A Modular Class of Fluorescent Difluoroboranes: Synthesis, Structure, Optical Properties, Theoretical Calculations and Applications for Biological Imaging

Article abstract of DOI:10.1002/chem.201601915

Ten borylated bipyridines (BOBIPYs) have been synthesized and selected structural modifications have been made that allow useful structure–optical property relationships to be gathered. These systems have been further investigated using DFT calculations and spectroscopic measurements, showing blue to green fluorescence with quantum yields up to 41 %. They allow full mapping of the structure to determine where selected functionalities can be implemented, to tune the optical properties or to incorporate linking groups. The best derivative was thus functionalised with an alkyne linker, which would enable further applications through click chemistry and in this optic, the stability of the fluorophores has been evaluated.

Full text of DOI:10.1002/chem.201601915

DOI: 10.1002/chem.201601915
Full Paper
&
Cell Imaging
A Modular Class of Fluorescent Difluoroboranes: Synthesis,
Structure, Optical Properties, Theoretical Calculations and
Applications for Biological Imaging
Sylvestre P. J. T. Bachollet,[a, b] Daniel Volz,[c] Bꢀla Fiser,[d] Stephan Mꢁnch,[b]
Franziska Rçnicke,[e] Jokin Carrillo,[a] Harry Adams,[a] Ute Schepers,[e] Enrique Gꢂmez-
Bengoa,[d] Stefan Brꢃse,*[b] and Joseph P. A. Harrity*[a]
Abstract: Ten borylated bipyridines (BOBIPYs) have been
synthesized and selected structural modifications have been
made that allow useful structure–optical property relation-
ships to be gathered. These systems have been further in-
vestigated using DFT calculations and spectroscopic meas-
urements, showing blue to green fluorescence with quan-
tum yields up to 41%. They allow full mapping of the struc-
ture to determine where selected functionalities can be im-
plemented, to tune the optical properties or to incorporate
linking groups. The best derivative was thus functionalised
with an alkyne linker, which would enable further applica-
tions through click chemistry and in this optic, the stability
of the fluorophores has been evaluated.
tural modifications.[8] However, despite the numerous achieve-
ments reached in this field, there is still a large interest in the
discovery of new fluorophore families that offer short synthesis
routes, facile structure modifications and good optical proper-
ties.[9]
Introduction
Fluorescent molecules have been of particular interest recently,
and they have found applications in numerous fields, such as
materials sciences,[1] laser dyes,[2] fluorescent switches,[3] che-
mosensors,[4] photodynamic therapy[5] and biological imaging.[6]
Organoboron compounds constitute a popular class of fluores-
cent molecules, and the well-established boron-dipyrrome-
thene (BODIPY) series are a widely employed sub-class because
of their excellent optical properties, including their high quan-
tum yields and their strong stability to physiological condi-
tions.[7] Another useful aspect of BODIPYs is that their emission
wavelength can be systematically fine-tuned via minimal struc-
During a program of research in our labs focused on cyclo-
addition approaches to benzene-[10] and heteroaromatic-boron-
ic acid derivatives,[11] we uncovered a novel intramolecular
cycloaddition paradigm that took advantage of the in situ for-
mation of a Lewis acid–base complex (Scheme 1).[12] We were
intrigued to find that products derived from this process
exhibited strong fluorescence. Specifically, although many dyes
contain the difluoroboryl group,[13] these are generally coordi-
nated to two heteroatoms. Four-coordinate organoboron com-
pounds based on N,C chelates are emerging as a useful sub-
set of this family of fluorescent molecules,[14] although only
a few reports on analogues possessing BF2 group have been
reported to-date.[15] As our cycloaddition chemistry offered the
opportunity to assemble bespoke aromatic difluoroboranes,
with the potential to vary substituents as a means of easily tai-
loring properties, we undertook an investigation of this novel
class of fluorescent compounds, in order to better understand
[a] S. P. J. T. Bachollet, Dr. J. Carrillo, H. Adams, Prof. J. P. A. Harrity
Department of Chemistry, University of Sheffield
Brook Hill, Sheffield, S3 7HF (UK)
E-mail: j.harrity@sheffield.ac.uk
[b] S. P. J. T. Bachollet, S. Mꢀnch, Prof. S. Brꢁse
Institute of Organic Chemistry, Karlsruhe Institut of Technology
Fritz-Haber-Weg 6, 76131 Karlsruhe (Germany)
E-mail: stefan.braese@kit.edu
[c] Dr. D. Volz
Cynora GmbH
Werner-von-Siemens-Straße 2–6, 76646 Bruchsal (Germany)
[d] B. Fiser, E. Gꢂmez-Bengoa
Departamento de Quꢃmica Organica I
Universidad del Paꢃs Vasco
Manuel Lardizabal 3, 20018 San Sebastian (Spain)
[e] F. Rçnicke, Prof. U. Schepers
Institute of Toxicology and Genetics
Karlsruhe Institute of Technology, Hermann-von-Helmholtz-Platz 1
76344 Eggenstein-Leopoldshafen (Germany)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201601915.
Scheme 1. Directed cycloaddition approach to fluorescent BOBIPYs.
Chem. Eur. J. 2016, 22, 1 – 10
1
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
the scope and origin of their photophysical properties. We
report herein the synthesis of a series of borylated bipyridines
(BOBIPYs), their photophysical properties and the use of DFT
calculations to account for the origin of their fluorescent be-
haviour.
4b,c and simple pyridyl analogue 4d were synthesized in
acceptable overall yields. The directed cycloadditions are
known to produce side-products bearing a dialkynylborane
unit.[17] In this regard, three BOBIPYs 5a–c were isolated that
each contained alkynylboranes and we therefore decided to
study the fluorescent properties of these compounds in order
to establish the influence of the substituents at boron. Overall
therefore, nine pyridyl boranes were generated within a simple
reaction scheme that allowed rapid diversity at all key points
in the fluorophore framework.
Results and Discussion
Synthesis route to BOBIPYs
We were able to grow suitable quality crystals of com-
pounds 4a,f and 5a for characterization by single-crystal X-ray
diffraction (Figure 1).[18] The structures confirmed the four-coor-
dinate boron center in each case, with the polycyclic core lo-
cated in one plane within an extended aromatic system. Only
the substituents at the boron and the phenyl ring originating
from the alkyne are located out of the main plane of the mole-
cules. The obtained structural information of 4 f has been used
to validate the DFT calculations and to interpret photo-physical
properties.
Our cycloaddition chemistry offered the opportunity to pre-
pare benzene[16] and pyridine[12] substituted difluoroboranes,
however, looking ahead to applications of our fluorescent mol-
ecules for cell imaging, we chose the latter compounds as the
polar heterocycle core offered the prospect of increased water
solubility versus a benzene derived analogue. Moreover, in an
effort to maximise our chances of obtaining a powerful fluoro-
phore, we first set out to synthesise a target with an extended
aromatic system. Accordingly, amidrazone 2a was prepared
from the corresponding nitrile and hydrazine, and condensa-
tion with 9,10-phenanthrenequinone afforded triazine 3a
(Scheme 2). Finally, cycloaddition with potassium phenylethy-
nyltrifluoroborate afforded the first BOBIPY 4a in good yield
over the three steps. In order to investigate the influence of
the conjugation pattern of the 3-borylated pyridine moiety,
BOBIPYs 4e and 4 f were also prepared in modest to good
yields. In addition, the structure of the Lewis basic donor was
varied and BOBIPYs containing isomeric isoquinoline donors
Photophysical Properties of BOBIPYs 4a–f, 5a–c
The photophysical properties of all new fluorophores are sum-
marized in Tables 1 and 2.[19] We note that excitation spectra
were measured by monitoring the peak emission of the re-
spective compounds and that emission spectra were obtained
by excitation at 365 nm for each sample. Owing to the high di-
Scheme 2. Synthesis of BOBIPYs 4a–f and 5a–c.
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
2
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Table 2. Excited state lifetime and estimation of radiative and non-radia-
tive decay constants kr and knr in dichloromethane.
Compound
t [ns]
F
kr [108 sꢀ1
]
knr [108 sꢀ1
]
4a
4b
4c
4d
4e
4 f
5a
5b
5c
2.6
2.0
1.8
1.6
3.7
2.1
1.4
7.1
1.4
0.33
0.08
0.23
0.12
0.41
0.13
0.13
0.11
0.11
1.2
0.4
1.3
0.8
1.1
0.6
0.9
0.2
0.8
2.6
1.0
4.3
5.5
1.6
4.2
6.2
1.2
6.3
ed. This set of equations is valid for a mono-exponential emis-
sion decay, for example, for fluorescence of organic mole-
cules.[21] According to the rule of Kasha and Vavilov,[22] F is in-
dependent of the observed wavelength in those cases and the
same applies for kr and t. Based on these assumptions, we cal-
culated values for kr and knr, which are given in Table 2.
The relative error on t is usually in the order of 25%, while
the F can be determined with an absolute error of ꢁ0.02. In
the worst case (4b) this means an additional relative error of
25%. Due to propagation of this uncertainty, the relative error
of kr is 35%. For knr, the relative error is roughly the same. With
this, it is possible to draw qualitative conclusions based on this
data.
Figure 1. ORTEPs of BOBIPY 4a (top left), 4 f (top right) and 5a (bottom).
Hydrogen atoms are omitted for clarity. Selected bond lengths [ꢅ] and
angles [8]: 4a: N1-B1 1.728(2), C26-B1 1.719(2), F1-B1 1.345(2), F2-B1 1.310(1);
N1-C9-C10-C26 ꢀ1.19(1), C8-C9-C10-N2 7.31(2); 4 f: N1-B1 1.657(3), C14-B1
1.620(3), F1-B1 1.384(2), F2-B1 1.387(3); N1-C9-C10-C14 ꢀ0.62(2), C8-C9-C10-
N2 ꢀ1.28(3); 5a: N1-B1 1.640(7), C26-B1 1.621(8), C46-B1 1.591(9), C48-B1
1.577(1); N1-C9-C10-C26 ꢀ1.21(7), C8-C9-C10-N2 ꢀ4.90(9).
As indicated in Table 2, the new BOBIPYs show promising
properties with PLQY values up to 41% and an emission color
from blue to green with ideal Stokes shifts. The different
trends, which can be related to the molecular structure, will be
discussed below. Additionally, BOBIPYs also show solid state lu-
minescence. This has been established for samples 4a, 4d and
4e. The emission spectra as well as the PLQY values differ from
solution measurements, as expected: 4a, 4d and 4e emit in
solid state at 490 nm, 450 nm and 504 nm, respectively, with
PLQY values of 0.24, 0.20 and 0.19. The overlap between ab-
sorption and emission spectra in solution suggests that com-
plex, intermolecular energy transfer processes occur in solid
state, where the chromophores are close to each other.
versity of the structures synthesized herein, the emission color
in dichloromethane solution varied between 420 and 473 nm,
the absorption spectra are significantly different for each sub-
stitution pattern and both Stokes shifts and photoluminescent
quantum yield (PLQY) cover a broad range. The emission spec-
tra were broad and ill-structured.
Table 2 contains the excited state decay time t for 4a–f and
5a–c. If there are no quenching factors (impurities, trace
amounts of solvent) present,[20] the PLQY F can be generally
formulated as a function of the decay lifetime t as F=krt,
with kr: radiative decay rate. Additionally, with the relation:
tꢀ1 = kr + knr, the non-radiative decay rate knr can be calculat-
DFT calculations
Overview of structural and electronic properties
In order to gain an insight into the origin of the fluorescence
and the effects that cause the surprising and significant differ-
ence between the optical properties of these closely related
compounds, DFT calculations were carried out. Fluorophores
4a–f and 5a–c, together with a BF2-containing BODIPY ana-
logue 6 (Figure 2), were optimized using the B3LYP function-
al[23] combined with the 6-31G(d) basis set[24] as implemented
in Gaussian 09 program package.[25] The level of theory B3LYP/
6-31G(d) was selected based on the fact that it was previously
successfully applied to calculate similar systems.[26]
Table 1. Basic spectroscopic data of fluorophores 4a–f and 5a–c in di-
chloromethane.
Compound labs [nm] e [mꢀ1 cmꢀ1
] lem [nm] Stokes shift [nm] PLQY F
4a
4b
4c
4d
4e
4 f
5a
5b
5c
388
377
392
373
421
376
400
381
408
22921
20497
13358
17901
40587
15396
25245
29729
17218
473
452
467
424
465
423
456
420
453
85
75
75
51
44
47
56
39
45
0.33
0.08
0.23
0.12
0.41
0.13
0.13
0.11
0.11
Furthermore, TD-DFT calculations were carried out at the
same level of theory. Solvent effects of dichloromethane and
water were taken into account for the optimizations at the
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
3
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Table 3. LUMO and HOMO energies with the corresponding energy gaps
calculated in gas phase, dichloromethane (DCM) and water.
HOMO [eV]
LUMO [eV]
Gap [eV]
Gas
DCM Water Gas
DCM Water Gas DCM Water
4a ꢀ5.77 ꢀ5.99 ꢀ6.02 ꢀ2.50 ꢀ2.56 ꢀ2.58 3.27 3.43 3.44
4b ꢀ5.68 ꢀ5.89 ꢀ5.92 ꢀ2.32 ꢀ2.24 ꢀ2.24 3.36 3.65 3.68
4c ꢀ5.79 ꢀ5.97 ꢀ5.99 ꢀ2.50 ꢀ2.56 ꢀ2.57 3.29 3.41 3.42
4d ꢀ5.80 ꢀ5.99 ꢀ6.02 ꢀ2.19 ꢀ2.23 ꢀ2.24 3.61 3.76 3.78
4e ꢀ5.85 ꢀ6.05 ꢀ6.08 ꢀ2.46 ꢀ2.58 ꢀ2.60 3.39 3.47 3.48
4 f ꢀ6.20 ꢀ6.34 ꢀ6.36 ꢀ2.35 ꢀ2.43 ꢀ2.44 3.85 3.91 3.92
5a ꢀ5.51 ꢀ5.91 ꢀ5.96 ꢀ2.45 ꢀ2.53 ꢀ2.55 3.06 3.38 3.41
5b ꢀ5.54 ꢀ5.84 ꢀ5.89 ꢀ2.28 ꢀ2.23 ꢀ2.22 3.26 3.61 3.67
5c ꢀ5.47 ꢀ5.89 ꢀ5.93 ꢀ2.44 ꢀ2.50 ꢀ2.52 3.03 3.39 3.41
Figure 2. General structure of BOBIPYs (4a–f, 5a–c) and BODIPY 6 used for
comparison of optical properties.
same level of theory by applying the conductor-like polarizable
continuum model. The intramolecular reorganization energies
(lhole and lelectron) of the studied compounds were also comput-
ed. The acquired X-ray structure of compound 4 f (c.f. Figure 1)
was compared to its gas phase optimized counterpart to verify
the selected level of theory from a structural point of view.
Root-mean-square deviation (RMSD) was calculated to measure
the difference between the experimental and calculated geo-
metries. Pleasingly, the RMSD value was found to be rather
small (0.224 ꢅ) proving confidence that the applied level of
theory was appropriate.[27]
arise from the fact that the rotation of these groups lead to
different minima in different phases, which changes the locali-
zation of the HOMOs. Nevertheless, the solvent phase results
are consistent and match the experimental conditions. For
these reasons, the following discussion will be focused on the
results obtained in dichloromethane.
Molecular orbital plots and orbital energies for BOBIPYs
4a–f, 5a–c and BODIPY 6 were calculated and the results are
summarized in Table 3 and molecular orbital plots of selected
structures presented in Figure 3.[28] The HOMO to LUMO excita-
tion is dominant in most of the BOBIPYs, in a similar manner
to BODIPY 6. There are two other excitations (HOMO-2 to
LUMO and HOMO to LUMO+1), which are also important in
some cases, and this issue will be discussed later. The HOMO
and LUMO orbital energies of gas and solvent phase calcula-
tions were compared and, in most cases, the energies
decreased when solvents were incorporated (Table 3).[28]
The orbital localization is very similar in both phases for
BODIPY 6 and 4a–f, but not for 5a–c (Figure 3).[28] Specifically,
5a–c are the biggest systems within the studied set of mole-
cules because of the two phenylacetylene groups substituting
the boron. The deviations between the orbital localizations
Structure–photophysics relationship
The impact of the molecular structure of the BOBIPYs on the
photophysical properties was next investigated. For conven-
ience, we will discuss structure-property relationships for three
variables (Figure 4), and have des-
ignated these the northern ring (I),
the borane (II) and the southern
ring (III). As highlighted earlier, the
synthetic approach employed pro-
vided a modular method for the
structural modification of both the
northern and southern rings, while
the substitution of the boron
Figure 4. Three easily modifia-
group results from the cycloaddi-
tion step.
ble positions in the general
skeleton of the BOBIPYs.
Figure 3. Molecular orbital plots and orbital energies of 4a, 5a and BODIPY 6 along their optimized structures (middle row) in gas phase and in solvents.
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
4
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Variation of the northern ring (I) and borane (II) on photophy-
sics properties
the BF2 compounds, the emission is red shifted, the PLQY is
much higher and the Stokes shifts are larger. Analysis of kr and
knr reveals that the radiative processes are similar, while non-ra-
The impact of the substitution pattern of the northern ring on
photophysical properties was assessed by comparing com-
pounds 4a–d and 5a–c. Moreover, the effect of varying the
nature of the borane could be observed by comparing each
pair 4a/5a to 4c/5c, where the substitution pattern around
the boron atoms is changed from B-F to B-(phenylacetylide)
(Figure 5). Figure 5a and 5b show the absorption and emission
spectra for these seven compounds.
diative processes seem to be more dominant, yielding an over-
all decrease of the PLQY. This suggests that the B-substituents
affect the excited molecules, either by opening up addition
quenching pathways (e.g., by vibronic quenching) or by steric
or electronic effects.
In all seven cases, several transitions of similar energy and
relative extinction are present in the absorption spectra, even
though they are poorly resolved for 4a/4c and 5a/5c. The
emission spectra show one broad, ill-structured emission in
each case, while the Stokes shifts are fairly large. This indicates
a certain charge-transfer character of the emissive transition.
Analysis of the photophysical properties of 4a–d and
5a–c suggests that either the HOMO or the LUMO is likely to
be localized on the quinoline/isoquinoline moiety, while the
BR2 units seems to have an indirect effect on the photophysi-
cal properties, caused by steric or electronic effects. A strong
charge-transfer character is likely.
DFT analysis of variation of the northern ring (I) and borane
(II)
In general, the LUMO is mainly localized on the heterocycle
next to the BF2 moiety, whereas the HOMO spreads over
almost the whole molecule with the exception of the borane
moiety (Figure 3).[28] The phenyl group does not contribute to
the LUMO and slightly contributes to the HOMO of these mol-
ecules, probably because its out of the plane conformation
offers less electronic coupling. The LUMO energies are the
same for 4a and 4c (ꢀ2.56 eV, in DCM), while there is a small
difference between their HOMO energies (0.02 eV). Thus, the
HOMO–LUMO gaps are also very close to each other (3.43 and
3.41 eV, for 4a and 4c, respectively). The calculated maximum
absorption wavelength of 4a and 4c are also close (416.3 and
418.7 nm) and both of them assigned to the S!S1 electronic
transition with an excitation energy of 2.98 and 2.96 eV,
respectively.[28]
Figure 5. Normalized absorption (left) and emission spectra (right) of com-
pounds 4a–d (a) and 5a–c (b) in CH2Cl2. The color of the ring systems indi-
cates the color of the respective spectra in (a) and (b).
The qualitative shape of the spectra is dominated by the
ring systems: 4b, 4d and 5b differ from the other compounds.
Their PLQY is considerably lower and the absorption spectra
are more structured and the low-energy bands associated with
the HOMO–LUMO transition are blue-shifted compared to 4a/
4c and 5a/5c, while the p–p* transitions around 260 nm are
not. The emission spectra are also blue shifted, suggesting that
the bandgap of 4b, 4d and 5b is larger. Analysis of the excit-
ed state lifetime reveals interesting trends: both radiative rates
kr and non-radiative rates knr are considerably lower. The low kr
values over-compensate the knr values, yielding a very low
overall PLQY.
Compared to 4a or 4c, 4b has higher LUMO (ꢀ2.24 eV) and
HOMO (ꢀ5.89 eV) energies and its energy gap is also wider
(3.65 eV). This is in good agreement with the fact that its mea-
sured emission wavelength is smaller. The phenanthrene
moiety does not contribute to the LUMO of 4b, which could
be one of the reasons why this compound has a lower quan-
tum yield than its isomers (4a and 4c). The calculated labs
(371.7 nm) of 4b with the strongest oscillator strength is in
very good agreement with the measured value (377.0 nm). Its
excitation wavelength is assigned to the S0!S2 electronic tran-
sition from HOMO to LUMO+1 where the excitation energy is
3.34 eV.[28]
The substituents attached at boron do not seem to affect
the absorption spectra too much—the spectra are almost su-
perimposable, which shows that photoexcitation initially leads
to similar excited states. This suggests that neither the HOMO,
nor the LUMO are localized on the BR2 units. On the other
hand, the emission spectra are significantly different. 4a/5a
and 4c/5c have similar energies, while 4b/5b are slightly
higher, showing that the main trends seem to be dictated by
the relative orientation of the quinoline/isoquinoline rings. For
The measured emission wavelength of 4d is even smaller
than 4b, which is in agreement with its bigger HOMO–LUMO
gap (3.76 eV). The measured and calculated absorption
maxima are very close to each other (Dl=4 nm) and assigned
to the S0!S1 transition, but the excitation energy is higher by
around 0.3 eV. Despite the contribution of the phenanthrene
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
5
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
moiety of 4d to its LUMO, the measured quantum yield is
smaller (0.12) than in the case of 4a–c, probably because of
the decreased area of delocalization when changing from
a (iso)quinoline to a pyridine as directing group. Finally, the
calculated labs values of 5a-c are in good agreement with the
measured ones. In these cases, the electronic transition with
the strongest oscillator strength is S0!S3 from HOMO-2 to
LUMO and HOMO to LUMO+1 for 5a/5c and 5b, respective-
ly.[28]
pounds analyzed in this study: besides the fact that the emis-
sion shows a slightly structured profile, the absorption spectra
shows an additional high energy band around 230 nm. Also,
the relative extinction coefficient of the low energy bands
around 420 nm is almost a factor of two higher than for all
other compounds in the study (see Table 1). These results
point towards a more localized transition for this compound,
especially compared to reference compound 4a. In summary,
variation of position III allows for an even stronger manipula-
tion of the optoelectronic properties than positions I and II.
Overall therefore, these findings are in agreement with the
photophysical measurements, as they show that the phenyla-
cetylene substituents affect the photophysical properties of
the molecules (different electronic transitions), but the position
of the quinoline/isoquinoline rings is more crucial.
DFT analysis of variation of the southern ring (III)
The HOMO of 4 f is less delocalised because of its structure,
which lowers its energy and thus results in the largest HOMO–
LUMO gap (3.91 eV) within the whole set of BOBIPYs. The high-
est quantum yield (0.41) was found in the case of 4e and it
also has the lowest reorganization energy for electron transfer
(lelectron =0.24 eV).[28] In 4e the whole structure (except the
phenyl group) is arranged in one plane and this helps the de-
localization, while in 4a there is a slight twist between the iso-
quinoline and the phenanthrene moiety (Figure 7), which re-
sults in a less electronically coupled system and a smaller
quantum yield than in 4e.
Variation of the southern ring (III) on photophysics properties
In order to examine the effect of the southern ring on photo-
physical properties, we recorded absorption and emission
spectra for compounds 4a, 4e and 4 f. The spectra are
shown in Figure 6, PLQY and lifetime data is collected in
Tables 1 and 2.
Both emission and absorption spectra in Figure 6 are essen-
tially unique to each compound, indicating that modification
of (III) has a strong impact on the frontier orbitals
and the electronic states. For 4 f, a rather localized p-
system is present, which yields relatively high emis-
sion energy in both absorption and emission spectra.
Although the PLQY is rather low (11%, Table 1). A
comparison of the kr and knr values of 4 f and 4a
shows significant differences for both radiative and
non-radiative processes. This, together with the
strong qualitative difference in the absorption spec-
tra, suggests that the electronic state of these two
systems differ greatly. For 4e, both emission and ab-
sorption spectra differ greatly from the other com-
Figure 7. Structural differences between the optimized 4e and 4a in dichloromethane.
Stability Studies
To assess the potential of our new BOBIPY series for biological
applications, we next studied its stability under different condi-
tions, using 4 f as reference. Similar structures have been
shown to be stable to nucleophilic substitution processes, pho-
tochemical reactions, temperature variations and in buffered
solutions.[14b] We thus focused on the stability of the fluorobor-
ane group as this had not been previously established. Sub-
mission to mild nucleophilic conditions that could be encoun-
tered during applications (namely, NaOH(aq) in THF at room
temperature) returned the fluorophore untouched, confirming
the hydrolytic stability of the BF2 moiety under these condi-
tions. In contrast however, subjecting 4 f to base in refluxing
THF returned the corresponding boronic acid 7 in high yield
(Scheme 3). Upon crystallization in ethanol, we were able to
obtain 8, which was characterized by X-ray crystallography
(Figure 8).
Interestingly, changing the boron-substituents from F to OH
or OEt has a much greater impact on the optical properties of
Figure 6. Normalized absorption (left) and emission spectra (right) of com-
pounds 4a, 4e and 4 f in CH2Cl2. The color of the depicted subunits in (a)
indicates the color of the respective spectra in (b).
these compounds than the change from F to arylacetylide. In
contrast to 4 f and 5a–c, both 7 and 8 exhibit quantum yield
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
6
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Tuning BOBIPYs for bioapplications
We believed that compound 4a showed the best overall com-
bination of useful photophysical properties and chemical ac-
cessibility, making it the most suitable BOBIPY to exploit for
cell uptake and imaging studies. The quantum efficiency of 4a
in solution is in the order of 35% and the emission color is in
the green region of the optical spectrum. The broad absorp-
tion band, which is well above 400 nm should allow for excita-
tion with blue light rather than UV light sources, which is man-
datory to perform fluorescence microscopy experiments with-
out risking photo-damage. Also, a rather high Stokes shift
allows for good distinction of the emitted light from the exci-
tation light, as well as distinction of the fluorophore from
auto-fluorescence of biological tissue.
Scheme 3. Synthesis of 7 and 8.
As indicated by the DFT calculations, further tuning of the
color seems feasible to allow for creation of yellow or red-emit-
ting compounds. This is best done by addition of electron-do-
nating groups to the southern heterocycle unit (III) of the mol-
ecule, where the HOMO is localized, and/or addition of elec-
tron-withdrawing groups to the northern ring (I), where the
LUMO is localized (Figure 9). With this, the optical bandgap
could be decreased. The phenyl ring attached to 4a is not
a significant part of either HOMO or LUMO, and can be used
to add further functionality (Figure 9), for example, solubilizing
groups or transporters, which can be used to facilitate uptake
in living cells.
Figure 8. ORTEP of BOBIPY 8. Hydrogen atoms are omitted for clarity. Select-
ed bond lengths [ꢅ] and angles [8]: N1-B1 1.762(2), C14-B1 1.628(3), O1-B1
1.425(3), O2-B1 1.420(2); N1-C9-C10-C14 2.95(2), C8-C9-C10-N2 2.64(3).
values lower than our limits of detection. We also compared
the absorption spectra for 4 f, 7 and 8.[28] Not surprisingly,
spectra for 7 and 8 are almost superimposable. While no signif-
icant changes apart from slight broadening was observed for
4a–c versus 5a–c, the low-energy band is blue-shifted for 7/8,
as are other features, which indicates that the electronic states
are affected by the introduction of the oxygen substituents.
The X-ray structures of 4 f and 8 were also used as model
structures for DFT calculations and were studied computation-
ally together with 7. The corresponding B-X distances were col-
lected and compared. The boron-carbon distances are similar
in each case, while the B-N distances are longer by ~0.1 ꢅ in 7
and 8.[28] HOMO and LUMO molecular orbital plots of 4 f, 7
and 8 were also compared. The localization of the orbitals is
very similar in each case, but the orbital energies are changed
as the substituents are changed from F to OH or OEt.[28] The or-
bital energies of 7 and 8 calculated in DCM are shifted towards
higher values by 0.15 and 0.12 eV for HOMO and by 0.28 and
0.22 eV for LUMO orbitals in the case of 7 and 8, respectively.
The HOMO-LUMO gaps are also increased by 0.13/0.1 eV for
7/8 and surpass 4 eV. Pais and co-workers have studied related
systems with isoquinoline, aryl-substituents and boronic ester
groups, and highlighted that planarization of the biaryl system
can have an adverse impact on fluorescence efficiency.[14c] In
this case, the authors presented hints for a temperature-
dependent ring-opening of the B-heterocycle. A similar process
could in principle take place in our case, since the analogous
twisted conformation exists for 7. However, we believe that
this is unlikely in practice as the energy difference is quite high
(DG=5.7 kcalmolꢀ1), favoring the structure in which B–N coor-
dination is preserved. The underlying structure-property rela-
tionships appear rather complex as boronic esters in this class
can show fluorescence with highly variable quantum yields.[29]
Figure 9. Modification of the fluorophore 4a allows for further color-tuning
or the addition of more functionalities.
As a proof-of-principle, we designed a BOBIPY bearing a free
alkyne function that would offer the opportunity to be cova-
lently attached to any substrate of interest by means of
a “click” reaction. Alkyne 9 was thus prepared and its cycload-
dition with triazine 3a afforded fluorophore 4g, equipped
with a free alkyne (Scheme 4).
The properties of 4a and 4g are very similar to each other,
which was expected, because the difference between their
structure is located on the electronically less coupled phenyl
group. The difference between their HOMO–LUMO gap is only
0.02 eV, while their emission wavelengths are identical
(473 nm). The absorption and emission spectra of the two
compounds 4a and 4g are almost superimposable,[28] while
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
7
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
vent polarity on their photophysical properties. Measurements
revealed that 10 emits around 510 nm in water, in the green
region, and maintains high PLQY values. The red shift com-
pared to the DCM-measurements of 4a and 4g is likely to be
a result of a more polar environment, which affects the emis-
sion by stabilization of the charge transfer transitions.
To examine the cell compatibly, human cervix carcinoma
(HeLa) cells were treated with peptoid 10 for 24 h and subse-
quently imaged with a confocal fluorescence microscope. The
results confirmed the accumulation of the compound in endo-
somal vesicles, indicating an endocytotic uptake (Figure 10).
With this method, the intracellular location of 10 could be de-
tected and this demonstrates the potential of BOBIPYs for bio-
logical imaging.
Scheme 4. Synthesis of 4g.
Table 4. Photophysical properties of 4g compared to 4a in dichlorome-
thane.
Compound
4a
4g
labs [nm]
e [mꢀ1 cmꢀ1
lem [nm]
Stokes shift [nm]
PLQY F
388
22921
473
85
0.33
2.6
389
22252
473
84
0.31
2.4
]
Figure 10. Fluorescent confocal microscopy of 10 in HeLa cells. Left: Emis-
sion at 458 nm, Excitation: at 458 nm; Middle: Brightfield; Right: Merge. For
further conditions see SI.
t [ns]
kr [108 sꢀ1
]
1.2
2.6
1.3
2.9
knr [108 sꢀ1
]
Conclusion
In conclusion, we have developed a new strategy for the
design, synthesis and optical tuning of a family of borylated bi-
pyridines (BOBIPYs). The photophysical properties of these
compounds have been measured, and the results interrogated
by DFT calculations. With respect to the tetravalent N,C-linked
boron atom of these BOBIPYs, the LUMO is located on the C,N-
linking bipyridine-type group, while the HOMO is located
mainly in the C-linking aryl moiety. The modular synthesis of
BOBIPYs offers a simple way to tune optical properties, while
substituents incorporated from the key cycloaddition step of
alkynyl trifluoroborates provide an effective means for the in-
corporation of these compounds onto various molecular scaf-
folds. Indeed, we have exemplified the potential applicability
of BOBIPYs for cellular imaging by carrying out conjugation to
a peptoid and imaging the compound within human cervix
carcinoma cells.
the high PLQY is maintained even after addition of the conju-
gated alkyne group (Table 4). Importantly, for other emissive
compounds, the addition of conjugated groups can lead to
significant differences.[30]
To demonstrate the suitability of 4g to track molecules in
vivo, we coupled it to a peptoid bearing an azide side chain
on solid phase in a sub-monomer approach using copper
iodide and N,N-diisopropylethylamine (DIPEA). This peptoid is
a hexameric compound with five aminobutyl- and one azido-
propyl-side chain and was designed to enhance the water-sol-
ubility of the final compound (Scheme 5). Peptoid 10 was char-
acterized in aqueous solution.[28] The insolubility of 10 in
organic solvents, coupled with to the insolubility of 4a and 4g
in water, prevented any direct comparison of the impact of sol-
Scheme 5. Synthesis of peptoid 10.
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
8
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Yusuf, K. Liu, F. Guo, R. A. Lalancette, F. Jꢃkle, Dalton Trans. 2016, 45,
4580–4587.
Experimental Section
[15] a) M. Frizler, I. V. Yampolsky, M. S. Baranov, M. Stirnberg, M. Gtschow,
Org. Biomol. Chem. 2013, 11, 5913–5921; b) M. S. Baranov, K. M. Sol-
ntsev, N. S. Baleeva, A. S. Mishin, S. A. Lukyanov, K. A. Lukyanov, I. V.
Yampolsky, Chem. Eur. J. 2014, 20, 13234–13241.
The procedures for the synthesis, optical properties, theoretical cal-
culations and applications for biological imaging are provided in
the Supporting Information.
[16] J. D. Kirkham, R. J. Butlin, J. P. A. Harrity, Angew. Chem. Int. Ed. 2012, 51,
6402–6405; Angew. Chem. 2012, 124, 6508–6511.
Acknowledgements
[17] D. F. P. Crꢀpin, J. P. A. Harrity, J. Jiang, A. J. H. M. Meijer, A.-C. M. A.
Nassoy, P. Raubo, J. Am. Chem. Soc. 2014, 136, 8642–8653.
[18] CCDC 1451469 (4a), 1451470 (4 f), 1451471 (5a) and 1451472 (8) con-
tain the supplementary crystallographic data for this paper. These data
are provided free of charge by The Cambridge Crystallographic Data
Centre.
This work was supported by the FP7 Marie Curie Actions of
the European Commissionvia the ITN ECHONET Network
(MCITN-2012-316379), GRK 2039and the BioInterfaces Interna-
tional Graduate School.
[19] Full spectra of all compounds are contained in the Supporting Informa-
tion.
[20] J. R. Lakowicz, Principles of Fluorescence Spectroscopy, 3rd ed., Springer,
New York, 2006.
[21] D. V. O’Connor, W. R. Ware, J. C. Andre, J. Phys. Chem. 1979, 83, 1333–
1343.
Keywords: borylated bipyridines · cell imaging · density
functional calculations · fluorophore · photophysics
[1] R. H. Friend, R. W. Gymer, A. B. Holmes, J. H. Burroughes, R. N. Marks, C.
Taliani, D. D. C. Bradley, D. A. D. Santos, J. L. Brꢀdas, M. Lçgdlund, W. R.
Salaneck, Nature 1999, 397, 121–128.
[22] IUPAC. Compendium of Chemical Terminology, 2nd ed. (the “Gold
Book”). Compiled by A. D. McNaught and A. Wilkinson. Blackwell Scien-
tific Publications, Oxford, 1997.
[2] M. Benstead, G. H. Mehl, R. W. Boyle, Tetrahedron 2011, 67, 3573–3601.
[3] A. Hçrner, D. Volz, T. Hagendorn, D. Fꢁrniss, L. Greb, F. Rçnicke, M.
Nieger, U. Schepers, S. Brꢃse, RSC Adv. 2014, 4, 11528–11534.
[4] N. Boens, V. Leen, W. Dehaen, Chem. Soc. Rev. 2012, 41, 1130–1172.
[5] a) A. Kamkaew, S. H. Lim, H. B. Lee, L. V. Kiew, L. Y. Chung, K. Burgess,
Chem. Soc. Rev. 2013, 42, 77–88; b) J. H. Gibbs, Z. Zhou, D. Kessel, F. R.
Fronczek, S. Pakhomova, M. G. H. Vicente, J. Photochem. Photobiol. B
2015, 145, 35–47.
[6] R. P. Haugland, Handbook of Fluorescent Probes and Research Chemicals,
6th ed., Molecular Probes, Eugene, OR (USA), 1996.
[7] A. Loudet, K. Burgess, Chem. Rev. 2007, 107, 4891–4932.
[8] M. Santra, H. Moon, M.-H. Park, T.-W. Lee, Y. K. Kim, K. H. Ahn, Chem. Eur.
J. 2012, 18, 9886–9893.
[9] a) Z. Li, X. Lv, Y. Chen, W.-F. Fu, Dyes Pigm. 2014, 105, 157–162; b) Y. Wu,
Z. Li, Q. Liu, X. Wang, H. Yan, S. Gong, Z. Liu, W. He, Org. Biomol. Chem.
2015, 13, 5775–5782; c) J. R. Zimmerman, O. Johntony, D. Steigerwald,
C. Criss, B. J. Myers, D. H. Kinder, Org. Lett. 2015, 17, 3256–3259.
[10] a) T. Stalling, W. R. R. Harker, A.-L. Auvinet, E. J. Cornel, J. P. A. Harrity,
Chem. Eur. J. 2015, 21, 2701–2704; b) A.-L. Auvinet, J. P. A. Harrity, G.
Hilt, J. Org. Chem. 2010, 75, 3893–3896; c) P. M. Delaney, D. L. Browne,
H. Adams, A. Plant, J. P. A. Harrity, Tetrahedron 2008, 64, 866–873.
[11] a) J. Comas-Barcelꢂ, R. S. Foster, B. Fiser, E. Gomez-Bengoa, J. P. A. Harri-
ty, Chem. Eur. J. 2015, 21, 3257–3263; b) J. Huang, S. J. F. Macdonald,
J. P. A. Harrity, Chem. Commun. 2009, 45, 436–438.
[23] A. D. J. Becke, J. Chem. Phys. 1993, 98, 1372–1377.
[24] G. A. Petersson, A. Bennett, T. G. Tensfeldt, M. A. Al-Laham, W. A. Shirley,
J. Mantzaris, J. Chem. Phys. 1988, 89, 2193–2218.
[25] Gaussian 09, Revision B.1, 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,
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.
[26] L. Yuan, W. Lin, Y. Yang, H. Chen, J. Am. Chem. Soc. 2012, 134, 1200–
1211.
[27] For more details on the comparison of calculated and measured struc-
tural properties and on calculations of molecular orbital plots and maxi-
mum absorption wavelengths, see Supporting Information.
[28] The complete list of molecular orbital plots; absorption spectra ob-
tained by TD-DFT calculations; calculated reorganisation energies; com-
pared absorption spectra and calculated and measured BꢀX bond
lengths; can be found in the Supporting Information.
[12] S. P. J. T. Bachollet, J. F. Vivat, D. C. Cocker, H. Adams, J. P. A. Harrity,
Chem. Eur. J. 2014, 20, 12889–12893.
[13] D. Frath, J. Massue, G. Ulrich, R. Ziessel, Angew. Chem. Int. Ed. 2014, 53,
2290–2310; Angew. Chem. 2014, 126, 2322–2342.
[29] V. F. Pais, H. S. El-Sheshtawy, R. Fernꢇndez, J. M. Lassaletta, A. Ros, U. Pi-
schel, Chem. Eur. J. 2013, 19, 6650–6661.
[14] a) A. C. Shaikh, D. S. Ranade, S. Thorat, A. Maity, P. P. Kulkarni, R. G. Gon-
nade, P. Munshi, N. T. Patil, Chem. Commun. 2015, 51, 16115–16118;
b) V. F. Pais, M. M. Alcaide, R. Lꢂpez-Rodrꢆguez, D. Collado, F. Nꢇjera, E.
Pꢀrez-Inestrosa, J. M. Lassaletta, R. Fernꢇndez, A. Ros, U. Pischel, Chem.
Eur. J. 2015, 21, 15369–15376; c) V. F. Pais, J. M. Lassaletta, R. F. ndez,
H. S. El-Sheshtawy, A. Ros, U. Pischel, Chem. Eur. J. 2014, 20, 7638–
7645; d) Y.-L. Rao, S. Wang, Inorg. Chem. 2011, 50, 12263–12274; e) M.
[30] D. Volz, M. Nieger, J. Friedrichs, T. Baumann, S. Brꢃse, Inorg. Chem.
Commun. 2013, 37, 106–109.
Received: April 25, 2016
Published online on && &&, 0000
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
9
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
Lighting the way: A family of borylated
bipyridines (BOBIPYs) fluorophores
offers tunable optical properties and
linker possibilities via a designed strat-
egy of synthesis. The photophysical
properties of these compounds have
been measured, and the results interro-
gated by DFT calculations, enabling the
mapping of the structure for further im-
plementations.
&
Cell Imaging
S. P. J. T. Bachollet, D. Volz, B. Fiser,
S. Mꢀnch, F. Rçnicke, J. Carrillo,
H. Adams, U. Schepers, E. Gꢂmez-Bengoa,
S. Brꢁse,* J. P. A. Harrity*
&& – &&
A Modular Class of Fluorescent
Difluoroboranes: Synthesis, Structure,
Optical Properties, Theoretical
Calculations and Applications for
Biological Imaging
&
&
Chem. Eur. J. 2016, 22, 1 – 10
www.chemeurj.org
10
ꢄ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!