High stokes shift anilido-pyridine boron difluoride dyes

Article abstract of DOI:10.1002/anie.201105228

To dye for: Three related families of fluorescent dyes, rigidified by boron difluoride and based on bidentate anilido-pyridyl donor ligands, are reported (see picture). The dyes show quantum yields up to 0.75 and improved Stokes shifts of >100 nm relative

Full text of DOI:10.1002/anie.201105228

Communications
DOI: 10.1002/anie.201105228
Fluorescence
High Stokes Shift Anilido-Pyridine Boron Difluoride Dyes**
Juan F. Araneda, Warren E. Piers,* Belinda Heyne,* Masood Parvez, and Robert McDonald
Fluorescent dyes are important tools for a variety of fields,
including nanoscience,[1] biological chemistry,[2] and solar
energy conversion.[3] Ideally, a dye possesses high absorption
coefficients and quantum yields, a large Stokes shift, tunable
absorption/emission profiles, and high chemical and photo-
chemical stability. Few compounds, or families of compounds,
have optimal attributes in all of these categories and thus,
there are many dyes available for researchers to choose
from.[2c] Despite this cornucopia of options, new dyes with a
specific set of properties are still of considerable interest.
One family of dyes that enjoys widespread use are the
dipyrrins rigidified by boron difluoride, or BODIPY dyes,[4]
exemplified by I. While the dipyrrin ligand[5] itself is non-
emissive, when complexed to the BF2 unit, the compounds are
high-quantum-yield emitters. The chromophore is largely
associated with the rigidified dipyrrin ligand, the BF2 unit
serving as a light-atom anchor for the fluorophore. The ligand
framework provides a versatile scaffold for chemical modi-
fication and there are hundreds of examples of this royal
family of dyes.
the lmax of absorption. Such assemblies, although effective, are
usually structurally complex and require significant synthetic
effort.
Scheme 1. Anilido-pyridine boron difluoride dyes.
A second, less explored strategy for improving Stokes
shifts in BF2-rigidified dyes relies on principles of ligand
design. The dipyrrin core is symmetrical about a C2 axis and
desymmetrization of the bidentate nitrogen ligand represents
a potential way to render ground and excited states more
energetically distinct. Recognizing that the C3N2 core of the
dipyrrin ligand is similar to that of the ubiquitous b-
diketiminato (“nacnac”) ligand framework II,[7] we postu-
lated that desymmetrization of this donor set (synthetically
less challenging than modification of the dipyrrin core) may
lead to effective boron-based dyes.[8] Indeed, a first-gener-
ation desymmetrized nacnac analog, the anilido-imine frame-
work (III) introduced by our group in 2003,[9] was employed
by Mu et al. in this context,[10] producing dyes with improved
Stokes shifts of around 70 nm over comparable BODIPY
dyes. Given the sensitivity of the imine moiety in III to
nucleophiles,[9] we have evolved the ligand design to the
anilido-pyridine scaffold generically depicted in IV. This
design renders the resulting BF2 complexes indefinitely air-
and moisture-stable in contrast to dyes III. Herein we report
the synthesis and properties of several BF2-rigidified dyes
based on this novel, unsymmetrical bidentate nitrogen donor
ligand motif.[11] Dyes IV are exceptionally photostable and
show high Stokes shifts of 90–120 nm.
One significant deficit in the properties of BODIPY dyes
is their generally low Stokes shift.[4b] Stokes shifts of > 80 nm
are desirable to minimize reabsorption of emitted photons;
BODIPY dyes typically exhibit shifts of 7–15 nm. Efforts to
address this issue involve the incorporation of the BODIPY
dye into an energy transfer cassette[6] in which the energy
absorbed by the BODIPY chromophore is transferred to a
second fluorophore (by some through-space or through-bond
mechanism) that emits at a wavelength well-separated from
[*] J. F. Araneda, W. E. Piers, B. Heyne, M. Parvez
Department of Chemistry, University of Calgary
2500 University Drive N.W., Calgary, Alberta, T2N 1N4 (Canada)
E-mail: wpiers@ucalgary.ca
Homepage: http://www.chem.ucalgary.ca/research/groups/
wpiers/
The range of compounds prepared is shown in Scheme 1.
The ligands were synthesized as their free anilines and
complexed to boron using standard, high-yielding methods
that involve treatment with BF3 in the presence of NEt3. The
key steps in any synthesis of ligands are indicated in
Scheme 1; full details are given in the Supporting Informa-
R. McDonald
Department of Chemistry, University of Alberta
11227 Saskatchewan Drive, Edmonton, Alberta, T6G 2G2 (Canada)
[**] Funding for this work was provided by the NSERC of Canada.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201105228.
12214
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 12214 –12217

tion. The parent anilido-pyridine ligands in compounds 1a–d
were prepared by Suzuki–Miyaura coupling[12] of 2-fluoro-
phenylboronic acid with the appropriate a-bromo pyridine
partner. Treatment of the resulting aryl fluoride with two
equivalents of LiN(H)Ar (Ar= Ph or 2,6-diisopropylphenyl)
gave the ligands in excellent yield. The more rigid bidentate
nitrogen donor in compounds 2a–b was prepared through
palladium-catalyzed Hartwig–Buchwald amination[13] of 10-
bromobenzo[h]quinoline using aniline (2a, 76%) or 2,6-
diisopropylaniline (2b, 46%). Finally, the ligands in com-
pounds 3a–b, based on the 9H-carbazole unit were prepared
using a Negishi coupling protocol[14] from the potassium salt
of the corresponding carbazole bromide and a 2-pyridylzinc
reagent[15] in 67–85% yield.
All eight new compounds were fully characterized by
multinuclear NMR spectroscopy and elemental analysis. All
except compound 3a were structurally characterized by X-ray
crystallography.[16] The 11B chemical shifts are in the range of 1
to 3 ppm and appear as well-defined triplets because of
coupling with the two 19F nuclei (1JBF ꢀ 27 Hz). The reso-
nances for the fluorine atoms appear at ꢁ129 to ꢁ141 ppm in
the 19F[17] NMR spectra, either as broad signals or 1:1:1:1
quartets when the coupling to boron is resolved. Figure 1
pounds 2, and the carbazolido compounds 3, where this
dihedral angle is essentially 08. Full details on these and the
structures of 1b–d and 2b can be found in the Supporting
Information.
None of the free ligands described above shows visible
fluorescence upon irradiation with a UV lamp, but all of the
difluoroboryl complexes are strongly emissive in both solu-
tion (CH2Cl2) and—qualitatively—in solid state. The solution
photophysical properties of these compounds are summarized
in Table 1, which also includes data for the BODIPY dye I for
Table 1: Absorption, fluorescence, and photophysical data for new dyes
and a typical BODIPY dye (I) for comparison.[a]
[b]
[c]
labs
[nm]
emax
[mꢁ1 cmꢁ1
lem
[nm]
f[d]
SS
[nm]
SS
tf
f
]
[cmꢁ1
]
[ns]
1a
1b
1c
1d
2a
2b
3a
3b
I
417
416
419
418
465
466
416
431
517
9529
9640
10065
10622
9114
5844
11252
9772
531
511
515
518
584
569
496
526
538
0.33
0.29
0.27
0.31
0.60
0.66
0.75
0.62
0.83
114
95
96
100
119
103
80
5148
4469
4449
4618
4382
3885
3877
4190
755
2.0
6.9
6.1
5.4
6.2
11.1
5.5
5.8
6.2
95
21
64000
[a] Recorded in deoxygenated anhydrous dichloromethane. [b] Longest
absorption maximum. [c] Emission maximum upon excitation at the
longest absorption maximum. [d] Absolute quantum yield determined by
calibrated integrating sphere systems.
Figure 1. Thermal ellipsoid (50%) diagrams of the molecular struc-
tures of 1a (A), 2a (B), and 3b (C). Selected bond lengths [ꢀ] and
angles [8] for 1a: B–N1 1.598(3), B–N2 1.498(3), N1–B–N2 106.5(2),
and C4–C5–C6–C7 ꢁ14.7(3). Selected bond lengths [ꢀ] and angles [8]
for 2a: B–N1 1.5942(12), B–N2 1.5053(14), N1–B–N2 109.99(8), and
C4–C5–C6–C7 3.2(1). Selected bond lengths [ꢀ] and angles [8] for 3b:
B–N1 1.608(5), B–N2 1.508(5), N1–B–N2 107.5(3), and C4–C5–C6–C7
0.0.
Figure 2. Normalized absorption and emission profiles of dye 2a,
showing the 119 nm Stokes shift.
comparison. Figure 2 shows the UV/Vis absorption and
emission profiles for compound 2a; the absorption profiles
for each dye match closely the excitation profiles obtained by
monitoring the wavelength of maximum emission. The parent
anilido-pyridine compounds 1a–d show absorption maxima
around 100 nm shifted to shorter wavelengths relative to I,
but emit with significantly improved Stokes shifts in the range
of 95–114 nm, albeit with moderate absolute quantum yields
of around 0.30. Surmising that the flexibility about the C5–C6
vector in these compounds may be providing nonradiative
pathways for excited-state relaxation, the more rigidified
shows thermal ellipsoid diagrams for 1a, 2a and 3b, along
with selected metrical data. In each family, the B–N1 distance
(pyridyl nitrogen) is around 0.1 ꢀ longer than the B–N2
lengths to the anilido nitrogen, emphasizing the metrical
asymmetry present in these molecules. In compounds 1a–d,
there is a considerable twist about the C4–C5–C6–C7 dihedral
angle in the ligand backbone (10–158) that flattens dramat-
ically in the more rigid benzo[h]quinolido-derived com-
Angew. Chem. Int. Ed. 2011, 50, 12214 –12217
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12215

Communications
ligands found in compounds 2 and 3 were designed to
gates formed through parallel stacking of molecules with a
“face-to-face” orientation in this medium. According to
Kasha’s exciton model, such aggregates are essentially non-
fluorescent. The large increase in emission intensity in the
presence of the membrane allows for an estimation of the
binding constant of 1.4 ꢁ 103 mꢁ1. Thus, this dye is compet-
itively membrane-specific, but with a large Stokes shift and
high photostability in comparison to BODIPY dyes.
In conclusion, a new family of BF2-rigidified dyes with
large Stokes shifts and high photostability has been prepared.
Their efficacy as probes for biological membranes has been
demonstrated using a liposome model. Modular ligand
syntheses will allow for attachment of a variety of groups
amenable to conjugation with proteins, lipids, and other
biological molecules[22] for applications in real biological
systems; these modifications are currently underway.
minimize this effect. Indeed, these families of dyes show a
doubled emission quantum yield relative to that of family 1,
with compound 3a giving a quantum yield comparable to that
of the BODIPY dye I. Significantly, dyes 2 and 3 retain the
high Stokes shifts found in compounds 1; as the representa-
tive absorption/emission profile for dye 2a (Figure 2) shows,
there is very little overlap between the absorption and
emission profiles for these dyes. Fluorescence lifetime
values are competitive with that found for I; the lifetime of
11.1 ns for 2b suggests that the benzo[h]quinolido-based dyes
may be useful in applications requiring longer fluorescence
lifetimes. One drawback of all of these dyes is the low molar
absorption coefficients; efforts towards increasing the absorp-
tivity of these dyes are underway.
Another significant advantage of the dye families
reported here in comparison to BODIPY and other com-
monly employed dyes is their exceptional photostability. Dye
samples (10ꢁ5 m in CH2Cl2) were irradiated at 420 nm at room
temperature and absorption spectra were recorded periodi-
cally. Under these conditions, BODIPY dye I was > 90%
photodegraded after 3 h, whereas rhodamine 101 bleached in
less than 2 h. In contrast, the absorption profile of 1a was
essentially unchanged after 10 h of irradiation. Dyes 2a,b
were photostable for at least 16 h, whereas the carbazole-
based dyes 3a,b retained around 80% of the intensity of
absorbance after this time period. Thus, in addition to air and
moisture stability, the anilido-pyridyl-based dyes do not
exhibit significant photodegradation over the course of a
day of continuous irradiation.
Received: July 26, 2011
Published online: October 21, 2011
Keywords: boron · fluorescent probes · liposome ·
.
photochemistry
[1] a) G. R. Whittell, M. D. Hager, U. S. Schubert, I. Manners, Nat.
Mater. 2011, 10, 176; b) A. Kaeser, A. P. H. J. Schenning, Adv.
Mater. 2010, 22, 2985; c) L. Maretti, P. S. Billone, Y. Liu, J. C.
Scaiano, J. Am. Chem. Soc. 2009, 131, 13972.
[2] a) M. S. T. Goncalves, Chem. Rev. 2008, 108, 190; b) J. M.
Baumes, J. J. Gassensmith, J. Giblin, J.-J. Lee, A. G. White,
W. J. Culligan, W. M. Leevy, M. Kuno, B. D. Smith, Nat. Chem.
2010, 2, 1025; c) R. P. Haugland, Handbook of Molecular Probes
and Research Products, 9th ed., Molecular Probes, OR, 2002.
[3] a) J.-H. Yum, P. Chen, M. Grꢂtzel, M. K. Nazeeruddin, Chem-
SusChem 2008, 1, 699; b) D. Kumaresan, R. P. Thummel, T.
Bura, G. Ulrich, R. Ziessel, Chem. Eur. J. 2009, 15, 6335.
[4] a) A. Loudet, K. Burgess, Chem. Rev. 2007, 107, 4891; b) R.
Ziessel, G. Ulrich, A. Harriman, New J. Chem. 2007, 31, 496;
c) G. Ulrich, R. Ziessel, A. Harriman, Angew. Chem. 2008, 120,
1202; Angew. Chem. Int. Ed. 2008, 47, 1184.
[5] T. E. Wood, A. Thompson, Chem. Rev. 2007, 107, 1831.
[6] A. Burghart, L. H. Thoresen, J. Chen, K. Burgess, F. Bergstrom,
L. B.-A. Johansson, Chem. Commun. 2000, 2203.
[7] L. Bourget-Merle, M. F. Lappert, J. R. Severn, Chem. Rev. 2002,
102, 3031.
[8] B. Qian, S. W. Baek, M. R. Smith III, Polyhedron 1999, 18, 2405.
[9] P. G. Hayes, G. C. Welch, D. J. H. Emslie, C. L. Noack, W. E.
Piers, M. Parvez, Organometallics 2003, 22, 1577.
[10] a) Y. Ren, X. M. Liu, H. Xia, L. Ye, Y. Mu, Eur. J. Inorg. Chem.
2007, 1808; b) X. M. Liu, Y. Ren, H. Xia, X. Fan, Y. Mu, Inorg.
Chim. Acta 2010, 363, 1441.
[11] NN bidentate ligands related to IV: a) Q. D. Liu, M. S. Mudadu,
R. Thummel, Y. Tao, S. N. Wang, Adv. Funct. Mater. 2005, 15,
143; b) S. B. Cortright, J. N. Johnston, Angew. Chem. 2002, 114,
355; Angew. Chem. Int. Ed. 2002, 41, 345; c) S. B. Cortright, J. C.
Huffman, R. A. Yoder, J. N. Coalter, J. N. Johnston, Organo-
metallics 2004, 23, 2238.
[12] a) N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457; b) A.
Suzuki, Angew. Chem. 2011, 123, 6854; Angew. Chem. Int. Ed.
2011, 50, 6722.
In contrast to the majority of BODIPY dyes,[18] the
photophysical properties of which are relatively insensitive to
solvent polarity, the anilido-pyridine supported dyes show
moderate hypsochromic shifts of the S0!S1 transition in more
polar solvents. For example, 3b gives a lmax of 452 nm in
hexane, but 426 nm in DMSO, suggesting that the excited
state is less polar than the ground state in these molecules.
The behavior of 3b was modeled using the Lippert–Mataga
equation,[19] supporting the notion that general solvent effects
are primarily responsible for the observed spectral shifts (see
the Supporting Information). The treatment indicates that the
dipole moment of the ground state is around 13.9 D greater
than that of the excited state, an observation consistent with
the desymmetrization of the molecule and a key feature of the
dyes that result in greater Stokes shifts.
Because of their planarity, hydrophobicity, and charge
neutrality, BODIPY dyes have been used as probes for
biological membrane function and dynamics.[20] The efficacy
of the anilido-pyridine dyes as probes for investigating
membrane properties was assessed in unilamellar liposomes
prepared
from
pure
dimyristoylphosphatidylcholine
(DMPC)[21] using dye 3a through a spectroscopic titration
technique. The efficient partitioning of the dye into the lipid
bilayer is demonstrated by a shift of the maximum emission
from 536 nm in water to 482 nm in the presence of the
liposomes, along with a significant increase in the relative
fluorescence intensity of around 70 fold. The very weak
intensity and large red shift in the emission band for 3a in
aqueous solution is likely due to the formation of H aggre-
[13] a) J. P. Wolfe, S. Wagaw, S. L. Buchwald, J. Am. Chem. Soc. 1996,
118, 7215; b) M. S. Driver, J. F. Hartwig, J. Am. Chem. Soc. 1996,
118, 7217.
12216 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 12214 –12217

[14] a) E. Negishi, A. O. King, N. Okukado, J. Org. Chem. 1977, 42,
1821; b) E. Negishi, Angew. Chem. 2011, 123, 6870; Angew.
Chem. Int. Ed. 2011, 50, 6738.
for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.
cam.ac.uk/data_request/cif.
[15] B. M. Coleridge, C. S. Bello, D. H. Ellenberger, A. Leitner,
Tetrahedron Lett. 2010, 51, 357.
[17] K. D. Conroy, W. E. Piers, M. Parvez, J. Organomet. Chem. 2008,
693, 834.
[16] Crystal data for 1a: C17H13BF2N2, MW= 294.10, monoclinic,
P21/c, a = 11.2101(4), b = 7.3817(4), c = 17.1219(9) ꢀ, a = 90, b =
[18] a) W. Qin, M. Baruah, M. Van der Auweraer, F. C. De Schryver,
N. L. Boens, J. Phys. Chem. A 2005, 109, 7371; b) W. Qin, T.
Rohand, M. Baruah, A. Stefan, M. V. der Auweraer, W. Dehaen,
N. Boens, Chem. Phys. Lett. 2006, 420, 562; c) A. Filarowski, M.
Kluba, K. Cieslik-Boczula, A. Koll, A. Kochel, L. Pandey, W. M.
De Borggraeve, M. Van der Auweraer, J. Catalan, N. Boens,
Photochem. Photobiol. Sci. 2010, 9, 996.
102.339(3), g = 908, V= 1384.10(11) ꢀ3, Z = 4, 1 = 1.411 gcmꢁ3
,
MoKa radiation, l = 0.71073 ꢀ, T= 173(2) K, 4290 measured
reflection, 3143 unique, min/max transmission = 0.9839 and
0.9939, R1 (I > 2s) = 0.0586, wR2 = 0.1152, GoF = 1.095, No. of
parameters = 195, final difference map within + 0.323 and
ꢁ0.258 eꢀ3. Crystal data for 2a: C19H13BF2N2, MW= 318.12,
[19] J. R. Lakowicz, Principles of Fluorescence Spectroscopy, 4th ed.,
Springer, 2006.
monoclinic,
P21/c,
a = 90,
a = 7.4793(2),
b = 20.1482(6),
g = 908,
, MoKa radiation, l =
c =
V=
10.2383(3) ꢀ,
b = 111.0840(3),
[20] a) J. Karolin, L. B. A. Johansson, L. Strandberg, T. Ny, J. Am.
Chem. Soc. 1994, 116, 7801; b) I. D. Johnson, H. C. Kang, R. P.
Haugland, Anal. Biochem. 1991, 198, 228; c) M. Dahim, N. K.
Mizuno, X.-M. Li, W. E. Momsen, M. M. Momsen, H. L. Brock-
man, Biophys. J. 2002, 83, 1511; d) I. A. Boldyrev, X. Zhai, M. M.
Momsen, H. L. Brockman, R. E. Brown, J. G. Molotkovsky, J.
Lipid Res. 2007, 48, 1518.
[21] B. Heyne, D. Brault, M. P. Fontaine-Aupart, S. Kohnen, F. Tfibel,
A. Mouithys-Mickalad, G. Deby-Dupont, P. Hans, M. Hoebeke,
Biochim. Biophys. Acta Gen. Subj. 2005, 1724, 100.
1439.57(7) ꢀ3, Z = 4, 1 = 1.468 gcmꢁ3
0.71073 ꢀ, T= 173(2) K, 12783 measured reflection, 3313
unique, min/max transmission = 0.9534 and 0.9823, R1 (I >
2s) = 0.0338, wR2 = 0.0948, GoF = 1.048, No. of parameters =
217, final difference map within + 0.283 and ꢁ0.206 eꢀ3. Crystal
data for 3b: C26H29BCl2F2N2, MW= 489.22, monoclinic, P21/m,
a = 11.4796(5), b = 6.9227(4), c = 15.4393(5) ꢀ, a = 90, b =
97.296(3), g = 908, V= 1217.02(10) ꢀ3, Z = 2, 1 = 1.335 gcmꢁ3
,
MoKa radiation, l = 0.71073 ꢀ, T= 173(2) K, 4108 measured
reflection, 2965 unique, min/max transmission = 0. 9650 and
0.9209 R1 (I > 2s) = 0.0746, wR2 = 0.1721, GoF = 1.057, No. of
parameters = 208, final difference map within + 0.726 and
ꢁ0.767 eꢀ3. CCDC 836295, 836296, 836297, 836298, 836299,
836290, 836291 contain the supplementary crystallographic data
[22] a) K. Krumova, G. Cosa, J. Am. Chem. Soc. 2010, 132, 17560;
b) M. F. Debets, S. S. van Berkel, J. Dommerholt, A. J. Dirks,
F. P. J. T. Rutjes, F. L. van Delft, Acc. Chem. Res. 2011, 44, 805-
815; c) E. Saxon, C. R. Bertozzi, Science 2000, 287, 2007.
Angew. Chem. Int. Ed. 2011, 50, 12214 –12217
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 12217