Effects of counter anions,: P -substituents, and solvents on optical and photophysical properties of 2-phenylbenzo [b] phospholium salts

Article abstract of DOI:10.1039/c7dt01839h

In this work, we investigated the effects of counter anions, P-substituents, and solvents on the optical and photophysical properties of 2-phenylbenzo[b]phospholium salts in solution. A series of 2-phenylbenzo[b]phospholium salts was prepared by P-alkylation or P-phenylation of 1,2-diphenylbenzo[b]phosphole followed by anion exchange reactions. X-ray crystallographic analyses of six benzo[b]phospholium salts showed that each phosphorus center has an onium nature with an essentially tetrahedral geometry. ¹H NMR and steady-state UV-vis absorption and fluorescence spectroscopic measurements of these phospholium salts revealed the pivotal role of counter-anion solvation. The observed results are discussed on the basis of the association-dissociation equilibrium between a contact ion pair (CIP) and a solvent-separated ion pair (SSIP) in solution. The hexafluorophosphates exist as SSIPs and emit intense fluorescence, irrespective of the P-substituents and solvents. In contrast, the iodides are present as SSIPs in methanol but exist as equilibrium mixtures of the two emitting species, SSIP and CIP, in dichloromethane. As a consequence, fluorescence intensities of the iodides varied significantly depending on the solvents, P-substituents, and solution concentrations. These findings were studied in more detail using time-resolved fluorescence spectroscopy and fluorescence titration measurements. The light-emitting properties of the 2-phenylbenzo[b]phospholium halides in the CIPs rely on heavy atom effects derived from the counter halide anions on the S₁ state of the adjacent cationic benzo[b]phosphole π-systems. The present study suggests that 2-arylbenzo[b]phospholium salts would be promising scaffolds for developing new phosphole-based ionic fluorophores that are capable of responding to external stimuli such as anionic species and solvents.

Full text of DOI:10.1039/c7dt01839h

Dalton
Transactions
View Article Online
View Journal
PAPER
Effects of counter anions, P-substituents, and
solvents on optical and photophysical properties
Cite this: DOI: 10.1039/c7dt01839h
of 2-phenylbenzo[b]phospholium salts†
a
b
b
b
Yoshinari Koyanagi, Shogo Kawaguchi, Kaori Fujii, Yoshifumi Kimura,
*
c
d
e
Takahiro Sasamori,
Norihiro Tokitoh
and Yoshihiro Matano
*
In this work, we investigated the effects of counter anions, P-substituents, and solvents on the optical and
photophysical properties of 2-phenylbenzo[b]phospholium salts in solution. A series of 2-phenylbenzo[b]
phospholium salts was prepared by P-alkylation or P-phenylation of 1,2-diphenylbenzo[b]phosphole fol-
lowed by anion exchange reactions. X-ray crystallographic analyses of six benzo[b]phospholium salts
showed that each phosphorus center has an onium nature with an essentially tetrahedral geometry.
1
H NMR and steady-state UV-vis absorption and fluorescence spectroscopic measurements of these phos-
pholium salts revealed the pivotal role of counter-anion solvation. The observed results are discussed on
the basis of the association–dissociation equilibrium between a contact ion pair (CIP) and a solvent-
separated ion pair (SSIP) in solution. The hexafluorophosphates exist as SSIPs and emit intense fluorescence,
irrespective of the P-substituents and solvents. In contrast, the iodides are present as SSIPs in methanol
but exist as equilibrium mixtures of the two emitting species, SSIP and CIP, in dichloromethane. As a con-
sequence, fluorescence intensities of the iodides varied significantly depending on the solvents,
P-substituents, and solution concentrations. These findings were studied in more detail using time-
resolved fluorescence spectroscopy and fluorescence titration measurements. The light-emitting pro-
perties of the 2-phenylbenzo[b]phospholium halides in the CIPs rely on heavy atom effects derived from
the counter halide anions on the S1 state of the adjacent cationic benzo[b]phosphole π-systems. The
present study suggests that 2-arylbenzo[b]phospholium salts would be promising scaffolds for developing
new phosphole-based ionic fluorophores that are capable of responding to external stimuli such as
anionic species and solvents.
Received 20th May 2017,
Accepted 26th June 2017
DOI: 10.1039/c7dt01839h
rsc.li/dalton
their use in organic functional materials such as organic
Introduction
1
2
light-emitting diodes, fluorescent imaging probes, organic
3
4
Recently, considerable effort has been focused on the develop- semiconductors and organic solar cells. In particular, 2-aryl-
ment of phosphole-based π-electron systems (π-systems) for benzo[b]phosphole π-systems have received considerable atten-
tion from both organic synthesis and materials chemistry
fields because of their chemically tunable, stilbene-like
5
a
π-frameworks bridged by a phosphorus atom. Several research
groups have recently highlighted the substituent effects of
2
Department of Fundamental Sciences, Graduate School of Science and Technology,
Niigata University, Nishi-ku, Niigata 950-2181, Japan
b
Department of Molecular Chemistry and Biochemistry, Faculty of Science and
-arylbenzo[b]phospholes on their fluorescence properties. For
Engineering, Doshisha University, Kyotanabe 610-0321, Japan.
E-mail: yokimura@mail.doshisha.ac.jp
example, introducing an electron-donating amino group to the
para position of 2-phenylbenzo[b]phosphole P-oxides provides
charge-transfer character to the π-systems, resulting in
enhanced ability to emit in the long-wavelength visible
region. Substituents on the fused benzene ring of benzo[b]
phosphole P-oxides also exert a significant influence on the
c
Graduate School of Natural Sciences, Nagoya City University, Mizuho-ku,
Nagoya 467-8501, Japan
d
Institute for Chemical Research, Kyoto University, Gokasho, Uji, Kyoto 611-0011,
2
,6
Japan
e
Department of Chemistry, Faculty of Science, Niigata University, Nishi-ku,
Niigata 950-2181, Japan. E-mail: matano@chem.sc.niigata-u.ac.jp
7
1
13
π–π* transition energies and fluorescence intensities. In con-
†
Electronic supplementary information (ESI) available: H/ C NMR spectra and
trast to the P-oxides, little attention has been paid to the fluo-
differential pulse voltammograms. CCDC 1545873–1545878. For ESI and crystal-
lographic data in CIF or other electronic format see DOI: 10.1039/c7dt01839h
rescence properties of 2-arylbenzo[b]phospholium salts con-
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
8
taining a phosphonio center in the π-framework. In 2015,
Yamaguchi, Fukazawa, and co-workers reported that the fluo-
rescence wavelength and intensity of P1 were sensitive to
Results and discussion
Synthesis and characterization
2
a
solvent polarity.
2-Phenylbenzo[b]phospholium iodides 3, 5, and 7 were pre-
In recent years, Baumgartner and co-workers have reported pared by P-alkylation or Pd-catalyzed P-phenylation of 1,2-
1
2
phosphole lipids consisting of dithieno[3,2-b:2′,3′-d]phospho- diphenylbenzo[b]phosphole 2, which was obtained by the
9
13
lium salts such as P2. These cationic fluorophores showed reduction of its P-oxide 1 with trichlorosilane (Scheme 1).
enhanced emission in the solid state and in solution, as a Anion exchange reactions of iodides 3, 5, and 7 with excess
6
result of aggregation, and the unique photophysical behavior NaPF gave the corresponding hexafluorophosphates 4, 6, and
of the aggregates has been rationalized by considering confor- 8, respectively, whereas the reaction of 3 with excess NaCl
mation changes of their P-substituents. Quite recently, we yielded chloride 9. The P-(2-hydroxyethyl) (P-HE) derivatives 7
independently reported optical and photophysical properties and 8 were slightly soluble in water. Compounds 3–9 were fully
1
13
31
of phosphonio-substituted fluorescent π-systems consisting of characterized using multinuclear ( H, C, and P) NMR and
1
0
α,α′-linked thiophene–phosphole–thiophene skeleton P3.
Notably, the fluorescence intensities of the iodide (X = I) (HRMS). In the P{ H} NMR spectra of 3–9 in CD
varied widely depending on the solvents, whereas those of the 10 mM), a singlet peak assignable to the cationic phosphorus
IR spectroscopy and high-resolution mass spectrometry
3
1
1
2
2
Cl (ca.
3
1
1
hexafluorophosphate (X = PF ) were consistently high irrespec- center was observed at δ 28.5 to 35.2 ppm. The P{ H} NMR
6
tive of the solvents. The observed results implied that an spectra of 4, 6, and 8 displayed an additional septet peak due
association–dissociation equilibrium between a solvent-separ- to the counter hexafluorophosphate ion at δ −144.4 to
1
ated ion pair (SSIP) and a contact ion pair (CIP) would play a −144.5 ppm ( J
= 711 to 712 Hz). The HRMS spectra of 3–9
P–F
crucial role in controlling the emission behavior of P3. It is showed an intense peak derived from the fragmented phos-
+
6
well known that both internal and external heavy atom pertur- pholium ion [M − X] (X = I, PF , Cl). The structures of 3–9 in
bations affect the rates of intersystem crossing processes the solid state and in solution will be discussed in the follow-
1
between the singlet and triplet states. Thus, the iodide ions in ing sections on the basis of X-ray crystallography and H NMR
the CIP of P3 (X = I) are likely to exert internal heavy atom spectroscopy results.
1
1
effects on the excited-state dynamics of its π-system.
However, the intrinsic effects of counter anions,
Structures
P-substituents, and solvents on the optical and photophysical Crystal structures of 3–8 were successfully characterized by
properties of 2-arylbenzo[b]phospholium salts have not been X-ray crystallography (Fig. 2 and Table 1). As shown in Fig. 2,
studied comprehensively. A clear understanding of these each phosphorus center has a distorted tetrahedral geometry
effects is still an important subject in this field. Here, we with average C–P–C bond angles of 109.2–109.4°. The iodide
report the first comprehensive study on the synthesis, struc- and hexafluorophosphate ions are spatially separated from the
tural, optical, and photophysical properties of a series of
2
-phenylbenzo[b]phospholium salts (Fig. 1).
Scheme 1 Synthesis of 3–9. Inset: The numbering of hydrogen atoms
(H , H and H ).
Fig. 1 Cationic phosphole-based π-systems.
a
b
c
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
Table 1 Selected bond parameters for 3–8
3
4
5
6
7
8
Bond lengths (Å)
a
b
c
d
e
f
1.350(2)
1.466(2)
1.406(2)
1.785(2)
1.811(2)
1.472(2)
1.349(3)
1.474(3)
1.397(3)
1.789(2)
1.801(2)
1.471(2)
1.340(6) 1.352(2)
1.477(6) 1.466(2)
1.398(6) 1.398(2)
1.795(4) 1.792(2)
1.815(4) 1.810(1)
1.473(6) 1.473(2)
1.348(2)
1.467(2)
1.409(2)
1.785(1)
1.812(2)
1.470(2)
1.349(2)
1.467(2)
1.403(2)
1.792(2)
1.807(2)
1.471(2)
C–P–C bond angles (°)
Range
94.25(8)– 97.70(9)– 94.6(2)– 94.53(8)– 94.21(8)– 94.07(7)–
116.92(9) 113.8(1) 115.6(2) 114.52(7) 114.22(7) 113.22(8)
Average
09.4
1
109.3
109.3
7.9
109.3
3.0
109.3
31.1
109.2
34.0
Torsion angles (°): a–f–g
0.8 20.5
3
The high planarity of the entire π-skeleton in 5 and 6 may be
partly attributed to steric repulsion of the 2-phenyl group
against the two P-phenyl groups. Within structure 3, the iodide
a
ion is located close to the peri-hydrogen (H in Scheme 1),
ortho-hydrogen (H ) and one of the P-methyl hydrogen atoms,
c
with inter-atomic distances of 2.96, 3.09, and 3.19 Å, respect-
1
4
ively. Within structure 5, however, the iodide ion is located
close to the other peri-hydrogen (H ) with an inter-atomic dis-
b
tance of 2.90 Å. These distances are shorter than the sum of
the van der Waals radii of hydrogen and iodine atoms (3.35 Å),
indicating that the iodide ion in 3 and 5 weakly interacts with
the above-mentioned hydrogens in the solid state.
In the structures of 7 and 8, the distances between the
phosphorus and hydroxy oxygen atoms (2.97 Å for 7, 2.95 Å for
8
(
) are shorter than the sum of their van der Waals radii
3.30 Å), reflecting intramolecular P⋯O coordination.
Furthermore, the relatively short OH⋯I distance in 7 (2.68 Å)
Fig. 2 Crystal structures of (a) 3, (b) 4, (c) 5, (d) 6, (e) 7, and (f) 8 (50% implies that there is a hydrogen-bonding interaction between
probability ellipsoids except for hydrogen atoms). Top (top) and side
the hydroxy group and the iodide ion.
To gain insight into the electrostatic interactions between
the cation and anion of 3–9 in solution, we compared the H
(
bottom) views.
1
NMR spectra of these compounds in two different solvents.
phosphorus center, indicating that 3–8 essentially have an Selected spectral features of 3–9 observed in CD Cl (ca.
2
2
onium nature. As summarized in Table 1, the C–C and C–P 10 mM) are as follows. (i) 9 vs. 3 vs. 4 (Fig. 3a–c and Fig. S1a–c
bond lengths in the phosphole ring (a–e) and the inter-ring in the ESI†): the H , H (for the numbering, see Scheme 1),
a
c
bond length (f) of these six compounds are very close to each and HMe protons of 3 and 9 are shifted largely downfield com-
other. Thus, the P-substituents do not significantly alter the pared with the corresponding protons of 4, whereas the other
bond order in the 2-phenylbenzo[b]phosphole π-system. The protons are shifted only slightly. The large downfield shifts of
inter-ring torsion angles (a–f–g in Table 1) of the P-phenyl the specific protons around the phosphorus center indicate
derivatives (1,1,2-triphenylbenzo[b]phospholium salts) 5 (7.9°) electrostatic interactions between the cationic phosphorus
and 6 (3.0°) are considerably small compared with those of the center and the counter halide anion in 3 and 9. This inter-
corresponding P-alkyl derivatives 3, 4, 7, and 8 (20.5–34.0°). action is more noticeable for 9 than for 3, implying that the
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Optical and photophysical properties
To reveal the effects of counter anions and P-substituents on
the optical properties of the 2-phenylbenzo[b]phospholium
salts, we measured steady-state UV-vis absorption and fluo-
rescence spectra of 3–9 in solution (Table 2 and Fig. 4). As
listed in Table 2, the absorption and emission maxima of 3–9
in CH Cl are λabs = 354–362 nm and λem = 446–453 nm,
2
2
respectively, whereas those in MeOH are λabs = 353–360 nm
and λem = 440–453 nm, respectively. Stokes shifts of 3–9
−
1
−1
(
5510–5820 cm ) are larger than that of P-oxide 1 (4930 cm
6
c
2 2
in CH Cl ). The influence of the counter anions on the λabs
and λem values is negligible or very small for all the 2-phenyl-
benzo[b]phosphole π-systems; i.e., the π–π* transition energies
and spectral shapes of 3, 5, and 7 are almost identical to those
of 4, 6, and 8, respectively. The absorption and emission
bands of 5 and 6 are slightly red-shifted compared with the
corresponding bands of 3 and 4; the optical HOMO–LUMO
gaps of 5 and 6 (ΔEop = 3.05–3.06 eV) are narrower than those
of 3 and 4 (ΔEop = 3.09–3.13 eV). This may reflect the differ-
ence in planarity of their π-frameworks; i.e., the 2-phenyl
group in 5/6 is less twisted against the phosphole ring than
that in 3/4 (Fig. 2a–d). The solvent dependence of the ΔEop
values of 5/6 (ΔΔEop = 0.01 eV) is small, implying that the con-
formational change of the π-framework in 5/6 is little affected
by the solvents. There are small differences in reduction poten-
tials (Ered) of 4, 6, and 8 (ΔEred = 0.05–0.08 V), indicating that
1
Fig. 3 H NMR spectra (δ = 9.0–7.2 ppm) of (a) 9, (b) 3, (c) 4, (d) 5, (e) 6, the LUMO energy levels of the present π-systems are slightly
(
2 2
f) 7, and (g) 8 in CD Cl (ca. 10 mM).
affected by the P-substituents (Table 2 and Fig. S3 in the ESI†).
To investigate the effects of counter anions, P-substituents,
and solvents on the fluorescence properties of the present
phosphorus center interacts with the chloride ion more
strongly than with the iodide ion. (ii) 5 vs. 6 (Fig. 3d and e):
2-phenylbenzo[b]phospholium salts, we measured the concen-
only the H
with the corresponding proton of 6, suggesting that the iodide
ion in 5 is present close to H in CD Cl . Two phenyl groups Table 2 Optical and electrochemical data for 3–9
b
proton of 5 is shifted largely downfield compared
b
2
2
on the phosphorus atom in 5 may sterically hinder access of
the iodide ion to the phosphorus center, as deduced from the
crystal structure of 5 (Fig. 2c). The regiospecific downfield shift
a
b
c
d
Compd
Solvent
CH Cl
λ
abs (log ε)
λ
em
ΔEop
Stokes shift
3
4
2
2
2
2
2
2
2
2
2
2
2
358 (3.99)
358 (4.03)
362 (3.88)
362 (3.90)
357 (3.92)
356 (3.94)
354 (3.91)
354 (n.d.)
354 (n.d.)
359 (n.d.)
360 (n.d.)
354 (n.d.)
353 (n.d.)
354 (n.d.)
354 (n.d.)
354 (n.d.)
447
446
453
453
450
449
447
442
442
452
453
445
444
440
445
444
3.09
3.09
3.05
3.05
3.09
3.09
3.10
3.13
3.13
3.06
3.06
3.12
3.12
3.13
3.12
3.12
5560
5510
5550
5550
5790
5820
5640
5620
5620
5730
5700
5780
5810
5520
5780
5730
e
CH
CH
CH
Cl
Cl
Cl
of the H proton also suggests the existence of anion–π or
b
5
6
anion–(H–C) interactions between the iodide ion and the posi-
tively charged benzo[b]phosphole π-framework. (iii) 7 vs. 8
e
7
8
2
CH Cl
e
CH
CH
2
Cl
Cl
(
Fig. 3f, g and Fig. S1d, e in the ESI†): two characteristic fea-
tures should be noted. First, the H and H protons of 7
appeared further downfield than the corresponding protons of
. The iodide ion in 7 may be located close to the phosphorus
center, similar to the case of 3. Second, the hydroxy proton of
is shifted about 2 ppm further downfield than that of 8. It is
likely that a hydrogen-bonding interaction between the iodide
ion and the hydroxy group in 7 is present even in CD Cl . The
spectral features of 3–9 in CD OD (Fig. S2, S16, and S18 in the
9
3
4
5
6
7
8
9
7
8
2
a
c
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
MeOH
8
7
H
H
2
O
O
2
2
2
3
a
Absorption maximum (in nm). Values in parenthesis are molar
extinction coefficients at λabs. n.d. = not determined. Fluorescence
2 2
ESI†) are quite different from those in CD Cl ; when the
b
P-substituents are the same, there is almost no difference in
c
maximum (in nm) excited at λabs
.
Optical HOMO–LUMO gap (in eV)
the spectral features among iodides (3, 5, 7), hexafluoropho- determined from the intersection of the normalized absorption and
d
−1
e
emission spectra. Δν = 1/λabs − 1/λem (in cm ). Reduction potentials
red) were measured in CH Cl with Bu NPF as the supporting elec-
sphates (4, 6, 8), and chloride (9). These data suggest that the
-phenylbenzo[b]phospholium salts exist as SSIPs in CD OD,
irrespective of the counter anions.
(E
2
2
4
6
2
3
trolyte. Ered = −1.57 (4), −1.52 (6), and −1.60 V (8) vs. ferrocene/ferroce-
+
nium couple (Fc/Fc ). All redox processes were irreversible.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
Table 3 Photophysical data for 3–8
a
b
c
d
/s−1
d
nr /s−1
Compd Solvent Abs
Φ
f
τ
f
/ns
k
r
k
e
e
7 f
7 f
3
CH
2
Cl
2
0.1
0.57
7.7 (0.96)
7.7 × 10
5.7 × 10
1
.9 (0.04)
g
7
7
4
5
6
7
CH
CH
CH
CH
2
2
2
2
Cl
2
Cl
2
Cl
2
Cl
2
0.1
0.1
0.1
0.1
0.75
0.80
0.87
0.36
8.2
8.5
9.3
9.1 × 10
9.4 × 10
9.4 × 10
5.2 × 10
3.0 × 107
2.4 × 10
7
g
7
7
1.4 × 10
e,g
e,g
7 f
7 f
7.5 (0.90)
.4 (0.10)
8.1 × 10
1
8
7
7
7
7
7
7
7
7
7
8
3
4
5
6
7
8
7
8
CH
2
Cl
2
0.1
0.89
0.65
0.65
0.81
0.76
0.68
0.68
7.8
7.9
8.3
1.1 × 10
8.2 × 10
7.8 × 10
8.9 × 10
7.8 × 10
8.1 × 10
8.1 × 10
8.2 × 10
8.1 × 10
1.4 × 107
4.4 × 10
MeOH 0.1
MeOH 0.1
MeOH 0.1
MeOH 0.1
MeOH 0.2
MeOH 0.2
h
7
4.2 × 107
9.1i
9.8
8.4
8.4
2.1 × 107
2.4 × 107
3.8 × 107
3.8 × 107
1.8 × 107
1.7 × 10
H
H
2
O
O
0.2
0.2
0.82 10.0
0.83 10.3
2
a
c
b
d
Absorbance at
λ
abs
.
ex abs
Fluorescence quantum yield. λ = λ .
Fluorescence lifetime. Radiative and nonradiative decay rate con-
r f f
stants (k and knr, respectively) determined from the Φ and τ values.
e
Values in parentheses are the relative amplitudes of each component.
Radiative and nonradiative decay rate constants determined by the
f
g
h
averaged lifetime weighted by the amplitude. Abs = 0.12. Abs = 0.17.
i
Abs = 0.15.
cases (3 and 7 in CH Cl ). The fluorescence decay profiles of 3
2
2
2 2
and 7 in CH Cl were fit with a double exponential function
with a small contribution of the faster decay component.
Fig. 4 UV-vis absorption (solid lines) and fluorescence (dotted lines)
spectra of (a) 3, 4, (b) 5, 6, and (c) 7, 8 in CH
ex) = λabs. Concentration dependence of fluorescence intensities of (d)
, 4, (e) 5, 6, and (f) 7, 8 in CH Cl . λex = λabs
2 2
Cl . Excitation wavelength
Thus, two emitting species with different τ values exist in
f
(
λ
CH
slightly decreased the k
, 6, and 8. (6) The P-methyl and P-phenyl groups have
different effects on the non-radiative pathways from the S
state; the k values of 4 are almost the same as those of 6,
whereas the knr values of 4 are about twice as much as those of
. The observed results (1)–(6) can be rationalized by consider-
2
Cl
2
. (5) Changing the solvent from CH
2 2
Cl to MeOH
3
2
2
.
r
values, but increased the knr values of
4
1
tration dependence of the fluorescence intensities of 3–8 in
CH Cl . As shown in Fig. 4d–f, iodides 3, 5, and 7 exhibited
r
2
2
clear concentration quenching of fluorescence in the range of
6
Abs = 0.1–0.5 (at λ ), whereas hexafluorophosphates 4, 6, and
ex
ing an association–dissociation equilibrium between two emit-
ting species, CIP and SSIP, in solution (eqn (1)):
8
showed only small concentration quenching. The fluo-
rescence intensities of 3, 5, and 7 are appreciably lower than
those of 4, 6, and 8 at any concentrations. The P-substituents
also have definite impacts on the quenching behavior; the
degree of fluorescence quenching is in the order 5 < 3 < 7.
These results will be discussed later.
ð1Þ
For deeper insights into the photophysical properties of
3
–8, we next measured fluorescence quantum yields (Φ
f
) and
Obviously, the fluorescence properties of 3–8 reflect the
fluorescence lifetimes (τ ) of these compounds (Table 3). The varying degrees of electrostatic interactions between the benzo
f
radiative and non-radiative decay rate constants (k
respectively) determined from the Φ and τ values are also solvent, MeOH (dielectric constant = 33), the counter anion
listed in Table 3. The results are summarized as follows. (1) should be fully solvated through hydrogen-bonding and dipole
The Φ values vary from 0.36 to 0.89 depending on the interactions, resulting predominantly in SSIP formation. This
P-substituents and solvents. (2) The Φ values of 4, 6, and 8 in explains well the first-order kinetics observed for 3–8 in
r
and knr
,
[b]phosphole cation and counter anions. In the polar protic
f
f
f
f
CH Cl are larger than those in MeOH, whereas the Φ values MeOH; the emitting species correspond to the solvent-separ-
2
2
f
of 3, 5, and 7 in CH
MeOH, the Φ and τ
close to those of 4, 6, and 8, respectively. In CH Cl , the Φ
2
Cl
2
are smaller than those in MeOH. (3) In ated benzo[b]phosphole cations in the S1 state. This is also
f
f
values of 3, 5, and 7 are the same as or inferred from the fact that the counter anions have no influ-
ence on the Φ and τ values in MeOH. In the less-polar,
2
2
f
f
f
values of the iodides are smaller than those of the hexafluoro- aprotic solvent, CH Cl (dielectric constant = 9), the CIP/SSIP
2
2
phosphates, as deduced from Fig. 4d–f. (4) The fluorescence ratio should vary depending on the counter anions as well as
decay curves of 3–8 obeyed first-order kinetics, except for two the solution concentration. In the cases of 4, 6, and 8 in
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
2 2
CH Cl , the SSIP should be the major species because the
non-nucleophilic hexafluorophosphate ion cannot access the
cationic phosphorus center. Indeed, the fluorescence decay
curves of these compounds obeyed first-order kinetics. The
relatively large Φ values of 4, 6, and 8 in CH Cl compared
f
2
2
with those in MeOH indicate the intrinsic difference in the sol-
vation effects between CH Cl and MeOH on the excited-state
of the 2-phenylbenzo[b]phospholium π-systems.
In the cases of 3, 5, and 7 in CH Cl , both the CIP and SSIP
should be present at equilibrium, and the τ and Φ values
2
2
2
2
f
f
would be influenced by interaction with the counter iodide
ion. This intrinsically nucleophilic anion exerts the heavy atom
effects on the S
resulting in acceleration of the intersystem crossing processes.
Thus, the static quenching of the S state of the cationic
1
states of the adjacent cationic π-systems,
1
π-systems would be more pronounced in CIPs than in SSIPs.
The shorter and longer lifetime components observed for the
iodides (3 and 7) in CH Cl probably correspond to CIPs and
2 2
SSIPs, respectively. This interpretation is supported by the
Fig. 5 Fluorescence titration measurements of 4 (25 μM) with Bu
X = Cl, I) in CH Cl at 25 °C. λex = 358 nm (λabs). Changes in the fluo-
rescence spectra (a, b) and fluorescence intensities (c) by the addition of
Bu NX.
4
NX
observation that the longer τ values obtained for 3 (7.7 ns)
f
(
2
2
and 7 (7.5 ns) are comparable to the τ
f
values observed for 4
Cl , increasing
(8.2 ns) and 8 (7.8 ns). For the iodides in CH
2
2
4
the concentration shifts the equilibrium position from the
SSIP to CIP, leading to fluorescence quenching, as shown in
Fig. 4d–f. Among the three iodides, the 2-HE derivative 7
exhibited the largest quenching in CH Cl , in which the OH⋯I
hydrogen-bonding interaction may contribute to shift the equi-
librium to the CIP side and to strengthen the heavy atom
effects of the iodide ion in the CIP. The photophysical behav-
ior of 5 is somewhat complicated. The fluorescence decay of 5
spectral change was observed in a fluorescence titration experi-
2
2
ment of 3 with Bu NCl (Fig. S4b in the ESI†). The λabs/λem and
4
Φ
f
values of [3 + Bu
4
NCl (excess)] are almost identical to those
= 0.14–0.15),
of [4 + Bu NCl (excess)] (λabs/λem = 355/442 nm, Φ
4
f
indicating that anion exchange reactions of 3/4 with Bu NCl
4
took place to form the same CIP of chloride 9 as the major
2 2 f
in CH Cl obeyed first-order kinetics with a τ value of 8.5 ns,
4
species. The conversion from 3 to 9 by the addition of Bu NCl
indicating that the SSIP is the predominant species at the con-
centration of the measurement (15 μM). At higher concen-
trations, however, the equilibrium is probably shifted to the
CIP side, as deduced from Fig. 4e. It should be noted that the
1
was confirmed by H NMR spectroscopy (Fig. S5 in the ESI†).
Furthermore, we measured the concentration dependence of
the halide ion on the decay curves of 4 and 9 in CH
2 2
Cl
(
Fig. S6 in the ESI†). At higher halide concentrations, the
7
−1
k
nr values observed for 3/4 (4.2–4.4 × 10
s
in MeOH) are
amplitude of the faster component (CIP) increased relative to
the slower component (SSIP). This result suggests that the
change in the ratio of these two emitting species contributes
to the decay kinetics of the fluorescent 2-phenylbenzo[b]phos-
pholium ion.‡
7
about twice as much as those observed for 5/6 (2.1–2.4 × 10
−
1
s
in MeOH). Although these particular substituent effects are
not clearly understood yet, the distorted π-framework in 3/4
Fig. 2a and b) might undergo conformational relaxation from
the S state more rapidly than does the planar π-framework in
(
1
5
/6 (Fig. 2c and d).
Reaction of 7 in alkaline solution
Fluorescence titration measurements
2
Compounds 7 and 8 are slightly soluble in H O, and the result-
ing aqueous solutions are highly fluorescent (Φ = 0.82–0.83).
Considering possible application to a water-soluble fluorescent
probe, we examined the chemical stability of 7 in aqueous
f
To evaluate the validity of the above interpretation, as well as
to compare the effects of counter halide ions on the fluo-
rescence properties of a 2-phenylbenzo[b]phospholium ion, we
performed fluorescence titration measurements on 4 in
media. Compound 7 was stable in neutral H O–MeOH (1 : 1)
2
2 3
solution but decomposed in the presence of K CO (ca. 0.08
2 2 4
CH Cl with a tetrabutylammonium halide (Bu NX; X = Cl, I)
additive. As shown in Fig. 5 and Fig. S4a in the ESI,† the fluo-
rescence of 4 was gradually quenched by increasing the ‡The spectral shapes of the normalized absorption and emission spectra of 3–8
in CH Cl did not change in the range of Abs = 0.1–0.5. In addition, the time-
resolved fluorescence spectroscopy of 3 in CH Cl did not show any transient
4
amount of Bu NX; at the high concentrations, the fluorescence
2
2
2
2
intensities were in the order (X =) Cl > I. These results probably
reflect the difference in the heavy atom effects of the respective
counter halide ions in the CIPs; the heavier iodine atom exerts
emissive species that would be assignable to an excimer or exciplex in the range
of Abs = 0.1–1. Therefore, we can rule out the other possible concentration-
quenching mechanisms caused by the chromophore–chromophore interaction
a greater heavy atom effect on the cationic π-system. A similar in the ground/excited states.
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
M) to give 2-phosphorylstilbene 10 as the main product (eqn Chemical shifts are reported in ppm as relative values vs. tetra-
1
1
13
31
(
2)). Thin-layer chromatography and H NMR spectroscopy methylsilane (for H and C) or phosphoric acid (for P).
showed that conversion from 7 to 10 proceeded cleanly and High-resolution mass spectra (HRMS) were obtained on a
quantitatively, while the isolated yield of 10 was 27% because Thermo Fisher Scientific EXACTIVE spectrometer. UV-vis
of concomitant loss during the chromatographic purification absorption and fluorescence spectra were measured at room
process. Although the instability of 7 prevents its use as a temperature on JASCO V-530 and EP-8300 spectrometers,
water-soluble fluorescent probe, quantitative conversion from respectively. All the excitation spectra were essentially identical
7
to 10 in alkaline media would give valuable information to the corresponding absorption spectra. Absolute fluo-
about the molecular design of chemically stable benzo[b] rescence quantum yields were measured on a Hamamatsu
phosphole-based ionic fluorescent materials.
Photonics Quantaurus-QY spectrometer. Redox potentials were
measured on a CH Instruments model 650E electrochemical
workstation using a glassy carbon working electrode, a plati-
+
num wire counter electrode, and an Ag/Ag [0.01 M AgNO
3
,
ð2Þ 0.1 M Bu NPF (MeCN)] reference electrode. The potentials were
4
6
+
calibrated with ferrocene/ferrocenium (Fc/Fc ). Compounds 1
and 2 were prepared according to the reported procedure.
6
c,15
Other chemicals and solvents were of reagent grade quality
and used without further purification. Thin-layer chromato-
graphy was performed with Alt. 5554 DC-Alufolien Kieselgel 60
F254 (Merck), and preparative column chromatography was
performed using silica gel (spherical, neutrality). All reactions
were performed under an argon or a nitrogen atmosphere.
Conclusions
We prepared seven kinds of 2-phenylbenzo[b]phospholium
salts to comprehensively study the effects of counter anions,
P-substituents, and solvents on the optical and photophysical Synthesis and characterization
properties of cationic 2-phenylbenzo[b]phosphole π-systems.
The target compounds were prepared by P-alkylation or
P-phenylation of 1,2-diphenylbenzo[b]phosphole followed by
anion exchange reactions and characterized by NMR spec-
troscopy, mass spectrometry, and X-ray crystallography. The
hexafluorophosphates are highly fluorescent, irrespective of
the P-substituents and solvents. In contrast, the iodides are
basically present in equilibrium between the solvent-separated
ion pair and contact ion pair, and their emission properties
vary considerably depending on the solvents, P-substituents,
solution concentrations, and amount of halide additives.
Steady-state and time-resolved fluorescence spectroscopy of
these emissive 2-phenylbenzo[b]phospholium salts revealed
that internal heavy atom effects of the counter halide ions play
3
2 2
. To a solution of 2 (0.197 g, 0.69 mmol) in CH Cl (10 mL)
was added iodomethane (0.65 mL, 10 mmol), and the reaction
mixture was stirred for 48 h at room temperature. The result-
ing mixture was evaporated under reduced pressure to obtain
a solid residue, which was recrystallized from CH Cl /hexane
to give 3 as a pale yellow solid (0.277 g, 94%): R = 0.3 (CH Cl /
MeOH = 10/1); mp 268–270 °C; H NMR (400 MHz, CD
2
2
f
2
2
1
2
Cl
2
,
1
7
8
1 mM): δ = 3.09 (d, JH–P = 14.4 Hz, 3H), 7.44–7.48 (m, 5H),
.59–7.64 (m, 1H), 7.66–7.72 (m, 2H), 7.78–7.85 (m, 3H),
.09–8.15 (m, 2H), 8.17 (d, JH–P = 36.8 Hz, 1H), 8.47–8.52
1
(
m, 1H) ppm, H NMR (400 MHz, CD OD, 10 mM): δ = 2.93
3
(
d, JH–P = 15.2 Hz, 3H), 7.42–7.48 (m, 3H), 7.55–7.60 (m, 2H),
7
8
.62–7.67 (m, 1H), 7.68–7.73 (m, 2H), 7.83–7.97 (m, 5H),
.01–8.06 (m, 1H), 8.40 (d, J = 37.2 Hz, 1H) ppm;
H–P
a pivotal role in controlling the fluorescence properties of adja- 13
1
C{ H} NMR (100 MHz, CDCl ): δ = 8.3 (d, J
= 49.4 Hz),
C–P
3
cent cationic benzo[b]phosphole π-systems in contact ion
pairs. Furthermore, the P-substituents have a definite impact
on the structures and excited-state dynamics of these cationic
π-frameworks. The present findings suggest that 2-arylbenzo[b]
phospholium salts would be promising scaffolds for the devel-
opment of new phosphole-based ionic fluorophores that are
capable of responding to external stimuli such as anionic
species and solvents.
1
15.8 (d, JC–P = 82.7 Hz), 123.8 (d, JC–P = 94.1 Hz), 126.6 (d,
JC–P = 6.8 Hz), 127.0 (d, JC–P = 9.1, Hz), 129.5 (d, JC–P = 11.4 Hz),
1
1
1
29.6 (d, JC–P = 78.9 Hz), 130.0 (s), 130.7 (s), 130.8 (d, JC–P =
3.7 Hz), 131.2 (d, JC–P = 12.2 Hz), 133.0 (d, JC–P = 11.4 Hz),
33.5 (d, JC–P = 12.1 Hz), 135.7 (d, JC–P = 3.0 Hz), 135.9 (d,
JC–P = 2.2 Hz), 143.2 (d, J
= 25.0 Hz), 144.0 (d, JC–P = 16.7 Hz)
C–P
31
1
2 2
ppm; P{ H} NMR (162 MHz, CD Cl , 10 mM): δ = 31.0 ppm;
31
1
P{ H} NMR (162 MHz, CD OD, 10 mM): δ = 31.4 ppm; HRMS
3
+
(
ESI): m/z = 301.1132 ([M – I] , calcd 301.1141).§
4
. A mixture of 3 (49 mg, 0.11 mmol), NaPF
6
(0.387 g,
2
.3 mmol), CH Cl (2.5 mL), and MeCN (1 mL) was stirred at
2
2
Experimental
General
room temperature. After 1 h, the mixture was washed with
All melting points were recorded on Yazawa micro melting
point apparatus and are uncorrected. H, C and P NMR
spectra were recorded on a Varian 400 MHz or 700 MHz
§
Although we could not obtain sufficient elemental analysis results because of
1
13
31
contamination of aerobic moisture and/or recrystallization solvents,
the purity of the obtained compounds could be confirmed by spectroscopic
spectrometer or a Bruker NanoBay 400 MHz spectrometer. analyses.
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
water (2.5 mL). NaPF
6
(0.387 g, 2.3 mmol) was then added, (m, 5H), 7.65–7.75 (m, 5H), 7.77–7.93 (m, 9H), 8.10 (d, J =
1
and the resulting mixture was stirred for 2 h. After addition 0.4 Hz, JH–P = 38.0 Hz, 1H) ppm; H NMR (400 MHz, CD OD,
3
of CH
3 × 10 mL), dried over Na
solid residue, which was then chromatographed on silica gel 1H), 8.49 (d, JH–P = 38.4 Hz, 1H) ppm; C{ H} NMR (100 MHz,
using CH Cl /MeOH (20/1) as an eluent. Compound 4 was iso- CDCl ): δ = 114.8 (d, JC–P = 83.5 Hz), 123.3 (d, JC–P = 95.6 Hz),
lated as a colorless solid (41 mg, 81%) by recrystallization from 126.9 (d, JC–P = 6.9 Hz), 128.2 (d, JC–P = 9.1, Hz), 129.3 (d, JC–P
CH Cl /hexane: R = 0.4 (CH Cl /MeOH = 10/1); mp 79–81 °C; 79.7 Hz), 129.9 (s), 130.1 (d, JC–P = 10.6 Hz), 130.7 (s), 131.2 (d,
2
Cl
2
(10 mL), the mixture was washed with water 10 mM): δ = 7.41–7.48 (m, 3H), 7.56–7.61 (m, 2H), 7.66–7.72
(
2
SO , and evaporated to obtain a (m, 1H), 7.73–7.78 (m, 4H), 7.90–7.99 (m, 8H), 8.09–8.14 (m,
4
1
3
1
2
2
3
=
2
2
f
2
2
−
1 1
IR (KBr): νmax = 837 (P–F) cm ; H NMR (400 MHz, CD
2
Cl
0 mM): δ = 2.67 (d, JH–P = 14.0 Hz, 3H), 7.40–7.48 (m, 5H), 11.4 Hz), 133.5 (d, JC–P = 11.3 Hz), 136.3 (d, JC–P = 3.1 Hz),
.60–7.66 (m, 1H), 7.68–7.73 (m, 2H), 7.79–7.91 (m, 6H), 8.10 136.5 (d, JC–P = 2.3 Hz), 144.0 (d, J = 25.0 Hz), 146.1 (d,
2
,
JC–P = 13.7 Hz), 131.3 (d, JC–P = 10.7 Hz), 131.5 (d, JC–P =
1
7
(
9
7
7
1
5
1
1
C–P
1
31
1
d, JH–P = 37.6 Hz, 1H) ppm, H NMR (400 MHz, CD
3
OD,
mM): δ = 2.91 (d, JH–P = 14.8 Hz, 3H), 7.41–7.47 (m, 3H), δ = −144.5 (septet, JP–F = 711 Hz), 28.6 ppm; P{ H} NMR
.55–7.60 (m, 2H), 7.61–7.67 (m, 1H), 7.68–7.73 (m, 2H), (162 MHz, CD OD, 10 mM): δ = −144.6 (septet, JP–F = 708 Hz),
2 2
JC–P = 16.7 Hz) ppm; P{ H} NMR (162 MHz, CD Cl , 10 mM):
3
1
1
3
+
6
.83–7.96 (m, 5H), 7.99–8.04 (m, 1H), 8.39 (d, JH–P = 37.6 Hz, 29.0 ppm; HRMS (ESI): m/z = 363.1290 ([M – PF ] , calcd
1
3
1
H) ppm; C{ H} NMR (100 MHz, CDCl
0.8 Hz), 115.6 (d, JC–P = 81.9 Hz), 123.4 (d, JC–P = 94.1 Hz),
26.5 (d, JC–P = 6.8 Hz), 127.4 (d, JC–P = 9.1, Hz), 129.4 (d, JC–P
3
): δ = 5.2 (d, JC–P
=
363.1297).
7. To a solution of 2 (0.226 g, 0.79 mmol) in toluene
(10 mL) was added 2-iodoethanol (0.20 mL, 2.6 mmol), and
=
1.4 Hz), 129.6 (d, JC–P = 78.9 Hz), 129.9 (s), 130.6 (s), 131.0 (d, the reaction mixture was refluxed for 15 h. The resulting
JC–P = 13.6 Hz), 131.0 (d, JC–P = 11.4 Hz), 131.1 (d, JC–P = 11.3 mixture was evaporated under reduced pressure to obtain a
Hz), 131.9 (d, JC–P = 12.1 Hz), 135.9 (d, JC–P = 3.1 Hz), 136.2 (d, solid residue, which was recrystallized from CH Cl /hexane to
C–P = 1.5 Hz), 143.7 (d, JC–P = 25.0 Hz), 144.5 (d, JC–P = 16.7 Hz) give 7 as a pale yellow solid (0.329 g, 91%): R = 0.2 (CH Cl
ppm; P{ H} NMR (162 MHz, CD Cl , 10 mM): δ = −144.4 MeOH = 10/1); mp 174–176 °C; H NMR (400 MHz, CD Cl ,
2
2
J
f
2
2
/
3
1
1
1
2
2
2
2
3
1
1
(
septet, JP–F = 712 Hz), 29.8 ppm; P{ H} NMR (162 MHz, 10 mM): δ = 2.97–3.06 (m, 1H), 3.83–3.92 (m, 1H), 4.10–4.20
CD OD, 10 mM): δ = −144.6 (septet, JP–F = 708 Hz), 31.4 ppm; (m, 1H), 4.25–4.37 (m, 1H), 4.83 (s, 1H), 7.40–7.46 (m, 5H),
HRMS (ESI): m/z = 301.1133 ([M − PF ] , calcd 301.1141).
3
+
7.60–7.67 (m, 3H), 7.75–7.83 (m, 3H), 8.01 (d, JH–P = 35.6 Hz,
6
1
5
. To a solution of 2 (0.275 g, 0.96 mmol) in xylene (5 mL) 1H), 8.02–8.08 (m, 1H), 8.27–8.31 (m, 1H) ppm; H NMR
was added iodobenzene (220 μL, 2.0 mmol) and Pd (dba) (400 MHz, CD OD): δ = 3.36–3.46 (m, 1H), 3.84–4.06 (m, 3H),
47 mg, 0.051 mmol), and the reaction mixture was refluxed 7.42–7.47 (m, 3H), 7.55–7.60 (m, 2H), 7.62–7.70 (m, 3H),
for 3 h. The resulting suspension was cooled to room tempera- 7.79–7.90 (m, 3H), 7.91–7.98 (m, 2H), 8.07–8.12 (m, 1H), 8.32
2
3
3
(
1
3
1
ture, and the precipitate was filtered, washed with hexane, and (d, J = 0.6 Hz, JH–P = 36.6 Hz, 1H) ppm; C{ H} NMR
dissolved in CH Cl . The resulting solution was evaporated (100 MHz, CDCl ): δ = 25.4 (d, J = 46.2 Hz), 55.6 (d, JC–P
=
2
2
3
C–P
under reduced pressure to obtain a solid residue, which was 6.9 Hz), 116.6 (d, JC–P = 81.2 Hz), 122.3 (d, JC–P = 92.6 Hz),
recrystallized from CH Cl /hexane to give 5 as a yellow solid 126.6 (d, JC–P = 6.1 Hz), 126.8 (d, JC–P = 9.1, Hz), 129.7 (d, JC–P =
2
2
(
0.381 g, 81%): R = 0.3 (CH Cl /MeOH = 10/1); mp 230–232 °C; 10.6 Hz), 129.9 (s), 130.7 (s), 130.7 (d, JC–P = 12.9 Hz), 131.0 (d,
f
2
2
1
H NMR (400 MHz, CD
2
Cl
2
, 10 mM): δ = 7.41–7.54 (m, 5H),
JC–P = 12.1 Hz), 131.2 (d, JC–P = 76.6 Hz), 132.9 (d, JC–P = 11.4
7
8
.66–7.76 (m, 5H), 7.81–7.93 (m, 7H), 7.94–7.99 (m, 1H), Hz), 133.9 (d, JC–P = 11.4 Hz), 135.4 (d, JC–P = 3.1 Hz), 135.8 (d,
1
.00–8.03 (m, 1H), 8.46 (d, JH–P = 38.4 Hz, 1H) ppm; H NMR JC–P = 2.3 Hz), 142.8 (d, J
= 16.7 Hz), 144.1 (d, JC–P = 23.6 Hz)
C–P
3
1
1
(400 MHz, CD
3
OD, 10 mM): δ = 7.41–7.48 (m, 3H), 7.56–7.61 ppm; P{ H} NMR (162 MHz, CD
2
2
Cl , 10 mM): δ = 35.2 ppm;
3
1
1
(m, 2H), 7.67–7.72 (m, 1H), 7.73–7.79 (m, 4H), 7.90–8.00 (m,
P{ H} NMR (162 MHz, CD
3
OD): δ = 35.2 ppm; HRMS (ESI):
+
8
H), 8.10–8.15 (m, 1H), 8.51 (d, J
= 38.4 Hz, 1H) ppm; m/z = 331.1242 ([M – I] , calcd 331.1246).
H–P
1
3
1
C{ H} NMR (100 MHz, CDCl
3
): δ = 114.7 (d, JC–P = 83.4 Hz),
6
8. A mixture of 7 (0.101 g, 0.22 mmol), NaPF (0.740 g,
1
23.2 (d, JC–P = 96.3 Hz), 126.9 (d, JC–P = 6.9 Hz), 128.5 4.4 mmol), and MeCN (5 mL) was stirred at room temperature.
(
(
(
d, JC–P = 78.9, Hz), 128.6 (d, JC–P = 9.1 Hz), 129.7 (s), 130.0 After 1 h, the reaction mixture was concentrated under
d, JC–P = 11.4 Hz), 130.5 (s), 131.1 (d, JC–P = 13.6 Hz), 131.3 reduced pressure to obtain a solid residue. After addition of
d, JC–P = 11.4 Hz), 131.4 (d, JC–P = 10.7 Hz), 133.5 (d, JC–P
=
2 2
CH Cl , the precipitate was filtered, and the filtrate was evapor-
1
1.4 Hz), 136.2 (d, JC–P = 3.1 Hz), 136.3 (d, JC–P = 2.3 Hz), 144.0 ated under reduced pressure. MeCN (5 mL) and NaPF6
3
1
1
(
(
(
d, JC–P = 25.0 Hz), 146.6 (d, JC–P = 16.7 Hz) ppm; P{ H} NMR (0.742 g, 2.3 mmol) were then added to the residue, and the
162 MHz, CD
162 MHz, CD OD, 10 mM): δ = 28.9 ppm; HRMS (ESI): m/z = filtration and concentration were repeated to obtain an oily
3
1
1
2 2
Cl , 10 mM): δ = 28.5 ppm; P{ H} NMR resulting mixture was stirred for 1 h at room temperature. The
3
+
3
63.1290 ([M – I] , calcd 363.1297).
. According to a similar procedure described for the syn- CH
thesis of 4, 6 was prepared from 5. Compound 6 was isolated isolated as a colorless solid (53 mg, 50%) by recrystallization
as a pale yellow solid in 71% yield based on 5: R = 0.5 from CH Cl /hexane: R = 0.3 (CH Cl /MeOH = 10/1); mp
/MeOH = 10/1); mp 218–220 °C; IR (KBr): νmax = 836 135–137 °C; IR (KBr): νmax = 844 (P–F) cm−
residue, which was chromatographed on silica gel using
6
2
2
Cl /MeOH (20/1 to 4/1) as eluents. Compound 8 was
f
2
2
f
2
2
1
1
(CH
2
Cl
2
;
H NMR
−
1 1
(P–F) cm ; H NMR (400 MHz, CD Cl , 10 mM): δ = 7.41–7.50 (400 MHz, CD Cl , 10 mM): δ = 2.73 (dd, J = 5.0, 5.0 Hz, 1H),
2
2
2
2
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
2
4
7
1
3
7
8
.98–3.07 (m, 1H), 3.60–3.70 (m, 1H), 3.91–4.03 (m, 1H), (m, 1H), 7.45–7.59 (m, 4H), 7.66 (d, J = 16.0 Hz, 1H), 7.70–7.79
3
1
1
.11–4.24 (m, 1H), 7.40–7.48 (m, 5H), 7.60–7.69 (m, 3H), (m, 4H) ppm; P{ H} NMR (162 MHz, CDCl ): δ = 37.2 ppm;
3
+
.77–7.89 (m, 5H), 8.01 (d, JH–P = 36.4 Hz, 1H), 8.02–8.06 (m, HRMS (ESI): m/z = 349.1340 ([M + H] , calcd 349.1352); UV-vis
1
H) ppm; H NMR (400 MHz, CD
3
OD): δ = 3.34–3.44 (m, 1H), (CH
2
2
Cl ): λmax = 309 nm (log ε = 4.22).
.82–4.05 (m, 3H), 7.40–7.47 (m, 3H), 7.55–7.60 (m, 2H),
.61–7.70 (m, 3H), 7.79–7.89 (m, 3H), 7.90–7.97 (m, 2H),
X-ray crystallography
1
3
1
.06–8.11 (m, 1H), 8.31 (d, JH–P = 36.8 Hz, 1H) ppm; C{ H} Single crystals used for X-ray crystallography were grown from
= 47.0 Hz), 56.3 (d, CH Cl –hexane (for 3), CH Cl –Et O (for 4 and 5), CHCl –
NMR (100 MHz, CDCl ): δ = 24.3 (d, J
3
C–P
2
2
2
2
2
3
2 2
JC–P = 6.9 Hz), 116.3 (d, JC–P = 81.2 Hz), 122.1 (d, JC–P = 93.3 toluene (for 6) or CH Cl –toluene (for 7 and 8). All measure-
Hz), 126.5 (d, JC–P = 6.8 Hz), 127.0 (d, JC–P = 8.3, Hz), 129.7 (d, ments were made on a Rigaku R-AXIS RAPID diffractometer
JC–P = 10.6 Hz), 130.0 (s), 130.7 (d, JC–P = 76.6 Hz), 130.7 (s), (for 3) or a Rigaku Saturn RIGAKU Saturn70 CCD system with
1
J
30.8 (d, JC–P = 12.9 Hz), 131.1 (d, JC–P = 12.1 Hz), 132.2 (d, a VariMax Mo Optic diffractometer (for 4–8) using multi-layer
C–P = 11.4 Hz), 132.8 (d, JC–P = 11.4 Hz), 135.6 (d, JC–P mirror monochromated Mo-Kα radiation (0.71070 Å) at 103 K
= 16.7 Hz), 144.2 (for 3–6 and 8) or 108 K (for 7). The data were corrected for
=
3
.0 Hz), 136.0 (d, JC–P = 2.3 Hz), 143.4 (d, J
C–P
3
1
1
(d, JC–P = 24.3 Hz) ppm; P{ H} NMR (162 MHz, CD
2
Cl
2
,
Lorentz and polarization effects. The structures were solved by
3
1
1
16
17
1
0 mM): δ = −144.5 (septet, JP–F = 712 Hz), 34.3 ppm; P{ H} using direct methods (SIR2011 or SIR2014 ) and refined
2
NMR (162 MHz, CD OD): δ = −144.6 (septet, J
3
3
= 708 Hz), by full-matrix least squares techniques against F using
3
P–F
+
18
19
5.2 ppm; HRMS (ESI): m/z = 331.1241 ([M – PF
6
] , calcd SHELXL2013 or SHELXL2014/7. The non-hydrogen atoms
were refined anisotropically and hydrogen atoms were refined
31.1246).
9
. A mixture of 3 (0.102 g, 0.238 mmol), brine, CH Cl , using the rigid model.
2 2
MeOH and MeCN was vigorously shaken in a separatory
funnel at room temperature. The organic phase was separated, 0.06 mm, monoclinic, P2
dried over Na SO , and evaporated to obtain a solid residue, Å, c = 10.65593(19) Å, β = 102.343(7)°, V = 1836.18(8) Å , Z = 4,
3 (CCDC 1545873†): C21
H18IP, MW = 428.25, 0.14 × 0.07 ×
1
/n, a = 9.19717(17) Å, b = 19.1790(4)
3
2
4
calcd = 1.549 g cm− , μ = 18.29 cm , collected 30 889,
3
−1
which was chromatographed on silica gel using CH
10/1 to 5/1) as eluents. Compound 9 was isolated as a color- independent 4200, parameters 209, R
less solid (77.7 mg, 97%) by recrystallization from CH Cl / 0.0211 (I > 2.0σ(I)), GOF = 1.145.
2
Cl
2
/MeOH
ρ
(
w
= 0.0523 (all data), R =
1
2
2
hexane: R
H NMR (400 MHz, CD
f
= 0.12 (CH
2
Cl
Cl
4.4 Hz), 7.43–7.49 (m, 5H), 7.58–7.69 (m, 3H), 7.71–7.80 (m, Å, c = 20.4281(3) Å, β = 94.8115(9)°, V = 1958.89(5) Å , Z = 4,
2
/MeOH = 10/1); mp 242–244 °C;
4 (CCDC 1545874†): C21
18 6 2
H F P , MW = 446.29, 0.11 × 0.06 ×
1
2
2
, 10 mM): δ = 3.28 (d, 3H, JH–P
=
0.05 mm, monoclinic, P2 /c, a = 11.5651(2) Å, b = 8.32080(10)
1
3
1
3
calcd = 1.513 g cm− , μ = 2.82 cm , collected 32 989, inde-
OD): δ = 2.93 (d, 3H, pendent 3843, parameters 334, R = 0.01003 (all data), R
JH–P = 15.2 Hz), 7.42–7.48 (m, 3H), 7.55–7.60 (m, 2H), 7.63–7.67 0.0410 (I > 2.0σ(I)), GOF = 1.085.
3
−1
H), 8.12 (d, 1H, JH–P = 36.0 Hz), 8.23–8.30 (m, 2H), 8.84–8.89
ρ
1
(m, 1H) ppm; H NMR (400 MHz, CD
3
w
1
=
(
1
m, 1H), 7.68–7.73 (m, 2H), 7.83–7.97 (m, 5H), 8.01–8.05 (m,
5 (CCDC 1545875†): C27
H), 8.40 (d, 1H, JH–P = 37.2 Hz) ppm; C{ H} NMR (100 MHz, 0.03 mm, monoclinic, P2 /n, a = 9.8866(2) Å, b = 21.0534(5) Å,
= 48.8 Hz), 116.5 (d, JC–P = 82.7 Hz), c = 12.0079(3) Å, β = 94.878(2)°, V = 2490.35(10) Å , Z = 4,
2
H22Cl IP, MW = 575.22, 0.08 × 0.07 ×
1
3
1
1
3
CDCl ): δ = 7.4 (d, J
3
C–P
calcd = 1.534 g cm− , μ = 15.77 cm , collected 48 571, inde-
3
−1
1
J
24.2 (d, JC–P = 94.0 Hz), 126.5 (d, JC–P = 6.8 Hz), 126.6 (d,
C–P = 9.1 Hz), 129.7 (d, JC–P = 78.9 Hz), 129.8 (d, JC–P = 11.4 pendent 5168, parameters 280, R
Hz), 130.0 (s), 130.6 (s), 130.7 (d, JC–P = 13.7 Hz), 131.3 (d, 0.0377 (I > 2.0σ(I)), GOF = 1.043.
C–P = 11.3 Hz), 132.9 (d, JC–P = 12.1 Hz), 133.8 (d, JC–P = 12.2 6 (CCDC 1545876†): C26
Hz), 135.4 (d, JC–P = 3.1 Hz), 135.6 (d, JC–P = 2.3 Hz), 143.0 (d, 0.15 mm, monoclinic, P2 /c, a = 9.1887(2) Å, b = 19.4913(3) Å,
ρ
w
= 0.0908 (all data), R
1
=
J
20 6 2
H F P , MW = 508.36, 0.27 × 0.20 ×
1
3
1
1
3
JC–P = 25.0 Hz), 143.9 (d, JC–P = 16.7 Hz) ppm; P{ H} NMR c = 13.1164(2) Å, β = 101.3225(14)°, V = 2303.42(7) Å , Z = 4,
3
1
1
−3
−1
(
(
(
162 MHz, CD
162 MHz, CD OD): δ = 31.4 ppm; HRMS (ESI): m/z = 301.1143 pendent 4396, parameters 307, R
[M – Cl] , calcd 301.1141). 0.0400 (I > 2.0σ(I)), GOF = 1.060.
0. A mixture of 7 (21 mg, 0.046 mmol), K 7 (CCDC 1545877†): C22 20IOP, MW = 458.25, 0.40 × 0.30 ×
.15 mmol), water (1 mL), and MeOH (1 mL) was stirred for 0.10 mm, monoclinic, P2 /n, a = 9.41250(10) Å, b = 22.5530(4)
h at room temperature. After addition of CH Cl (10 mL) and Å, c = 10.2457(2) Å, β = 116.7384(11)°, V = 1942.39(6) Å , Z = 4,
2
Cl
2
, 10 mM): δ = 31.7 ppm; P{ H} NMR
ρ
calcd = 1.466 g cm , μ = 2.50 cm , collected 35 987, inde-
3
w
= 0.0963 (all data), R
1
=
+
1
2
CO
3
(21 mg,
H
0
2
1
3
2
2
calcd = 1.567 g cm− , μ = 17.38 cm , collected 50 099, inde-
3
−1
water (10 mL), the aqueous layer was separated and extracted
with CH Cl (3 × 10 mL). The combined organic extracts were pendent 4015, parameters 227, R
washed with water, dried over Na SO , and evaporated to 0.0168 (I > 2.0σ(I)), GOF = 1.103.
ρ
2
2
w
= 0.0430 (all data), R
1
=
2
4
obtain a solid residue. Compound 10 was isolated as a color-
less solid (4.4 mg, 27%) by recrystallization from CH Cl /
2 2
8 (CCDC 1545878†): C22
0.07 mm, triclinic, P ˉ1 , a = 9.0781(4) Å, b = 10.6021(5) Å, c =
= 1171 11.4141(4) Å, α = 76.150(3)°, β = 87.971(3)°, γ = 86.907(3)°,
20 6 2
H F OP , MW = 476.32, 0.35 × 0.28 ×
hexane: R = 0.5 (CH Cl /MeOH = 10/1); IR (KBr): ν
f
2
2
max
PvO) cm−
1
1
3
−3
−1
(
3
; H NMR (400 MHz, CDCl ): δ = 2.49–2.58 (m, V = 1064.79(8) Å , Z = 2, ρcalcd = 1.486 g cm , μ = 2.68 cm ,
1
w
H), 2.69–2.79 (m, 1H), 3.84 (dd, J = 6.2, 6.2 Hz, 1H), 3.97–4.07 collected 17 292, independent 4068, parameters 281, R =
(
m, 2H), 6.90 (d, J = 16.0 Hz, 1H), 7.23–7.35 (m, 5H), 7.37–7.42 0.0923 (all data), R = 0.0370 (I > 2.0σ(I)), GOF = 1.073.
1
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.

View Article Online
Paper
Dalton Transactions
Fluorescence lifetime measurements
H. Kaji, Y. Matano and H. Imahori, Chem. – Eur. J., 2009,
1
5, 10000–10004; (f) Y. Matano, A. Saito, T. Fukushima,
The fluorescence lifetimes of 3, 5, 7 and 8 were measured in
CH Cl , MeOH and H O at room temperature using a TCSPC
2 2 2
Y. Tokudome, F. Suzuki, D. Sakamaki, H. Kaji, A. Ito,
K. Tanaka and H. Imahori, Angew. Chem., Int. Ed., 2011, 50,
system. The excitation wavelength was 374 nm (pulsed diode
laser). The fluorescence lifetimes of 4 and 6 were measured
using a streak camera. The excitation wavelength was 360 nm.
The excitation pulse was generated by making the second
harmonic pulse of the output of an optical parametric amplifier
8
016–8020; (g) Y. Matano, H. Ohkubo, Y. Honsho, A. Saito,
S. Seki and H. Imahori, Org. Lett., 2013, 15, 932–935;
h) T. Higashino, T. Yamada, T. Sakurai, S. Seki and
(
H. Imahori, Angew. Chem., Int. Ed., 2016, 55, 12311–12315.
(a) H. Tsuji, K. Sato, Y. Sato and E. Nakamura, Chem. –
Asian J., 2010, 5, 1294–1297; (b) Y. Matano, A. Saito,
Y. Suzuki, T. Miyajima, S. Akiyama, S. Otsubo,
E. Nakamoto, S. Aramaki and H. Imahori, Chem. – Asian J.,
4
(720 nm) operated by an amplified Ti:sapphire laser system.
Acknowledgements
2012, 7, 2305–2312; (c) Y. Matano, Y. Hayashi, H. Nakano
and H. Imahori, Heteroat. Chem., 2014, 25, 533–547;
(d) Y. Matano, H. Ohkubo, T. Miyata, Y. Watanabe, Y. Hayashi,
T. Umeyama and H. Imahori, Eur. J. Inorg. Chem., 2014,
This work was partially supported by JSPS KAKENHI (Grant
number 25288020 and 15 K13762) and the Naito foundation.
X-ray diffraction analysis of 3 was conducted in NAIST, sup-
ported by the Ministry of Education, Culture, Sports, Science
and Technology (MEXT), Japan (Nanotechnology Platform
Program, Synthesis of Molecules and Materials). We thank
Prof. Hiroko Yamada and Prof. Naoki Aratani (NAIST) for their
kind assistance.
1620–1624; (e) Y. Matano, Chem. Rec., 2015, 15, 636–650.
5 For reviews and accounts, see: (a) R. A. Aitkin, Sci. Synth.,
2001, 10, 789–808; (b) Y. Matano and H. Imahori, Org.
Biomol. Chem., 2009, 7, 1258–1271; (c) A. Fukazawa and
S. Yamaguchi, Chem. – Asian J., 2009, 4, 1386–1400;
(
5
d) B. Wu and N. Yoshikai, Org. Biomol. Chem., 2016, 14,
402–5416.
6
(a) A. Fukazawa, Y. Ichihashi, Y. Kosaka and S. Yamaguchi,
Chem. – Asian J., 2009, 4, 1729–1740; (b) Y. Unoh,
K. Hirano, T. Satoh and M. Miura, Angew. Chem., Int. Ed.,
2013, 52, 12975–12979; (c) Y. Matano, Y. Motegi,
S. Kawatsu and Y. Kimura, J. Org. Chem., 2015, 80, 5944;
(d) E. Yamaguchi, A. Fukazawa, Y. Kosaka, D. Yokogawa,
S. Irle and S. Yamaguchi, Bull. Chem. Soc. Jpn., 2015, 88,
1545–1552; (e) Z. Zhuang, F. Bu, W. Luo, H. Peng, S. Chen,
R. Hu, A. Qin, Z. Zhao and B. Z. Tang, Chem. – Eur. J., 2017,
23, 2542–2547.
Notes and references
1
For example, see: (a) C. Fave, T.-Y. Cho, M. Hissler,
C.-W. Chen, T.-Y. Luh, C.-C. Wu and R. Réau, J. Am. Chem.
Soc., 2003, 125, 9254–9255; (b) T. Baumgartner and
R. Réau, Chem. Rev., 2006, 106, 4681–4727 (correction:
Chem. Rev., 2007, 107, 303); (c) M. Hissler, C. Lescop and
R. Réau, C. R. Chimie, 2008, 11, 628–640; (d) H. Tsuji,
K. Sato, Y. Sato and E. Nakamura, J. Mater. Chem., 2009, 19,
3
364–3366; (e) M. G. Hobbs and T. Baumgartner,
Eur. J. Inorg. Chem., 2007, 3611–3628; (f) D. Joly,
D. Tondelier, V. Deborde, B. Geffroy, M. Hissler and
R. Réau, New J. Chem., 2010, 34, 1603–1611; (g) M. P. Duffy,
P.-A. Bouit, B. Geffroy, D. Tondelier and M. Hissler,
Phosphorus, Sulfur Silicon Relat. Elem., 2015, 190, 845–853
and references therein.
(a) E. Yamaguchi, C. Wang, A. Fukazawa, M. Taki, Y. Sato,
T. Sasaki, M. Ueda, N. Sasaki, T. Higashiyama and
S. Yamaguchi, Angew. Chem., Int. Ed., 2015, 54, 4539–4543;
7 B. Wu, M. Santra and N. Yoshikai, Angew. Chem., Int. Ed.,
2014, 53, 7543–7546.
8 (a) A. Fukazawa, H. Yamada and S. Yamaguchi, Angew.
Chem., Int. Ed., 2008, 47, 5582–5585; (b) A. Fukazawa,
H. Yamada, Y. Sasaki, S. Akiyama and S. Yamaguchi, Chem.
– Asian J., 2010, 5, 466–469; (c) A. Fukazawa, E. Yamaguchi,
E. Ito, H. Yamada, J. Wang, S. Irle and S. Yamaguchi,
Organometallics, 2011, 30, 3780–3879; (d) S. Arndt,
M. M. Hansmann, P. Motloch, M. Rudolph, F. Rominger
and A. S. K. Hashm, Chem. – Eur. J., 2017, 23, 2542–2547.
9 (a) Y. Ren, W. H. Kan, M. A. Henderson, P. G. Bomben,
C. P. Berlinguette, V. Thangadurai and T. Baumgartner,
J. Am. Chem. Soc., 2011, 133, 17014–17026; (b) Y. Ren and
T. Baumgartner, J. Am. Chem. Soc., 2011, 133, 1328–1340;
(c) Y. Ren, W. H. Kan, V. Thangadurai and T. Baumgartner,
Angew. Chem., Int. Ed., 2012, 51, 3964–3968; (d) Y. Ren and
T. Baumgartner, Inorg. Chem., 2012, 51, 2669–2678; (e) Y. Ren
and T. Baumgartner, Dalton Trans., 2012, 41, 7792–7800;
(f) Y. Ren, F. Biegger and T. Baumgartner, J. Phys. Chem. C,
2013, 117, 4748–4758; (g) Z. Wang and T. Baumgartner,
Chem. Rec., 2015, 15, 199–217, and references therein.
2
(
b) C. Wang, A. Fukazawa, M. Taki, Y. Sato, T. Higashiyama
and S. Yamaguchi, Angew. Chem., Int. Ed., 2015, 54, 15213–
5217; (c) M. Taki, H. Ogasawara, H. Osaki, A. Fukazawa,
1
Y. Sato, K. Ogasawara, T. Higashiyama and S. Yamaguchi,
Chem. Commun., 2015, 51, 11880–11883.
(a) H. Tsuji, K. Sato, L. Ilies, Y. Itoh, Y. Sato and
E. Nakamura, Org. Lett., 2008, 10, 2263–2265;
3
(
b) Y. Matano, T. Miyajima, T. Fukushima, H. Kaji,
Y. Kimura and H. Imahori, Chem. – Eur. J., 2008, 14, 8102–
115; (c) Y. Dienes, M. Eggenstein, T. Karpati,
8
T. C. Sutherland, L. Nyulaszi and T. Baumgartner, Chem. –
Eur. J., 2008, 14, 9878–9889; (d) H. Tsuji, K. Sato, Y. Sato
and E. Nakamura, J. Mater. Chem., 2009, 19, 3364–3366; 10 Y. Koyanagi, Y. Kimura and Y. Matano, Dalton Trans., 2016,
e) A. Saito, T. Miyajima, M. Nakashima, T. Fukushima, 45, 2190–2200.
(
Dalton Trans.
This journal is © The Royal Society of Chemistry 2017

View Article Online
Dalton Transactions
Paper
1
1 The effects of cation–anion interactions in cationic poly-
methine dye molecules on their ability to generate singlet
oxygen have been studied. See: (a) M. P. Samtsov,
the sum of van der Waals radii of the atoms participating
are apparent. See: E. Hosten, T. Gerber and R. Betz,
Z. Kristallogr. – New Cryst. Struct., 2012, 227, 331–332.
E. S. Voropai, K. N. Kaplevskii and D. G. Mel’nikov, J. Appl. 15 Y. Hayashi, Y. Matano, K. Suda, Y. Kimura, Y. Nakao and
Spectrosc., 2008, 75, 692–699; (b) M. P. Samtsov, H. Imahori, Chem. – Eur. J., 2012, 18, 15972–15983.
D. S. Tarasau, A. S. Stasheuski, K. N. Kaplevsky and 16 M. C. Burla, R. Caliandro, M. Camalli, B. Carrozzini,
E. S. Voropay, J. Appl. Spectrosc., 2014, 81, 214–221, and
references therein.
2 T. Butters and W. Winter, Chem. Ber., 1984, 117, 990–1002.
G. L. Cascarano, C. Giacovazzo, M. Mallamo, A. Mazzone,
G. Polidori and R. Spagna, J. Appl. Crystallogr., 2012, 45,
357.
1
1
3 (a) T. Sanji, K. Shiraishi, T. Kashiwabara and M. Tanaka, 17 M. C. Burla, R. Caliandro, B. Carrozzini, G. L. Cascarano,
Org. Lett., 2008, 10, 2689–2692; (b) A. Fukazawa, M. Hara,
T. Okamoto, E.-C. Son, C. Xu, K. Tamao and S. Yamaguchi,
Org. Lett., 2008, 10, 913–916.
C. Cuocci, C. Giacovazzo, M. Mallamo, A. Mazzone and
G. Polidori, J. Appl. Crystallogr., 2015, 48, 306.
18 G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam.
Crystallogr., 2008, 64, 112.
1
4 In the crystal of methyltriphenylphosphonium iodide,
C–H⋯I contacts whose range falls by more than 0.2 Å below 19 G. M. Sheldrick, University of Göttingen, Germany, 2014.
This journal is © The Royal Society of Chemistry 2017
Dalton Trans.