J. Cadet, T. Douki, J-L. Ravanat and P. Di Mascio, Photochem. Photobiol. Sci.
, 2009, 8, 903
M. R. Solomon, J. Sivaguru, S. Jockusch, W. Adam and N. J. Turro, Photochem. Photobiol. Sci.
, 2009, 8,
912
F. Manjón, D. García-Fresnadillo and G. Orellana, Photochem. Photobiol. Sci.
, 2009, 8, 926
E. Alarcón, A. M. Edwards, A. Aspée, C. D. Borsarelli and E. A. Lissi, Photochem. Photobiol. Sci.
, 2009, 8,
933
P. A. Hoijemberg, S. D. Karlen, C. N. Sanramé, P. F. Aramendía and M. A. García-Garibay, Photochem.
Photobiol. Sci.
, 2009, 8, 961
N. Quici and M. I. Litter, Photochem. Photobiol. Sci.
, 2009, 8, 975
Photochem. Photobiol.
Sci.
, 2009, 8, 985
T. S. Corbitt, L. Ding, E. Ji, L. K. Ista, K. Ogawa, G. P. Lopez, K. S. Schanze and D. G. Whitten, Photochem.
Photobiol. Sci.
, 2009, 8, 998
1
2
1 ∆ g
, Photochem.
Photobiol. Sci.
, 2009, 8, 1024
Photochem. Photobiol. Sci.
, 2009, 8, 1040

a
c
b
c b
c b
c
Received 12th March 2009, Accepted 27th April 2009
First published as an Advance Article on the web 12th May 2009
DOI: 10.1039/b905054j
The solvatochromic photophysical properties of two fluorescent, cyano-substituted BODIPY dyes—8-(4-bromophenyl)-3,4,4,5-tetracyano-4-bora-3a,4a-diazas -indacene ( 4CN ) and
8-(4-bromophenyl)-3,5-dicyano-4,4-difluoro-4-bora-3a,4a-diazas -indacene ( 2CN )—have been studied in various solvents by UV–vis spectrophotometry and steady-state and time-resolved fluorometry.
These two BODIPY analogues have comparable photophysical properties, implying that displacement of F by CN at boron has a negligible effect. Both compounds have high fluorescence quantum yields U f
(0.65–0.90 for 4CN and 0.63–0.88 for 2CN ) in the solvents studied and display mono-exponential fluorescence decay profiles in nonprotic solvents. A new, generalized treatment of the solvent effect based on four mutually independent, empirical solvent scales (dipolarity, polarizability, acidity, and basicity of the medium) indicates that solvent polarizability and, to a lesser degree, solvent (di)polarity are crucial factors causing the solvent-dependent shifts of the UV–vis absorption and fluorescence emission. The rate constants of radiative deactivation ( k f
) are nearly independent of the nonprotic solvent [ k f
= (1.4
± 0.1) ¥ 10 8 s 1 for 4CN and (1.5
± 0.2) ¥ 10 8 s 1 for 2CN ]. Both compounds undergo a color change in polar aprotic solvents (acetone, acetonitrile, and N , N -dimethylformamide), which can be stopped by addition of HClO
4
. The kinetics of this color change indicates that the decomposition of these cyano-substituted BODIPY compounds is complex.
The first 4,4-difluoro-4-bora-3a,4a-diazas -indacene (better known under the commercial name “BODIPY” or difluoro bo ron dipy rromethene) dye was reported in 1968.
1 Since then, BODIPYbased dyes have become widely used fluorescent probes in materials science, medical diagnostics, and biotechnology.
2 The syntheses, functionalization reactions, and spectroscopic properties of boradiazaindacene derivatives have been discussed in several excellent reviews.
3–5 Two valuable, recent, innovative functionalization reactions of the BODIPY core are: (i) S
N
Ar reactions of 3-chloro or 3,5-dichloroBODIPYs, 6–10 and (ii) nucleophilic displacement of fluoride atoms from boron.
9,11–16 Burgess et al.
have shown that nucleophilic displacement of F by CN and S
N
Ar reactions with cyanide can compete.
9 The reaction of dichloroBODIPY 1 with trimethylsilyl cyanide (TMSCN, as cyanide source) in the presence of tin tetrachloride occurred selectively at the C–Cl bonds to give the dicyanide 2CN , but boron trifluoride also promoted displacement of the B–F bonds in addition to substitution of Cl at the 3,5-positions to produce the tetracyanide 4CN (Scheme 1).
9 a Faculty of Chemistry, University of Wroclaw, F. Joliot-Curie 14, 50-383,
Wrocław, Poland b Department of Chemistry, Texas A & M University, Box 30012, College
Station, TX 77842-3012, USA. E-mail: burgess@tamu.edu; Fax: +1 979
845 1881; Tel: +1 979 845 4345 c Department of Chemistry, Katholieke Universiteit Leuven, Celestijnenlaan 200f-bus 02404, 3001, Leuven, Belgium. E-mail: Noel.Boens@chem.
kuleuven.be; Fax: +32 16 327 990; Tel: +32 16 327 497
† This article was published as part of the themed issue in honour of Esther
Oliveros.
‡ Electronic supplementary information (ESI) available: Fig. S1–S7. See
DOI: 10.1039/b905054j
Scheme 1 Syntheses of compounds 2CN and 4CN .
Since 2CN and 4CN are the first members of the BODIPY family with uncommon cyano substituents at the 3,5-positions, it is worth investigating the spectroscopic and photophysical properties of these new compounds with electron-withdrawing functions and
1006 | Photochem. Photobiol. Sci.
, 2009, 8 , 1006–1015 This journal is © The Royal Society of Chemistry and Owner Societies 2009

to compare them with other difluoroboron dipyrromethene dyes.
Furthermore, it is interesting to look into the effect on the photophysics of the additional displacement of F by CN at boron in 4CN .
In this work, the solvent-dependent photophysical/spectroscopic characteristics of 4CN and 2CN have been investigated by steady-state UV–vis absorption and emission spectroscopy as well as time-resolved fluorometry. From these experiments, we could determine the position of the spectral maxima [ l abs
(max), l ex
(max), and l em
(max)], the full width at half height of the maximum of the absorption band (fwhm abs
), the Stokes shifts ( D ¯ ), the fluorescence quantum yields ( U f
), the fluorescence decay times
( t ), and the rate constants of radiative ( k f
) and nonradiative ( k nr
) deactivation in the case of mono-exponential fluorescence decays.
We used single- and multi-parameter expressions to describe the solvent effect on the position of the spectral maxima and Stokes shifts of 4CN and 2CN . The color change reaction of 4CN and
2CN in polar aprotic solvents is also discussed.
2.1.
Synthesis
The fluorescent dyes 8-(4-bromophenyl)-3,4,4,5-tetracyano-4bora-3a,4a-diazas -indacene ( 4CN ) and 8-(4-bromophenyl)-3,5dicyano-4,4-difluoro-4-bora-3a,4a-diazas -indacene ( 2CN ) were synthesized according to a published procedure (Scheme 1).
9
2.2.
Spectroscopic properties of 4CN
A selection of UV–vis absorption and fluorescence emission spectra of 4CN dissolved in several solvents of varying polarity/polarizability is depicted in Fig. 1. The photophysical/spectroscopic properties of 4CN are compiled in
Table 1. The maximum of the main absorption band [ l abs
(max)] of
4CN —attributed to the 0–0 vibrational band of a strong S
0
→ S
1 transition—is located between 511 nm (in methanol) and 517 nm
(in cyclohexane and chloroform). This narrow, typical BODIPYlike main absorption band 17 of 4CN is seen in all solvents, except
Fig. 1 (A) Absorption spectra of 4CN in different solvents normalized to 1.0. (B) Corresponding normalized fluorescence emission spectra
( l ex
=
488 nm). Because all the spectra have similar shapes and for better clarity, only a limited number of spectra are shown.
Table 1 Photophysical properties of 4CN in several solvents
# a Solvent SdP b l abs
(max)/nm l em
(max)/nm D ¯ /cm 1 fwhm abs
/cm 1 U f c t
1 d /ns t
2 d /ns k f
/10 8 s 1 k nr
/10 7 s 1
1 Cyclohexane
2 1,4-Dioxane
0.000
0.312
3 Dibutyl ether 0.175
4 Diisopropyl ether 0.324
5
6
7
Diethyl ether
Chloroform
Ethyl acetate
8 THF
9 1-Octanol
10
11
1-Pentanol
2-Propanol
12 Acetone + H + e
13 Methanol
14 DMF + H + e
15 MeCN + H + e
0.385
0.614
0.603
0.634
0.454
0.587
0.808
0.907
0.904
0.977
0.974
514
515
514
512
511
511
514
511
517
516
515
513
513
517
512
527
527
526
523
523
523
528
523
527
529
526
523
523
527
524
480
442
444
410
449
449
516
449
367
476
406
373
373
367
447
823
724
725
735
693
745
796
758
641
743
681
673
693
655
756
0.79
5.55
0.87
5.75
0.80
5.87
0.77
5.33
0.86
6.11
0.90
6.05
0.77
5.99
0.84
5.72
0.75
5.81
0.81
5.89
0.74
5.85
0.90
5.92
0.80
6.17
0.70
4.94
0.65
6.04
0.14
1.44
1.91
1.42
1.52
1.34
1.45
1.41
1.48
1.29
1.47
1.26
1.50
3.78
2.26
3.58
4.31
2.29
1.82
3.84
2.80
4.44
1.86
1.42
1.09
6.07
5.79
a The solvents are numbered according to increasing dielectric constant e of the pure solvents.
determined by excitation at 488 nm with fluorescein in 0.1 M NaOH
0.08.
d aq as reference ( U f b Catal´an solvent dipolarity parameter.
= 0.92). The standard uncertainties on U
For the time-resolved fluorescence measurements, the samples were excited at 500 nm. The standard errors on t
1 and t
2 f c values were are between 0.002 and are
≤
15 ps.
e Acidified by
HClO
4
.
U f
This journal is © The Royal Society of Chemistry and Owner Societies 2009 Photochem. Photobiol. Sci.
, 2009, 8 , 1006–1015 | 1007

the polar aprotic ones. Indeed, in acetone, acetonitrile, and
N , N -dimethylformamide (DMF), 4CN undergoes a (possibly fast) reaction. However, addition of HClO
4 to these solvents inhibits or stops this reaction. The exact mode of action of HClO
4 is unknown. In these solvents, acidified by HClO by + H +
4
(indicated in Table 1), the typical narrow main absorption band of BODIPY can be observed. As an example the UV–vis absorption spectrum of 4CN in acetone with added HClO
4 is displayed in Fig. 1A. This reaction will be discussed further in section 2.4. The full width at half height of the maximum of the main absorption band (fwhm abs
) increases from 641 cm 1 cyclohexane to 823 cm 1 in in tetrahydrofuran (THF). A shoulder at the short wavelength side can be observed—located between
484 and 487 nm, the position of which is nearly independent of the solvent—which is assigned to the 0–1 vibrational band of the
S
0
→ S
1 transition. Furthermore, a weak absorption band is found between 356 and 367 nm, depending on the solvent. This broad and weak absorption band is assigned to the S
0
→ S
2 transition.
The fluorescence excitation spectra match the absorption spectra in all solvents studied and, moreover, l abs
(max) = l ex
(max).
The solvent-dependent range of the absorption maxima l abs
(max) of 4CN is generally red-shifted compared to 3,5unsubstituted BODIPY ( ~ 490– ~ 500 nm), and similar to that of 3,5-dimethylBODIPY ( ~ 505– ~ 520 nm), 3,5-dichloroBODIPY
( ~ 505– ~ 530 nm), 3,5-dimethoxyBODIPY ( ~ 510– ~ 520 nm), and
3,5-diphenoxyBODIPY ( ~ 510– ~ 520 nm).
6,17,18 The cited values are slightly sensitive to substituents at other positions of the BODIPY core, e.g.
the meso -substituent. Compared to 4CN , difluoroboron dipyrromethenes with 3,5-diaryl, 3,5-diethenylaryl, 3,5-dianilino, and 3,5-diethynylaryl substituents have bathochromic solventdependent l abs
(max) ranges.
17
The fluorescence emission spectra of 4CN in all the solvents tested, except the polar aprotic ones (see section 2.4), display the typical BODIPY features, 17 i.e.
a narrow, slightly Stokes-shifted band of mirror image shape of the S
0
→ S
1 transition (Fig. 1B).
The maximum emission wavelengths l em
(max) lie in a very narrow wavelength range, namely between 523 and 529 nm. In all solvents a shoulder is observed at the red wavelength side, located between
552 and 560 nm. The fluorescence quantum yield, U f
, of 4CN is high (0.65–0.90) in all solvents studied. These spectroscopic characteristics of 4CN are observed in apolar aprotic, polar protic (alcohols), and polar aprotic solvents (acetone, acetonitrile, and DMF) acidified by HClO
4
. However, in these pure, polar aprotic solvents a color change occurs, which will be discussed in section 2.4.
The range of the solvent-dependent emission maxima l em
(max) of 4CN follows the trend of l abs
(max). (i) l em
(max) values are blue-shifted compared to those of 3,5-diaryl, 3,5-diethenylaryl,
3,5-dianilino, and 3,5-diethynylaryl boradiazaindacene dyes.
(ii) l em
(max) values are comparable to those of 3,5-dimethyl, 3,5dichloro, 3,5-dimethoxy, and 3,5-diphenoxy substituted BODPY.
(iii) l em
(max) values are bathochromically-shifted in comparison to 3,5-unsubstituted difluoroboron dipyrromethenes.
6,17,18
Fluorescence decay traces of 4CN in different solvents were collected as a function of emission wavelength. The results of the time-resolved fluorescence experiments are also listed in Table 1.
The time-resolved fluorescence displayed single-exponential decay kinetics in all the solvents studied—except in some alcohols—with lifetimes ranging from 4.94 ns in acidified DMF to 6.11 ns in diethyl ether. In most alcohols, biexponential fluorescence decays were observed. In methanol, two decay times were found (
6.17 ns, t
2 t
1
= 1.91 ns) independent of the emission wavelength l
= em used for collecting the fluorescence decay traces. The amplitudes a
1 and a
2 associated with the decay times t
1 and t
2
, respectively, varied with l em
. For methanol, the ratio a
2
/ a
1 increased with increasing l em
: from 0.03 at 515 nm, over 0.05 at 525 nm, to 0.12
at 560 nm and 0.16 at 575 nm. Wavelength dependent ratios a
2 were also found for 1-pentanol: a
2
/ a
1
/ a
= 0.04 at 515 nm, 0.24 at
1
525 nm, 0.25 at 560 nm, and 0.03 at 575 nm. For 1-octanol, a
2 is small at each l em
: a
2
/ a
1
/ a
= 0.01 at 515 nm, 0.05 at 525 nm, 0.05
1 at 560 nm and 0.04 at 575 nm. The dependence of the short decay time, t
2
, on solvent viscosity is contrary to what would be expected in case of solvent relaxation (through dipolar interaction or hydrogen bond formation).
18 Moreover, the observed wavelength dependence of a
2
/ a
1 in protic solvents and, more specifically, the absence in all cases of negative pre-exponential factors a
2 at long emission wavelengths excludes one excited species being formed from the other one, but rather suggests the presence of a second
“independent” emitting species. The combination of the small a
2 and relatively small t
2
( i.e.
the small contribution of this species to the stationary emission spectra) excludes construction of reliable emission spectra of this species.
When the fluorescence decays were monoexponential, the rate constants of radiative ( k f
) and radiationless ( k nr
) deactivation were calculated from the measured fluorescence quantum yield ( U f
) and lifetime ( t ) according to eqn (1): k f
= U f
/ t (1a) k nr
= (1 U f
)/ t (1b)
The data in Table 1 show that k f
(nonprotic) solvent [ k f is nearly independent of the
= (1.4
± 0.1) ¥ 10 8 s 1 ]. The rate constants for radiationless deactivation solvent independent [ k nr k nr are relatively small and practically
= (4 ± 1) ¥ 10 7 s 1 ]. The small values of k nr corroborate a trend found for other boradiazaindacene dyes, where it was found that electron donating substituents increase k nr
, especially in polar solvents.
18
2.3.
Solvatochromism of 4CN
Solvent-dependent spectral shifts are often interpreted in terms of the Lippert–Mataga equation (eqn (2)), which describes the solvatochromic Stokes shift D ¯ (expressed in wavenumbers) as a function of the change of the dipole moment D m ge
= m e
m g of
4CN upon excitation. The validity of eqn (2) can be checked by using various solvents with different dielectric constants ( e ) and refractive indices ( n ) and by plotting D ¯ as a function of D f = f ( e ) f ( n 2 ).
19,20
D n =
2 D f
4 p e
0 hca
3
( m e
m g
)
2 + constant (2a) f ( e ) = ( e 1)/(2 e + 1) and f ( n 2 ) = ( n 2 1)/(2 n 2 + 1) (2b)
[
In eqn (2), D ¯ = ¯ abs between absorption [ ¯
em
¯ em is the solvatochromic shift (in cm
= 1/ l abs
1
(max)] and fluorescence emission
)
¯ em
= 1/ l em
(max)] maxima, h is Planck’s constant, c is the velocity of light, e
0 is the permittivity of vacuum, and a represents the radius of the cavity in which the solute resides.
m g and m e denote the
1008 | Photochem. Photobiol. Sci.
, 2009, 8 , 1006–1015 This journal is © The Royal Society of Chemistry and Owner Societies 2009

dipole moments of 4CN in ground and excited state, respectively.
The use of eqn (2) is limited to transitions where the excited state reached after excitation is also the emissive state (hence for S
0
→ S
1 excitation and equal dipole moments for the Franck–Condon and relaxed states) and where the excited-state dipole moment is independent of solvent polarity.
The Lippert–Mataga plot of the Stokes shift D ¯ versus D f for
4CN for the solvents of Table 1 is represented in Fig. S1 (ESI‡).
There is a poor linear relationship [correlation coefficient r = 0.335
with slope = (1.6
± 1.3) ¥ 10 2 cm 1 , intercept = (4.0
± 0.2) ¥ 10 2 cm 1 ] between D ¯ and D f . The small slope implies that the permanent dipole (if any) of 4CN does not differ much between the ground and excited state. The straight line in Fig. S1 represents the average value of D ¯ .
The solvent effect on ¯ abs
, ¯ em
, and D ¯ can also be described mathematically by a multilinear expression (eqn (3a)): y = y
0
+ a A + b B + c C + d D (3a) where y
0 stands for the physicochemical property of interest in the gas phase; a , b , c , and d are adjustable regression coefficients that reflect the sensitivity of the physicochemical property y (here n abs
, ¯ em
, D ¯ ) in a given solvent to the solvent properties A, B, C, and D. In the literature there are several solvent scales reported to describe the solvent dependence of y . Kamlet and Taft 21 proposed the p *, a , and b parameters to characterize, respectively, the polarity/polarizability, the acidity, and the basicity of a solvent
(eqn (3b)). Recently, Catal´an proposed a generalized treatment of the solvent effect based on a set of four empirical, independent solvent scales.
22 SdP, 22 SP, 23 SA, 24,25 and SB 26 characterize the dipolarity, polarizability, acidity, and basicity, respectively, of a certain solvent (eqn (3c)). The Kamlet–Taft solvatochromic parameters a , b , and p * parameters are taken from ref. 27, and the Catal´an SA and SB parameters come from ref. 24–26. The
Catal´an solvent polarizability parameters SP are collected from ref. 23. The new SdP solvent parameters can be found in ref. 22 and are also compiled in Table 1. The { SdP, SP, SA, and SB } values for dibutyl ether and diisopropyl ether were not available in the literature and have been determined especially for this study.
The { SdP, SP, SA, and SB } parameter values of Bu
2
O are 0.175,
0.672, 0.000, and 0.637, respectively. The corresponding { SdP, SP,
SA, and SB } values for i Pr
2
O are 0.324, 0.625, 0.000, and 0.657, respectively.
y = y
0
+ a a a + b b b + c p * p * (Kamlet–Taft) (3b) y = y
0
+ a
SA
SA + b
SB
SB + c
SP
SP + d
SdP
SdP (Catal´an, new) (3c)
Table 2 compiles the estimated regression coefficients y
0
, a , b , c , d (see eqn (3)) and correlation coefficients ( r ) for the multiple n abs
) and fluorescence n em
) maxima and the Stokes shift D ¯ of 4CN according to eqn (3b) and (3c) for the solvents of Table 1. The analysis of n abs data (Table 1) according to Kamlet–Taft (eqn (3b)) using a solvent scale in which solvent (di)polarity and polarizability effects are combined in the single parameter p * shows a poor fit, as assessed by the value of r (0.555) and the large standard errors on the estimated parameters a a
, b b
, and c p * as quality-of-fit criteria. Conversely, use of the new Catal´an solvent parameter set
(eqn (3c)), where (di)polarity and polarizability effects are split, gives a perfect multilinear fit to ¯ abs
( r = 0.988). Because this multiple linear regression of ¯ abs for a
SA and b
SB gave relatively small estimates with comparatively large associated standard errors, we decided to perform the multilinear analysis according to eqn (3c) with solvent dipolarity (SdP) and polarizability (SP) as only solvent scales. This analysis gave a very good fit too
( r = 0.972). Comparison of the simple linear regressions of ¯ abs versus SP and of ¯ abs versus SdP showed that solvent polarizability
( r = 0.732) is a somewhat more important factor than solvent
(di)polarity ( r = 0.680) influencing the solvent-dependent shifts of the absorption band. However, the better fit of ¯ abs as a function of { SP, SdP } (see above) demonstrates that solvent (di)polarity cannot be disregarded. For ¯ em
, the new Catal´an (eqn (3c)) solvent scales produced definitely the best results ( r = 0.947 for Catal´an versus 0.205 for Kamlet–Taft). The fit according to eqn (3c) with
{ SP, SdP } as independent variables (hence neglecting solvent acidity and basicity), which is reported in Table 2, has a correlation coefficient ( r = 0.934) that is close to that for the fit with { SA, SB,
SdP, and SP } as solvent scales ( r = 0.947). Therefore, as for n abs
, solvent acidity and basicity are not responsible for the shift of ¯ em
.
Evaluation of the simple linear regressions of ¯ em versus SP and of ¯ em versus SdP showed that the solvent-dependent shifts of the emission are influenced more by solvent polarizability ( r = 0.886)
This journal is © The Royal Society of Chemistry and Owner Societies 2009 Photochem. Photobiol. Sci.
, 2009, 8 , 1006–1015 | 1009 r
0.988
0.972
0.732
0.680
0.934
0.886
0.344
0.735

Fig. 2 Plot of the absorption, ¯ abs n abs em
.
n em
, maxima of 4CN as a function of f ( n 2 ). The numbers refer to the solvents of Table 1. The straight than by solvent (di)polarity ( r = 0.344). The linearity of the plots
(Fig. 2) of ¯ abs versus f ( n 2 ) ( r = 0.875) and of n em versus f ( n 2 ) ( r =
0.808) confirms that van der Waals and excitonic interactions with a polarizable solvent are primarily responsible for the experimental solvent-dependent shifts.
28 The slopes of the plots of absorption maxima ¯ abs
[( 4.5
± 0.9) ¥ 10 3
[( 4.8
± 0.7) ¥ 10 3 cm 1 cm
] versus f ( n 2
1 ] and emission maxima
) are almost identical.
¯ em
The multilinear fit of the small Stokes shifts D ¯ according to Kamlet–Taft ( r = 0.817) was better than to Catal´an ( r = n abs n em the Stokes shift
D ¯ will not be influenced by the solvent polarizability. Hence, the Catal´an solvent scales, which explicitly take into account this solvent property, lose their major advantage when the solvent dependence of D ¯ is considered. The minor solvent dependence of
D ¯ fails to be expressed accurately by any solvent scale (Lippert–
Mataga, Kamlet–Taft, or Catal´an). This is related to the very small difference between the dipole moments in ground and excited state.
2.4.
Reaction of 4CN in the polar aprotic solvents acetone, acetonitrile, and DMF
The preparation of solutions of 4CN in different solvents and the subsequent recording of their UV–vis absorption and fluorescence excitation and emission spectra went without a glitch until we started making a solution of 4CN in DMF. The original, intense, red color of the solution of 4CN in DMF changed almost immediately to pale yellow. The resulting UV–vis absorption and fluorescence emission spectra of 4CN in DMF are shown
[ in Fig. 3. It is evident that these broad blue-shifted absorption l abs
(max) = 477 nm] and fluorescence [ l em
(max) = 485 nm] bands are quite different from the narrow bandwidth spectra of
Fig. 3 Normalized absorption spectra of 4CN in DMF (blue solid line) and in DMF + H + (black solid line). Corresponding normalized fluorescence emission spectra of line) and in DMF + H + ( l ex
4CN in DMF ( l ex
=
430 nm, blue dotted
= 488 nm, black dotted line).
classical BODIPY dyes 17 and which are also found for 4CN in, e.g.
chloroform, 1,4-dioxane, or ethyl acetate (see Fig. 1). At the same time as the original BODIPY-like bands have gone, this very fast reaction in DMF produces new, broad, hypsochromic absorption and emission bands, which are associated with the formation of new chromophoric species. Similar, though more gradual, color changes were also observed in acetone and acetonitrile. It appears that the reaction is limited to the polar aprotic, hydrogen bond accepting solvents acetone (dielectric constant e = 20.7), DMF
1010 | Photochem. Photobiol. Sci.
, 2009, 8 , 1006–1015 This journal is © The Royal Society of Chemistry and Owner Societies 2009

( e = 36.7), and acetonitrile ( e = 37.5). In nonpolar aprotic (up to e = 7.5 for THF) and protic (alcohol) solvents (up to e = 33.1
for methanol) this reaction does not occur. The reaction can be prevented from occurring if one adds HClO
4
(at ~ 1 mM) to the polar aprotic solvent before the preparation of the 4CN solutions.
Fig. 3 displays the UV–vis absorption and fluorescence spectra of 4CN in DMF in the absence and presence of HClO
4
. The spectra in DMF + HClO features, 17
4 display the characteristic BODIPY i.e.
a narrow, strong absorption band around 510 nm and a shoulder at the short wavelength side in addition to a slightly Stokes-shifted emission band of mirror image shape. The
4CN absorption and emission spectra in acetone with added
HClO
4 are shown in Fig. 1 for comparison. A related, fast color change has been reported for the boron dipyrromethene dye PM650 (8-cyano-4,4-difluoro-1,2,3,5,6,7-hexamethyl-4-bora-
3a,4a-diazas -indacene) in N , N -dimethylamides.
29,30 The change of the PM650 color was not observed in acetone, although the solution became nearly transparent after several days. PM650 has an electron-withdrawing cyano group at the 8-position of the
BODIPY core. It was proposed that the color change is related to the presence of the electron-withdrawing cyano group in the chromophoric p -system of the BODIPY dye and the nature of solvents characterized by high electron-donor ability and low proton-donor capacity. The exact nature of the decomposition reaction has not been elucidated.
30 One should note that in some easily reduced 3-indolocarbocyanine dyes a similar (although reversible) color change has been observed in alcohols and acetone.
31
To obtain a deeper insight in the change of color of 4CN in polar, aprotic solvents, the UV–vis absorption and fluorescence excitation and emission spectra of the dye have been recorded as a function of the ageing time, i.e.
the time just after sample preparation. However, because the color change of 4CN in DMF is so fast (it starts immediately during sample preparation), we decided to follow the course of the reaction in pure acetone ( i.e.
without added HClO
4
) instead. Fig. 4A displays the absorption spectra of
4CN in acetone at several ageing times. We used a sample solution with an initial absorbance of the main absorption band higher than 0.1, because this allowed us to follow the reaction kinetics for longer times than samples with absorbances under 0.1. Indeed, samples with low absorbance ( < 0.1) have (evidently) low 4CN concentrations, so that the distinctive BODIPY-like bands have almost vanished when one starts following the reaction kinetics of such samples. Conversely, samples with higher concentration (and hence higher absorbance) let one still observe the characteristic
BODIPY-like absorption bands at early ageing times. At the beginning, the BODIPY-like main absorption band at 511 nm
(assigned to 0–0 vibrational transition of the S the broad band at 362 nm (S
0
0 the shoulder at 485 nm (0–1 vibrational transition of S
0
→ S
1
→ S
2
→ S
1 transition),
), and transition) are clearly visible.
The initial absorption spectra of 4CN in acetone (Fig. 4A) are nearly identical to those in acetone with added HClO
4
(Fig. 1A).
With time, however, the intensity of the bands at 511, 485, and
362 nm diminishes while the absorption band with maxima at 409 and 426 nm intensifies. Two clear isosbestic points, at 378 and
452 nm, can be observed. Fig. 4B shows the time-evolution of the fluorescence excitation spectra of 4CN in acetone observed at 550 nm. Two pseudo-isoemissive points, at 395 and 447 nm, are clearly visible. The development of the different bands as a
Fig. 4 (A) Absorption spectra of 4CN in acetone for different ageing times: 5 ¢ (a), 10 ¢ (b), 15 ¢ (c), 20 ¢ (d), 25 ¢ (e), 30 ¢ (f), 40 ¢ (g), 60 ¢ (h), 90 ¢
(i), 110 ¢ (j). (B, C) Fluorescence excitation spectra of 4CN in acetone for different ageing times. For (B) (
10 ¢ (b), 15 ¢ (c), 20 ¢ (d), 25 ¢ l
(e), 35 ¢ em
= 550 nm) the ageing times are: 5 ¢ (a),
(f), 45 ¢ (g), 55 ¢ (h), 65 ¢ (i). For (C) ( l em
=
477 nm) the ageing times are: 5 ¢ (a), 10 ¢ (b), 15 ¢ (c), 20 ¢ (d), 25 ¢ (e), 30 ¢ (f),
35 ¢ (g), 40 ¢ (h), 45 ¢ (i), 50 ¢ (g), 55 ¢ (h), 60 ¢ (i). (D) Fluorescence emission spectra ( l ex
=
360 nm) of 4CN in acetone for different ageing times: 5 ¢ (a),
10 ¢ (b), 15 ¢ (c), 20 ¢ (d), 25 ¢ (e), 30 ¢ (f), 35 ¢ (g), 45 ¢ (h), 50 ¢ (i), 55 ¢ (j), 70 ¢ (k),
90 ¢ (l).
This journal is © The Royal Society of Chemistry and Owner Societies 2009 Photochem. Photobiol. Sci.
, 2009, 8 , 1006–1015 | 1011

function of time is similar to what is observed in the absorption spectra. In Fig. 4C, the excitation spectra at 477 nm are displayed at different ageing times. Here, the band with maxima at 407 and 427 nm increases, whereas BODIPY-like bands are missing.
Fig. 4D presents the evolution with ageing time of the fluorescence emission spectra due to excitation at 360 nm. The main emission band [with l em
(max) = 523 nm], which diminishes as a function of time, corresponds to that found in acetone with added HClO
4
(Fig. 1B), and is attributable to BODIPY. The extra, hypsochromic emission band [with l em
(max) = 478 nm] increases with time. A pseudo-isoemissive point at 505 nm is observed.
The kinetics of the chemical reaction of 4CN in acetone has been analyzed from the time-evolution of the disappearance of the BODIPY-like absorption bands (monitored at the maximum,
511 nm, and at the shoulder, 485 nm) and the appearance of the hypsochromic absorption band, monitored at 427 nm (Fig. 4A).
Plots of the absorbance A , A 1 , and log A as a function of ageing time all showed a consistent picture: there is a rapid decrease of the
BODIPY-like absorbance at 511 and 485 nm and a concomitant rapid increase of the 427 nm absorbance, followed by a leveling off at longer ageing times. The time-evolution of the absorbances
A
511nm and A
485nm could be described excellently by a biexponential, decreasing function (eqn (4); Fig. S2, ESI‡). The absorbance
A
427nm as a function of ageing time could be fitted equally well by a biexponential, increasing function (eqn (4); Fig. S2, ESI‡).
However, the estimated, apparent rate constants k
1 and k
2 for decline of A
511nm and A
485nm and rise of A
427nm differ significantly.
This demonstrates that the formation of the hypsochromic band at 427 nm is not simply due to the disappearance of the original
BODIPY absorption. Analogous results were found for the timeevolution of the fluorescence signals F
478nm and F
523nm at 478 and 423 nm, respectively (Fig. 4D): the rate constants for rise of F
478nm and decline of F
523nm do not match. These exploratory kinetic data clearly illustrate that the decomposition of 4CN is complex. An array of characterization techniques ( 1 H NMR, mass spectrometry, HPLC) will have to be utilized for the elucidation of the chemical nature of the decomposition products of 4CN in acetone. However, because only a few mg of 4CN were available, we could only use fluorescence and absorption spectroscopy to monitor the reaction progress. Hence, for the moment, the structure of the decomposition product(s) is still unknown. It must be emphasized that the compounds formed in the reaction of PM650 in DMF, reported in the literature, could not be characterized, although 1 H NMR, mass spectrometry, HPLC, absorption spectrophotometry, and fluorometry were applied.
30 y ( t ) = y
0
+ a
1 exp( k
1 t ) + a
2 exp( k
2 t ) (4)
2.5.
Spectroscopic/photophysical properties of 2CN as a function of solvent
The spectroscopic/photophysical characteristics of 2CN are analogous to those of 4CN , so they are not discussed in detail. A selection of UV–vis absorption and fluorescence emission spectra of 2CN in a few solvents is shown in Fig. S3 (ESI‡). The fluorescence excitation spectra match the absorption spectra in all solvents studied.
l abs
(max) and l em
(max) of 2CN are redshifted by 3–5 nm compared to the corresponding values of
4CN . The absorption and fluorescence emission spectra display perfect mirror symmetry, indicating that the main absorption band corresponds to the S
0
→ S
1 transition. The photophysical properties of 2CN as a function of solvent are compiled in Table 3.
Compound 2CN also has high fluorescence quantum yields U f
(0.63–0.88).
The near-invariance of the Stokes shift D ¯ with D f (Fig. S4,
ESI‡) indicates that the dipole moment of 2CN does not change noticeably between ground and excited state.
Table 4 lists the estimated regression coefficients y
0
, a , b , c , d
(see eqn (3)) and correlation coefficients ( r ) for the multiple linear n abs n em
, and D ¯ of 2CN according to eqn
(3b) and (3c) for the solvents of Table 3. Analogous to the results for 4CN , the analyses of the ¯ abs n em data according to Kamlet–
Taft (eqn (3b)) gave inferior fits than according to Catal´an.
n abs versus SP and of n abs versus SdP showed that solvent polarizability ( r = 0.741) is a more crucial factor than solvent (di)polarity ( r = 0.602) effecting
Table 3 Photophysical properties of 2CN in several solvents
# a Solvent l abs
(max)/nm l em
(max)/nm l ex
(max)/nm D ¯ /cm 1 fwhm abs
/cm 1 U f b t
1 c /ns t
2 c /ns k f
/10 8 s 1 k nr
/10 7 s 1
1 Cyclohexane
2 1,4-Dioxane
520
519
3 Dibutyl ether 518
4 Diisopropyl ether 516
5
6
7
Diethyl ether
Chloroform
Ethyl acetate
8 THF
9 1-Octanol
10
11
1-Pentanol
2-Propanol
12 Acetone + H + d
13 Methanol
14 DMF + H + d
15 MeCN + H + d
516
521
516
517
519
518
516
515
515
517
514
530
532
530
527
527
532
528
530
531
530
528
527
526
532
527
520
519
518
516
516
521
516
517
519
518
516
515
514
517
514
362
471
437
405
405
397
440
475
435
437
440
442
406
545
480
651
737
696
696
712
668
741
752
715
734
706
766
748
793
766
0.69
0.81
0.69
0.73
0.67
0.85
0.70
4.64
5.43
5.18
4.73
5.16
5.7
5.15
0.79
4.96
0.85
5.54
0.73
5.36
0.71
5.18
0.88
4.97
0.66
5.09
0.70
5.01
0.63
5.33
0.19
0.16
0.62
1.37
1.49
1.50
1.34
1.55
1.30
1.49
1.36
1.59
1.77
1.40
1.18
6.68
3.50
5.98
5.71
6.40
2.63
5.82
4.23
2.41
5.64
6.94
a The solvents are numbered according to increasing dielectric constant e .
NaOH aq as reference ( U f
= 0.92). The standard uncertainties on U samples were excited at 500 nm. The standard errors on t
1 and t
2 f b U f values were determined by excitation at 488 nm with fluorescein in 0.1 M are between 0.002 and 0.08.
c are
≤
10 ps.
d Acidified by HClO
For the time-resolved fluorescence measurements, the
4
.
1012 | Photochem. Photobiol. Sci.
, 2009, 8 , 1006–1015 This journal is © The Royal Society of Chemistry and Owner Societies 2009

the solvent-dependent shifts of the absorption band. However, the n abs according to eqn (3c) as a function of { SP, SdP } ( r = 0.930) and the relatively large estimated d
SdP value demonstrate that solvent dipolarity cannot be disregarded.
n em according to eqn (3c) with { SP, SdP } as independent variables, which is reported in Table 4, has a correlation coefficient ( r = 0.969) that is close to that for the fit with
{ SA, SB, SdP, and SP } as solvent scales ( r = 0.985). Consequently, n em n abs
, solvent acidity and basicity do not influence the position n em versus SP and n em versus SdP showed that the solvent-dependent shifts of the emission are determined more by solvent polarizability ( r = 0.943) than by solvent (di)polarity ( r = 0.277). The linearity of the plots n abs n em versus f ( n 2 ) (Fig. S5, ESI‡) corroborates that van der Waals and excitonic interactions with a polarizable solvent are largely responsible for the solvent-dependent shifts.
n abs
[( 4.4
± 0.8) ¥ 10 3 cm 1 n em
28 The slopes
[( 4.9
± 0.6) ¥
10 3 cm 1 ] as function of f ( n 2 ) are almost equal and similar to those found for 4CN . The multilinear fit of the small Stokes shifts D ¯ of
2CN according to Catal´an ( r = 0.860) was somewhat poorer than to Kamlet–Taft ( r = 0.913). Both fits were better than for 4CN , however.
Analysis of the fluorescence decay traces of 2CN , measured as a function of emission wavelength, shows that single exponentials are found for all solvents, except the alcohols. In alkanols, two decay times are necessary for an acceptable fit to the decay curves.
The amplitudes a
1 and a
2
, associated with t
1 methanol, the ratio a
2
/ a
1 and t
2
, respectively, are dependent on the emission wavelength l em
. For example, in decreases from 0.13 at 515 nm to
0.06 at 530 nm and 0.05 at 560 nm, and finally to 0.03 at the longest wavelength (575 nm). The same trend of small, decreasing a
2
/ a
1 versus l em was observed for the other alkanols (2-propanol,
1-pentanol, and 1-octanol) and opposes the trend found for
4CN .
Using the lifetime ( t ) and quantum yield ( U f
) data, values for k f and k nr were calculated according to eqn (1). The rate constants for
[ fluorescence of 2CN are nearly identical in all (nonprotic) solvents k f
= (1.5
± 0.2) ¥ 10 8 s 1 ] and are comparable to those measured for 4CN . Like for 4CN , small k nr values were found for 2CN , which are practically independent of the solvent: k nr
= (5 ± 2) ¥ 10 7 s 1 .
The color change of 4CN in the polar aprotic solvents acetone, acetonitrile, and DMF is also found for 2CN (Fig. S6, ESI‡).
The time-evolution of the absorption and fluorescence spectra in acetone is displayed in Fig. S7 (ESI‡).
We have studied the spectroscopic/photophysical properties of
4CN and 2CN , two BODIPY dyes obtained by substitution reactions of the corresponding dichloroBODIPY 1 . These BOD-
IPY analogues have similar spectroscopic and photophysical characteristics, implying that displacement of F by CN at boron has a negligible effect. Their Stokes shifts are near-independent of the solvent, indicating that their dipole moments are virtually the same in the ground and excited state. A new, generalized description of the solvent effect, proposed by Catal´an and based on four mutually independent, empirical scales (SdP: dipolarity,
SP: polarizability, SA: acidity, and SB: basicity of the medium) indicates that solvent polarizability and, to a lesser extent, solvent polarity are the crucial factors for expressing the solvatochromic shifts of the UV–vis absorption and fluorescence emission. The fluorescence quantum yields are high for both dyes. In nonprotic solvents a single fluorescence lifetime is measured, whereas two decay times are found in most alcohols. A color change of
4CN and 2CN in polar aprotic solvents is observed. The timeevolution of this reaction in acetone can be monitored by UV– vis absorption spectrophotometry and fluorescence excitation and emission spectroscopy. The available kinetic data indicate that the decomposition reaction of cyano-substituted BODIPY derivatives is complex.
4.1.
Steady-state UV–vis absorption and fluorescence spectroscopy
Dilute solutions of 4CN or 2CN in different solvents were prepared by dissolving the dry, powdered dye in the appropriate solvent so that the absorbance at 488 nm was under 0.1 at 1 cm optical path length (corresponding to a dye concentration in the m M range).
This journal is © The Royal Society of Chemistry and Owner Societies 2009 Photochem. Photobiol. Sci.
, 2009, 8 , 1006–1015 | 1013 r
0.974
0.930
0.741
0.603
0.969
0.943
0.277
0.860

Perchloric acid 70% (Riedel-deHa¨en, ACS Reagent grade) and all the spectroscopic quality solvents for the (steady-state and timeresolved) spectroscopic measurements were used without further purification. For the preparation of HClO
4 acidified solutions of acetone, acetonitrile, and DMF, 1 m L of perchloric acid 70% was added per 7.5 mL of solvent. The final concentration of HClO
4 in the solution amounts to approximately 1 mM (9.3
¥ 10 4 mol L 1 ).
UV–vis absorption spectra were recorded on a Perkin Elmer
Lambda 40 UV–vis spectrophotometer. For the corrected steadystate excitation and emission spectra, a SPEX Fluorolog was used.
Freshly prepared samples in 1 cm quartz cells were utilized to perform all UV–vis absorption and emission measurements. For the determination of the relative fluorescence quantum yields ( U f
) in solution, only dilute solutions with an absorbance below 0.1
at the excitation wavelength (488 nm) were used. Fluorescein in
0.1 M NaOH aq
( U f
= 0.92) 32 was used as fluorescent reference for the fluorescence quantum yield determinations. The U f values reported in Tables 1 and 3 are the averages of three, fully independent measurements (all at absorbances under 0.1). The standard uncertainties on the U f values were between 0.02 and
0.08 for both 4CN and 2CN . In all cases, correction for the solvent refractive index was applied. All spectra were recorded at 20
◦
C using non-degassed samples.
associated pre-exponential factors a i
. The final curve-fitting was done by global ( i.e.
simultaneous) analysis in which decay traces recorded at all four emission wavelengths were described by a
(multi)exponential decay function with linked (global) decay times t i and local pre-exponentials a i
. The quality of the fit was judged for each fluorescence decay trace separately as well as for the global fluorescence decay surface. All reported curve fittings had c 2 values below 1.1.
The authors are thankful to the Flemish Ministry of Science and Technology for a fellowship to KS-B through the Bilateral
Scientific and Technological Cooperation Program (Grant No.
BIL05/16). The ‘Instituut voor de aanmoediging van innovatie door Wetenschap en Technologie in Vlaanderen’ (IWT) is acknowledged for grant ZWAP 04/007 and for a fellowship to
LP. BELSPO (Belgium) and the University Research Fund of
K.U.Leuven are acknowledged for grants IAP VI/27 and GOA
2006/2, respectively. Support for this work was also provided by
The National Institutes of Health (HG 01745 and GM 72041) and by the Robert A. Welch Foundation. Prof. J. Catal´an (Universidad
Aut ´onoma de Madrid, Spain) is thanked for the determination of
SdP, SP, SA, and SB parameters of dibutyl ether and diisopropyl ether.
4.2.
Time-resolved fluorescence spectroscopy
For each BODIPY sample in a specific solvent, fluorescence decay traces were recorded at four emission wavelengths by the single-photon timing method.
procedures 34
33 Details of the experimental have been described elsewhere. The samples were excited at a repetition rate of 8.10 MHz with 500 nm light using the frequency-doubled output from an OPO pumped by a picosecond
Ti:Sapphire laser. Fluorescence decay histograms were collected under magic angle in 4096 channels using 1 cm quartz cells. The absorbance at the excitation wavelength was always below 0.1.
All lifetime measurements were performed at 20
◦
C using freshly prepared, non-degassed samples. Histograms of the instrument response functions (using a LUDOX scatterer) and sample decays were recorded until they typically reached 10 4 counts in the peak channel. The measured width at half height of the maximum of the instrument response function was ~ 40 ps.
The fitting parameters were determined by nonlinear leastsquares estimation by minimizing the global reduced chi-square
( c 2 g
): c g
2 = q l
i
w li
( y o li
y c li
) / v (5) where the index l sums over q experiments and the index i sums over the appropriate channel limits for each individual experiment.
y o li and y c li denote respectively the observed and calculated (fitted) values corresponding to the i th channel of the l th experiment, and w li is the corresponding statistical weight.
n represents the number of degrees of freedom for the entire multidimensional fluorescence decay surface.
The statistical criteria to judge the quality of the fit included both graphical and numerical tests and have been described previously.
35 The decays were analyzed first individually by a
(multi)exponential decay law in terms of decay times t i and their
1 A. Treibs and F.-H. Kreuzer, Justus Liebigs Ann. Chem.
, 1968, 718 ,
208–223.
2 R. P. Haugland, The Handbook. A Guide to Fluorescent Probes and
Labeling Technologies , Molecular Probes, Inc., Eugene, Oregon, USA,
10th edn, 2005.
3 R. Ziessel, G. Ulrich and A. Harriman, New J. Chem.
, 2007, 31 , 496–
501.
4 A. Loudet and K. Burgess, Chem. Rev.
, 2007, 107 , 4891–4932.
5 G. Ulrich, R. Ziessel and A. Harriman, Angew. Chem., Int. Ed.
, 2008,
47 , 1184–1201.
6 M. Baruah, W. Qin, N. Basari´c, W. M. De Borggraeve and N. Boens,
J. Org. Chem.
, 2005, 70 , 4152–4157.
7 M. Baruah, W. Qin, R. A. L. Vall´ee, D. Beljonne, T. Rohand, W. Dehaen and N. Boens, Org. Lett.
, 2005, 7 , 4377–4380.
8 T. Rohand, M. Baruah, W. Qin, N. Boens and W. Dehaen, Chem.
Commun.
, 2006, 266–268.
9 L. Li, B. Nguyen and K. Burgess, Bioorg. Med. Chem. Lett.
, 2008, 18 ,
3112–3116.
10 ¨ Tetrahedron Lett.
, 2008, 49 , 1413–1416.
11 R. Ziessel, C. Goze and G. Ulrich, Synthesis , 2007, 936–949.
12 C. Goze, G. Ulrich and R. Ziessel, Org. Lett.
, 2006, 8 , 4445–4448.
13 C. Goze, G. Ulrich and R. Ziessel, J. Org. Chem.
, 2007, 72 , 313–322.
14 G. Ulrich, C. Goze, S. Goeb, P. Retailleau and R. Ziessel, New J. Chem.
,
2006, 30 , 982–986.
15 L. Bonardi, G. Ulrich and R. Ziessel, Org. Lett.
, 2008, 10 , 2183–2186.
16 C. Tahtaoui, C. Thomas, F. Rohmer, P. Klotz, G. Duportail, Y. Mely,
D. Bonnet and M. Hibert, J. Org. Chem.
, 2007, 72 , 269–272.
17 The following represents a non-exhaustive list of BODIPY papers with spectroscopic data: ( a ) E. Vos de Wael, J. A. Pardoen, J. A. van
Koeveringe and J. Lugtenburg, Recl. Trav. Chim. Pays-Bas , 1977, 96 ,
306–309; ( b ) T. L ´opez Arbeloa, F. L ´opez Arbeloa, I. L ´opez Arbeloa,
I. Garc´ıa-Moreno, A. Costela, R. Sastre and F. Amat-Guerri, Chem.
Phys. Lett.
, 1999, 299 , 315–321; ( c ) K. Rurack, M. Kollmannsberger and J. Daub, New J. Chem.
, 2001, 25 , 289–292; ( d ) K. Rurack, M.
Kollmannsberger and J. Daub, Angew. Chem., Int. Ed.
, 2001, 40 , 385–
387; ( e ) J. Ba ˜nuelos Prieto, F. L ´opez Arbeloa, V. Mart´ınez Mart´ınez,
T. Arbeloa L ´opez, F. Amat-Guerri, M. Liras and I. L ´opez Arbeloa,
Chem. Phys. Lett.
, 2004, 385 , 29–35; ( f ) W. Qin, M. Baruah, M. Van der Auweraer, F. C. De Schryver and N. Boens, J. Phys. Chem. A ,
1014 | Photochem. Photobiol. Sci.
, 2009, 8 , 1006–1015 This journal is © The Royal Society of Chemistry and Owner Societies 2009

2005, 109 , 7371–7384; ( g ) W. Qin, M. Baruah, A. Stefan, M. Van, der
Auweraer and N. Boens, ChemPhysChem , 2005, 6 , 2343–2351; ( h ) W.
Qin, T. Rohand, M. Baruah, A. Stefan, M. Van der Auweraer, W.
Dehaen and N. Boens, Chem. Phys. Lett.
, 2006, 420 , 562–568; ( i ) W.
Qin, M. Baruah, W. M. De Borggraeve and N. Boens, J. Photochem.
Photobiol., A , 2006, 183 , 190–197; ( j ) Z. Dost, S. Atilgan and E. U.
Akkaya, Tetrahedron , 2006, 62 , 8484–8488; ( k ) Z. Li and R. Bittman,
J. Org. Chem.
, 2007, 72 , 8376–8382; ( l ) W. Qin, T. Rohand, W. Dehaen,
J. N. Clifford, K. Driessen, D. Beljonne, B. Van Averbeke, M. Van der
Auweraer and N. Boens, J. Phys. Chem. A , 2007, 111 , 8588–8597; ( m ) Z.
Ekmekci, M. D. Yilmaz and E. U. Akkaya, Org. Lett.
, 2008, 10 , 461–
464; ( n ) L. Li, J. Han, B. Nguyen and K. Burgess, J. Org. Chem.
, 2008,
73 , 1963–1970.
18 ( a ) M. Baruah, W. Qin, C. Flors, J. Hofkens, R. A. L. Vall´ee, D.
Beljonne, M. Van der Auweraer, W. M. De Borggraeve and N. Boens,
J. Phys. Chem. A , 2006, 110 , 5998–6009; ( b ) T. Rohand, J. Lycoops,
S. Smout, E. Braeken, M. Sliwa, M. Van der Auweraer, W. Dehaen,
W. M. De Borggraeve and N. Boens, Photochem. Photobiol. Sci.
, 2007,
6 , 1061–1066; ( c ) W. Qin, M. Baruah, M. Sliwa, M. Van der Auweraer,
W. M. De Borggraeve, D. Beljonne, B. Van Averbeke and N. Boens,
J. Phys. Chem. A , 2008, 112 , 6104–6114.
19 E. Lippert, Z. Naturforsch., A: Astrophys., Phys. Phys. Chem.
, 1955,
10 , 541–545.
20 N. Mataga, Y. Kaifu and M. Koizumi, Bull. Chem. Soc. Jpn.
, 1955, 28 ,
690–691; N. Mataga, Y. Kaifu and M. Koizumi, Bull. Chem. Soc. Jpn.
,
1956, 29 , 465–470.
21 M. J. Kamlet and R. W. Taft, J. Am. Chem. Soc.
, 1976, 98 , 377–
383.
22 J. Catal´an, J. Phys. Chem. B , 2009, 113 , 5951–5960.
23 J. Catal´an and H. Hopf, Eur. J. Org. Chem.
, 2004, 4694–4702.
24 J. Catal´an, in Handbook of solvents , ed. G. Wypych, ChemTec Publishing, Toronto, 2001, pp. 583–616.
25 J. Catal´an and C. D´ıaz, Liebigs Ann./Recl.
, 1997, 1941–1949.
26 J. Catal´an, C. D´ıaz, V. L ´opez, P. P´erez, J. L. G. de Paz and J. G.
Rodriguez, Liebigs Ann.
, 1996, 1785–1794.
27 Y. Marcus, Chem. Soc. Rev.
, 1993, 22 , 409–416.
28 D. Pevenage, D. Corens, W. Dehaen, M. Van, der Auweraer and F. C.
De Schryver, Bull. Soc. Chim. Belg.
, 1997, 106 , 565–572.
29 F. L ´opez Arbeloa, J. Ba ˜nuelos Prieto, V. Mart´ınez Mart´ınez, T.
Arbeloa L ´opez and I. L ´opez, Arbeloa, ChemPhysChem , 2004, 5 , 1762–
1771.
30 J. Ba ˜nuelos Prieto, T. Arbeloa, M. Liras, V. Mart´ınez Mart´ınez and F.
L ´opez Arbeloa, J. Photochem. Photobiol., A , 2006, 184 , 298–305.
31 M. Van der Auweraer, M. Van, den Zegel, F. C. De, Schryver, N. Boens and F. Willig, J. Phys. Chem.
, 1986, 90 , 1169–1175.
32 J. Olmsted, J. Phys. Chem.
, 1979, 83 , 2581–2584.
33 ( a ) D. V. O’Connor and D. Phillips, Time-correlated Single Photon
Counting , Academic Press, New York, 1984; ( b ) M. vandeVen, M.
Ameloot, B. Valeur and N. Boens, J. Fluoresc.
, 2005, 15 , 377–413;
( c ) N. Boens, W. Qin, N. Basari´c, J. Hofkens, M. Ameloot, J. Pouget,
J.-P. Lef`evre, B. Valeur, E. Gratton, M. vandeVen, N. D. Silva, Y.
Engelborghs, K. Willaert, A. Sillen, G. Rumbles, D. Phillips, A. J. W. G.
Visser, A. Van, Hoek, J. R. Lakowicz, H. Malak, I. Gryczynski, A. G.
Szabo, D. T. Krajcarski, N. Tamai and A. Miura, Anal. Chem.
, 2007,
79 , 2137–2149.
34 L. Crovetto, A. Orte, E. M. Talavera, J. M. Alvarez-Pez, M. Cotlet, J.
Thielemans, F. C. De Schryver and N. Boens, J. Phys. Chem. B , 2004,
108 , 6082–6092.
35 M. Van den Zegel, N. Boens, D. Daems and F. C. De Schryver, Chem.
Phys.
, 1986, 101 , 311–335.
This journal is © The Royal Society of Chemistry and Owner Societies 2009 Photochem. Photobiol. Sci.
, 2009, 8 , 1006–1015 | 1015