
1998, 67, 509. (f) Matz, M. V.; Fradkov, A. F.; Labas, Y. A.; Savitsky,
A. P.; Zaraisky, A. G.; Markelov, M. L.; Lukyanov, S. A. Nat. Biotechnol.
1999, 17, 969. (g) Hastings, J. W.; Morin, J. G., In Green Fluorescent
Protein; Chalfie, M., Kain, S., Eds.; Wiley-Liss: New York, 1998; pp
17−41. (h) Sullivan, K. F.; Kay, S. A. Green Fluorescent Proteins;
8201
dx.doi.org/10.1021/jo2012384|J. Org. Chem. 2011, 76, 8189−8202

The Journal of Organic Chemistry
Article
(21) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A. Jr.; Vreven, T.;
Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.;
Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.;
Petersson, G. A.; Ehara, M.; Toyota, K.; Hada, M.; Fukuda, R.;
Hasegawa, J.; Ishida, M.; Nakajima, T.; Kitao, O.; Nakai, H.; Honda,
Y.; Nakatsuji, H.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.;
Adamo, C.; Jaramillo, J.; Cammi, R.; Pomelli, C.; Gomperts, R.;
Stratmann, R. E.; Ochterski, J.; Ayala, P. Y.; Morokuma, K.; Salvador,
P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.;
Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari,
K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.;
Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.;
Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.;
Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.
Gaussian 03, version A.1; Gaussian, Inc., Pittsburgh, PA, 2002.
(22) Ahlrichs, R.; Bar, M.; Haser, M.; Horn, H.; Kolmel, C. Chem.
Phys. Lett. 1989, 162, 165.
8202
dx.doi.org/10.1021/jo2012384|J. Org. Chem. 2011, 76, 8189−8202