BODIPY-Based Oligo(ethylene glycol) Dendrons as Fluorescence Thermometers: When Thermoresponsiveness Meets Intramolecular Electron/Charge Transfer

Article abstract of DOI:10.1002/chem.201404292

The temperature-dependent photophysical properties of a series of 4,4-difluoro-4-bora-3a,4a-diaza-s-indacene (BODIPY) derivatives with different oligo(ethylene glycol) (OEG) dendrons were investigated. Weak fluorescence emission was observed for these BOD

Full text of DOI:10.1002/chem.201404292

DOI: 10.1002/chem.201404292
Full Paper
&
Bioimaging
BODIPY-Based Oligo(ethylene glycol) Dendrons as Fluorescence
Thermometers: When Thermoresponsiveness Meets
Intramolecular Electron/Charge Transfer**
Hua Wang,[a] Yongquan Wu,[b] Pan Tao,[c] Xing Fan,[c] and Gui-Chao Kuang*[a]
Abstract: The temperature-dependent photophysical prop-
erties of a series of 4,4-difluoro-4-bora-3a,4a-diaza-s-ind-
acene (BODIPY) derivatives with different oligo(ethylene
glycol) (OEG) dendrons were investigated. Weak fluores-
cence emission was observed for these BODIPY derivatives
in dilute solution with low viscosity. BDP-G0 and BDP-G1-
TEG exhibit a high quantum yield in viscous glycerol solu-
tions, contrary to the moderate and little fluorescence en-
hancement for BDP-G1 and BDP-G2 under the same condi-
tions. The photoinduced electron transfer (PET) may have
quenched the fluorescence, as supported by calculation. In-
terestingly, the thermoresponsive BODIPY derivatives show
heat-induced luminescence enhancement with
a high
signal-to-noise ratio and their emission maxima are depen-
dent on the structures of branched tri(ethylene glycol) moi-
eties. Finally, preliminary studies on the BODIPY derivatives
as intracellular fluorescence indicators in living HeLa cells
were carried out.
Introduction
dendrimers, which display thermoresponsive properties, have
been demonstrated,[6] the emission intensities of these ther-
mometers decrease in the heating process, which is not de-
sired as fluorescence turn-on type of sensors. Therefore, it is
challenging to develop thermoresponsive dendrimers, which
show heat-induced fluorescence intensities.
Molecular fluorescence thermometers with a long wavelength
emission and biocompatible properties are powerful tools as
sensors in biological imaging application.[1] Among various
fluorescence thermometers, the compounds showing strong
fluorescence intensities are good candidates due to their high
signal-to-noise ratio. The most well-reported molecular fluores-
cence thermometers are based on viscosity or polarity sensi-
tive dyes, whose quenching pathway is impaired during the
thermo-dehydration process, which restores their fluorescence
emission.[2] Various kinds of polymeric fluorescence thermome-
ters have been applied in intracellular temperature measure-
ment.[3] However, to realize their widespread biological applica-
tion and to overcome several drawbacks such as illumination
brightness, excitation source fluctuations, and photobleach-
ing,[4] researchers turn their attention to investigate the fluores-
cence properties of thermoresponsive dendrimers due to their
advantages derived from a uniform structure and a tunable
functionality.[5] Although a few works referring to luminescent
4,4-Difluoro-4-bora-3a,4a-diaza-s-indacene (BODIPY, which is
abbreviated as BDP in the text) dyes bare excellent features
such as high extinction coefficients, a long wavelength emis-
sion (ꢀ500 nm), high fluorescence quantum yields, and
narrow emission bandwidths.[7] BDP derivatives therefore have
been adopted in fluorescence sensors and biotechnology. In
fact, a BDP core with a phenyl substituent at the meso position
has been extensively investigated. The spectroscopic proper-
ties could be tuned by the substituents on the BDP cores and
on the benzene moieties.[8] On one hand, the fluorescence in-
tensity would be restored for the meso-phenyl BDP derivative
without ortho substituents in viscous solutions, which usually
slow down intramolecular rotation.[9] These BDP derivatives
were utilized to measure the microviscosity in live cells by de-
termining their fluorescence life time values.[10] On the other
hand, the fluorescence properties are dependent on the sol-
vents polarity, a twisted intramolecular charge-transfer (TICT)
process at polar solvents would be involved for the BDP deriv-
atives containing donor and acceptor units.[11]
[a] H. Wang, Prof. Dr. G.-C. Kuang
Department of Polymer Materials, Shanghai University
Nanchen Road 333, 200444, Shanghai (P.R. China)
E-mail: gckuang@shu.edu.cn
Thermoresponsive BDP derivatives will not only show a spe-
cial photophysical behavior but also provide a convenient way
to alter their fluorescence properties.[2c,12] It is of our interest to
develop a thermoresponsive polymer bearing BDP fluoro-
phores to gain a molecular insight into the thermo-reversible
phase separation.[13] Though several water-soluble amphiphilic
BDP dyes with thermoresponsive branched tri(ethylene glycol)
moieties have been reported, attentions have focused on their
[b] Dr. Y. Wu
Department of Chemistry, Fudan University
Handan Road 220, 200433, Shanghai (P.R. China)
[c] P. Tao, Prof. Dr. X. Fan
College of Chemistry and Chemical Engineering, Chongqing University
Shazheng Road 174, 400044, Chongqing (P.R. China)
[**] BODIPY=4,4-difluoro-4-bora-3a,4a-diaza-s-indacene.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201404292.
Chem. Eur. J. 2014, 20, 16634 – 16643
16634
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
photodynamic therapy[14] and photophysical properties,[15] The
thermoresponsive properties of this class of compounds and
their applications in live cell imaging have not been reported.
Our group is interested in the synthesis of oligo(ethylene
glycol) (OEG) dendrimers/dendronized polymers and the inves-
tigation of their thermoresponsive properties.[13,16] In this work,
a series of amphiphilic BDP derivatives (Scheme 1) were pre-
pared and their thermoresponsive and photophysical proper-
ties were studied. The mechanisms responsible for the fluores-
cence enhancement in aqueous solution during the heating
process were examined. In addition, the potential of the BDP
derivatives as intracellular fluorescent indicators for bioimaging
in HeLa cells was evaluated by laser scanning fluorescence mi-
croscopy (LSFM).
Scheme 2. Synthetic procedure for the BDP derivatives. a) trifluoroacetic
acid (TFA), pyrrole, RT, 18 h. b) p-Chloranil, dichloromethane, ꢁ58C–RT.
c) TEA, BF3·OEt2, RT. d) POCl3, DMF, 1,2-dichloethane, K2CO3, 0–608C, 6 h
(78%).
Scheme 1. Structures of the BDP derivatives studied in the present work.
Results and Discussion
Design and Synthesis of the BDP derivatives
BDP-Gn (n=0, 1, and 2), BDP-G1-TEG, and BDP-G1-Me were
prepared similarly. These BDP derivatives were synthesized by
following well-reported procedures (Scheme 2).[7] Formyl-modi-
fied OEG dendrons followed a classic synthetic pathway, that is
acid-catalyzed condensation with pyrrole, then oxidation with
p-chloranil, and finally complexation with boron trifluoride
etherate, to afford the target BDP derivatives. Three model
compounds BDP-G1-Me, BDP-G1-CHO, and BDP-G1-TEG were
prepared as well. These new compounds were characterized
by NMR spectroscopy and high-resolution mass spectrometry
(see the Supporting Information).
Figure 1. Normalized absorption and emission spectra of a) BDP-G1 (2.0 mm)
in various solvents and b) the BDP derivatives (2.0 mm) in chloroform.
lex =490 nm.
cating no significant donor–acceptor interaction in the ground
state. The absorption and emission peak values and the rela-
tive ffl values are summarized in the Table 1. These BDP deriv-
atives could be divided to three types depending on the ffl
values: 1) BDP-G0 and BDP-G1-TEG with one-folded OEG sub-
stituent on the phenyl ring (close to the BDP fluorophore) are
considered as the first type, both show a weak emission in low
Spectroscopic properties of the BDP derivatives in
dilute solution
The absorption and emission spectra of the BDP derivatives in
various solvents were collected (Figure 1). The spectral profiles
are almost identical and show less than 8 nm shift in various
solvents (except for BDP-G2 in glycerol and BDP-G1-CHO), indi-
Chem. Eur. J. 2014, 20, 16634 – 16643
www.chemeurj.org
16635
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ure 2a). BDP-G1 demonstrated a similar spectral pro-
file and a moderate intensity enhancement, as shown
in Figure 2b. In contrast, BDP-G2 only showed
a small intensity enhancement and even a new band
in a solution containing a high fraction of glycerol
(Figure S1 in the Supporting Information), which is
consistent to the pale emission when the volume
fraction of glycerol exceeds 80% (Figure S2 in the
Supporting Information). The fluorescence enhance-
ment in the viscous glycerol solution was quantita-
tively verified by the quantum yield ffl (Figure 3). The
ffl value of BDP-G0 increased dramatically and
reached 46% (90 vol% fraction glycerol), which is
much higher than those for BDP-G1 (3.3%) and BDP-
G2 (0.7%) under the same conditions. Considering
that the intramolecular rotation is restricted in the
viscous glycerol solution, some other pathways
might be responsible for the low ffl values of BDP-
G1 and BDP-G2 in glycerol.
Table 1. Spectroscopic data for the BDP derivatives.[a]
Comp. Solvent[b] labs lem
ffl
Comp.
Solvent labs lem
ffl
[nm] [nm]
[nm] [nm]
BDP-G0 HEX
PEN
498
500
497
500
495
494
500
499
501
500
503
499
497
503
501
507
501
503
509 0.026 BDP-G1-TEG HEX
497 509
499 511
498 510
0.019
0.030
0.036
0.039
0.196
0.013
0.018
0.392
0.363
0.381
0.408
0.322
0.109
0.082
0.007
0.012
0.008
0.003
512 0.036
510 0.032
512 0.040
512 0.466
510 0.013
515 0.019
512 0.014 BDP-G1-Me
513 0.023
511 0.023
517 0.013
518 0.033
510 0.001
519 0.002
PEN
SIO[c]
SIO[c]
CHCl3
GLY[d]
ACN
CHCl3 500 514
GLY[d]
ACN
497 513
495 509
DMSO
BDP-G1 HEX
PEN
DMSO 500 516
HEX
PEN
SIO[c]
501 508
502 510
500 508
SIO[c]
CHCl3
GLY[d]
CHCl3 504 512
GLY[d]
ACN
501 510
498 506
ACN
DMSO
BDP-G2 HEX
PEN
DMSO 502 511
511 0.008 BDP-G1-CHO HEX
497 509
499 511
496 509
516 0.002
515 0.013
517 0.014
PEN
SIO[c]
SIO[c]
CHCl3
GLY[d]
CHCl3 499 515
503 521/600 0.007
GLY[d]
ACN
491 507 <0.001
494 509 <0.001
The emission peak at the long-wavelength band
for BDP-G2 in a high fraction of glycerol solution
might be ascribed to the emission of a charge-trans-
fer excited state (Figure S3b in the Supporting Infor-
mation).[18] The excitation spectrum revealed that the
long-wavelength band is from the BDP maximum ab-
sorption (Figure S3a in the Supporting Information).
This emission band is strongly dependent on the sol-
vents polarity and viscosity. Considering that the in-
ACN
DMSO
498
503
513 0.001
521 0.002
DMSO 499 524
0.001
[a] Quantum yields (ffl) were measured by using fluorescein as a reference (ffl =0.79
in 0.1m NaOH). [b] ACN=acetonitrile, PEN=pentadecane, HEX=hexane, GLY=glycer-
ol, and SIO=silicon oil. [c] The data determined in silicon oil/chloroform (90:10, v/v),
h=400 MPas. [d] The data determined in a glycerol/water mixture (90:10, v/v), h=
215 MPas.
viscosity solvents and restore their fluorescence intensity in
the viscous glycerol. Because the intramolecular rotation is the
main pathway to impair the fluorescence emission,[9] their
emission intensity would be enhanced in the polar glycerol so-
lution, 2) the second types are BDP-G1 and BDP-G2, which
bear three-folded OEG branches and display negligible fluores-
cence emission in common solvents and show somewhat en-
hancement in the viscous nonpolar silicon oil. Although the in-
tramolecular rotation is restricted in viscous glycerol, the fluo-
rescence emission might be quenched by other pathways, and
3) BDP-G1-Me shows high quantum yields (ffl >0.10) in most
of the organic solvents, whereas BDP-G1-CHO exhibits only
little detectable fluorescence emission (ffl <0.01) in all organic
solvents. Therefore, both BDP derivatives showing solvents in-
dependence are ascribed to the third types. For BDP-G1-Me,
the rotation of the single bond between the benzene moiety
and BDP is greatly restricted due to the methyl substituents on
the BDP core. Therefore, it is reasonable that BDP-G1-Me show
high ffl values in various solvents.[9a] For BDP-G1-CHO, the in-
tramolecular rotation and the enhanced electron effect might
account for its low ffl value. All the photophysical measure-
ments determined in dilute solution served for the investiga-
tion in concentrated aqueous solution at various temperatures,
which will be discussed below.
tramolecular rotation is restricted when dissolving BDP-G2 in
the viscous glycerol or silicon oil, respectively, we can compare
the solvents polarity effect on the emission spectra (Figure S3b
in the Supporting Information). The emission spectrum show
higher intensity in silicon oil than the one in glycerol without
a charge-transfer emission band. Therefore, the charge transfer
impaired the quantum yield could be blocked for BDP-G2 in
low polarity solvents. In contrast, the fluorescence of BDP-G1
only shows little change in both solvents.
Photoinduced electron transfer (PET) quenches the fluores-
cence of BDP-G1 and BDP-G2 in viscous glycerol solutions. Due
to the electron-donating ability of the OEG substituent, it is
plausible that the three-folded OEG dendron donates an elec-
tron to the acceptor BDP fluorophore in the excited state, and
thus impairing the fluorescence emission.[8a] Figure 4 illustrates
the PET process based on the HOMO energy level calculation
of both BDP acceptors and benzene moiety donors at the
B3LYP/6-31G level. The HOMO level of G1 is calculated to be
ꢁ5.85 eV, which is higher than that of BDP (ꢁ5.97 eV). There-
fore, it is plausible that the PET can impair the fluorescence
emission of BDP-G1 and BDP-G2 in glycerol. In contrast, the
four-methyl-substituted BDP displays higher HOMO level than
the BDP without a substituent due to the electron-donating
ability of the methyl group (ꢁ5.38 eV). Therefore, BDP-G1-Me
shows a high ffl value in various solvents because the PET
pathway is not significant. BDP-G1-CHO is expected to exhibit
a weak emission due to a strong PET effect (Figure 4b).
The viscosity-dependent fluorescence intensities of the BDP
derivatives were examined by fluorescence spectroscopy. BDP-
G0 exhibited a negligible fluorescence intensity in pure water
and showed progressively enhancement upon gradually in-
creasing the fraction of glycerol in the solvent mixture (Fig-
The feasibility of the intramolecular PET can be determined
by the thermodynamic free energy, which is dependent on the
Chem. Eur. J. 2014, 20, 16634 – 16643
www.chemeurj.org
16636
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 2. Fluorescence spectra of a) BDP-G0 and b) BDP-G1 in glycerol/water
mixtures with different fractions of glycerol, the concentration of the BDP
derivatives was 2.0 mm. The dotted and solid lines were acquired in pure
water and a glycerol/water mixture (9:1, v/v). lex =490 nm.
Figure 4. a) HOMO level of the BDP fluorophore and the benzene moieties.
b) Schematic PET process in dilute solution.
DGPET ¼ EoxðdÞꢁEredðaÞꢁE0,0ꢁðe2=edÞ
ð1Þ
Where DGPET is the Gibbs free energy of the PET, Eox(d) and
Ered(a) are the oxidation potential of the benzene moieties and
the reduction potential of the BDP, respectively, E0,0 is the exci-
tation energy of the BDP fluorophore, e is the dielectric con-
stant of the solvent, and d is the distance between the charges
in the separated state.
Nagano and coworkers has demonstrated that, with the po-
larity of solvents increasing, the oxidation potential of the ben-
zene moiety decreased and the reduction potential of the BDP
increased.[8a] The E0,0 values show a negligible change in the
various solvents (except for BDP-G1-CHO), whereas the last
term, e2/ed, is usually too small to be taken into consideration.
Therefore, a PET from the donor moiety to the acceptor BDP is
favored due to the decreased free energy in high polarity sol-
vents, which accounts for our observation that both BDP-G1
and BDP-G2 display low quantum yield in polar glycerol
solution.
Figure 3. Plots of the quantum yield versus the glycerol fraction for the BDP
~
&
*
derivatives ( =BDP-G0, =BDP-G1, and =BDP-G2) in glycerol/water
mixtures.
polarity of the solvents. According to the Rehm–Weller equa-
tion [Eq. (1)],[19] the free energy of a PET can be estimated by
the comparison of the redox potentials of both the BDP fluoro-
phore and the benzene moiety.
Chem. Eur. J. 2014, 20, 16634 – 16643
www.chemeurj.org
16637
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Spectroscopic properties of the BDP derivatives in aqueous
solution
There is a delicate balance for molecules to have thermores-
ponsive properties. BDP-G1, BDP-G1-TEG, and BDP-G2 are ther-
moresponsive, that is, these above-mentioned BDP derivatives
are water soluble at low temperature and their aqueous solu-
tions turn into opaque at elevated temperatures. Taking BDP-
G1 as an example, due to the dehydration and hydration of
OEG groups, the transparent yellow solution at room tempera-
ture (258C) turned turbid after heating (508C) and changed
back after cooling (Figure 5). This thermoresponsive process is
reversible and could be repeated for multiple times. In con-
trast, the aqueous solutions of BDP-G0, BDP-G1-Me, and BDP-
G1-CHO are not thermoresponsive.
Figure 6. Plots of the transmittance versus the temperature for aqueous
solutions of a) BDP-G1 and b) BDP-G2. Heating (solid lines) and cooling
(dotted lines) rate is 0.2 Kminꢁ1
.
Figure 5. Aqueous solution of BDP-G1 (0.04 wt%) at 25 and 508C.
perature was lower than their Tcp values but presented drastic
fluorescence emission after heating (Figures S10a and b in the
Supporting Information). The color of their solution changed
from green to yellow (under l=365 nm UV excitation). In con-
trast, BDP-G1-TEG did not show too much changes in bright-
ness as well as color during the heating process (Figure S10c
in the Supporting Information). The photophysical properties
from the apparent emission intensity and the color of the solu-
tions might be ascribed to the different molecule structure
and the excited molecules reactivity during the heating
process.
Turbidity tests were performed by UV/Vis measurements to
investigate the thermoresponsive behavior of BDP-G1 and
BDP-G2 in detail (for BDP-G1-TEG, see the Supporting Informa-
tion). As shown in Figure 6, the turbidity curves plotted tem-
peratures were assembled and the apparent phase transition
temperatures (Tcps) (the temperature reached at 50% of the ini-
tial transmittance at l=700 nm) plotted concentrations were
compared. Regarding the thermoresponsive properties, there
are two important points worthy of note: 1) The Tcp values are
concentration dependent. As the concentration of BDP-G1 de-
creased from 0.12 to 0.04 wt%, the Tcp values significantly in-
creased from 19 to 378C, accompanied by a broader phase-
transition range. In contrast, BDP-G2 only shows six degree
changes in the same concentration range. We presume that
the overall hydrophilicity for small molecules BDP-G1 will be
influenced much more than that of BDP-G2 by concentration.
2) The Tcp values are generation dependent. Besides the smaller
Tcp value changes, BDP-G2 showed a much higher Tcp value
than BDP-G1 at the same concentration, which could be as-
cribed to the high hydrophilicity of the G2 dendron (Figure S5
in the Supporting Information).
The photophysical properties during the thermoresponsive
process were monitored by fluorescence spectra at various
temperatures. The emission intensities of BDP-G1 and BDP-G2
increased by 22- and 61-fold at their corresponding maximum
bands, respectively (Figures 7a and b). As described in the pre-
vious studies, two pathways, intramolecular rotation and PET,
could quench the fluorescence emission of BDP-G1 and BDP-
G2 in dilute solution. The increased microviscosity during the
heating process might suppress the free rotation between the
dipyrromethane framework and the meso-phenyl ring.[9] That
is, the hydrophobic BDP fluorophore that stays in a free rota-
tion state at low temperature (<Tcp) would be wrapped by the
dehydrated OEG groups at high temperature (>Tcp). Thus, the
intramolecular rotation is restricted.[2c] In addition, the hydro-
philic microenvironment around the BDP fluorophore would
change to be hydrophobic due to the dehydration of the OEG
It is more interesting that BDP-G1, BDP-G1-TEG, and BDP-G2
behave as heat-induced fluorescence thermometers. Both BDP-
G1 and BDP-G2 showed very faint fluorescence when the tem-
Chem. Eur. J. 2014, 20, 16634 – 16643
www.chemeurj.org
16638
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
groups after heating.[16d] That is, the polarity around the BDP is
greatly decreased. Thus, the PET pathway is impaired.[8a] Both
effects work in concert to restore the fluorescence emission of
BDP-G1 and BDP-G2 during the heating process.
Noteworthy to mention here is that the emission bands of
both molecular thermometers in the locally excited state (LE)
undergo bathochromic shifts during the heating process. BDP-
G1 underwent a bathochromic shift of 13 nm from 535 to
548 nm, whereas BDP-G2 showed a bathochromic shift of
7 nm (see the Supporting Information), which is attributed to
the donor–acceptor interaction between the benzene moieties
and the BDP fluorophore during the heating process in polar
aqueous solution.[20] In addition, there is a new emission band
at long wavelength (around l=600 nm) observed for BDP-G2
after the temperature exceeded 458C. This new band is attri-
buted to the TICT state, which is well reported in the BDP de-
rivatives containing an electron donor and an electron accep-
tor connected by a single bond (Scheme 3).[11,20] An aqueous
Scheme 3. Photophysical processes for an aqueous solution of BDP-G2.
solution of BDP-G1 did not show the long-wavelength emis-
sion band form a TICT state after heating, this might be attri-
buted to the dehydrated OEG branches, which cannot stabi-
lized the TICT state. In contrast, BDP-G1-TEG shows less intensi-
ty enhancement (about five-fold) during the thermoresponsive
process, accompanied with a little change at the emission
maximum (Figure 7c). Therefore, the three-folded OEG branch-
es behaved as electron-donor account for both the emission
shift and the intensity enhancement. Motivated by these re-
sults, we determined the fluorescence properties of the model
compounds BDP-G1-Me (Figure 7d) and BDP-G1-CHO (Fig-
ure S12 in the Supporting Information) in concentrated aque-
ous solution (c=4.00ꢁ10ꢁ4 m) at different temperatures. Their
emission intensities decreased during heating process.
Figure 7. Temperature-varied fluorescence spectra of a) BDP-G1
(c=4.14ꢁ10ꢁ4 m), b) BDP-G2 (c=4.06ꢁ10ꢁ4 m), c) BDP-G1-TEG (c=3.
41ꢁ10ꢁ4 m), and d) BDP-G1-Me (c=4.00ꢁ10ꢁ4 m). The insets are plots of
fluorescence intensity at a) 547, b) 550, c) 541, and d) 516 nm versus
temperature. lex =490 nm.
Chem. Eur. J. 2014, 20, 16634 – 16643
www.chemeurj.org
16639
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Bioimaging
the fluorescence intensity in live cells. Two possibilities may ac-
count for the fluorescence difference: 1) BDP-G1 and BDP-G2
might have a poor cell permeability due to an increased size
of the dendrons, whereas BDP-G0 passed the cell membrane
smoothly due to its smaller size. 2) The fluorescence intensities
of the BDP derivatives followed the above order are attribut-
able to their lipophilicity, sensor BDP-G0 with high lipophilicity
preferentially localizes in hydrophobic domains of cellular or-
ganelles. The first possibility is excluded due to the observa-
tion that the fluorescent BDP dyes with branched tri(ethylene
glycol) arms still show a strong fluorescence intensity in
cells.[14,21] BDP-G0 is likely to stay in the endocytotic vesicles
and to show enhanced fluorescence intensity in this confined
environments.[9a]
BDP-Gn (n=0, 1, and 2) show little toxicity as determined by
using a standard methyl thiazolyl tetrazolium (MTT, Sigma–Al-
drich) assay in HeLa cell lines. HeLa cells were incubated with
the BDP derivatives up to 50 mm in phosphate-buffered saline
(PBS). Over the course of the test, no apparent toxicity of the
BDP derivatives was observed based on no significant differen-
ces in the cell morphologies, which is in agreement with a pre-
vious report that the BDP sensors have good chemical stability
and generally low toxicity.[13] As shown in Figure S13 in the
Supporting Information, the cellular viabilities were determined
to be higher than 80% after 24 h incubation (except for BDP-
G2 at 50 mm). Therefore, the toxicity of the BDP derivatives at
low micromolar loading concentrations (ꢂ10 mm) is insignifi-
cant over the time scale of the imaging tests.
Conclusion
A practical application of the BDP derivatives in live HeLa
cells bioimaging was developed by LSFM. The untreated HeLa
cells displayed negligible background fluorescence. The cells
were first incubated in the PBS in the presence of BDP-Gn (n=
0, 1, and 2) for 30 min or 4 h. After replacement of the media
to remove the extracellular probe molecules, the cells were
studied under LSFM. Interestingly, BDP-G0 showed very strong
intracellular fluorescence (Figure 8b). The overlay of a bright-
field image and the confocal luminescence showed that the lu-
minescence is localized in the cytoplasm, not in the nucleus or
the cell membrane (Figure 8c). In contrast, BDP-G1 and BDP-
G2 displayed weak and almost no intracellular fluorescence, re-
spectively. (Figures 8e and h) Therefore, the intracellular fluo-
rescence intensity shows the same tendency as that in glycer-
ol/water mixtures, that is, the high generation dendron impairs
In summary, amphiphilic BDP derivatives with different hydro-
philic substituents were successfully prepared. BDP-G1, BDP-
G1-TEG, and BDP-G2 displayed thermoresponsive properties
and enhanced fluorescence intensities after heating. This is the
first report referred to thermoresponsive dendrimers of which
the fluorescence intensity could be enhanced at elevated tem-
perature. Detailed investigations revealed that intramolecular
rotation and PET quench pathway were impaired in the heat-
ing process. Furthermore, these compounds were used in
aqueous solutions as luminescent dyes for imaging in living
HeLa cells. The hydrophilic dendrons were needed for the sol-
ubility in aqueous solution, whereas lipophilic BDP-G0 perme-
ated the cells with ease. Therefore, preliminary cell image stud-
ies revealed that the intracellular fluorescence intensity is lipo-
philicity dependent. Further applications of other fluorescent
probes in cell imaging are underway and will be reported in
due course.
Experimental Section
Materials
Trifluoroacetic acid was purchased from ACROS. Pyridinium chloro-
chromate (PCC), pyrrole, p-chloranil, boron trifluoride diethyl ether-
ate (BF3·OEt2) were purchased from TCI. Dichloromethane and 1,2-
dichloroethane was dried over CaH2. Tetrahydrofuran (THF) was
predried over sodium and then heated to reflux over LiAlH4 before
use. Triethylamine was dried over sodium hydroxide (NaOH) pel-
lets. Other reagents and solvents were purchased at reagent grade
and used without further purification. All synthetic steps were run
under a nitrogen atmosphere. Macherey–Nagel precoated thin-
layer chromatography (TLC) plates (silica gel 60 G/UV254, 0.25 mm)
were used for analysis. Silica gel 60M (Macherey–Nagel, 0.040–
0.063 mm, 200–300 mesh) was used as the stationary phase for
column chromatography. All samples were dried thoroughly under
vacuum prior to analytical measurements to remove strongly ad-
hering solvent molecules.
General instrumentation and measurements
Figure 8. a,d,g) Bright-field images and b,e,h) confocal luminescence of HeLa
cells incubated with BDP-G0, BDP-G1, and BDP-G2 in PBS for 30 min at
258C. c,f,i) Overlay images of the bright-field and the confocal images.
1H and 13C NMR spectra were recorded on a Bruker AV 500 (1H:
500 MHz, 13C: 125 MHz) spectrometer, and chemical shifts are re-
Chem. Eur. J. 2014, 20, 16634 – 16643
www.chemeurj.org
16640
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ported as d values in [ppm] relative to internal tetramethylsilane
(Me4Si). High-resolution MALDI/TOF MS analyses were performed
on Ionspec Ultra instruments. UV/Vis turbidity measurements were
carried out on a PE UV/Vis spectrophotometer (Lambda 35)
equipped with a thermo-controlled bath. An aqueous solution of
the sensor was placed in the spectrophotometer (path length
1 cm) and heated or cooled at a rate of 0.2 Kminꢁ1. The absorp-
tions of the solution at l=700 nm were recorded every 5 s. The
cloud point (Tcp) is determined as the one at which the transmit-
tance at l=700 nm has reached 50% of its initial value. The fluo-
rescence spectra were measured on Horiba Jobin Yvon Fluorolog-3
equipped with a Peltier temperature controller. Confocal imaging
of cells was performed with a modified Olympus FV1000 laser
scanning confocal microscope equipped a continuous-wave NIR
laser operating at l=980 nm (Connet Fiber Optics, China). The vis-
cosity was determined at a speed of 75 revolutions per minute by
the DV-79 Series digital viscometer (Shanghai Nirun Intelligent
Technology Co., Ltd. China).
role-H), 7.07 (d, J=8.8 Hz, 2H; CH), 7.53 (d, J=8.8 Hz, 2H; CH),
7.92 ppm (s, 2H; pyrrole-H); 13C NMR (CDCl3): d=61.71, 67.64,
69.53, 70.31, 70.86, 72.54, 114.67, 118.31, 126.50, 131.40, 132.41,
134.81, 143.41, 147.37, 161.25 ppm; HRMS (MALDI/TOF): m/z calcd
for C21H23BN2O4F2Na: 438.1656 [M+Na]+; found 438.1647.
BDP-G2
BDP-G2 was isolated as an orange viscous liquid (0.30 g, 49% over
1
two steps). H NMR (CD2Cl2): d=1.14–1.18 (m, 27H; CH3), 3.45–3.86
(m, 144H; CH2), 4.08–4.27 (m, 24H; CH2), 4.41 (d, J=12.4 Hz, 6H;
CH2), 6.58 (d, J=6.6 Hz, 8H; CH), 6.87 (s, 2H; pyrrole-H), 7.07 (d, J=
3.8 Hz, 2H; pyrrole-H), 7.90 ppm (s, 2H; pyrrole-H); 13C NMR
(CD2Cl2): d=15.00, 66.38, 68.66, 69.10, 69.46, 69.49, 69.63, 69.68,
69.79, 69.81, 70.41, 70.50, 70.52, 70.56, 70.57, 70.59, 70.65, 70.70,
70.76, 72.28, 72.62, 73.00, 73.04, 106.68, 110.43, 118.53, 128.78,
131.62, 134.01, 134.75, 137.45, 140.78, 143.77, 147.25, 152.54 ppm;
HRMS (MALDI/TOF): m/z calcd for C126H209BN2O48F2Na: 2589.3960
[M+Na]+; found 2589.3964.
Synthesis
BDP-G1-TEG
BDP-G1
BDP-G1-TEG was isolated as an orange viscous liquid (20 mg,
10%). 1H NMR (CDCl3): d=1.20 (t, J=6.3 Hz, 9H; CH3), 3.51–3.92
(m, 46H; CH2), 4.14 (s, 6H; CH2), 4.45 (s, 2H; CH2), 6.55 (d, J=
13.8 Hz, 4H; CH), 6.96 (s, 2H; pyrrole-H), 7.05 (d, J=7.8 Hz, 2H;
CH), 7.53 (d, J=7.7 Hz, 2H; pyrrole-H),7.91 ppm (s, 2H; pyrrole-H);
13C NMR (CDCl3): d=15.18, 27.21, 29.32, 29.52, 29.55, 29.62, 29.70,
29.78, 31.91, 55.07, 66.66, 67.92, 68.91, 69.40, 69.64, 69.84, 70.57,
70.59, 70.73, 70.61, 70.87, 70.95, 72.37, 73.30, 74.89, 107.30, 114.71,
118.30, 118.32, 126.47, 131.39, 132.41, 133.75, 134.83, 137.83,
143.43, 147.40, 152.63, 161.35 ppm; HRMS (MALDI/TOF): m/z calcd
for C52H77BN2O16F2Na: 1056.5267 [M+Na]+; found 1056.5262.
A solution of p-chloranil (0.41 g, 1.65 mmol) in dichloromethane
(10 mL) was added dropwise to a solution of the corresponding al-
dehyde (1.24 g, 1.65 mmol, Scheme S1 in the Supporting Informa-
tion) in dichloromethane (20 mL) at ꢁ58C. The reaction mixture
was stirred at ꢁ58C for 30 min and then the temperature was
risen to RT. The mixture was stirred at RT for another 13 h before
triethylamine (TEA) (5.00 g, 49.41 mmol) and BF3·OEt2 (7.00 g,
49.32 mmol) were added. The solution was stirred at RT for 5 h
before partitioned between brine and dichloromethane. The or-
ganic portion was dried over MgSO4 before concentration in
vacuo. The residue was purified on a silica column by using di-
chloromethane/MeOH (from 50:1 to 30:1) as the eluent to afford
compound BDP-G1 as an orange viscous liquid (0.90 g, 69% over
two steps). 1H NMR (CDCl3): d=1.17–1.21 (m, 9H; CH3), 3.47–3.75
(m, 30H; CH2), 3.83–3.88 (m, 6H; CH2), 4.18 (t, J=4.7 Hz, 4H; CH2),
4.27 (t, J=4.9 Hz, 2H; CH2), 6.54–6.56 (m, 2H; pyrrole-H), 6.83 (s,
2H; CH), 7.01 (d, J=4.1 Hz, 2H; pyrrole-H), 7.92 ppm (s, 2H; pyr-
role-H); 13C NMR (CDCl3): d=8.77, 15.10, 47.34, 66.55, 66.57, 69.11,
69.61, 69.71, 69.74, 70.50, 70.55, 70.59, 70.61, 70.76, 72.58, 110.57,
118.52, 128.52, 131.53, 134.69, 140.80, 143.87, 147.02, 152.48 ppm;
HRMS (MALDI/TOF): m/z calcd for C39H59BN2O12F2Na: 818.4066
[M+Na]+; found 818.4057.
BDP-G1-CHO
POCl3 (1 mL) was added dropwise to a vigorously stirred anhydrous
solution of DMF (1 mL) which was kept in an ice bath under N2.
The resulting pale yellow viscous liquid was allowed to stir at room
temperature for additional 30 min. To this, a solution of BDP-G1
(440 mg, 0.55 mmol) in 1,2-dichloroethane (30 mL) was then slowly
introduced and the resultant brown solution was heated at 608C
for 3 h. The reaction mixture was cooled to RT and poured into an
ice-cold saturated NaHCO3 solution and stirred for 1 h. This mixture
was extracted with dichloromethane (2ꢁ100 mL) and dried over
anhydrous Na2SO4. The solvent was evaporated in vacuo. The resi-
due was purified on a silica column by using dichloromethane/
MeOH (from 100:1 to 70:1) as the eluent to afford compound BDP-
BDP-G1-Me
1
1
BDP-G1-Me was isolated as an orange solid (0.89 g, 55%). H NMR
G1-CHO as an orange viscous liquid (0.35 g, 78%). H NMR (CDCl3):
(CD2Cl2): d=1.14–1.19 (m, 9H; CH3), 1.54 (s, 6H; CH3), 2.50 (s, 6H;
CH3), 3.45–3.83 (m, 36H; CH2), 4.12 (s, 4H; CH2), 4.22 (s, 2H; CH2),
6.02 (s, 2H; pyrrole-H), 6.57 ppm (s, 2H; CH). 13C NMR (CD2Cl2): d=
8.64, 14.10, 14.29, 14.95, 47.43, 66.40, 68.87, 69.50, 69.65, 69.70,
70.37, 70.52, 70.53, 70.64, 72.64, 107.13, 121.08, 129.90, 131.24,
141.39, 143.35, 153.55, 155.45 ppm; HRMS (MALDI/TOF): m/z calcd
for C43H67BN2O12F2Na: 874.4692 [M+Na]+; found 874.4683.
d=1.16–1.21 (m, 9H; CH3), 3.48–3.88 (m, 36H; CH2), 4.20 (s, 4H;
CH2), 4.30 (s, 2H; CH2), 6.72 (s, 1H; pyrrole-H), 6.84 (s, 2H; CH), 7.22
(s, 1H; pyrrole-H), 7.37 (s, 1H; pyrrole-H), 8.14 (s, 1H; pyrrole-H),
8.26 (s, 1H; pyrrole-H), 9.87 ppm (s, 1H; CH); 13C NMR (CDCl3): d=
15.24, 66.72, 66.75, 69.50, 69.84, 69.88, 69.94, 70.76, 70.80, 70.81,
70.97, 72.90, 76.90, 77.15, 77.41, 110.95, 121.49, 128.04, 128.83,
131.87, 134.55, 135.05, 135.12, 135.14, 136.83, 141.85, 142.72,
149.14, 149.27, 152.89, 184.97 ppm; HRMS (MALDI/TOF): m/z calcd
for C40H59BN2O13F2Na: 846.4019 [M+Na]+; found 846.4006.
BDP-G0
The synthesis of BDP-G0 was started from the reported aldehyde[17]
and isolated as an orange viscous liquid (0.65 g, 45%). 1H NMR
(CDCl3): d=3.56–3.60 (m, 1H; OH), 3.63–3.65 (m, 2H; CH2), 3.72–
3.77 (m, 4H; CH2), 3.93 (t, J=4.7 Hz, 2H; CH2), 4.24 (t, J=4.7 Hz,
2H; CH2), 6.54–6.56 (m, 2H; pyrrole-H), 6.97 (d, J=4.1 Hz, 2H; pyr-
Quantum yield of the fluorescence
Fluorescence quantum yields were determined by
method by using a solution of fluorescein (fs =0.79, 0.1m NaOH)
as a reference.[22] The excitation wavelength was l=490 nm. The
a relative
Chem. Eur. J. 2014, 20, 16634 – 16643
www.chemeurj.org
16641
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[1] C. Pietsch, U. S. Schubert, R. Hoogenboom, Chem. Commun. 2011, 47,
8750–8765.
fluorescence quantum yields were calculated based on
Equation (2).
[2] a) K. Okabe, N. Inada, C. Gota, Y. Harada, T. Funatsu, S. Uchiyama, Nat.
Commun. 2012, 3, 1–9; b) T. Tsuji, S. Yoshida, A. Yoshida, S. Uchiyama,
Anal. Chem. 2013, 85, 9815–9823; c) D. P. Wang, R. Miyamoto, Y. Shir-
aishi, T. Hirai, Langmuir 2009, 25, 13176–13182.
2
2
ð2Þ
fx ¼ fs ꢃ ðAsFxnx Þ=ðAxFsns Þ
where As and Ax are the absorbance values of the reference and
samples solutions at their respective excitation wavelengths, Fs and
Fx are the corresponding integrated fluorescence intensities, and n
is the refractive index of the solvent of the samples (nx) or of the
reference (ns). The absorbance of the samples and the references
at their respective excitation wavelengths was kept below 0.1. The
refractive index of the glycerol/water solvent of the sample (nx)
was calculated according to Equation (3).[23]
[3] Y. Hiruta, M. Shimamura, M. Matsuura, Y. Maekawa, T. Funatsu, Y. Suzuki,
E. Ayano, T. Okano, H. Kanazawa, ACS Macro Lett. 2014, 3, 281–285.
[4] C. Y. Chen, C. T. Chen, Chem. Commun. 2011, 47, 994–996.
[5] R. R. Ramireddy, K. R. Raghupathi, D. A. Rorres, S. Thayumanavan, New J.
Chem. 2012, 36, 340–349.
[6] a) D. W. Chang, L. M. Dai, J. Mater. Chem. 2007, 17, 364–371; b) D. W.
Chang, I.-Y. Jeon, J.-B. Baek, L. M. Dai, Chem. Commun. 2010, 46, 7924–
7926; c) D. W. Chang, H.-J. Choi, S.-M. Jung, L. M. Dai, J.-B. Baek, J.
Mater. Chem. 2012, 22, 13365–13373.
[7] a) A. Loudet, K. Burgess, Chem. Rev. 2007, 107, 4891–4932; b) G. Ulrich,
R. Ziessel, A. Harriman, Angew. Chem. 2008, 120, 1202–1219; Angew.
Chem. Int. Ed. 2008, 47, 1184–1201; c) N. Boens, V. Leen, W. Dehaen,
Chem. Soc. Rev. 2012, 41, 1130–1172; d) L. Yuan, W. Lin, K. Zheng, L. He,
W. Huang, Chem. Soc. Rev. 2013, 42, 622–661; e) M. Baruah, W. Qin, N.
nx ¼ nsþ½0:0011625 ꢃ glycerolðm mꢁ1 %Þ ꢃ 1ꢄ
where 1 is specific gravity of glycerol.
ð3Þ
´
Basaric, W. M. D. Borggraeve, N. Boens, J. Org. Chem. 2005, 70, 4152–
Cytotoxicity of the BDP derivatives
4157; f) M. Benstead, G. A. Rosser, A. Beeby, G. H. Mehl, R. W. Boyle, Pho-
tochem. Photobiol. Sci. 2011, 10, 992–999.
The cytotoxicity was measured by performing methyl thiazolyl tet-
razolium (MTT) assays on the HeLa cells lines, which were provided
by the Institute of Biochemistry and Cell Biology (Chinese Academy
of Sciences). The HeLa cells were grown in RPMI 1640 (Roswell
Park Memorial Institute’s Medium) supplemented with 10% FBS
(fetal bovine serum) at 378C under 5% CO2. The BDP derivatives at
10, 20, 30, 40, and 50 mm concentration were added to the wells of
the treatment group. The cells were incubated for 24 h at 378C
under 5% CO2 and then MTT was added to the wells for test.
[8] a) H. Sunahara, Y. Urano, H. Kojima, T. Nagano, J. Am. Chem. Soc. 2007,
129, 5597–5604; b) Y. Gabe, Y. Urano, K. Kikuchi, H. Kojima, T. Nagano, J.
Am. Chem. Soc. 2004, 126, 3357–3367; c) T. Matsumoto, Y. Urano, T.
Shoda, H. Kojima, T. Nagano, Org. Lett. 2007, 9, 3375–3377; d) M. Isik, T.
Ozdemir, I. S. Turan, S. Kolemen, E. U. Akkaya, Org. Lett. 2013, 15, 216–
219; e) J. J. Shie, Y. C. Liu, Y. M. Lee, C. Lim, J. M. Fang, C. H. Wong, J.
Am. Chem. Soc. 2014, 136, 9953–9961.
[9] a) M. K. Kuimova, G. Yahioglu, J. A. Levitt, K. Suhling, J. Am. Chem. Soc.
2008, 130, 6672–6673; b) I. Lꢂpez-Duarte, T. T. Vu, A. Izquierdo, J. A.
Bull, M. K. Kuimova, Chem. Commun. 2014, 50, 5282–5284; c) Z. G.
Yang, Y. X. He, J. H. Lee, N. Park, M. Suh, W. S. Chae, J. F. Cao, X. J. Peng,
H. Jung, C. Kang, J. S. Kim, J. Am. Chem. Soc. 2013, 135, 9181–9185;
d) M. A. H. Alamiry, A. C. Benniston, G. Copley, K. J. Elliott, A. Harriman,
B. Stewart, Y.-G. Zhi, Chem. Mater. 2008, 20, 4024–4032.
Live cell imaging
HeLa cells were placed on 14 mm glass coverslips and allowed to
adhere for 12 h. The cells were washed with PBS and then incubat-
ed solely with BDP-G0, BDP-G1, and BDP-G2 in PBS (pH 7.4) at
258C for 30 min, 4 h and 4 h, respectively. Cell imaging was then
carried out after washing the cells with PBS. The excitation wave-
length is l=488 nm, and the emission wavelength is l=500–
600 nm.
[10] L. Wang, Y. Xiao, W. M. Tian, L. Z. Deng, J. Am. Chem. Soc. 2013, 135,
2903–2906.
[11] a) E. Lager, J. Liu, A. Aguilar-Aguilar, B. Z. Tang, E. PeÇa-Cabrera, J. Org.
Chem. 2009, 74, 2053–2058; b) Y. Hong, J. W. Y. Lam, B. Z. Tang, Chem.
Commun. 2009, 4332–4353; c) R. R. Hu, E. Lager, A. Aguilar-Aguilar, J. Z.
Liu, J. W. Y. Lam, H. H. Y. Sung, I. D. Williams, Y. C. Zhong, K. S. Wong, E.
PeÇa-Cabrera, B. Z. Tang, J. Phys. Chem. C 2009, 113, 15845–15853.
[12] a) R. Parꢃs, I. Quijada-Garrido, O. Garcꢃa, M. Liras, Macromolecules 2011,
44, 80–86; b) A. Nagai, K. Kokada, J. Miyake, Y. Cyujo, J. Polym. Sci. Part
A 2010, 48, 627–634; c) M. Liras, J. M. Carcꢃa-Garcꢃa, I. Quijada-Garrida,
A. Gallardo, R. Parꢃs, Macromolecules 2011, 44, 3739–3745.
[13] S. Li, K. Liu, G. C. Kuang, T. Masuda, A. Zhang, Macromolecules 2014, 47,
3288–3296.
[14] S. Atilgan, Z. Ekmekci, A. L. Dogan, D. Guc, E. U. Akkaya, Chem.
Commun. 2006, 4398–4400.
Computational methods
All gas-phase calculations were carried out by using Gaussian 09
suite of programs.[24] The B3LYP[25] hybrid functional in conjunction
with the 6-31G(d) basis set[26] is utilized to optimize all geometries
fully and to perform the harmonic vibrational analyses for confirm-
ing minima (all real frequencies).
[15] a) S. L. Zhu, J. T. Zhang, G. K. Vegesna, F.-T. Luo, S. A. Green, H. Y. Liu,
Org. Lett. 2011, 13, 438–441; b) S. L. Zhu, J. T. Zhang, G. K. Vegesna, R.
Pandey, F.-T. Luo, S. A. Green, H. Y. Liu, Chem. Commun. 2011, 47, 3508–
3510; c) S. L. Zhu, J. G. Zhang, G. Vegesna, A. Tiwari, F.-T. Luo, M. Zeller,
R. Luck, H. H. Li, S. Green, H. Y. Liu, RSC Adv. 2012, 2, 404–407.
[16] a) W. Li, A. Zhang, K. Feldman, P. Walde, A. D. Schlꢄter, Macromolecules
2008, 41, 3659–3667; b) W. Li, A. Zhang, A. D. Shclꢄter, Chem. Commun.
2008, 5523–5525; c) W. Li, D. L. Wu, A. D. Schlꢄter, A. Zhang, J. Polym.
Sci. Part A 2009, 47, 6630–6640; d) M. J. N. Junk, W. Li, A. D. Schlꢄter, G.
Wegner, H. W. Spiess, A. Zhang, D. Hinderberger, Angew. Chem. 2010,
122, 5818–5823; Angew. Chem. Int. Ed. 2010, 49, 5683–5687.
[17] W. Wang, A. D. Q. Li, Bioconjugate Chem. 2007, 18, 1036–1052.
[18] a) M. Kollmannsberger, K. Rurack, U. Resch-Genger, J. Daub, J. Phys.
Chem. A 1998, 102, 10211–10220; b) K. Rurack, M. Kollmannsberger, U.
Resch-Genger, J. Daub, J. Am. Chem. Soc. 2000, 122, 968–969.
[19] D. Rehm, A. Well, Isr. J. Chem. 1970, 8, 259–271.
Acknowledgements
This work was financially supported by the National Natural
Science Foundation of China (Nos. 21034004, 21204047, and
21374058). We thank Prof. Afang Zhang (Shanghai University),
Prof. Anjun Qin (South China University of Technology), and
Prof. Xiaohua Chen (Chongqing University) for their help, and
Dr. Hongmei Deng for her assists with the NMR measurements.
Keywords: BODIPY · dendrimers · fluorescence · imaging
agents · thermoresponsiveness
[20] Z. R. Grabowski, K. Rotkiewicz, W. Rettig, Chem. Rev. 2003, 103, 3899–
4301.
Chem. Eur. J. 2014, 20, 16634 – 16643
www.chemeurj.org
16642
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[21] G. K. Vegesna, S. R. Sripathi, J. T. Zhang, S. T. Zhu, W. L. He, F.-T. Luo, W. J.
Jahng, M. Frost, H. Y. Liu, ACS Appl. Mater. Interfaces 2013, 5, 4107.
[22] J. Q. Umberger, V. K. LaMer, J. Am. Chem. Soc. 1945, 67, 1099–1109.
[23] D. Basker, Analyst 1978, 103, 185–186.
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg,
S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cio-
slowski, D. J. Fox, Gaussian Inc., Wallingford CT, 2009.
[25] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652; b) C. Lee, W. Yang,
R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[24] Gaussian 09, Revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Ko-
bayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyen-
gar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B.
Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
[26] a) A. D. McLean, G. S. Chandler, J. Chem. Phys. 1980, 72, 5639–5648;
b) R. Krishnan, J. S. Binkley, R. Seeger, J. A. Pople, J. Chem. Phys. 1980,
72, 650–654; c) T. Clark, J. Chandrasekhar, G. W. Spitznagel, P. von R.
Schleyer, J. Comput. Chem. 1983, 4, 294–301; d) M. J. Frisch, J. A. Pople,
J. S. Binkley, J. Chem. Phys. 1984, 80, 3265–3269.
Received: July 8, 2014
Published online on October 21, 2014
Chem. Eur. J. 2014, 20, 16634 – 16643
www.chemeurj.org
16643
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim