Colorimetric and ratiometric fluorescent chemosensor based on diketopyrrolopyrrole for selective detection of thiols: An experimental and theoretical study

Article abstract of DOI:10.1021/jo201487m

A colorimetric and ratiometric fluorescent thiol probe was devised with diketopyrrolopyrrole (DPP) fluorophore. The probe gives absorption and emission at 523 and 666 nm, respectively. In the presence of thiols, such as cysteine, the absorption and emissi

Full text of DOI:10.1021/jo201487m

Article
pubs.acs.org/joc
Colorimetric and Ratiometric Fluorescent Chemosensor Based on
Diketopyrrolopyrrole for Selective Detection of Thiols: An
Experimental and Theoretical Study
Ling Deng,† Wenting Wu,† Huimin Guo,† Jianzhang Zhao,*,† Shaomin Ji,† Xin Zhang,† Xiaolin Yuan,‡
and Chunlei Zhang‡
†State Key Laboratory of Fine Chemicals, School of Chemical Engineering, Dalian University of Technology, E-208 West Campus, 2
Ling-Gong Road, Dalian 116024, P. R. China
‡Center Laboratory, Affiliated Zhongshan Hospital of Dalian University, Dalian 116001, P. R. China
S
* Supporting Information
ABSTRACT: A colorimetric and ratiometric fluorescent thiol
probe was devised with diketopyrrolopyrrole (DPP) fluo-
rophore. The probe gives absorption and emission at 523 and
666 nm, respectively. In the presence of thiols, such as
cysteine, the absorption and emission band shifted to 479 and
540 nm, respectively. Correspondingly, the color of the probe
solution changed from purple to yellow, and the fluorescence
changed from red to yellow. The emission intensity at 540 nm
was enhanced by 140-fold. The Stokes shift of probe 1 (107
nm) is much larger than the unsubstituted DPP fluorophore
(56 nm). Mass spectral analysis demonstrated that besides the expected Michael addition of thiols to the CC bonds, the CN
groups of the malonitrile moieties also react with thiols to form 4,5-dihydrothiazole structure. Probe 1 was used for fluorescence
imaging of intracellular thiols. In the presence of thiols, both the green and red channel of the microscopy are active. With
removal of the intracellular thiols, signal can only be detected through the red channel; thus, ratiometric bioimaging of
intracellular thiols was achieved. The ratiometric response of probe 1 was rationalized by DFT calculations. Our complementary
experimental and theoretical studies will be useful for design of ratiometric/colorimetric molecular probes.
photon absorption materials,33,34 but its application in
1. INTRODUCTION
fluorescent molecular probes is very rare.35,36 (3) Some new
Thiols such as cysteine (Cys), homocysteine, glutathione,
sensing mechanisms for thiol probes, such as Michael addition,
peptides, and proteins are important for biological science;
have been developed, but most of these probes use coumarin as
thus, selective detection of thiols is important.1 Fluorescence
the fluorophore;37−40 thus, much room is left to fully explore
molecular probes have been developed for this purpose because
this new sensing mechanism with other fluorophores to
of the high sensitivity and high temporal and spatial resolution
of this method.1−10 Several factors are crucial for successful
improve the photophysical properties of the probes, such as
achieving longer emission wavelength, etc. Furthermore, all of
molecular probe design, such as the appropriate binding site
the thiol probes based on Michael addition are colorimetric or
(recognition site), the fluorophore, and the fluorescence
fluorescence OFF−ON probes,16,19,39,41 and no thiol probes
have been developed to combine the ratiometric and
colorimetric fluorescence transductions.
transduction mechanism, etc.11−14 Various fluorophores have
been used for fluorescence thiol sensors,1 such as rhodamine,15
fluorescein,16 and BODIPY, etc.4,5,17,18 The typical fluorescence
transduction mechanism is photoinduced electron transfer
(PET).1 However, we noticed some challenges for the
development of fluorescent thiols probes.1 (1) Most of the
thiols probes show a fluorescent OFF−ON switching effect.
However, ratiometric or colorimetric probes are more desired
in order to eliminate the influence of probe concentration on
quantification.19 (2) The fluorophores used for thiol probes are
limited to coumarin, fluorescein, rhodamine, or BODIPY.1
Probes based on other fluorophores were rarely reported. On
the other hand, diketopyrrolopyrrole (DPP) is a robust
chromophore20−26 and has been widely used in electromaterials
such as organic semiconductors,27 photovoltaics,28−32 and two-
In order to address the above limitations, herein we use DPP
as the fluorophore for thiol probes. In order to design the
ratiometric/colorimetric probe, we devised probe 1 (Scheme
1), in which DPP was used as the fluorophore and the
malononitrile moiety was introduced. Previously, DPP was only
used as the fluorophore for F− anion and pH sensors.35,36 To
the best of our knowledge, this is the first molecular probe with
DPP as the fluorophore to recognize bioactive small organic
molecules. The probe is red-emitting. In the presence of thiols,
Received: July 22, 2011
Published: October 18, 2011
© 2011 American Chemical Society
9294
dx.doi.org/10.1021/jo201487m|J. Org. Chem. 2011, 76, 9294−9304

The Journal of Organic Chemistry
Article
a
Scheme 1. Synthesis of the Thiol Probe 1
a
Key: (i) Na, FeCl3, tert-amyl alcohol, 100 °C, 24 h, 80%; (ii) potassium tert-butoxide, 1-bromobutane, DMF, 60 °C, 12 h, 40%; (iii) THF, 2 M HCl,
60 °C, 2 h 75%; (iv) malononitrile, Al2O3, CH2Cl2, rt, 30 min, 78%.
such as Cys, yellow emission was observed and the emission
intensity was greatly intensified. Furthermore, the solution
color of the probe changed from purple to yellow. Thus, the
probe simultaneously displays the colorimetric and ratiometric
signal transduction. The probe is highly selective toward thiols,
and the sensing mechanism, i.e., multiaddition of the thiol
molecules to the CC double bond and the −CN group, was
demonstrated by mass spectral analysis. The photophysical
properties of the probe, such as UV−vis absorption and
fluorescence emission, as well as the ratiometric and the
colorimetric signal transduction, were fully rationalized by DFT
calculations on the excited states of the fluorophores. We found
that the phenyl substituent is in π-conjugation with the DPP
fluorophore core; this finding will be useful for future design of
molecular probes based on DPPs.
2. RESULTS AND DISCUSSION
Figure 1. UV−vis absorption spectra of probe 1 and the unsubstituted
2.1. Design and Synthesis of Probe 1 Based on DPP
Fluorophore. To the best of our knowledge, DPP has rarely
been used as a fluorophore for fluorescent molecular
probes.35,36 Recently, it was shown that the DPP-based
chromophore can be used for detection of fluoride anions.35
Inspired by elegant development of the thiol probes based on
Michael addition,1,16,19 herein we devised probe 1 (Scheme 1),
with the intention of obtaining a ratiometric probe because
addition of thiols to the CC double bonds will reduce the
electron withdrawal effect of the malonitrile moiety; thus, the
UV−vis absorption and emission of chromophore may move to
fluorophore DPP (1.0 × 10−5 mol dm−3). In CH3CN/H2O mixed
solvent (4/1, v/v), 25 °C.
a shorter wavelength. 4-Cyanobenzaldehyde was used as the
starting material for the preparation, with the aldehyde group
protected by ethylene glycol. Deprotection by HCl leads to the
intermediate with biformyl structure. Condensation with
malonitrile, under the catalysis of Al2O3, gave the probe 1,
which was obtained in satisfying yield (78%) (Scheme 1). It
9295
dx.doi.org/10.1021/jo201487m|J. Org. Chem. 2011, 76, 9294−9304

The Journal of Organic Chemistry
Article
Figure 2. Emission spectra and excitation spectra of probe 1 and DPP.
1.0 × 10−5 mol dm−3 of the compounds in CH3CN/H2O (v/v= 4/1),
25 °C.
Figure 4. UV−vis absorption of probe 1 before and after addition of
Cys. In mixed solvent of CH3CN/H2O (4:1, v/v). c (probe 1) = 1.0 ×
10−5 mol dm−3, c (Cys) = 2.0 × 10−3 mol dm−3, 37 °C.
should be pointed out that Al2O3 is an efficient catalyst for the
condensation reaction.
The Stokes shift of probe 1 is 107 nm, much larger than the
Stokes shift of DPP (56 nm).
2.2. Electronic Spectroscopy of the Probe. The UV−
vis absorption spectra of the probe and the unsubstituted
fluorophore DPP were compared (Figure 1). DPP shows
absorption at 461 nm (ε = 16100 M−1 cm−1). For probe 1,
however, the absorption band is red-shifted to 523 nm (ε =
13500 M−1 cm−1). The red-shift of the absorption band is due
to the extended π-conjugation framework of probe 1 vs DPP.
The absorption band in the visible region is due to the π−π*
transition localized on DPP fluorophore. This assignment is
supported by the theoretical calculation based on the density
functional theory (DFT, see the later section).
The emission and excitation of the compounds were studied
(Figure 2). DPP shows an excitation peak at 467 nm and an
emission band at 523 nm, respectively. Interestingly, probe 1
shows much red-shifted excitation and emission bands at 559
and 666 nm, respectively. The emission quantum yield of probe
1 (ΦF = 1.5%, in CH3CN/H2O (4:1, v/v)) is much smaller
than that of DPP (ΦF = 75% in CH3CN/H2O (4:1, v/v)), but
the emission of probe 1 is much stronger in ether (ΦF = 65%).
It is known that the emission wavelength of a D−π−A
chromophore is sensitive to the polarity of the solvent; i.e., the
emission wavelength will show a red-shift in more polar
solvents.6 Such a polarity-dependent emission wavelength is
detrimental to the application of ratiometric fluorescence
probes in vivo, for which the polarity of the microenvironment
is uncontrollable. Thus, the absorption and the emission of
probe 1 in solvents with different polarity were studied (Figure
3). The UV−vis absorption spectra show minor variations in
different solvents (Figure 3a). For the fluorescence spectra,
generally the emission intensity decreased in polar solvents;6
however, the emission band did not show any shift. The UV−
vis absorption and the fluorescence of the fluorophore DPP
were also studied, and similar results were observed (Figure
S10, Supporting Information). This solvent polarity-independ-
ent emission wavelength will be ideal for construction of
ratiometric fluorescent molecular probes.
2.3. Spectroscopic Response of Probe 1 to Thiols.
The change of the UV−vis absorption of probe 1 in the
Figure 3. UV−vis absorption and fluorescence spectra of probe 1 in different solvents: (a) UV−vis absorption and (b) fluorescence spectra of 1 (λex
= 496 nm). c (probe 1) = 1.0 × 10−5 mol dm−3, 25 °C.
9296
dx.doi.org/10.1021/jo201487m|J. Org. Chem. 2011, 76, 9294−9304

The Journal of Organic Chemistry
Article
Figure 5. Fluorescence emission spectra of probe 1 in the presence of thiols in different solvents: (a) emission of probe of 1 before and after addition
of Cys (λex = 490 nm). In mixed solvent of CH3CN/H2O (4:1, v/v). (b) emission spectra (λ ex = 490 nm) of probe 1 before and after addition of 3-
mercaptopropionic acid (MPA). In CH3CN. c (Probe 1) = 1.0 × 10−5 mol dm−3, c (Cys or 3-mercaptopropionic acid) = 2.0 × 10−3 mol dm−3, 37
°C.
Along with the change of absorptions, the color of probe 1
solution was changed from purple to yellow. Compound 1 can
be described as a colorimetric probe. A similar response was
found for mercaptopropanoic acid and homocysteine (Figure
S11, Supporting Information). No response was found for 1
toward nonthiol substrates, such as glutamic acid and ascorbic
acid, etc.
Table 1. Photophysical Parameters of the Probe 1 upon
Reaction with Cys
λabs
(nm)
λex
(nm)
λem
(nm)
a
ε/M−1 cm−1
ΦF
τ (ns)
1
523
479
1.35 × 104
1.22 × 104
559
515
666
540
0.015
0.177
6.25
4.37
1 + Cys
a
Quantum yield of fluorescence in CH3CN/H2O (4:1, v/v), with
The emission changes of the probe in the presence of Cys
were studied (Figure 5a). Probe 1 gives weak red emission at
666 nm. In the presence of Cys, a new emission band at 540
nm developed, and the emission intensity at 540 nm was
enhanced by 140-fold. Correspondingly, the emission color of
the solution changed from deep red to light-yellow. Thus, 1 is a
ratiometric probe. To the best of our knowledge, thiol probes
that show the simultaneous ratiometric and colorimetric spectral
changes are rarely reported,39 and this is the first ratiometric
fluorescent thiol probe that is based on fluorophore other than
coumarin. Recently, ratiometric thiol probes based on coumarin
chromophore were reported, but the emission band was at blue
region.19,42 Although colorimetric thiol probes based on
coumarin fluorophore were reported,37,39 the probe showed a
fluorescence OFF−ON effect, not ratiometric fluorescence
changes. Probe 1 shows two emission bands in the presence of
Cys (Figure 5a), which is proposed on the basis of the reaction
product with thiol (for the putative mechanism, see Figure 7).
We noticed that the red emission of probe 1 in protic
solvents is weak, but the emission in the presence of Cys is
much stronger (Figure 5a). However, we propose that this
result does not necessarily mean that the probe can not be used
in biological media because the probe may enter the nonprotic
hydrophobic micropocket, and then the red-emission and the
yellow emission may become more balanced. Balanced
emission at two different wavelengths is useful for the
ratiometric sensing. Therefore, as a proof of concept, we
studied the recognition of thiols with probe 1 in nonprotic
solvents (acetonitrile, 3-mercaptopropionic acid was used in
stead of Cys, due to the low solubility of Cys in CH3CN)
(Figure 5b). More balanced emission at 558 and 663 nm was
observed.
quinine sulfate as the standard (Φ = 0.54 in 0.05 M H2SO4).
Figure 6. Ratiometric response of probe 1 to different analytes with
emission intensiy ratio at 540 and 666 nm. In mixed solvent of
CH3CN/H2O (4:1, v/v). c (probe 1) = 1.0 × 10−5 mol dm−3, 37 °C.
Inset: recognition kinetics of the probe in the presence of Cys.
presence of Cys was studied (Figure 4). In the presence of Cys,
the absorption band at 523 nm decreased and a new absorption
band at 479 nm developed. Moreover, the absorption band of
probe 1 at 337 nm decreased, and a new band at 275 nm
developed. The blue-shift of the absorption bands are attributed
tentatively to the addition reaction of Cys to the CC bond of
the probe, thus reducing of the π-conjugation framework.
The change of the photophysical properties of probe 1 in the
presence of Cys were presented in Table 1. It was found the
9297
dx.doi.org/10.1021/jo201487m|J. Org. Chem. 2011, 76, 9294−9304

The Journal of Organic Chemistry
Article
Figure 7. Proposed recognition mechanism of probe 1 toward thiols. (a) Addition of thiols to the probe, demonstrated by Cys. The molecular
structures were derived from the mass spectral analysis of the reaction mixtures. (b) TOF-HRMS of 1 upon addition of Cys after 360 min. In
CH3CN/H2O (4:1, v/v). c(probe 1) = 1.0 × 10−5 mol dm−3, c(Cys) = 2.0 × 10−3 mol dm−3, 37 °C. C1 (1 + 4Cys), [(C46H50N8O10S4 − 2H)2−/2]
calcd 500.1193, found 500.1183; C2 (1 + 3Cys), (C43H43N7O8S3 ) calcd 881.2335, found 881.2283; C3 (1 + 2Cys), [(C46H36N6O6S2 − H)−] calcd
−
calcd 760.2138, found 760.2109.
fluorescence quantum yield was increased by 11-fold in the
presence of Cys. The blue shift of the emission band is up to
126 nm, which leads to vivid color change of the emission.
The ratiometric responses of the probe toward different
analytes are summarized in Figure 6. Probe 1 gives a strong
response toward small thiol molecules, such as Cys,
mercaptopropionic acid, and homocysteine. No significant
responses were found for other analytes. For example, the
fluorescence enhancement of probe 1 in the presence of
cysteine is ca. 140-fold. The ratiometric response (I540nm/I666nm
values, where I540nm and I666nm are the emission intension at 540
and 666 nm in the presence of thiols) is up to 6.8 for Cys. For
other biologically related molecules, such as alanine, etc., the
response is much smaller (Figure 6). Remarkable responses
were observed in the presence of thiols such as 3-
mercaptopropionic acid or homocysteine. The detection limit
of probe 1 for Cys is determined as 6.02 × 10−7 mol dm−3.
2.4. Sensing Mechanism of Probe 1 Toward Thiols.
Initially, we anticipated the normal Michael addition reaction of
thiols with probe 1, i.e., addition of the thiols to the CC
double bonds.37−39 However, mass spectral analysis of the
reaction mixture of the probes with Cys reveals a rather
complicated mixture (Figure 7). On the basis of the mass
spectra, we propose that up to four Cys molecules can be added
to the probe molecule; thus, we propose that the CN group
also reacts with Cys. A similar reaction was observed for a
polythiophene thiol probe.43
2.5. Rationalization of the Colorimetric and Ratio-
metric Fluorescence Response of Probe 1 toward
Thiols. Recently, theoretical calculations have been used for
9298
dx.doi.org/10.1021/jo201487m|J. Org. Chem. 2011, 76, 9294−9304

The Journal of Organic Chemistry
Article
a
Table 2. Selected Parameters for the Vertical Excitation (UV−vis Absorptions) and the Fluorescence Emission of Probe 1
TDDFT//B3LYP/6-31G(d)
b
c
d
e
electronic transitions
excitation energy (eV)
f
composition
CI
absorption
S0→S1
S0→S7
1.88 (660 nm)
3.39 (366 nm)
0.8781
0.9568
H → L
H-6 → L
H-4 → L
H-2 → L
H-1 → L+1
L → H
0.7082
0.1193
0.3613
0.4753
0.1157
0.7101
f
emission
S1→S0
1.45 (852 nm)
1.1787
a
Electronic excitation energies (eV) and oscillator strengths (f), configurations of the low-lying excited states of the probe 1. Based on the optimized
geometries of the ground state and singlet excited state (S1). Acetonitrile was used as solvent in all of the calculations. Calculated by TDDFT at the
b
B3LYP/6-31G(d) level. Only selected excited states were considered. The numbers in parentheses are the excitation energy in wavelength.
c
d
e
Oscillator strength. H stands for HOMO and L stands for LUMO. Only the main configurations are presented. Coefficient of the wave function
f
for each excitations. The CI coefficients are in absolute values. Note the calculation of the emission by TDDFT methods gives the information for
the S0→S1 transition.
Furthermore, we found that the biphenyl moieties in the
probe are in π-conjugation with the DPP core.
First, the ground-state geometry of probe 1 was optimized.
The phenyl ring is tilted by about 29° toward the DPP core.
The malonitrile moiety is coplanar with the phenyl moiety. The
UV−vis absorption of probe 1 was calculated with the TDDFT
method based on the optimized S0 state geometry. The
calculated absorption of probe 1 is at 659 nm, and HOMO→
LUMO is the main component of the S0→S1 transition (Table
2). HOMO is localized on the DPP core, and the LUMO is
more distributed on the malonitrile moiety. Thus, the S0→S1
transition can be described as an intramolecular charge-transfer
(ICT) process. The discrepancy between the calculated
absorption wavelength and the experimental results is in line
with the known fact that the DFT calculations usually
underestimate the excitation energy for the transitions with
remarkable ICT character.49,50
In order to study the emission of probe 1, the geometry of
the lowest lying singlet excited state (S1 state, which is
responsible for fluorescence, Kasha’s rule) was optimized. The
excitation (i.e., UV−vis absorption, the left two columns) and the
Figure 8. Frontier molecular orbitals (MOs) involved in the vertical
HOMO/LUMO at the S1 geometry were presented (Figure 8;
note that the MOs will change at different geometries). We
found that there is geometry relaxation at the S1 state compared
to the S0 state geometry. For example, the dihedral angle
between the DPP core and the peripheral phenyl ring is 29° at
the S0 state. At the S1 state, however, the dihedral angle
decreased to 21°. Furthermore, the LUMO energy level at the
S1 state is much lower than that at the S0 state, and the HOMO
energy level at the S1 state is higher than that at S0 state
geometry. The emission transition is basically a CT process;
thus, the discrepancy between the calculated emission wave-
length (852 nm) and the experimental observations (666 nm)
is within expectations.49,50
Similarly, the molecular geometry, the UV−vis absorption
and fluorescence emission of the adduct of the probe with Cys
were studied (Figure 9 and Table 3). The peripheral phenyl
ring of the adduct takes a similar dihedral angle toward the DPP
core of 31°. The calculated absorption wavelength of the probe
1−Cys adduct is at 497 nm, which is very close to the
experimentally observed 479 nm (Figure 2). We found that
compared to the probe, the S0→S1 transition of the thiol adduct
is more localized on the DPP chromophore core.
emission (right column) of probe 1. Acetonitrile was used as solvent
for the calculations. The vertical excitation are based on the optimized
ground-state geometry (S0 state, Franck−Condon principle) and the
emission-related calculations were based on the optimized excited state
(S1 state, Kasha’s rule) at the B3LYP/6-31G(d)/level using Gaussian
09W. Note the energy levels of the HOMO and LUMO at the S1 state
are different from that at the S0 state. IC stands for internal conversion,
and CT stands for conformation transformation. Excitation and
radiative processes are marked as solid lines, and the nonradiative
processes are marked by dotted lines. Only selected MOs of the
transitions are presented; for details, see Table 2.
the study of molecular probes.4−7,44−48 Previously, we used
density functional theory (DFT) and time-dependent DFT
(TDDFT) calculations for study of the fluorescence trans-
duction of molecular probes.4−7,46−48 In principle, the new
concept we developed is to use excited state, rather than the
molecular orbitals, to study the absorption/emission properties
of the molecular probes. Herein we used DFT and TDDFT
methods for study of the spectral transduction of probe 1 upon
recognition of thiols, i.e., the colorimetric and the ratiometric
fluorescence response of the probes in presence of thiols. We
found that the blue-shift of UV−vis absorption and the
fluorescence in the presence of thiols can be predicted by the
TDDFT calculations, based on the ground-state and the
excited-state geometries of the compounds, respectively.
The geometry of the S1 state of the probe 1−Cys adduct was
optimized, and we found a geometry relaxation at the S1 state;
i.e., the dihedral angle between the phenyl ring and the DPP
core decreased from 31° (ground state) to 19°. Thus, the
9299
dx.doi.org/10.1021/jo201487m|J. Org. Chem. 2011, 76, 9294−9304

The Journal of Organic Chemistry
Article
Figure 9. Frontier molecular orbitals (MOs) involved in the vertical excitation (i.e., UV−vis absorption, the left two columns) and emission (right
column) of the 1-Cys adduct (C-1 in Figure 7, R groups are simplified as H to reduce computation time). Acetonitrile was used as the solvent. The
vertical excitation related calculations were based on the optimized ground-state geometry (S0 state, Franck−Condon principle); the emission related
calculations were based on the optimized excited state (S1 state, Kasha’s rule); IC stands for internal conversion and CT stands for conformation
transformation. Excitation and radiative processes are marked as solid lines and the nonradiative processes are marked by dotted lines. Only selected
MOs are presented; for details see Table 3. All calculations were carried out at the B3LYP/6-31G(d) level using Gaussian 09W.
Table 3. Selected Parameters for the Vertical Excitation (UV−vis Absorptions) and the Fluorescence Emission of the Probe 1−
a
Cys Adduct
TDDFT//B3LYP/6-31G(d)
b
c
d
e
electronic transitions
excitation energy (eV)
f
composition
CI
absorption
S0→S1
2.50 (497 nm)
4.55 (273 nm)
0.6605
0.4790
H → L
0.7071
0.1218
0.1604
0.4519
0.4565
0.7105
S0→S19
H-10 → L
H-5 → L+1
H → L+6
H → L+8
L → H
f
emission
S1→S0
1.97 (628 nm)
0.8456
a
Electronic excitation energies (eV) and oscillator strengths (f), configurations of the low-lying excited states. Calculated by TDDFT//B3LYP/6-
31G(d), based on the optimized ground-state geometries and the optimized singlet excited state (S1). Acetonitrile was employed as solvent in all of
b
c
the calculations. Only selected excited states were considered. The numbers in parentheses are the excitation energy in wavelength. Oscillator
strength. H stands for HOMO and L stands for LUMO. Only the main configurations are presented. Coefficient of the wave function for each
excitations. The CI coefficients are in absolute values. Note that the calculation of the emission by TDDFT methods gives the information of S0→S1
d
e
f
transition.
molecular backbone is more coplanar at the excited state. The
calculated emission wavelength is 628 nm. Despite the
discrepancy between the theoretically predicted value and the
experimental result, the trend is clear; that is, with addition of
Cys to the probe, the UV−vis absorption and emission were
predicted to move to the higher energy region of the spectrum,
in line with the experimental results. Furthermore, the frontier
molecular orbitals indicate that the phenyl groups are in π-
conjugation with the DPP chromophore. These results may be
helpful for design ratiometric fluorescent molecular probes and
probes or fluorescent dyes based on DPP core.
also studied the absorption and the fluorescence emission
properties of the unsubstituted fluorophore DPP (Figure 10
and Table 4). The theoretically predicted absorption band at
477 nm is in good agreement with the experimental observation
at 461 nm. The calculated emission band at 600 nm is red-
shifted by ca. 80 nm compared to the experimental results.
Despite of the discrepancy between the calculation and the
experimental results, the DFT/TDDFT calculations clearly
predicted the blue shift of the absorption and fluorescence of
probe 1 upon interaction with thiols, i.e., the colorimetric and
ratiometric fluorescence changes. Previously, the fluorescence
OFF−ON switching effect was predicted by DFT/TDDFT
calculations.4−7,44,51 To the best of our knowledge, however,
In order to validate the DFT calculations on the UV−vis
absorption and the fluorescence of the probe 1−Cys adduct, we
9300
dx.doi.org/10.1021/jo201487m|J. Org. Chem. 2011, 76, 9294−9304

The Journal of Organic Chemistry
Article
active, and a yellow image was observed (Figure 11h). With
removal of the intracellular thiols by N-methylmaleimide, only
the red channel is active (Figure 11j) and red emission was
detected (Figure 11i). It should be pointed out that ratiometric
fluorescent thiol probes are rarely reported.19,42 The known
ratiometric thiol probes gives emission change at shorter
wavelength range than probe 1. For example, emission changed
from green to blue color in the presence of thiols.19,42
2.6. Conclusions. A colorimetric and ratiometric fluores-
cent thiol probe was designed with diketopyrrolopyrrole (DPP)
as the fluorophore, which shows absorption at 523 nm and a
deep red emission at 666 nm. The Stokes shift of the probe
(107 nm) is larger than that of the diphenyl DPP fluorophore
core (56 nm). In the presence of thiols, the absorption and the
emission peaks show 44 and 126 nm blue-shift, respectively.
The emission intensity at 540 nm was enhanced by 140-fold in
the presence of Cys. Correspondingly, the color of the probe
solution changed from purple to yellow, and the fluorescence
changed from deep red to yellow upon interaction with thiols.
Mass spectra analysis demonstrated that the recognition is
based on the Michael addition of thiols to the CC bond, as
well as the −CN groups. To the best of our knowledge, this is
the first DPP-based molecular probe to recognize small
bioactive organic molecules, and this is also the first thiol
probe based on the malonitrile/Michael addition mechanism
that show the simultaneous colorimetric and the ratiometric
fluorescence changes. The colorimetric and the ratiometric
fluorescence changes of the probe 1 upon reaction with thiols
were fully rationalized by DFT/TDDFT calculations, based on
optimized ground-state (S0) and excited-state (S1) geometries
(calculation of the absorption and emission wavelengths).
Furthermore, we found the emission wavelength of DPP
fluorophore is independent of the solvent polarity, which is
ideal for fluorophore to construct ratiometric fluorescent
molecular probes. The probe was used for fluorescence imaging
of intracellular thiols. Both green and red channels can be used
for detection of intracellular cells. Without intracellular thiols,
only red channel is active. Our complementary experimental
and theoretical study will be useful for design of ratiometric and
colorimetric molecular probes.
Figure 10. Frontier molecular orbitals (MOs) involved in the vertical
excitation (i.e., UV−vis absorption, the left two columns) and emission
of diphenyl diketopyrrolopyrrole (DPP) (right column). Acetonitrile
was used as the solvent for the calculations. The vertical excitation
related calculations were based on the optimized ground-state
geometry (S0 state, Franck-Condon principle), and the emission
related calculations were based on the optimized excited state (S1 state,
Kasha’s rule). Note that the energy levels of the HOMO and LUMO
at S1 state are different from that at the S0 state. IC stands for internal
conversion, and CT stands for conformation transformation.
Excitation and radiative processes are marked as solid lines, and the
nonradiative processes are marked by dotted lines. Only selected MOs
are presented; for details, see Table 4. Calculations were carried out at
the B3LYP/6-31G(d)/ level using Gaussian 09W.
this is the first time that the colorimetric and the ratiometric
fluorescence transduction of the molecular probes were studied
by the DFT/TDDFT calculations.
The probe 1 was used for fluorescent bioimaging of
intracellular thiols (Figure 11). We found that with incubation
of MDA-231 cells with probe 1, both the green channel (Figure
11e) and the red channel (Figure 11f) of the microscope are
Table 4. Selected Parameters for the Vertical Excitation (UV−vis Absorptions) and the Fluorescence Emission of Diphenyl
a
Diketopyrrolopyrrole (DPP).
TDDFT//B3LYP/6-31G(d)
b
c
d
e
electronic transitions
excitation energy (eV)
f
composition
CI
absorption
S0→S1
S0→S8
2.60 (477 nm)
4.11 (302 nm)
0.4710
0.3862
H →L
0.7079
0.5602
0.2922
0.2640
0.1586
0.1053
0.2178
0.6747
0.7109
H-6 → L
H-4 → L
H-2 → L
H → L+4
H-6 → L
S0→S12
S1→S0
4.82 (257 nm)
2.07 (600 nm)
0.3567
H-1 → L+1
H → L+4
L → H
f
emission
a
Electronic excitation energies (eV) and oscillator strengths (f), configurations of the low-lying excited states of DPP, based on the optimized
ground-state and the singlet excited-state geometries. Acetonitrile was employed as solvent in all the calculations). Calculated at TDDFT//B3LYP/
b
6-31G(d) level with Gaussian 09W. Only selected excited states were considered. The numbers in parentheses are the excitation energy in
c
d
e
wavelength. Oscillator strength. H stands for HOMO, and L stands for LUMO. Only the main configurations are presented. Coefficient of the
f
wave function for each excitations. The CI coefficients are in absolute values. Note the calculation of the emission by TDDFT methods gives the
information of S0→S1 transition.
9301
dx.doi.org/10.1021/jo201487m|J. Org. Chem. 2011, 76, 9294−9304

The Journal of Organic Chemistry
Article
Figure 11. DIC and fluorescence images of living MDA-231 cells. Top panel (a−d): images of the cells without the probe 1. (a) Fluorescence image
of MDA-231 cells from the green channel; (b) fluorescence image of MDA-231 cells from the red channel; (c) DIC image of MDA-231 cells; (d)
overlay of images a, b and c. Middle panel (e−h): fluorescence imaging of the intracellular thiols with probe 1, MDA-231 cells incubated with 10 μM
probe 1 for 60 min at 37 °C. (e) Fluorescence image from the green channel after incubation; (f) fluorescence image from the red channel after
incubation; (g) DIC image of cells after incubation; (h) overlay of images e−g. Bottom panel (i−l): fluorescence images of the MDA-231 cells
pretreated with 1 mM N-methylmaleimide for 60 min at 37 °C and then incubated with 10 μM probe 1 for 60 min at 37 °C. (i) Fluorescence image
from the green channel; (j) fluorescence image of from the red channel; (k) DIC image and (l) overlay of images i−k.
After the mixture was cooled to room temperature, dichloromethane
(DCM) was added, and then the mixture was washed with water. The
organic layer was dried over anhydrous Na2SO4, and the solvent was
removed under reduced pressure. The product was purified by column
3. EXPERIMENTAL SECTION
3.1. General Methods. NMR spectra were recorded on a Bruker
400 MHz spectrometer (CDCl3 as solvents, TMS as standard, δ = 0.00
ppm). High-resolution mass spectra (HRMS) were determined on a
LC/Q-TOF MS system. Melting points were measured on an X-4
microscopic melting point meter. Fluorescence spectra were measured
on a RF5301 PC spectrofluorometer. Fluorescence lifetimes were
measured with a OB920 phosphorescence/fluorescence lifetime
spectrometer. Absorption spectra were recorded on a UV2550 UV/
chromatography (silica gel; DCM/ethylacetate, 10/1, v/v) to give 3 as
1
an orange solid (1.4 g, 40%): mp 217.7−219.1 °C; H NMR (400
MHz, CDCl3) δ 7.82 (d, 4H, J = 8.0 Hz), 7.64 (d, 4H, J = 8.0 Hz),
5.89 (s, 2H), 4.06−4.17 (m, 8H), 3.74 (t, 4H, J = 7.6 Hz), 1.52−1.60
(m, 4H), 1.20−1.30 (m, 4H), 0.84 (t, 6H, J = 7.2 Hz); EI-HRMS
+
(C32H36N2O6 ) calcd 544.2573, found 544.2571.
vis spectrophotometer and an Agilent 8453 UV/vis near-IR
spectrophotometer.
3.3. Synthesis of 2,5-Dibutyl-3,6-bis(4-formylphenyl)-
pyrrolo[3,4-c]pyrrole-1,4-dione (2).52 A mixture of 3 (1.2 g, 2.2
3.2. 3,6-Bis[4-1,3-dioxolan-2-ylphenyl]pyrrolo[3,4-c]pyrrole-
1,4-dione (4).52 Under argon atmosphere, sodium (1.10 g, 48.0
mmol), FeCl3 (0.05 g), and dry tert-amyl alcohol (24 mL) were mixed.
The mixture was heated to 100 °C until the sodium disappeared
completely. Then 4-(1,3-dioxolan-2-yl)benzonitrile (4.2 g, 24 mmol)
was added. A solution of diisopropyl succinate (1.94 g, 9.6 mmol) in
dry tert-amyl alcohol (10 mL) was added dropwise during 1 h at 100
°C. After the solution was allowed to stand for 24 h at 100 °C, acetic
acid (10 mL) was added. After the mixture was allowed to stand for 1
h, the precipitate was filtered and washed with hot water and hot
methanol. The red solid was dried under vacuum: yield 4.15 g, 80%;
mmol), THF (20 mL), and HCl (2 M, 10 mL) was stirred for 2 h at
60 °C. After being cooled to room temperature, DCM (60 mL) was
added, and the mixture was washed with water and dried over Na2SO4.
After filtration, the solvent was removed under reduced pressure, and
the crude product was purified by column chromatography (silica gel;
DCM/ethyl acetate, 10/1, v/v) to give compound 2 (0.75 g, 75%): mp
1
184.4−186.1 °C; H NMR (400 MHz, CDCl3) δ 10.10 (s, 2H), 8.03
(d, 4H, J = 8.0 Hz), 7.97 (d, 4H, J = 8.0 Hz), 3.78 (t, 4H, J = 7.6 Hz),
1.51−1.59 (m, 4H), 1.21−1.31 (m, 4H), 0.84 (t, 6H, J = 7.2 Hz); EI-
+
HRMS (C28H28N2O4 ) calcd 456.2049, found 456.2052.
3.4. Synthesis of 2,5-Dibutyl-3,6-bis(4-malonitrilephenyl)-
pyrrolo[3,4-c]pyrrole-1,4-dione (Probe 1). Compound 2 (75 mg,
0.16 mmol), malonitrile (32.5 mg, 0.49 mmol), aluminum oxide
(Catalyst. 80 mg), and dry DCM (10 mL) were mixed. The mixture
was stirred at room temperature for 30 min. After filtration, the solvent
was evaporated, and the crude product was purified by column
chromatography (silica gel; DCM/ethylacetate, 15/1, v/v) to give
+
mp 161.5−162.7 °C. EI-HRMS (C24H20N2O6 ) calcd 432.1321, found
432.1324. No 1H NMR spectrum was measured due to the poor
solubility of the compound in common organic solvents.
3.2. 2,5-Dibutyl-3,6-bis[4-1,3-dioxolan-2-ylphenyl]pyrrolo-
[3,4-c]pyrrole-1,4-dione (3).52 A mixture of 4 (2.8 g, 6.5 mmol),
potassium tert-butoxide (1.6 g, 13.9 mmol), and dry DMF (80 mL)
was heated to 60 °C, 1-bromobutane (6.5 g, 48 mmol) in DMF (20
mL) was added slowly, and the mixture was stirred at 60 °C for 12 h.
1
probe 1 as a purple solid (68.9 mg, 78%): mp 279.5−281.1 °C; H
NMR (400 M Hz, CDCl3) δ 8.07 (d, 4H, J = 8.4 Hz), 8.00 (d, 4H, J =
9302
dx.doi.org/10.1021/jo201487m|J. Org. Chem. 2011, 76, 9294−9304

The Journal of Organic Chemistry
Article
8.4 Hz), 7.81 (s, 2H), 3.79 (t, 4H, J = 7.2 Hz), 1.58−1.60 (m, 4H),
1.26−1.31 (m, 4H), 0.87 (t, 6H, J = 7.2 Hz); 13C NMR (100 MHz,
CDCl3) δ 162.4, 158.1, 147.3, 133.3, 133.1, 131.3, 129.8, 113.5, 112.5,
(3) Tang, B.; Yin, L.; Wang, X.; Chen, Z.; Tong, L.; Xu, K. Chem.
Commun. 2009, 5293−5295.
(4) Guo, H.; Jing, Y.; Yuan, X.; Ji, S.; Zhao, J.; Li, X.; Kan, Y. Org.
Biomol. Chem. 2011, 9, 3844−3853.
+
111.9, 84.8, 42.3, 31.8, 20.2, 13.8; EI-HRMS (C34H28N6O2 ) calcd
552.2274, found 552.2267. Anal. Calcd for C34H28N6O2 (0.25H2O):
C, 73.30; H, 5.15; N, 15.08. Found: C, 73.39; H, 5.003; N, 15.05.
3.5. Theoretical Calculations. All of the calculations were based
on density functional theory (DFT) with B3LYP functional and 6-
31G(d) basis set. Acetonitrile was used as solvent in all calculations
(PCM model). The UV−vis absorption of the compounds (vertical
excitation) were calculated with the TDDFT method based on the
optimized ground-state geometry (S0 state, Franck−Condon princi-
ple). For the fluorescence emission, the emission wavelength were
calculated based on the optimized excited states geometry (S1 state,
Kasha’s rule) by the TDDFT method. All of the calculations were
performed with Gaussian 09W.53
(5) Shao, J.; Guo, H.; Ji, S.; Zhao, J. Biosens. Bioelectron. 2011, 26,
3012−3017.
(6) Ji, S.; Yang, J.; Yang, Q.; Liu, S.; Chen, M.; Zhao, J. J. Org. Chem.
2009, 74, 4855−4865.
(7) Ji, S.; Guo, H.; Yuan, X.; Li, X.; Ding, H.; Gao, P.; Zhao, C.; Wu,
W.; Wu, W.; Zhao, J. Org. Lett. 2010, 12, 2876−2879.
(8) Lin, W.; Long, L.; Yuan, L.; Cao, Z.; Chen, B.; Tan, W. Org. Lett.
2008, 10, 5577−5580.
(9) Bouffard, J.; Kim, Y.; Swager, T. M.; Weissleder, R.; Hilderbrand,
S. A. Org. Lett. 2008, 10, 37−40.
(10) Zhang, R.; Yu, X.; Ye, Z.; Wang, G.; Zhang, W.; Yuan, J. Inorg.
Chem. 2010, 49, 7898−7903.
3.6. Fluorescence Microscopy and Cell-Uptake Studies.
MDA-231 cells were grown in RPMI-1640 medium in an atmosphere
of 5% CO2, 95% air at 37 °C. Cells were incubated on 96-well plate
and allowed to adhere for 24 h. After 24 h, the cells were washed with
PBS (phosphate buffered saline) and then incubated at 37 °C in the
presence of probe 1 (10 μM, PBS, pH 7.4) for 60 min. For the control
experiment, the MDA-231 cells were pretreated with 1 mM N-
methylmaleimide in PBS solution (pH 7.4) at 37 °C for 60 min to
eliminate all of the free thiols within the cells. After the cells were
washed with PBS buffer three times, the maleimide-pretreated cells
were incubated at 37 °C in the presence of probe 1 (10 μM, PBS, pH
7.4) for 60 min. Luminescence imaging was performed after the cells
were washed three times with PBS buffer. The luminescence images
were obtained using a Nikon ECLIPSE-Ti confocal laser scanning
microscopy with a 20× objective lens. Green and red luminescence
was excited at 488 nm argon laser. The differential interference
contrast (DIC) and the fluorescence was captured, digitized and
processed with the EZ-C1 Image Examiner software. The images for
the thiol-dependent activation and corresponding maleimide controls
were taken with identical settings. The wavelength range of the filters
used for the microcopy are 515−560 nm for the green channel and
605−680 nm for the red channel, respectively.
(11) Wright, A. T.; Anslyn, E. V. Chem. Soc. Rev. 2006, 35, 14−28.
(12) James, T. D. Top. Curr. Chem. 2007, 277, 107−152.
(13) Kim, D.; Chung, Y.; Jun, M.; Ahn, K. H. J. Org. Chem. 2009, 74,
4849−4854.
(14) Ryu, D.; Park, E.; Kim, D.; Yan, S.; Lee, J. Y.; Chang, B.; Ahn, K.
H. J. Am. Chem. Soc. 2008, 130, 2394−2395.
(15) Li, H.; Fan, J.; Wang, J.; Tian, M.; Du, J.; Sun, S.; Sun, P.; Peng,
X. Chem. Commun. 2009, 5904−5906.
(16) Chen, X.; Ko, S.; Kim, M. J.; Shin, I.; Yoon, J. Chem. Commun.
2010, 2751−2753.
(17) Matsumoto, T.; Urano, Y.; Shoda, T.; Kojima, H.; Nagano, T.
Org. Lett. 2007, 9, 3375−3377.
(18) Li, X.; Qian, S.; He, Q.; Yang, B.; Li, J.; Hu, Y. Org. Biomol.
Chem. 2010, 8, 3627−3630.
(19) Kim, G.; Lee, K.; Kwon, H.; Kim, H. Org. Lett. 2011, 13, 2799−
2801.
(20) Burckstummer, H.; Weissenstein, A.; Bialas, D.; Wurthner, F. J.
̈
̈
̈
Org. Chem. 2011, 76, 2426−2432.
(21) Zhang, G.; Liu, K.; Li, Y.; Yang, M. Polym. Int. 2009, 58, 665−
673.
̌ ́ ̌
(22) Lunak, S. Jr.; Vynuchal, J.; Vala, M.; Havel, L.; Hrdina, R. Dyes
Pigments 2009, 82, 102−108.
(23) Kuwabara, J.; Yamagata, T.; Kanbara, T. Tetrahedron 2010, 66,
3736−3741.
ASSOCIATED CONTENT
■
S
* Supporting Information
(24) Celik, S.; Ergun, Y.; Alp, S. J. Fluoresc. 2009, 19, 829−835.
DFT calculation details, 1H and 13C NMR data of the
compounds, photophysical data of the probe, and z-matrix of
the calculated compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
̌ ̌ ́
(25) Vala, M.; Vynuchal, J.; Toman, P.; Weiter, M.; Lunak, S. Jr. Dyes
Pigments 2010, 84, 176−182.
(26) Stas, S.; Sergeyev, S.; Geerts, Y. Tetrahedron 2010, 66, 1837−
1845.
(27) Suraru, S.; Zschieschang, U.; Klauk, H.; Wurthner, F. Chem.
̈
Commun. 2011, 1767−1769.
AUTHOR INFORMATION
■
(28) Wienk, M. M.; Turbiez, M.; Gilot, J.; Janssen, R. A. J. Adv.
Mater. 2008, 20, 2556−2560.
(29) Chen, G.; Chiang, C.; Kekuda, D.; Lan, S.; Chu, C.; Wei, K. J.
Polym. Sci., Part A: Polym. Chem. 2010, 48, 1669−1675.
(30) Woo, C. H.; Beaujuge, P. M.; Holcombe, T. W.; Lee, O. P.;
Corresponding Author
*E-mail: zhaojzh@dlut.edu.cn. Group homepage: http://
finechem.dlut.edu.cn/photochem.
́
Frechet, J. M. J. J. Am. Chem. Soc. 2010, 132, 15547−15549.
ACKNOWLEDGMENTS
■
(31) Falzon, M.; Zoombelt, A. P.; Wienk, M. M.; Janssen, R. A. J.
Phys. Chem. Chem. Phys. 2011, 13, 8931−8939.
We thank the NSFC (20972024, 21073028, and 21103015),
the Royal Society (UK) and NSFC (China−UK Cost-Share
Science Networks, 21011130154), the Fundamental Research
Funds for the Central Universities (DUT10ZD212 and
DUT11LK19), the Ministry of Education (SRFDP-
200801410004 and NCET-08-0077), the State Key Laboratory
of Fine Chemicals (KF0802), and the Education Department of
Liaoning Province (2009T015) for financial support.
(32) Kanimozhi, C.; Balraju, P.; Sharma, G. D.; Patil, S. J. Phys. Chem.
B 2010, 114, 3095−3103.
(33) Guo, E.; Ren, P.; Zhang, Y.; Zhang, H.; Yang, W. Chem.
Commun. 2009, 5859−5861.
(34) Jiang, Y.; Wang, Y.; Hua, J.; Qu, S.; Qian, S.; Tian, H. J. Polym.
Sci., Part A: Polym. Chem. 2009, 47, 4400−4408.
(35) Qu, Y.; Hua, J.; Tian, H. Org. Lett. 2010, 12, 3320−3323.
(36) Yamagata, T.; Kuwabara, J.; Kanbara, T. Tetrahedron Lett. 2010,
51, 1596−1599.
(37) Kwon, H.; Lee, K.; Kim, H. Chem. Commun. 2011, 1773−1775.
(38) Yi, L.; Li, H.; Sun, L.; Liu, L.; Zhang, C.; Xi, Z. Angew. Chem.
Int. Ed. 2009, 48, 4034−4037.
(39) Jung, H.; Ko, K.; Kim, G.; Lee, A.; Na, Y.; Kang, C.; Lee, J.;
Kim, J. Org. Lett. 2011, 13, 1498−1501.
REFERENCES
■
(1) Chen, X.; Zhou, Y.; Peng, X.; Yoon, J. Chem. Soc. Rev. 2010, 39,
2120−2135.
(2) Xiong, L.; Zhao, Q.; Chen, H.; Wu, Y.; Dong, Z.; Zhou, Z.; Li, F.
Inorg. Chem. 2010, 49, 6402−6408.
9303
dx.doi.org/10.1021/jo201487m|J. Org. Chem. 2011, 76, 9294−9304

The Journal of Organic Chemistry
Article
(40) Lin, W.; Yuan, L.; Cao, Z.; Feng, Y.; Long, L. Chem.Eur. J.
2009, 15, 5096−5103.
(41) Huo, F.; Sun, Y.; Su, J.; Chao, J.; Zhi, H.; Yin, C. Org. Lett.
2009, 11, 4918−4921.
(42) Yuan, L.; Lin, W.; Yang, Y. Chem. Commun. 2011, 47, 6275−
6277.
(43) Yao, Z.; Bai, H.; Li, C.; Shi, G. Chem. Commun. 2011, 47, 7431−
7433.
(44) Kowalczyk, T.; Lin, Z.; Voorhis, T. V. J. Phys. Chem. A 2010,
114, 10427−10434.
(45) Larkin, J. D.; Fossey, J. S.; James, T. D.; Brooks, B. R.; Bock, C.
W. J. Phys. Chem. A 2010, 114, 12531−12539.
(46) Han, F.; Chi, L.; Liang, X.; Ji, S.; Liu, S.; Zhou, F.; Wu, Y.; Han,
K.; Zhao, J.; James, T. D. J. Org. Chem. 2009, 74, 1333−1336.
(47) Zhang, X.; Chi, L.; Ji, S.; Wu, Y.; Song, P.; Han, K.; Guo, H.;
James, T. D.; Zhao, J. J. Am. Chem. Soc. 2009, 131, 17452−17463.
(48) Zhang, X.; Wu, Y.; Ji, S.; Guo, H.; Song, P.; Han, K.; Wu, W.;
Wu, W.; James, T. D.; Zhao, J. J. Org. Chem. 2010, 75, 2578−2588.
(49) Wu, C.; Tretiak, S.; Chernyak, V. Y. Chem. Phys. Lett. 2007, 433,
305−311.
(50) Dreuw, A.; Head-Gordon, M. J. Am. Chem. Soc. 2004, 126,
4007−4016.
(51) Gabe, Y.; Urano, Y.; Kikuchi, K.; Kojima, H.; Nagano, T. J. Am.
Chem. Soc. 2004, 126, 3357−3367.
(52) Jin, Y.; Xu, Y.; Liu, Y.; Wang, L.; Jiang, H.; Li, X.; Cao, D. Dyes
Pigm. 2011, 90, 311−318.
(53) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.;et al. Gaussian 09W
Revision A.1, Gaussian Inc., Wallingford, CT, 2009.
9304
dx.doi.org/10.1021/jo201487m|J. Org. Chem. 2011, 76, 9294−9304