Synthesis, spectroscopy, crystal structure determination, and quantum chemical calculations of BODIPY dyes with increasing conformational restriction and concomitant red-shifted visible absorption and fluorescence spectra

Article abstract of DOI:10.1002/asia.201000248

Starting from the conforma-tionally unconstrained compound 3, 5-di-(2-bromophenoxy)-4, 4-difluoro-8-(4-methylphenyl)-4-bora-3a, 4a-diaza-s-in-dacene (1), two BODIPY dyes (2 and 3) with increasingly rigid conformations were synthesized in outstanding total

Full text of DOI:10.1002/asia.201000248

FULL PAPERS
DOI: 10.1002/asia.201000248
Synthesis, Spectroscopy, Crystal Structure Determination, and Quantum
Chemical Calculations of BODIPY Dyes with Increasing Conformational
Restriction and Concomitant Red-Shifted Visible Absorption and
Fluorescence Spectra
Volker Leen,[a] Wenwu Qin,*[b] Wensheng Yang,[b] Jie Cui,[b] Chan Xu,[b] Xiaoliang Tang,[b]
Weisheng Liu,[b] Koen Robeyns,[a] Luc Van Meervelt,[a] David Beljonne,[c]
Roberto Lazzaroni,[c] Claire Tonnelꢀ,[c] Noꢁl Boens,[a] and Wim Dehaen*[a]
Abstract: Starting from the conforma-
tionally unconstrained compound 3,5-
di-(2-bromophenoxy)-4,4-difluoro-8-(4-
methylphenyl)-4-bora-3a,4a-diaza-s-in-
dacene (1), two BODIPY dyes (2 and
3) with increasingly rigid conforma-
tions were synthesized in outstanding
total yields through palladium cata-
lyzed intramolecular benzofuran for-
mation. Restricted bond rotation of the
phenoxy fragments leads to dyes 2 and
3, which absorb and fluoresce more in-
tensely at longer wavelengths relative
to the unconstrained dye 1. Reduction
of the conformational flexibility in 2
and 3 leads to significantly higher fluo-
rescence quantum yields compared to
those of 1. X-ray diffraction analysis
shows the progressively more extended
planarity of the chromophore in line
with the increasing conformational re-
striction in the series 1!2!3, which
explains the larger red shifts of the ab-
sorption and emission spectra. These
conclusions are confirmed by quantum
chemical calculations of the lowest
electronic excitations in 1–3 and dyes
of related chemical structures. The
effect of the molecular structure on the
visible absorption and fluorescence
emission properties of 1–3 has been ex-
amined as a function of solvent by
means of the new, generalized treat-
ment of the solvent effect (J. Phys.
Chem. B 2009, 113, 5951–5960). Sol-
vent polarizability is the primary factor
responsible for the small solvent-de-
pendent shifts of the visible absorption
and fluorescence emission bands of
these dyes.
Keywords: BODIPY
· conforma-
tional mobility · crystal structure ·
fluorescent dyes · solvent effect
Introduction
has been put in the conversion of their favorable spectro-
scopic properties to longer wavelengths. The search for
bright fluorophores that absorb and emit in the red visible
(Vis) to the near infrared (NIR) spectral range continues
Since the appearance of difluoroboron dipyrromethene[1,2]
(or BODIPY[3]) dyes in the late 1960s, considerable effort
[c] Prof. Dr. D. Beljonne, Prof. Dr. R. Lazzaroni, C. Tonnelꢀ
Laboratory for Chemistry of Novel Materials
Universitꢀ de Mons
[a] Dr. V. Leen, Dr. K. Robeyns, Prof. L. Van Meervelt, Prof. N. Boens,
Prof. W. Dehaen
Department of Chemistry
Place du Parc 20, 7000 Mons (Belgium)
Katholieke Universiteit Leuven
Celestijnenlaan 200f—bus 02404, 3001 Leuven (Belgium)
Fax : (+32)16-327990
E-mail: Wim.Dehaen@chem.kuleuven.be
Supporting information for this article contains the synthetic proce-
dures and product characterization methods, experimental details on
steady-state spectroscopy and crystal structure determination, and or-
bital representations of the HOMO and LUMO of BODIPY dyes 1–
3, 5, 8–10, and is available on the WWW under http://dx.doi.org/
10.1002/asia.201000248
[b] Prof. Dr. W. Qin, W. Yang, J. Cui, C. Xu, Dr. X. Tang, Prof. Dr. W. Liu
Key Laboratory of Nonferrous Metal Chemistry and
Resources Utilization of Gansu Province
and State Key Laboratory of Applied Organic Chemistry
College of Chemistry and Chemical Engineering
Lanzhou University
Lanzhou 730000 (China)
Fax : (+86)931-8912582
E-mail: qinww@lzu.edu.cn
2016
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 2016 – 2026

Scheme 1. Transition metal catalyzed benzofuran formation.
ceaselessly. Stable dyes with sharp absorption and fluores-
cence emission bands in the red or NIR region of the spec-
trum, combined with high molar absorption coefficients e(l)
and high fluorescence quantum yields Ff, may find extensive
use in many different fields, such as optical engineering, an-
alytical chemistry, biological in vivo imaging and sensing ap-
plications, and materials science.[3,4]
restricted bond rotations, we designed a convenient and
highly efficient two-step synthetic route to BODIPY deriva-
tives fused with rigid benzofuran units. The advantages of
our synthetic strategy are two-fold: i) this synthesis elimi-
nates the tedious preparation of ring-fused pyrroles, ii)
through this sequential ring formation reaction, a nonsym-
metric dye is easily accessible.
In recent years, a substantial number of research publica-
tions have focused on the syntheses, functionalization reac-
tions, and spectroscopic/photophysical properties of 4,4-di-
fluoro-4-bora-3a,4a-diaza-s-indacene dyes. Extension of the
conjugated system is the most widely used method to obtain
BODIPY dyes with spectra shifted to longer wavelengths.[2]
Conveniently, this can be established by substitution of the
BODIPY core with aromatic groups.[5,6] 3,5-Diaryl substitut-
ed BODIPY dyes can show a bathochromic shift of over
100 nm, but generally have low to moderately high fluores-
cence quantum yields Ff, owing to nonradiative decay
through rotational relaxation of the aromatic substituents.
The introduction of alkenyl substituents produces even
larger bathochromic shifts and the rigidity of the double
bond generally results in higher Ff values.[6–8] Alkyne substi-
tution at the 3(,5)-position(s) produces interesting BODIPY
fluorophores with high quantum yields and large red
shifts.[6,9]
Annulation, that is, the building of a ring onto some start-
ing molecule, can be used to extend the ring system and
reduce the nonradiative decay through rotational relaxation
and hence to enhance fluorescence. Building of heterocyclic
rings onto the aromatic rings attached to the BODIPY core
can even lead to NIR emitting dyes. Burgess et al. described
conformationally constrained, aryl-substituted BODIPY
dyes.[10] Ono et al. reported the synthesis and optical proper-
ties of BODIPY dyes fused with rigid bicyclo rings.[11] Goeb
and Ziessel prepared benzo fused BODIPY dyes substituted
with 3,5-dithienyl groups and with alkynylaryl fragments at
the boron center.[12] A series of bright fluorophores named
Keio Fluors, based on BODIPY, were reported by Suzuki
et al.[13,14] Shen, You, Rurack, and co-workers described
phenanthrene-fused[15] and dihydronaphthalene-fused[16]
BODIPY dyes. The preparation of all these reported, rigid
BODIPY dyes required lengthy multi-step syntheses of the
ring-fused pyrroles as starting materials. Intrigued by these
reports on fluorescent dyes with extended conjugation and
In the present work, we prepared three 4,4-difluoro-4-
bora-3a,4a-diaza-s-indacene dyes (1–3, Scheme 1) with in-
creasingly rigid conformations caused by benzofuran ring
formation. We investigated their spectroscopic and photo-
physical characteristics in a large number of solvents by
UV/Vis spectrophotometry and steady-state fluorescence
spectroscopy. These experiments allowed us to determine
the wavelengths of the spectral maxima [labs
lem(max)], the full width at half height of the maximum of
the absorption (fwhmabs and fluorescence emission
ACHUTNGTRNENUG(max), lexACHTUNGTRENNUNG(max),
AHCTUNGTRENNUNG
)
ꢀ
(fwhmem) bands, the Stokes shifts (Dn), and the fluorescence
quantum yields (Ff). We used single- and multi-parameter
expressions to describe the solvent effect on the position of
the spectral maxima and Stokes shifts of 1–3. In addition, X-
ray diffraction analysis of 1–3 was performed and revealed
an increasingly more extensive planarity of the chromo-
phore in line with reduction of conformational flexibility. Fi-
nally, quantum chemical calculations were performed to un-
derstand the spectroscopic properties of 1–3 and related,
previously described ring-fused BODIPY dyes.
Results and Discussion
Synthesis
As reported in our earlier work,[17–19] 3,5-dichlorinated
BODIPY dyes (such as 4, Scheme 2) can be substituted effi-
ciently by a variety of nucleophiles. In a previous report, we
described the synthesis of compound 1 (with o-bromophe-
noxy groups at the 3,5-positions of BODIPY) (Scheme 2).[20]
Nucleophilic aromatic substitution of 3,5-dichloro-BODIPY
4 with 2-bromophenol and 2-iodophenol takes place readily
and in near quantitative yield, furnishing the desired starting
compounds 1 and 1a in gram scale.
Proper placement of a halogen (X=Br or I) atom on the
phenyl ring could make the system susceptible to benzofur-
an ring formation. Indicative of the plausible reactivity of
Chem. Asian J. 2010, 5, 2016 – 2026
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2017

W. Qin, W. Dehaen et al.
FULL PAPERS
rated in 32% yield from a mixture of starting material 1 and
dye 3. The use of a nickel catalyst under similar conditions
did not result in the desired ring formation. Also, the use of
2-iodophenol instead of 2-bromophenol did not produce
drastic changes. As expected, reactions were faster when
using iodinated dye 1a, and the isolated yield was slightly
higher.
Crystallographic Structure
Scheme 2. Nucleophilic substitution of dichloro-BODIPY 4 with ortho-
halogen phenol nucleophiles. The numbering of the BODIPY core is also
indicated.
As shown in Figure 1, the boron center of 2 adopts a tetra-
hedral configuration with bond angles of ꢀ1108 and bond
ꢁ
ꢁ
distances of 1.38 ꢂ (B F) and 1.55–1.56 ꢂ (B N). In the
BODIPY ring system, the two planar pyrrole subunits and
the boron atom constitute a rigid plane, in which the maxi-
mum deviation of non-hydrogen atoms is 0.024 ꢂ. The two
fluorine atoms are equidistant above and under the mean
plane of the BODIPY core, and the F-B-F plane is almost
the 2,6-positions of BODIPY in palladium chemistry, is its
use for Heck-type reactions described by Burgess et al.[21]
This transition metal catalyzed oxidative ring formation has
found widespread use in the synthesis of heteroaromatic sys-
tems.[22] In these reactions, the metal of choice is mostly pal-
ladium.
Initial experiments with Pd0 sources proved the viability
of our premise: the closed products were being formed after
prolonged heating (7 days) and in low yield. Often these re-
actions are carried out in polar solvents, like N,N-dimethyl-
formamide (DMF), N,N-dimethylacetamide (DMA), or N-
methylpyrrolidone (NMP), to increase the reaction rate. Al-
though a fast reaction was observed in these solvents, the
products had limited stability under these conditions. As op-
timized reaction conditions, we used PdII precatalysts and
the long reaction times could be reduced to two days
(Table 1). Even with microwave irradiation (entry 10, MW),
the reaction still took 8 h to go to completion. This micro-
wave procedure led to the ring-closed product 3 in excellent
yield. Interestingly, stopping the reaction before completion
allowed us to isolate the intermediate BODIPY dye 2.
There is no significant difference in reactivity between the
first ring formation and the second: compound 2 was sepa-
Figure 1. Stick representation of compound 2 (top) and compound 3
(bottom). To illustrate the planarity of the ring systems a bottom view is
placed at the right. Created with PyMol, (DeLano, W.L. The PyMOL
Molecular Graphics System (2002) from http://www.pymol.org).
Table 1. Reaction conditions of transition metal catalyzed annulation.
perpendicular (89.98) to the
Catalyst
Solvent
Ligand
Base
t
T [8C])
Yield [%][a]
plane of the ring system. The
tolyl group at the meso-position
makes an angle of 62.098 with
the BODIPY plane, which is in
the range of most BODIPY de-
rivatives (40.3–908), but is
smaller than the average value
[Pd
[Pd
A
N
DMF
DMF
DMF
P(Ph)3
P(Ph)3
dba
K2CO3
K2CO3
K2CO3
Na2CO3
K2CO3
K2CO3
KOAc
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
K2CO3
18 h
2 h
4 h
RT
100
100
140
100
80
110
110
110
130
110
110
120
0
37
16
60
32
0
35
23
55
88
69
74[b]
0
N
N
E
xylene[d]
dioxane
DMF
P
N
48 h
96 h
12 h
48 h
48 h
72 h
8 h (MW)[c]
48 h
36 h
96 h
[Pd
[Pd
[Pd
[Pd
[Pd
[Pd
[Pd
[Pd
P(Ph)3
Bu4NCl
P(Ph)3
toluene
toluene
toluene
toluene
toluene
toluene
xylene[d]
PACTHNGUTRENU(NG OPh)3
of 74.58 found in the CSD.[23]
A
DPPE
P(Ph)3
P(Ph)3
P(Ph)3
DPPP
remarkable structural feature is
that all of the atoms in the ben-
zofuran fragment also lie in the
BODIPY plane, the deviation
is within 0.054–0.271 ꢂ. In addi-
tion, the 2-bromophenoxy sub-
stituent is inclined relative to
[NiACHTUNGTRENNUNG
[a] Yield of isolated product of 3 starting from a reaction with 0.1 mmol 1. [b] Yield of isolated product of 3
starting from 0.1 mmol of iodinated compound 1a. [c] Irradiated at 150 W and 1308C. [d] A mixture of xylenes
is used. dba=dibenzylideneacetone, DPPE= 1,2-bis(diphenylphosphino)ethane, DPPP=1,3-bis(diphenylphos-
phino)propane.
2018
www.chemasianj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 2016 – 2026

Increasing Conformational Restriction in BODIPY Dyes
the BODIPY plane, and the dihedral angle between both is
58.808.
For dye 3 (Figure 1), the entire aromatic system is flat:
the deviation from planarity for the boron atom, when con-
sidering the entire aromatic system, is 0.19 ꢂ. The p-tolyl
group makes an angle of 55.48 with the BODIPY system.
The structure of 1 was confirmed by X-ray analysis, how-
ever the crystals diffract poorly and the structure will not be
discussed here. The ORTEP representation of dye 1 (as well
as 2 and 3) is shown in Figure S1 in the Supporting Informa-
tion for reference purposes.
Spectroscopic Properties
Figure 2. Normalized absorption and fluorescence emission profiles of 1–
3 in diethyl ether.
Dyes 1–3 are strongly colored solids with a metallic luster
that form intensely colored solutions with a bright fluores-
cence when irradiated (Fig-
ure S2 in the Supporting Infor-
Table 2. Spectroscopic/photophysical properties of 1 in several solvents. The solvents are numbered according
to increasing refractive index n.
mation).
The UV/Vis absorption and
fluorescence emission spectra
of dissolved in limited
ꢀ
Dn
Solvent
labs
(max)
lem
(max)
fwhmabs
[cmꢁ1
fwhmem
[cmꢁ1
Ff
[nm]
[nm]
A
]
G
]
N
]
1
a
1
methanol
512
511
511
512
513
514
513
513
515
515
516
515
516
516
515
517
518
518
516
518
528
529
530
530
531
531
532
532
533
532
533
534
533
532
534
533
535
537
536
536
592
666
702
663
661
623
696
696
656
620
618
691
618
583
691
581
613
683
723
648
726
806
992
766
724
721
806
909
718
757
715
754
754
676
794
712
710
671
790
710
926
956
848
919
990
848
946
842
845
979
842
839
876
949
975
805
833
833
902
902
0.41ꢂ0.03
0.29ꢂ0.01
0.38ꢂ0.01
0.29ꢂ0.02
0.27ꢂ0.01
0.58ꢂ0.01
0.38ꢂ0.02
0.48ꢂ0.02
0.61ꢂ0.04
0.29ꢂ0.01
0.64ꢂ0.05
0.47ꢂ0.02
0.44ꢂ0.02
0.48ꢂ0.03
0.29ꢂ0.02
0.71ꢂ0.02
0.63ꢂ0.01
0.58ꢂ0.03
0.38ꢂ0.03
0.47ꢂ0.01
number of solvents have been
reported before.[20] The number
of solvents used in the analysis
of the solvatochromism of 1
was extended to twenty in the
current investigation. We also
used twenty solvents in the
analysis of the solvatochromism
of 2 and 3.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
acetonitrile
diethyl ether
acetone
ethyl acetate
2-propanol
butanenitrile
dibutyl ether
1-butanol
THF
1-pentanol
1,4-dioxane
dichloromethane
cyclohexane
DMF
Compound
1 displays the
typical absorption features of
classic BODIPY dyes in all sol-
vents studied (Figure 2): that is,
a narrow absorption band with
1-decanol
chloroform
tetrachloromethane
DMSO
a
maximum in the l=511–
toluene
518 nm range, irrespective of
the solvent employed. This ab-
sorption band is assigned to the
!
S1 S0 transition, while an addi-
tional, considerably weaker, broad absorption band, ob-
served at the short wavelength side, is attributed to the
S2 S0 transition.
The absorption and fluorescence emission spectra of 2
and 3 (Figure 2; Figure S3 and S4 in the Supporting Infor-
mation) are of similar shape as those of similar, previously
described difluoroboron dipyrromethene dyes.[24] Introduc-
tion of the benzofuran ring in 2 causes a large bathochromic
!
The main absorption band of 1 is hardly affected by sol-
vent polarity: The maximum being slightly blue-shifted
when the solvent is changed from toluene (l=518 nm) to
acetonitrile or diethyl ether (l=511 nm), which is consistent
with the general behavior of BODIPY chromophores.[24]
Derivative 1 also shows the typical emission features of
BODIPY, these are, a narrow, lightly Stokes-shifted band of
mirror image shape, and fluorescence bands that are blue-
shifted with decreasing solvent polarizability (from l=
537 nm in CCl4 to l=528 nm in methanol). The fluorescence
quantum yields Ff are 0.27–0.71. Compiled in Table 2 are
the spectroscopic and photophysical data of 1 as a function
of solvent.
(ꢀ30 nm) shift in both the absorption [labs
ACHTUNGTREN(NGNU max)] and emis-
sion [lem(max)] maxima compared to 1 (Table 3). The ab-
AHCTUNGTRENNUNG
sorption and emission maxima of the most rigid structure 3
are further red-shifted by approximately 30 nm compared
with 2 (Table 4). The progressively more extended planarity
of the chromophore in the series 1!2!3 accounts for the
increasing bathochromic shifts of labsACHTUNTRGENNUG(max) and lemACHTUNGTRENNUNG(max).
The fact that the absorption band positions of 1–3 do not
show any particular trend as a function of solvent polarity
suggests that the absorbing state of the dyes is weakly dipo-
lar.
Chem. Asian J. 2010, 5, 2016 – 2026
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2019

W. Qin, W. Dehaen et al.
FULL PAPERS
Table 3. Spectroscopic/photophysical properties of 2 in several solvents. The solvents are numbered according
to increasing refractive index n.
The increasingly restricted
bond rotation in the ring-fused
systems 2 and 3 explains the no-
ticeably higher quantum yields
Ff of 3 (Ff ꢃ0.5 in all 20 sol-
vents studied) and 2 (Ff ꢃ0.5 in
18 out of 20 solvents) compared
to those of 1 (Ff ꢃ0.5 in 6 out
of 20 solvents).
ꢀ
Dn
Solvent
labs
(max)
lem
(max)
fwhmabs
[cmꢁ1
fwhmem
[cmꢁ1
Ff
[nm]
[nm]
A
]
G
]
N
]
1
2
3
4
5
6
7
8
methanol
544
544
546
544
547
547
544
549
549
547
549
549
550
551
547
551
553
553
549
553
562
563
563
563
562
563
564
565
565
563
565
565
566
564
567
566
567
567
567
569
589
620
553
620
488
520
652
516
516
520
516
516
514
418
645
481
446
446
578
508
918
954
776
918
844
805
990
733
802
873
800
800
831
661
1011
794
791
656
1003
788
821
785
758
788
755
725
845
722
722
720
750
717
747
662
809
717
682
714
742
742
0.57ꢂ0.02
0.59ꢂ0.04
0.62ꢂ0.05
0.57ꢂ0.02
0.63ꢂ0.01
0.64ꢂ0.05
0.70ꢂ0.01
0.57ꢂ0.01
0.56ꢂ0.02
0.50ꢂ0.01
acetonitrile
diethyl ether
acetone
ethyl acetate
2-propanol
butanenitrile
dibutyl ether
1-butanol
The molar absorption coeffi-
cients emax of 1–3 at the absorp-
tion maximum labsACTHUNRGTNE(NUG max) were
9
10
11
12
13
14
15
16
17
18
19
20
THF
1-pentanol
1,4-dioxane
dichloromethane
cyclohexane
DMF
0.71ꢂ0.07 determined in toluene, ethyl
0.55ꢂ0.03
0.59ꢂ0.01
0.69ꢂ0.04
0.42ꢂ0.02
0.78ꢂ0.01
0.58ꢂ0.04
0.63ꢂ0.02
0.53ꢂ0.15
0.48ꢂ0.01
acetate,
and
methanol
(Table 5). The corresponding
values of the oscillator strength
f calculated according to Equa-
tion (1),[4] in which n is the sol-
vent index of refraction and n is
the wavenumber (in cmꢁ1), are
also compiled in Table 5.
1-decanol
chloroform
tetrachloromethane
DMSO
toluene
Z
4:32 ꢄ 10ꢁ9
ð1Þ
Table 4. Spectroscopic/photophysical properties of 3 in several solvents. The solvents are numbered according
to increasing refractive index n.
ꢀ
ꢀ
f ¼
eðnÞdn
n
ꢀ
Solvent
labs
(max)
lem
(max)
Dn
fwhmabs
[cmꢁ1
fwhmem
[cmꢁ1
Ff
Progressive reduction of the
flexibility of the chromophore
is reflected by the considerably
higher emax, f, and Ff values.
The brightness, defined as the
product of the molar absorption
coefficient e(l) at the excitation
wavelength l and the fluores-
cence quantum yield [e(l)ꢃ
Ff],[25] intensifies considerably
by restricting the conformation-
al flexibility. The maximum
values of the brightness [i.e.,
[nm]
[nm]
N
]
G
]
N
]
1
2
3
4
5
6
7
8
methanol
576
578
578
577
579
580
579
582
582
580
582
583
583
583
580
584
586
588
587
587
590
592
593
593
593
593
593
594
594
595
596
596
597
593
597
597
599
598
599
602
412
409
438
468
408
378
408
347
347
435
404
374
402
289
491
373
370
284
341
424
787
876
687
816
750
716
842
654
742
745
711
768
740
589
836
676
732
641
767
767
710
824
794
764
875
791
789
786
786
768
807
735
866
783
805
754
890
799
0.63ꢂ0.03
0.58ꢂ0.02
0.69ꢂ0.03
0.58ꢂ0.04
0.61ꢂ0.05
0.62ꢂ0.06
0.64ꢂ0.01
0.77ꢂ0.07
0.58ꢂ0.17
0.56ꢂ0.08
0.62ꢂ0.05
0.62ꢂ0.14
0.63ꢂ0.07
0.65ꢂ0.06
0.51ꢂ0.05
0.65ꢂ0.05
0.63ꢂ0.19
0.59ꢂ0.12
0.75ꢂ0.01
0.58ꢂ0.08
acetonitrile
diethyl ether
acetone
ethyl acetate
2-propanol
butanenitrile
dibutyl ether
1-butanol
9
10
11
12
13
14
15
16
17
18
19
20
THF
1-pentanol
1,4-dioxane
dichloromethane
cyclohexane
DMF
for e(l)=emax at labsACHTNUTGRNEUNG(max)] for
1-decanol
1–3 in toluene, ethyl acetate,
and methanol are given in
Table 5.
The average absorption and
emission spectral bandwidths of
chloroform
tetrachloromethane
DMSO
1346
699
toluene
1–3 are quite narrow [fwhmabs
=
(7.6ꢂ0.8)ꢃ102 cmꢁ1
for
1,
Excitation by light yields fluorescence emission spectra
with mirror image shape of the absorption spectra. The
(8.4ꢂ1.0)ꢃ102 cmꢁ1 for 2, and (7.7ꢂ1.5)ꢃ102 cmꢁ1 for 3;
fwhmem =(9ꢂ6)ꢃ102 cmꢁ1 for 1, (7.6ꢂ0.5)ꢃ102 cmꢁ1 for 2,
and (7.9ꢂ0.4)ꢃ102 cmꢁ1 for 3]. Increasing the rigidity of the
systems 2 and 3—by benzofuran formation and by increas-
ing the planarity of the chromophore—diminishes the de-
grees of vibrational freedom relative to 1.
ꢀ
Stokes shifts Dn are small for 1–3 and decrease with dimin-
ishing conformational mobility (n˜ =651ꢂ43 cmꢁ1 for 1, n˜ =
533ꢂ66 cmꢁ1 for 2, and 390ꢂ52 cmꢁ1 for 3) (Tables 2–4).
The fluorescence excitation spectra of 1–3 match the absorp-
tion spectra in all cases and, moreover, lex
(max)=labs
E
These spectral/photophysical properties can be compared
to those of conformationally restricted BODIPY dyes pub-
lished in the literature (Table 6).[10–14,16] The symmetric, con-
strained dye 5 (Scheme 3) of Burgess and co-workers[10] with
2,6-dioxy benzofuran rings should be contrasted to com-
The observation that the emission band positions of 1–3 do
not exhibit any distinct trend as a function of solvent polari-
ty implies that emission occurs from the weakly dipolar, re-
laxed Franck–Condon excited state of the dyes.
2020
www.chemasianj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 2016 – 2026

Increasing Conformational Restriction in BODIPY Dyes
Table 5. Spectroscopic/photophysical properties of 1–3.
lem
furan ring) are slightly blue-shifted relative to those of 3
(with benzofuran rings). Conversely, labs(max) and lem(max)
ACHTUNGTREN(NUNG max) of 8 (with a single methyl substituent on each
Dye Solvent lem
G
f
Ff
Max. brightness
[Lmolꢁ1 cmꢁ1
[nm]
]
]
A
ACHTUNGTRENNUNG
1
2
3
MeOH 512
N
0.307 0.41 17220
0.226 0.27 12150
0.248 0.47 16450
0.362 0.57 39330
0.359 0.63 45360
0.390 0.48 20160
of 9 are located over 650 nm. Dye 9 can be regarded as a
3,5-distyryl substituted BODIPY (such as 10–11)[6–8] with
oxygen atom bridge linkers that preclude free rotation of
the ethenyl fragments. The absence of a meso-substituent
and hence limited conformational flexibility in 8 and 9 leads
to very high quantum yields Ff. Fusion of dihydronaphtha-
lene groups (with their 1,2-positions) to the 1,2- and 6,7-po-
sitions of the BODIPY core (compounds 12a–c)[16] and to
the 2,3- and 5,6-positions (compounds 13a,b)[10] leads to
dyes with red-shifted absorption
AcOEt 513
toluene 518
MeOH 544
AcOEt 547
toluene 553
MeOH 576
AcOEt 579
toluene 587
and
fluorescence
emission
Table 6. Spectroscopic/photophysical properties of (ring-fused) BODIPY dyes from the literature discussed
here.
maxima comparable to classic
difluoroboron dipyrromethenes.
Compounds 13 have more bath-
ochromically shifted spectra
than 12, indicative of a more ef-
ficient delocalization in 13. The
decreasing Ff values in 12a–c
are in line with the increasing
conformational mobility of the
meso-substituents (H, Me, and
Ph). The higher Ff value of 13b
in comparison to 13a arises
from the restricted rotational
freedom of the 8-substituent
imposed by methyl groups at
ꢀ[a]
Dye
labs
(max)
lem
N
Dn
emax
[Lmolꢁ1 cmꢁ1
]
Ff
Solvent
Reference
[nm]
[nm]
A
]
3
5
6
7
8
9
10
11
587
637
600
727
579
652
630
633
562
562
573
634
619
602
647
609
780
583
661
642
646
569
599
609
647
629
424
243
246
936
118
209
297
318
219
155000
0.58
0.34
0.99
0.20
0.96
0.90
0.96
0.83
0.55
0.44
0.23
0.38
0.72
toluene
CHCl3
THF
CH2Cl2
CHCl3
CHCl3
cyclohexane
THF
THF
THF
THF
MeOH
MeOH
this study
[10]
[11b]
[12]
[13,14]
[14]
[6]
[7]
[16]
[16]
[16]
151000
[c]
90000
202000
314000
138000[b]
[c]
[c]
[c]
[c]
12a (R=H)
12b (R=Me)
12c (R=Ph)
13a (R=H)
13b (R=Me)
1099
1032
317
126 250
145 750
[10]
[10]
257
ꢀ
ꢀ
[a] Dn is calculated from reported labs
(max) and lem
G
[b] emax value determined in current study. [c] Not available.
pound 3 with 3,6-dioxy benzo-
furan rings. The most striking
difference between these two
dyes is the larger red shifts of
labsACHTUNGTRENNUNG(max) and lemACHTUNGTRENNUNG(max) for dye
5 which absorbs and emits light
ꢀ50 nm more to the red rela-
tive to dye 3. This large differ-
ence reflects the improved con-
jugation of the 2,6-dioxy deriva-
tives compared to the new 3,5-
dioxy dyes. The benzo-fused
compounds 6 and 7 also have
bathochromically shifted labs
-
A
ACHTUNGTRENNUNG
3. BODIPY 7 has a low Ff
value, owing to the rotational
mobility of the 3,5-dithienyl
fragments. The Keio Fluors 8
and 9 of Suzuki et al.[13,14] (with
2,6-dioxy furan rings as in 5,
Scheme 3) have extremely high
emax compared to 3, 5, and 7.
ꢀ
The Stokes shifts Dn of 8 and 9
are very small. labsACTHNUGTRNEUN(G max) and Scheme 3. Chemical structures of (ring-fused) BODIPY dyes from the literature discussed here.
Chem. Asian J. 2010, 5, 2016 – 2026
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2021

W. Qin, W. Dehaen et al.
FULL PAPERS
the 1,7-positions. To understand all these experimental ob-
servations on ring-fused BODIPY compounds, we per-
formed quantum chemical calculations, which are reported
in the next section.
For 1, the HOMO has no weight on the substituents (bro-
mophenoxy rings). If the bromophenoxy groups of 1 are re-
placed by either one or two benzofuran rings in 2 and 3, re-
spectively, the HOMO shows significant contributions on
the substituent moiety thereby explaining the predicted and
measured bathochromic and hyperchromic shifts.
Quantum Chemical Calculations
Compounds 3 and 5
All dyes studied have a ring-fused BODIPY core. For all
compounds investigated, ground-state geometry optimiza-
tion was first performed at the semi-empirical AM1 level.
These geometries were subsequently used as input for excit-
ed-state calculations at the AM1/SCI level (in the gas
phase). The main goal of the theoretical work is to unravel
the relationships between the nature and position of the
substituents grafted on the BODIPY core and the resulting
spectral properties, namely excitation energy and optical ab-
sorption cross-section.
It is instructive to compare 3 and 5 as these only differ by
the position of the benzofuran rings on the BODIPY core:
Compound 3 presents 3,5-dioxy benzofuran rings whereas 5
features 2,6-dioxy benzofuran rings (see Scheme 3).
Experimentally, it has been found that 5 was more red-
shifted than 3, reflecting its improved conjugation. The theo-
retical results confirm this observation (Table 7).
Looking at the transition wavelengths (Table 7), we find
labs
G
perimental result. In the case of 5, we find a bigger differ-
ence with a transition at l=599 nm, approximately 38 nm
less than the experimental one. Again, although the theoret-
ical values of the absorption maxima and of the spectral
shifts are underestimated for both compounds, the qualita-
tive trend is nicely reproduced by the calculations, with 5
Compounds 1–3
To understand the effect of conformational constraints on
the optical properties of BODIPY dyes, we performed cal-
culations on molecules 1–3. Starting from 1, the rigidity of
the structure has been increased by the introduction of ben-
zofuran rings instead of bromophenoxy groups (1!2!3,
see Figure 1). The results of the AM1/SCI calculations are
presented in Table 7.
showing
a
calculated bathochromic shift of ꢀ37 nm
(ꢀ0.14 eV) compared to 3 (against the experimental value
of ꢀ50 nm).
When comparing the frontier
orbitals, a major visible differ-
Table 7. Calculated (AM1/SCI) (left) and measured (right) photophysical properties of compounds 1–3, 5, 8–
10. To convert the theoretical value of the oscillator strength f (for the gas phase) to values in a solvent with
ence can be noticed for the
LUMO: in 3, the LUMO has a
vanishingly small weight on the
benzofuran rings, while these
contribute significantly to the
LUMO wavefunction in 5. This
is confirmed by the computed
transition densities (that pro-
vide a local map for the transi-
tion dipole and the excited-
state localization) for the
lowest electronic excitation
(Figure 3). The transition is
ꢀ
ꢁ
2
.
n2 þ2
[26]
refractive index n, one has to multiply the theoretical f value by n
3
theoretical values (in the gas phase)
experimental values
Transition
Transition
Oscillator
Transition
Transition
Oscillator
energy [eV]
wavelength [nm]
strength f
energy [eV]
wavelength [nm]
strength f
1
2
3
5
8
9
10
2.382
2.303
2.206
2.069
2.152
2.023
2.090
521
538
562
599
576
613
593
0.7515
0.8525
1.0182
0.7912
0.8544
1.2316
0.8995
2.396[a]
2.245[a]
2.114[a]
1.949[b]
2.144[b]
1.904[b]
1.933[c]
518[a]
553[a]
587[a]
637[b]
579[b]
652[b]
642[c]
0.248[a]
0.390[a]
0.802[a]
–
–
–
–
[a] Data in toluene. [b] Data in chloroform. [c] Data in cyclohexane.
The increase in rigidity leads to a bathochromic shift of
labs
G
one bromophenoxy group is replaced by a benzofuran ring
(in 2). If a second benzofuran ring is introduced (in 3), the
bathochromic shift reaches ꢀ24 nm (ꢀ0.10 eV) in compari-
son to 2. The calculations reproduce well the observed spec-
tral shifts, even though the AM1/SCI values are systemati-
cally underestimated (see Table 7).
The experimental oscillator strengths, f, extracted from
Equation (1), are in line with our theoretical findings: the
oscillator strength f increases when the 3,5-substituents are
frozen in a more rigid structure, as a result of the extended
p delocalization.
Differences between the frontier orbitals of the three
compounds are only noticeable in the case of the HOMO.
Figure 3. Transition densities of compounds 3, 5, 8, and 9.
2022
www.chemasianj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 2016 – 2026

Increasing Conformational Restriction in BODIPY Dyes
almost completely localized on the BODIPY core in 3, yet
one can see small but non- negligible contributions on the
benzofuran rings in 5, consistent with the extended conjuga-
tion. In a simple valence bond picture, the phenyl units in 5
are connected to the central pyrrole rings of the BODIPY
core in ꢄN-aꢅ positions, while they are grafted in ꢄN-bꢅ posi-
tions for 3, thereby explaining the more efficient delocaliza-
tion in 5.
regularly appear to be unsuitable because that specific
single parameter is so dependent on the particular probe
used to construct the single-parameter scale concerned that
it fails to predict the behavior of other solutes with consider-
ably different properties from those of the probe. A multi-
parameter approach is often preferable and has been ap-
plied successfully to various physicochemical parameters.[4]
The solvent effect can be split into two different types of
contributions, namely specific interactions (i.e., described as
localized donor–acceptor interactions involving specific orbi-
tals or acid–base interactions involving hydrogen bonding)
and nonspecific interactions (i.e., those arising from the sol-
vent acting as a dielectric continuum).
Compounds 8–10
Compounds 8 and 9 differ by the substituent on the furan
rings, whereas 9 and 10 are both 3,5-distyryl compounds but
9 is conformationally restricted through O-linked bridges.
From the spectroscopic data above, it has been found that 9
presents a larger red shift compared to 8 and 10.
The solvent effect on the physicochemical observable y
can be expressed by the multi-linear Equation (2a):
The theoretical labsACTHNUGRTENUNG(max) (Table 7) obtained for 8 is in
good agreement with the experimental value (underestimat-
ed by only ꢀ3 nm). For 9 and 10, the computed transition
wavelengths are underestimated by a larger amount, of
about 40 nm, compared to the experimental results, l=652
and 630 nm, respectively. However, both experimental and
theoretical results follow the same tendency and show that
preventing rotation by the presence of O-linked bridges is
an efficient way to increase the oscillator strength.
y ¼ y0þa Aþb Bþc Cþd D
ð2aÞ
in which y0 denotes the physicochemical property of interest
in the gas phase; a, b, c, and d are adjustable coefficients
that reflect the sensitivity of the physicochemical property y
in a given solvent to the various {A, B, C, D} solvent param-
eters. The physicochemical observables y analyzed in this
ꢀ
paper are the absorption maxima nabs =1/labsACTHUNGTRENNUNG
ꢀ
emission maxima nem =1/lemACTHUNGRTENNGUN
We can first compare the frontier orbitals of 8 and 10 to
disentangle the influence of the presence of O-linked
bridges. There are no obvious differences neither between
the HOMO and LUMO orbitals of the two compounds nor
in terms of the transition densities (not shown). It thus ap-
pears that extending the conjugation from 8 to 10 through
substitutions at the 3,5 positions of the BODIPY core is not
very effective in raising the oscillator strength and/or in low-
ering the excitation energy. We attribute this to the fact that
the extension of the p system then occurs along a direction
that is almost perpendicular to the polarization of the lowest
electronic excitation.
By far, a much more dramatic improvement is seen in 9.
Although the frontier molecular orbitals of 8 and 9 look
very similar and are in both cases mostly confined on the
central BODIPY core, the transition densities extend signifi-
cantly over the external phenyl rings in 9 (Figure 3), which
explains the largest oscillator strength computed for that
compound. This effect arises from multi-configurational
mixing of the HOMO–LUMO excitation with electronic
configurations involving deeper molecular orbitals with
large weights on the outer phenyl rings.
Dn ¼ nabs ꢁ nem (all in cmꢁ1).
ꢀ
ꢀ
ꢀ
In the literature there are several solvent scales reported
to describe the solvent dependence of y. Kamlet and Taft[27]
proposed the a, b, and p* parameters to characterize, re-
spectively, the acidity, the basicity, and the polarity/polariza-
bility of a solvent [Eq. (2b)].
*
y ¼ y0 þ aa a þ bb b þ cpꢅ
p
ðKamlet ꢁ TaftÞ
ð2bÞ
Recently, Catalꢆn proposed a generalized treatment of
the solvent effect based on a set of four empirical, mutually
independent solvent scales.[28] The dipolarity, polarizability,
acidity, and basicity of a certain solvent are characterized by
the parameters SdP,[28] SP,[29] SA,[30,31] and SB,[32] respectively
[Eq. (2c)].
ꢁ
y ¼ y0þaSA SAþbSB SBþcSP SPþdSdP SdP ðCatalanÞ
ð2cÞ
The Kamlet–Taft solvatochromic {a, b, p*} parameters are
taken from Ref. [33], whereas the Catalꢆn SA and SB pa-
rameter values come from Refs [30–32]. The Catalꢆn solvent
polarizability parameters SP are collected from Ref. [29].
The new SdP solvent dipolarity parameters can be found in
Ref. [28].
Representations of the HOMO and LUMO of the com-
pounds are available in Figure S9 in the Supporting Infor-
mation.
Compiled in Table 8 are the estimated regression coeffi-
cients y0, aSA, bSB, cSP, dSdP (Catalꢆn) and their respective
standard errors, and the correlation coefficients (r) for the
multiple linear regression analyses of the maxima of absorp-
Solvatochromism
Solvent-dependent spectral shifts are often analyzed in
terms of a single parameter (i.e., by using a global single-pa-
rameter scale or one describing only nonspecific solvent ef-
fects arising from the solvent acting as a dielectric continu-
um). However, empirical single-parameter solvent scales
ꢀ
ꢀ
tion (nabs) and fluorescence emission (nem) and the Stokes
ꢀ
shift Dn of 1–3 according to Equation (2c) for the solvents
of Tables 2–4.
Chem. Asian J. 2010, 5, 2016 – 2026
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2023

W. Qin, W. Dehaen et al.
FULL PAPERS
Table 8. Estimated coefficients (y0, a, b, c, d; in cmꢁ1) and correlation coefficient (r) for the multiple linear re-
The large (negative) estimat-
ed cSP values compared to the
{aSA, bSB, dSdP} estimates in the
ꢀ
ꢀ
ꢀ
gression analysis of the absorption (nabs) and fluorescence emission maxima (nem), and the Stokes shift (Dn) of
1–3 as a function of the Catalꢆn {SA, SB, SP, SdP} solvent scales for the 20 solvents listed in Tables 2–4.
Compound
y0
aSA
bSB
cSP
dSdP
r
ꢀ
analysis of nabs of 1–3 according
ꢀ
1
nabs
20238ꢂ105
19631ꢂ86
606ꢂ114
ꢁ199ꢂ58
ꢁ19ꢂ47
ꢁ180ꢂ63
ꢁ177ꢂ67
ꢁ60ꢂ50
ꢁ117ꢂ72
ꢁ128ꢂ59
ꢁ12ꢂ59
ꢁ115ꢂ72
14ꢂ35
ꢁ1223ꢂ141
ꢁ1254ꢂ115
31ꢂ154
110ꢂ26
81ꢂ22
29ꢂ29
204ꢂ30
39ꢂ23
165ꢂ33
148ꢂ27
46ꢂ27
102ꢂ33
0.932
0.953
0.619
0.935
0.900
0.816
0.952
0.926
0.683
to Equation (2c) and the rela-
tively large standard errors on
aSA, bSB, and dSdP in comparison
to those on cSP (Table 8) indi-
cate that the small change of
ꢀ
nem
ꢁ36ꢂ29
50ꢂ38
20ꢂ40
ꢁ1ꢂ31
20ꢂ43
1ꢂ36
ꢀ
Dn
ꢀ
nabs
2
3
18928ꢂ121
18325ꢂ91
603ꢂ130
ꢁ1139ꢂ162
ꢁ904ꢂ122
ꢁ235ꢂ174
ꢁ1439ꢂ144
ꢁ1239ꢂ142
ꢁ200ꢂ174
ꢀ
nem
ꢀ
Dn
ꢀ
nabs
18129ꢂ107
17660ꢂ106
469ꢂ130
ꢀ
nabs may reflect principally a
ꢀ
nem
ꢁ28ꢂ36
28ꢂ43
minor change in polarizability
of the environment of the chro-
mophore. Indeed, the multi-
linear fit according to Equa-
ꢀ
Dn
Use of the Catalꢆn solvent parameter set {SA, SB, SP,
SdP} [Eq. (2c)], for which solvent (di)polarity and polariza-
tion (2c) in which solvent polarizability is disregarded (i.e.,
with {SA, SB, SdP} as independent variables) is very poor
(r=0.451 for 1, 0.677 for 2, and 0.525 for 3), signifying that
solvent polarizability is of the utmost importance. The anal-
yses according to Equation (2c) with three independent var-
iables including SP and SdP (namely {SA, SP, SdP}, {SB, SP,
SdP}) always gave good fits (0.874<r<0.952), implying that
the absorption maxima are hardly influenced by the acidity
and basicity of the medium. The analyses which disregarded
solvent dipolarity (i.e., Equation (2c) with {SA, SB, SP} as
independent variables) were somewhat less acceptable (r=
0.846 for 1, 0.706 for 2, 0.847 for 3). All these analyses
point to solvent polarizability as the major parameter deter-
mining the position of the absorption maxima. However, es-
pecially for 2, dipolarity of the medium cannot be disregard-
ꢀ
bility effects are split, gives excellent fits to nabs (0.932 for 1,
0.935 for 2, and 0.952 for 3). To visualize the goodness-of-fit
ꢀ
of nabs as a function of the Catalꢆn solvent parameters {SA,
ꢀ
SB, SP, SdP}, we plotted the nabs values calculated, according
to Equation (2c), by using the estimated values of y0, aSA
,
ꢀ
bSB, cSP, dSdP versus the corresponding experimental nabs
values for dyes 1–3 (Figures S5a, S6a, and S7a in the Sup-
ꢀ
porting Information). The analyses of the nabs data of 1–3 ac-
cording to Kamlet–Taft [Eq. (2b)] using the {a, b, p*} sol-
vent scales showed poor fits, as assessed by the values of r
(0.315 for 1, 0.486 for 2, 0.324 for 3) and the large standard
errors on the estimated {aa, bb, cp*} coefficients as quality-of-
fit criteria (Table S1 in the Supporting Information).
ꢀ
Excellent fits were found for the multilinear analyses of
ed. This is corroborated by the linear regression of nabs
ꢀ
the nem data of 1–3 according to Equation (2c) (r values
versus SP which is the most unsatisfactory for 2 (r=0.679
for 2 versus 0.803 for 1 and 0.839 for 3). Dyes 1 and 3 are
symmetric and it can be assumed that the nonsymmetric dye
2 will have a somewhat larger dipole than 1 and 3, which
were 0.953 for 1, 0.900 for 2, and 0.926 for 3). The plots of
nem were calculated, according to Equation (2c), by using
the estimated values of y0, aSA, bSB, cSP, dSdP versus the corre-
ꢀ
ꢀ
ꢀ
sponding experimental nem values of 1–3 are shown in Fig-
can explain the larger sensitivity of nabs of 2 to solvent dipo-
ure S5b, S6b, and S7b (Supporting Information). Contrary to
the superb results obtained by using the Catalꢆn solvent
larity.
The relatively large (negative) cSP estimates and the com-
paratively large standard errors on aSA, bSB, and dSdP in com-
parison to those on cSP (Table 8) point to solvent polarizabil-
ꢀ
scales, the nem data of 1–3 could not be described adequately
by the Kamlet–Taft {a, b, p*} solvent parameters [Eq. (2b)],
as judged by the low r values (0.317 for 1, 0.325 for 2, 0.369
for 3) and the large standard errors on the estimated {aa, bb,
cp*} coefficients (Table S1 in the Supporting Information).
It is remarkable that the small solvent-dependent shifts
ꢀ
ity as the major factor affecting nem of 1–3. The analyses of
ꢀ
nem according to Equation (2c) with {SA, SB, SP}, {SB, SP,
SdP}, and {SA, SP, SdP}, as independent variables, all gave
high-quality fits (for 1 r=0.908, 0.953, and 0.948, respective-
ly; for 2 r=0.879, 0.890, and 0.900, respectively; for 3 r=
0.911, 0.926, and 0.923, respectively). The common inde-
pendent variable in all these analyses is SP (solvent polariz-
(all nabs measured lie within n˜ =264 cmꢁ1 for 1, n˜ =299 cmꢁ1
ꢀ
for 2 and n˜ =354 cmꢁ1 for 3; all nem measured lie within n˜ =
ꢀ
317 cmꢁ1 for 1, n˜ =218 cmꢁ1 for 2, and n˜ =338 cmꢁ1 for 3)
can be described so accurately by the Catalꢆn solvent scales.
The advantage of this new, generalized treatment of the sol-
vent effect is that it allows one to split up the relative contri-
butions of dipolarity, polarizability, acidity, and basicity of
the medium. Hence, it is informative to determine which
solvent parameter is primarily responsible for the slight sol-
ꢀ
ability). Conversely, the analysis of nem according to Equa-
tion (2c), in which SP was disregarded, gave low r-values
(0.433 for 1, 0.399 for 2, 0.367 for 3). Therefore, solvent po-
larizability is the most crucial solvent property influencing
ꢀ
ꢀ
nem. Additional evidence that the position of nem is con-
trolled largely by solvent polarizability comes from the
ꢀ
ꢀ
ꢀ
vatochromic shifts of nabs and nem of 1–3. It must be noted
that in the Kamlet–Taft approach, the solvent scale p* com-
bines solvent (di)polarity and polarizability effects, and
hence, this methodology can never be used to disentangle
solvent polarizability and (di)polarity effects.
linear regressions of nem versus SP (r=0.907 for 1, 0.872 for
2, 0.909 for 3). In these linear regressions, the lowest r
values are found for 2, implying that solvent polarizability
alone does not determine the position of nem. The fact that
ꢀ
ꢀ
solvent polarizability is the principal factor affecting nabs and
2024
www.chemasianj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 2016 – 2026

Increasing Conformational Restriction in BODIPY Dyes
tury Excellent Talents in University (NCET-09-0444). This study was sup-
ported in part by the Key Program of National Natural Science Founda-
tion of China (20931003). The collaboration between Leuven and Mons
is supported by the Science Policy Office of the Belgian Federal Govern-
ment (BELSPO-IAP project 6/27). Research in Mons is also supported
by FNRS/FRFC and Rꢀgion Wallonne (Programme d’excellence OPTI2-
MAT). DB is a Research Director of the Fonds National de la Recherche
Scientifique (FNRS, Belgium).
ꢀ
nem of 1–3 can explain the failure of the Kamlet–Taft ap-
proach because solvent polarizability is not treated as a sep-
arate parameter in this methodology.
ꢀ
The substandard fits of the small Stokes shift Dn of 1–3 as
a function of the solvent parameters {SA, SB, SP, SdP}—as
substantiated by rather low r values (0.619 for 1, 0.816 for 2,
and 0.684 for 3)—are no surprise because the only solvent
parameter that really matters (i.e., polarizability) has an
ꢀ
ꢀ
ꢀ
identical effect on nabs and nem and hence will not affect Dn.
[1] A. Treibs, F.-H. Kreuzer, Liebigs Ann. Chem. 1968, 718, 208–223.
[2] 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.
[3] R. P. Haugland, The Handbook. A Guide to Fluorescent Probes and
Labeling Technologies, 10th ed., Invitrogen, Eugene, OR, 2005.
[4] B. Valeur, Molecular Fluorescence. Principles and Applications,
Wiley-VCH, Weinheim, Germany, 2002.
[5] a) L. H. Thoresen, H. Kim, M. B. Welch, A. Burghart, K. Burgess,
Synlett 1998, 1276–1278; b) A. Burghart, H. Kim, M. B. Welch,
L. H. Thoresen, J. Reibenspies, K. Burgess, F. Bergstrçm, L. B.-ꢂ.
Johansson, J. Org. Chem. 1999, 64, 7813–7819.
[6] T. Rohand, W. Qin, N. Boens, W. Dehaen, Eur. J. Org. Chem. 2006,
4658–4663.
[7] R. Rurack, M. Kollmannsberger, J. Daub, New J. Chem. 2001, 25,
289–292.
[8] Z. Dost, S. Atilgan, E. U. Akkaya, Tetrahedron 2006, 62, 8484–8488.
[9] V. Leen, E. Braeken, K. Luckermans, C. Jackers, M. Van der Au-
weraer, N. Boens and W. Dehaen, Chem. Commun. 2009, 4515–
4517.
[10] J. Chen, J. A. Burghart, A. Derecskei-Kovacs, K. Burgess, J. Org.
Chem. 2000, 65, 2900–2906.
[11] a) M. Wada, S. Ito, H. Uno, T. Murashima, N. Ono, T. Urano, Y.
Urano, Tetrahedron Lett. 2001, 42, 6711–6713; b) Z. Shen, H. Rçhr,
K. Rurack, H. Uno, M. Spieles, B. Schulz, G. Reck, N. Ono, Chem.
Eur. J. 2004, 10, 4853–4871.
Conclusions
We have demonstrated that the combination of nucleophilic
aromatic substitution of 3,5-dichloro-4,4-difluoro-8-(4-tolyl)-
4-bora-3a,4a-diaza-s-indacene with o-bromophenol, followed
by palladium catalyzed benzofuran ring formation, is a pow-
erful and facile method for the formation of BODIPY dyes
with bathochromically shifted visible absorption and fluores-
cence spectra with considerably higher e(l), f, and Ff values.
Dyes 1, 2, and 3 have increasingly restricted conformational
flexibility. Their UV/Vis absorption and fluorescence emis-
sion spectra have been determined as a function of solvent.
These BODIPY derivatives show small Stokes shifts and
narrow absorption and emission bands. The small solvent-
dependent shifts of the absorption and emission bands of 1–
3 are primarily determined by solvent polarizability. The ab-
sorption and emission maxima of 2 are shifted to the red by
ꢀ30 nm in comparison to the unconstrained structure 1. The
most rigid structure 3 displays an additional bathochromic
shift in both the absorption and emission spectra of ꢀ30 nm
compared to 2. These results are also supported by quantum
chemical calculations which showed that the increase of con-
formational constraints leads to larger bathochromic shifts.
The crystal structures of the three dyes have been deter-
mined and show an increasing planarity of the chromophore
in line with a reduction of the conformational flexibility.
[12] S. Goeb, R. Ziessel, Org. Lett. 2007, 9, 737–740.
[13] K. Umezawa, Y. Nakamura, H. Makino, D. Citterio, K. Suzuki, J.
Am. Chem. Soc. 2008, 130, 1550–1551.
[14] K. Umezawa, A. Matsui, Y. Nakamura, D. Citterio, K. Suzuki,
Chem. Eur. J. 2009, 15, 1096–1106.
[15] A. B. Descalzo, H.-J. Xu, Z.-L. Xue, K. Hoffmann, Z. Shen, M. G.
Weller, X.-Z. You, K. Rurack, Org. Lett. 2008, 10, 1581–1584.
[16] Y.-W. Wang, A. B. Descalzo, Z. Shen, X.-Z. You, K. Rurack, Chem.
Eur. J. 2010, 16, 2887–2903.
´
[17] M. Baruah, W. Qin, N. Basaric, W. M. De Borggraeve, N. Boens, J.
Org. Chem. 2005, 70, 4152–4157.
[18] M. Baruah, W. Qin, R. A. L. Vallꢀe, D. Beljonne, T. Rohand, W.
Dehaen, N. Boens, Org. Lett. 2005, 7, 4377–4380.
[19] T. Rohand, M. Baruah, W. Qin, N. Boens, W. Dehaen, Chem.
Commun. 2006, 266–268.
[20] T. Rohand, J. Lycoops, S. Smout, E. Braeken, M. Sliwa, M. Van der
Auweraer, W. Dehaen, W. M. De Borggraeve, N. Boens, Photochem.
Photobiol. Sci. 2007, 6, 1061–1066.
[21] C. Thivierge, R. Bandichhor, K. Burgess, Org. Lett. 2007, 9, 2135–
2138.
[22] G. Zeni, R. Larock, Chem. Rev. 2006, 106, 4644–4680.
[23] The Cambridge Structural Database: a quarter of a million crystal
structures and rising. F. H. Allen, Acta Crystallogr. Sect. B 2002, 58,
380–388 [DOI: 10.1107/S0108768102003890].
Experimental Section
The synthetic route, materials and solvents used and the experimental
procedures for 1) the spectroscopic measurements, 2) the crystal structure
determinations, and 3) the quantum chemical calculations can be found
in the Supporting Information.
CCDC 755091 (1), 755092 (2), and 755093 (3) contain the supplementary
crystallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre at
www.ccdc.cam.ac.uk/data_request/cif.
[24] The following represents a non-exhaustive list of BODIPY papers
with spectroscopic/photophysical data; a) E. Vos de Wael, J. A. Par-
doen, J. A. van Koeveringe, J. Lugtenburg, Recl. Trav. Chim. Pays-
Bas 1977, 96, 306–309; b) T. Lꢇpez Arbeloa, F. Lꢇpez Arbeloa, I.
Lꢇpez Arbeloa, I. Garcꢈa-Moreno, A. Costela, R. Sastre, F. Amat-
Guerri, Chem. Phys. Lett. 1999, 299, 315–321; c) A. Costela, I.
Garcꢈa-Moreno, C. Gomez, R. C. Sastre, F. Amat-Guerri, M. Liras,
F. Lꢇpez Arbeloa, J. BaÇuelos Prieto, I. Lꢇpez Arbeloa, J. Phys.
Chem. A 2002, 106, 7736–7742; d) F. Lꢇpez Arbeloa, J. BaÇuelos
Acknowledgements
The Instituut voor de aanmoediging van innovatie door Wetenschap en
Technologie in Vlaanderen (IWT) is acknowledged for a fellowship to
VL. We thank the Fonds voor Wetenschappelijk Onderzoek (FWO), the
Ministerie voor Wetenschapsbeleid and the K.U.Leuven for continuing fi-
nancial support. WQ thanks the Scientific Research Fund for Introducing
Talented Persons to Lanzhou University and the Program for New Cen-
Chem. Asian J. 2010, 5, 2016 – 2026
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemasianj.org
2025

W. Qin, W. Dehaen et al.
FULL PAPERS
Prieto, V. Martꢈnez Martꢈnez, T. Arbeloa Lꢇpez, I. Lꢇpez Arbeloa,
ChemPhysChem 2004, 5, 1762–1771; e) J. BaÇuelos Prieto, F. Lꢇpez
Arbeloa, V. Martꢈnez Martꢈnez, T. Arbeloa Lꢇpez, F. Amat-Guerri,
M. Liras, I. Lꢇpez Arbeloa, Chem. Phys. Lett. 2004, 385, 29–35;
f) W. Qin, M. Baruah, M. Van der Auweraer, F. C. De Schryver, N.
Boens, J. Phys. Chem. A 2005, 109, 7371–7384; g) W. Qin, T.
Rohand, M. Baruah, A. Stefan, M. Van der Auweraer, W. Dehaen,
N. Boens, Chem. Phys. Lett. 2006, 420, 562–568; h) Z. Li, R. Bitt-
man, J. Org. Chem. 2007, 72, 8376–8382; i) Z. Ekmekci, M. D.
Yilmaz, E. U. Akkaya, Org. Lett. 2008, 10, 461–464; j) L. Li, J. Han,
B. Nguyen, K. Burgess, J. Org. Chem. 2008, 73, 1963–1970; k) E.
Fron, E. Coutino-Gonzalez, L. Pandey, M. Sliwa, M. Van der Au-
weraer, F. De Schryver, J. Thomas, Z. Dong, V. Leen, M. Smet, W.
Dehaen, T. Vosch, New J. Chem. 2009, 33, 1490–1496.
[26] R. A. L. Vallꢀe, M. Van der Auweraer, F. C. De Schryver, D. Bel-
jonne, M. Orrit, ChemPhysChem 2005, 6, 81–91.
[27] a) M. J. Kamlet, R. W. Taft, J. Am. Chem. Soc. 1976, 98, 377–383;
b) M. J. Kamlet, J. L. M. Abboud, R. W. Taft, J. Am. Chem. Soc.
1977, 99, 6027–6038.
[28] J. Catalꢆn, J. Phys. Chem. B 2009, 113, 5951–5960.
[29] J. Catalꢆn, H. Hopf, Eur. J. Org. Chem. 2004, 4694–4702.
[30] J. Catalꢆn in Handbook of Solvents, (Ed.: G. Wypych), ChemTec
Publishing, Toronto, 2001, pp. 583–616.
[31] J. Catalꢆn, C. Dꢈaz, Liebigs Ann. 1997, 1941–1949.
[32] J. Catalꢆn, C. Dꢈaz, V. Lꢇpez, P. Pꢀrez, J. L. G. de Paz, J. G. Rodri-
guez, Liebigs Ann. 1996, 1785–1794.
[33] Y. Marcus, Chem. Soc. Rev. 1993, 22, 409–416.
Received: April 6, 2010
Published online: August 2, 2010
[25] S. E. Braslavsky, Pure Appl. Chem. 2007, 79, 293–465.
2026
www.chemasianj.org
ꢁ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Asian J. 2010, 5, 2016 – 2026