Photoinduced charge separation of phenothiazine-platinum-naphthalene diimide triads linked by twisted phenylene bridges

Article abstract of DOI:10.1007/s11164-012-0642-6

Two triads (i.e., 3PTZ-Pt-MNDI and 10PTZ-Pt-MNDI) consisting of 3-phenothiazine (3PTZ) or 10-phenothiazine (10PTZ), bipyridine-diacetylide platinum complex (Pt), and naphthalene diimide (MNDI) chromophores linked by highly twisted biphenylene spacers have

Full text of DOI:10.1007/s11164-012-0642-6

Res Chem Intermed (2013) 39:185–204
DOI 10.1007/s11164-012-0642-6
Photoinduced charge separation of phenothiazine–
platinum–naphthalene diimide triads linked by twisted
phenylene bridges
•
Ryoji Sugimura Shuichi Suzuki
•
•
Masatoshi Kozaki Kazutoshi Keyaki
•
•
Koichi Nozaki Hironori Matsushita
•
•
Noriaki Ikeda Keiji Okada
Received: 3 October 2011 / Accepted: 5 December 2011 / Published online: 10 July 2012
Ó Springer Science+Business Media B.V. 2012
Abstract Two triads (i.e., 3PTZ–Pt–MNDI and 10PTZ–Pt–MNDI) consisting of
3-phenothiazine (3PTZ) or 10-phenothiazine (10PTZ), bipyridine–diacetylide
platinum complex (Pt), and naphthalene diimide (MNDI) chromophores linked by
highly twisted biphenylene spacers have been prepared. The formation and decay of
the charge-separated (CS) states in toluene were studied by use of picosecond and
nanosecond laser photolysis via selective excitation of the Pt moiety. The time
required for formation of the CS state, PTZ?–Pt–MNDI-, from PTZ–3Pt*–MNDI
was determined to be sCS = 280 ps for 3PTZ?–Pt–MNDI- and sCS = 230 ps for
10PTZ?–Pt–MNDI-. The lifetimes of the CS states were determined to be
sCR1 = 75 ns (95 %) and sCR2 = 285 ns (5 %) for 3PTZ?–Pt–MNDI- and
sCR = 830 ns for 10PTZ?–Pt–MNDI-. Formation and decay of the CS states are
discussed in terms the Marcus theory and the spin-correlated radical pair
mechanism.
R. Sugimura ꢀ S. Suzuki ꢀ M. Kozaki ꢀ K. Okada (&)
Department of Chemistry, Graduate School of Science, Osaka City University, Sumiyoshi-ku,
Osaka 558-8585, Japan
e-mail: okadak@sci.osaka-cu.ac.jp
K. Keyaki ꢀ K. Nozaki (&)
Department of Chemistry, Graduate School of Science and Engineering, University of Toyama,
Gofuku, Toyama 930-8555, Japan
e-mail: nozaki@sci.u-toyama.ac.jp
H. Matsushita ꢀ N. Ikeda (&)
Department of Macromolecular Science and Engineering, Graduate School of Science and
Technology, Kyoto Institute of Technology, Sakyo-ku, Kyoto 606-8585, Japan
e-mail: ikeda@kit.ac.jp
123

186
R. Sugimura et al.
Keywords Photoinduced electron transfer ꢀ Phenothiazine ꢀ Naphthalene diimide ꢀ
Platinum complex
Introduction
Photoinduced charge separation is a fundamental and crucial process for energy
conversion from light to storable electric or chemical energy. In bacteria, a natural
photosynthetic system enables very fast (within 150 fs) energy transfer to a special
pair (i.e., SP: Bchl2). Electron transfer to bacteriopheophytin in the L unit (BPhL) then
results in the formation of a charge-separated (CS) state, Bchl?2 BPhL-, in 3 ps and
ubiquinone QA in 200 ps to efficiently yield Bchl?2 QA-. In this CS state, the charges are
?
-
˚
separated by 35 A in the protein matrix. Bchl2 QA has a long lifetime of 100 ms in the
absence of QB [1]. To date, a number of artificial systems featuring multi-
chromophoric donor–acceptors have been designed and synthesized to obtain long-
lived CS states [2]. However, these systems do not always produce long-lived CS
states. One promising strategy for obtaining long-lived CS states is the spin-control
approach [3], which uses a chromophore with ultrafast intersystem crossing. Metal
complexes containing Ru, Ir, and Pt emit intense phosphorescence at room
temperature because of strong spin–orbit coupling (SOC) and are expected to be
good sensitizers for generation of triplet CS states [4–11]. These triplet CS states
recombine slowly to the singlet ground states because of the spin-forbidden nature of
the transformation. We recently reported that MTA–Pt–MNDI (1), which features a
bipyridine–diacetylide platinum complex (Pt) linked with a dimethoxydimethyl-
triphenylamine (MTA) moiety as the donor and a naphthalene diimide (MNDI)
moiety as the acceptor, produced a long-lived CS state with high efficiency
(sCS & 1 ls, UCS = 0.96) [12]. In this complex, MTA and MNDI are bound to
the Pt atom through highly twisted phenylene linkers which enable rapid electron
(hole) shift (MTA?–Pt-–MNDI (or MTA–Pt?–MNDI-) ? MTA?–Pt–MNDI-) but
reduce back-electron transfer (i.e., MTA?–Pt-–MNDI (or MTA–Pt?–MNDI-) ?
MTA–Pt–MNDI) in the initially formed geminate ion pair. As a result, the long-
separated final ion pair MTA?–Pt–MNDI- is efficiently produced. The final CS state
was shown to be a spin-correlated radical pair (SCRP) state [12] with a precursor
MTA–3Pt*–MNDI state that is generated by the SOC-induced ultrafast intersystem
crossing (ISC) of MTA–1Pt*–MNDI. In this study, we apply this methodology to
phenothiazine (PTZ)-based triads; i.e., 3-phenothiazine (3PTZ)–Pt–MNDI (2) and
10-phenothiazine (10PTZ)–Pt–MNDI (3). Compounds 2 (Eox1: ?0.21 V vs. Fc?/Fc in
CH2Cl2) and 3 (Eox1: ?0.25 V) have similar oxidation potentials to 1 (Eox1: ?0.16 V),
suggesting they undergo similar photoinduced electron-transfer reactions. Com-
pounds 2 and 3 are positional isomers and have very similar oxidation potentials,
because of the common PTZ moieties and similar distances between PTZ and MNDI.
Hence, the forces driving charge recombination in the CS states in these positional
isomers are similar. However, the rates of charge recombination (CR) to the respective
ground states were found to be very different. This paper describes the details of the
CS and CR processes for the photoexcitation of 2 and 3 using the picosecond and
nanosecond transient absorption method.
123

Photoinduced charge separation
187
tBu
t
Bu
tBu
t Bu
tBu
tBu
N
N
N
N
N
N
Pt
Pt
Pt
Me
O
Me
Me
Me
Me
Me
Me
Me
Me
O
O
N
N
N
Me
Me
Me
Me
Me
Me
Me
Me
Me
MeO
O
O
O
N
N
O
S
O
O
N
N
N
S
N
C8H17
C8H17
C8H17
O
O
O
OMe
MTA-Pt-MNDI (1)
3PTZ-Pt-MNDI (2)
10PTZ-Pt-MNDI (3)
Experimental
General procedure
1H NMR spectra were recorded with a Jeol JNM-LA400 spectrometer. FAB-MS
and HRMS spectra were recorded with Jeol JMS-AX-500 and JMS-AX-700
spectrometers, respectively. Infrared spectra were recorded on KBr plates with a
Shimadzu FTIR-8700 spectrometer. Melting points were measured with a hot-stage
apparatus and are uncorrected. Absorption and luminescence spectra were measured
with a Jasco V-750 UV–visible spectrometer and a Shimadzu RF-5300 PC
fluorophotometer. Silica gel 60 (100–200 mesh) and aluminium oxide 90
(70–230 mesh) were used for column chromatography. All commercially available
compounds were reagent grade and used without further purification. Toluene and
THF were dried and distilled over sodium. CH2Cl2, diisopropylamine (iPr2NH),
N-ethyldiisopropylamine (iPr2NEt), and o-dichlorobenzene were dried and distilled
over calcium hydride. 3-Amino-10-phenyl-10H-phenothiazine (4) [13], 4,40-diiodo-
2,20,6,60-tetramethylbiphenyl (8) [12, 14], N-(4-ethynyl-2,6-dimethylphenyl)-N0-(n-octyl)
naphthalene-1,8:4,5-tetracarboxydiimide (12) [12], 4,40-di-tert-butyl-2,20-bipyri-
dine platinum dichloride (13) [15], and N-(4-bromo-2,6-dimethylphenyl)-N0-
(n-octyl)naphthalene-1,8:4,5-tetracarboxydiimide (23) [12] were prepared by
methods described in the literature. The 10-(40-tert-butylphenyl)-10H-phenothiazine
(22) was prepared from phenothiazine (16) and 1-bromo-4-tert-butylbenzene by the
cross coupling reaction using Pd2(dba)3-tBuOK in toluene.
Synthesis of 3-[(1E)-3,3-diethyltriaz-1-en-1-yl]-10-phenyl-10H-phenothiazine (5)
3-Amino-10-phenyl-10H-phenothiazine (4) (1.12 g, 3.86 mmol) and CH2Cl2
(20 mL) were placed in a 50-mL two-necked flask. The solution was cooled to
0 °C in an ice bath. BF3ꢀOEt2 (0.98 mL, 7.72 mmol) was added to the flask and the
mixture was stirred for 40 min under nitrogen. A solution of tert-butyl nitrite
(0.69 mL, 5.79 mmol) in CH2Cl2 (3 mL) was slowly added to the flask, and the
mixture was stirred for 1 h. K2CO3 (2.67 g, 19.3 mmol) and Et2NH (2.4 mL) were
added to the flask, and the mixture was stirred at room temperature for 20 h. The
mixture was filtered to remove insoluble materials which were washed with a small
amount of CH2Cl2. The filtrate was poured into water. The organic layer was
separated, and washed with water, dried over Na2SO4, filtered, and concentrated
under reduced pressure. The resulting mixture was separated and purified by
123

188
R. Sugimura et al.
alumina column chromatography using hexane–CH2Cl2 (2:1 v/v) as eluent, to give
triazene 5 (1.28 g, 89 %) as an orange solid. 5: C22H22N4S; MW 374.50; mp:
1
100–102 °C; H NMR (400 MHz, DMSO-d6) d (ppm): 7.66 (t, J = 7.3 Hz, 2H),
7.52 (t, J = 7.3 Hz, 1H), 7.40 (d, J = 7.1 Hz, 2H), 7.08 (dd, J = 7.6, 1.4 Hz, 1H),
7.05 (d, J = 2.2 Hz, 1H), 6.96–6.84 (m, 3H), 6.16 (t, J = 8.8 Hz, 2H), 3.68 (m,
4H), 1.16 (s (br), 6H); IR (KBr/cm-1) 2974, 1589, 1495, 1462, 1391, 1352, 1304,
1236, 1192, 1105, 893, 812, 772, 743, 698, 638, 544; MS (FAB?) = 374.19 [M?].
High-resolution MS: found m/z 374.1567; calcd for C22H22N4S m/z 374.1565.
Synthesis of 3-iodo-10-phenyl-10H-phenothiazine (6)
Triazene 5 (1.63 g, 4.35 mmol) was placed in a screw-capped tube. The tube was
flushed with nitrogen. CH3I (8 mL) and I2 (2.21 g, 8.70 mmol) were added, and the
tube was sealed and heated at 55 °C for 2 h. CH2Cl2 was added, and the mixture and
washed with aqueous NaHSO3 and water. The organic layer was dried over Na2SO4,
filtered, and concentrated under reduced pressure. The mixture was separated and
purified by alumina column chromatography using hexane–CH2Cl2 (5:1 v/v) as
eluent, to give 6 (1.33 g, 76 %) as a pale yellow solid. 6: C18H12INS; MW 401.26;
1
mp: 107–109 °C; H NMR (400 MHz, DMSO-d6) d (ppm): 7.68 (t, J = 7.3 Hz,
2H), 7.56 (t, J = 7.6 Hz, 1H), 7.42 (d, J = 7.3 Hz, 2H), 7.38 (d, J = 2.2 Hz, 1H),
7.23 (dd, J = 8.6, 2.2 Hz, 1H), 7.07 (dd, J = 7.3, 1.7 Hz, 1H), 6.93 (dt, J = 7.6,
1.7 Hz, 1H), 6.86 (dt, J = 7.6, 1.2 Hz, 1H), 6.10 (dd, J = 7.8, 1.2 Hz, 1H), 5.89
(d, J = 8.6 Hz, 1H); IR (KBr/cm-1) 1583, 1558, 1491, 1456, 1439, 1377, 1302,
1258, 1240, 1130, 1069, 1042, 870, 851, 808, 781, 746, 696, 629, 538. High-
resolution MS: found m/z 400.9730; calcd for C18H12INS m/z 400.9735.
Synthesis of 10-phenyl-3-(4,4,5,5-tetramethyl-1,3,2-dioxaborolan-2-yl)-10H-
phenothiazine (7)
Compound 6 (1.25 g, 3.12 mmol), bis(pinacolato)diboron (1.19 g, 4.67 mmol), [1,10-
bis(diphenylphosphino)ferrocene]dichloropalladium(II) (255 mg, 0.312 mmol), and
KOAc (919 mg, 9.36 mmol) were dissolved in DMSO (20 mL) in a two-necked flask.
The mixture was stirred at 80 °C for 1 h under nitrogen. After cooling to room
temperature, the mixture was filtered through a Celite pad. The Celite pad was washed
with EtOAc. The combined filtrate was washed with water (ca 100 mL), dried over
Na2SO4, and concentrated under reduced pressure. The resulting mixture was
separated and purified by silica gel column chromatography using hexane–CH2Cl2
(2:1 v/v) as eluent, to give 7 (997 mg, 80 %) as a pale yellow solid. 7: C24H24BNO2S;
1
MW 401.33; mp: 156–158 °C; H NMR (400 MHz, DMSO-d6) d (ppm): 7.70 (t,
J = 7.3 Hz, 2H), 7.57 (t, J = 7.6 Hz, 1H), 7.42 (d, J = 7.3 Hz, 2H), 7.23 (d,
J = 1.5 Hz, 1H), 7.18 (dd, J = 8.0, 1.5 Hz, 1H), 7.05 (dd, J = 7.3, 1.7 Hz, 1H), 6.91
(dt, J = 7.6, 1.7 Hz, 1H), 6.85 (dt, J = 7.6, 1.2 Hz, 1H), 6.09 (d, J = 8.0 Hz, 1H),
6.07 (dd, J = 7.8, 1.2 Hz, 1H), 1.25 (s, 12H); IR (KBr/cm-1) 2980, 2932, 1603, 1589,
1576, 1491, 1470, 1441, 1404, 1383, 1354, 1302, 1258, 1144, 1105, 962, 866, 824,
775, 750, 698, 667, 629, 536; MS (FAB?) = 401.20 [M?]. High-resolution MS: found
m/z 401.1616; calcd for C24H24BNO2S m/z 401.1621.
123

Photoinduced charge separation
189
Synthesis of 3-(40-iodo-2,20,6,60-tetramethylbiphenyl-4-yl)-10-phenyl-10H-
phenothiazine (9)
Compound 7 (450 mg, 1.12 mmol), 4,40-diiodo-2,20,6,60-tetramethylbiphenyl (8)
(1.55 g, 3.36 mmol), Cs2CO3 (730 mg, 2.24 mmol), and Pd(OAc)2 (12.6 mg,
0.056 mmol) were dissolved in DMF (20 mL) in a two-necked flask. The solution
was stirred at 110 °C for 1 h under nitrogen. After cooling to room temperature, the
mixture was filtered through a Celite pad. The Celite pad was washed with CH2Cl2.
The combined filtrate was concentrated under reduced pressure. The resulting
residue was separated and purified by alumina column chromatography using
hexane–CH2Cl2 (6:1 v/v) as eluent, to give 9 (516 mg, 76 %) as a pale yellow solid.
1
9: C34H28INS; MW 609.56; mp: 219–221 °C; H NMR (400 MHz, DMSO-d6) d
(ppm): 7.70 (t, J = 7.3 Hz, 2H), 7.58 (t, J = 7.6 Hz, 1H), 7.56 (s, 2H), 7.47 (d,
J = 7.3 Hz, 2H), 7.43 (d, J = 2.2 Hz, 1H), 7.39 (s, 2H), 7.28 (dd, J = 8.6, 2.2 Hz,
1H), 7.10 (dd, J = 7.3, 1.5 Hz, 1H), 6.94 (dt, J = 7.6, 1.5 Hz, 1H), 6.87 (dt,
J = 7.6, 1.2 Hz, 1H), 6.19 (d, J = 8.6 Hz, 1H), 6.14 (dd, J = 7.3, 1.2 Hz, 1H), 1.86
(s, 6H), 1.81 (s, 6H); IR (KBr/cm-1) 1576, 1493, 1466, 1439, 1377, 1312, 1259,
999, 866, 816, 745, 698, 540; MS (FAB?) = 609.22 [M?]. High-resolution MS:
found m/z 609.0989; calcd for C34H28INS m/z 609.0987.
Synthesis of 10-phenyl-3-{2,20,6,60-tetramethyl-40-[(trimethylsilyl)ethynyl]biphenyl-
4-yl}-10H-phenothiazine (10)
Compound 9 (450 mg, 0.738 mmol), PdCl2(PPh3)2 (51.8 mg, 0.0738 mmol), CuI
(14.1 mg, 0.0738 mmol), THF (5 mL), and iPr2NEt (2.5 mL) were placed in a
screw-capped tube. The tube was flushed with nitrogen. Trimethysilylacetylene
(0.125 mL, 0.886 mmol) was added. The tube containing the mixture was sealed
and heated at 70 °C for 6 h. The solvents were evaporated under reduced pressure.
The residue was separated and purified by silica gel column chromatography using
hexane–CH2Cl2 (6:1 v/v) as eluent, to give 10 (380 mg, 89 %) as a pale yellow
1
solid. 10: C39H37NSSi; MW 579.87; pale yellow solid; mp: 209–211 °C; H NMR
(400 MHz, DMSO-d6) d (ppm): 7.71 (t, J = 8.1 Hz, 2H), 7.58 (t, J = 7.3 Hz, 1H),
7.47 (d, J = 7.3 Hz, 2H), 7.43 (d, J = 2.2 Hz, 1H), 7.40 (s, 2H), 7.29 (s, 2H),
7.29–7.27 (m, 1H), 7.10 (dd, J = 7.3, 1.2 Hz, 1H), 6.94 (dt, J = 7.6, 1.2 Hz, 1H),
6.87 (dt, J = 7.3, 1.2 Hz, 1H), 6.19 (d, J = 8.8 Hz, 1H), 6.14 (dd, J = 8.1, 1.2 Hz,
1H), 1.85 (s, 6H), 1.83 (s, 6H), 0.24 (s, 9H); IR (KBr/cm-1) 2957, 2918, 2152, 1593,
1578, 1493, 1464, 1439, 1377, 1310, 1250, 1160, 1020, 1001, 854, 748, 698, 648,
540; MS (FAB?) = 579.36 [M?]. High-resolution MS: found m/z 579.2407; calcd
for C39H37NSSi m/z 579.2416.
Synthesis of 2-(40-ethynyl-2,20,6,60-tetramethylbiphenyl-4-yl)-10-phenyl-10H-
phenothiazine (11)
Compound 10 (292 mg, 0.504 mmol) was dissolved in THF (15 mL). To this
solution, 1.0 M THF solution of Bu4N?ꢀF- (0.63 mL, 0.63 mmol) was added and
the mixture was stirred at room temperature for 15 min. CH2Cl2 (ca 50 mL) was
123

190
R. Sugimura et al.
added to the mixture and washed with water. The organic layer was separated, dried
over Na2SO4, filtered, and concentrated. The mixture was separated by alumina
column chromatography using hexane–CH2Cl2 (6:1 v/v) as eluent, to give 11
(251 mg, 98 %) as a yellow powder. 11: C36H29NS; MW 507.69; yellow solid; mp:
1
229–231 °C; H NMR (400 MHz, DMSO-d6) d (ppm): 7.71 (t, J = 8.0 Hz, 2H),
7.58 (t, J = 7.3 Hz, 1H), 7.47 (d, J = 7.3 Hz, 2H), 7.43 (d, J = 2.2 Hz, 1H), 7.40
(s, 2H), 7.29 (s, 2H), 7.29–7.27 (m, 1H), 7.10 (dd, J = 7.3, 1.2 Hz, 1H), 6.94 (dt,
J = 7.4, 1.2 Hz, 1H), 6.87 (dt, J = 7.4, 1.2 Hz, 1H), 6.19 (d, J = 8.3 Hz, 1H), 6.14
(d, J = 8.0 Hz, 1H), 4.11 (s, 1H), 1.86 (s, 6H), 1.84 (s, 6H); IR (KBr/cm-1) 3292,
2914, 1593, 1576, 1491, 1466, 1439, 1377, 1312, 1265, 1244, 1001, 866, 816, 775,
743, 710, 698, 650, 613, 540. High-resolution MS: found m/z 507.2021; calcd for
C36H29NS m/z 507.2021.
Synthesis of 3PTZ–Pt–MNDI (2)
Compound 11 (100 mg, 0.197 mmol), 12 (100 mg, 0.197 mmol), 13 (105 mg,
0.197 mmol), and CuI (3.8 mg, 0.020 mmol) were placed in a two-necked flask.
CH2Cl2 (14 mL) and iPr2NH (0.7 mL) were added and the flask was flushed with
nitrogen. The mixture was stirred at room temperature for 18 h. The solvents were
evaporated under reduced pressure. The resulting mixture was separated and
purified by alumina column chromatography to give 14 (58.2 mg, 20 %, eluent:
hexane–CH2Cl2 1:1 v/v), 2 (99.7 mg, 34 %, eluent: CH2Cl2), and 15 (58.9 mg,
20 %, eluent: CH2Cl2–EtOAc 10:1 v/v). Analytically pure samples were obtained
by reprecipitation from CH2Cl2 with ether. 14: C90H80N4PtS2; MW 1476.84; brown
1
solid; mp: [300 °C; H NMR (400 MHz, CDCl3) d (ppm): 9.81 (d, J = 6.1 Hz,
2H), 7.95 (d, J = 2.0 Hz, 2H), 7.65–7.59 (m, 6H), 7.50 (t, J = 7.6 Hz, 2H), 7.43
(dd, J = 8.3, 1.5 Hz, 4H), 7.34 (s, 4H), 7.32 (d, J = 2.2 Hz, 2H), 7.22 (s, 4H), 7.11
(dd, J = 8.5, 2.2 Hz, 2H), 7.04 (dd, J = 7.1, 1.7 Hz, 2H), 6.87–6.79 (m, 4H), 6.23
(d, J = 8.5 Hz, 2H), 6.19 (dd, J = 8.0, 1.4 Hz, 2H), 1.94 (s, 12H), 1.87 (s, 12H),
1.45 (s, 18H); IR (KBr/cm-1) 3061, 2965, 2916, 2870, 2100, 1593, 1547, 1493,
1464, 1441, 1418, 1375, 1308, 1252, 1204, 1155, 1130, 1074, 1001, 866, 814, 775,
745, 698, 600, 542, 419. High-resolution MS: found m/z 1475.5458; calcd for
C90H80N4PtS2 m/z 1475.5472. 3PTZ–Pt–MNDI (2): C86H81N5O4PtS; MW 1475.74;
1
brown solid; mp: ca 274 °C (dec); H NMR (400 MHz, CDCl3) d (ppm): 9.80 (d,
J = 6.1 Hz, 1H), 9.74 (d, J = 6.1 Hz, 1H), 8.83 (d, J = 7.6 Hz, 2H), 8.80 (d,
J = 7.6 Hz, 2H), 7.95 (d, J = 2.0 Hz, 2H), 7.65–7.61 (m, 4H), 7.50 (t, J = 7.6 Hz,
1H), 7.44–7.42 (m, 4H), 7.34 (s, 2H), 7.32 (d, J = 2.0 Hz, 1H), 7.23 (s, 2H), 7.12
(dd, J = 8.6, 2.2 Hz, 1H), 7.04 (dd, J = 7.1, 2.0 Hz, 1H), 6.87–6.79 (m, 2H), 6.23
(d, J = 8.6 Hz, 1H), 6.19 (dd, J = 8.0, 1.2 Hz, 1H), 4.22 (t, J = 7.6 Hz, 2H), 2.09
(s, 6H), 1.94 (s, 6H), 1.88 (s, 6H), 1.74 (m, 2H), 1.46–1.28 (m, 28H), 0.88 (t,
J = 6.4 Hz, 3H); IR (KBr/cm-1) 2959, 2924, 2856, 2104, 1709, 1670, 1618, 1580,
1547, 1464, 1418, 1369, 1339, 1308, 1248, 1192, 1155, 1130, 1076, 1024, 1001,
976, 868, 849, 824, 793, 770, 745, 700, 602, 419. High-resolution MS: found m/
z 1474.5665; calcd for C86H81N5O4PtS m/z 1474.5657.
123

Photoinduced charge separation
191
Synthesis of 10-(40-iodo-2,20,6,60-tetramethylbiphenyl-4-yl)-10H-phenothiazine (17)
Phenothiazine (16) (200 mg, 1.00 mmol), 4,40-diiodo-1,10,3,30-tetramthylbiphenyl
(8) (1.39 g, 3.01 mmol), copper powder (127 mg, 2.00 mmol), K2CO3 (553 mg,
4.00 mmol), and 18-crown-6-ether (26 mg, 0.098 mmol) were placed in a 50-mL
two-necked flask. o-Dichlorobenzene (5 mL) was added to the flask and the mixture
was heated under reflux for 4 h under nitrogen. After cooling to room temperature,
the mixture was filtered through a Celite pad. The Celite pad was washed with
CH2Cl2 (*30 mL). The filtrate was concentrated under reduced pressure. The
residue was separated and purified by silica gel column chromatography using
hexane–CH2Cl2 (10:1 v/v) as eluent to give compound 17 (344 mg, 64 %) as a pale
1
yellow solid. 17: C28H24INS; MW 533.47; mp: 220–222 °C; H NMR (400 MHz,
DMSO-d6) d (ppm): 7.62 (s, 2H), 7.24 (s, 2H), 7.07 (dd, J = 7.4, 1.6 Hz, 2H), 6.98
(dt, J = 8.2, 1.6 Hz, 2H), 6.86 (dt, J = 7.4, 1.2 Hz, 2H), 6.27 (dd, J = 8.2, 1.2 Hz,
2H), 1.91 (s, 12H); IR (KBr/cm-1); 1589, 1574, 1489, 1464, 1443, 1377, 1310,
1246, 1232, 1128, 1047, 1001, 856, 827, 800, 741, 652. High-resolution MS: found
m/z 533.0673; calcd for C28H24INS m/z 533.0674.
Synthesis of 10-{2,20,6,60-tetramethyl-40-[(trimethylsilyl)ethynyl]biphenyl-4-yl}-
10H-phenothiazine (18)
Compound 17 (150 mg, 0.281 mmol), PdCl2(PPh3)2 (20 mg, 0.028 mmol), CuI
(5.4 mg, 0.028 mmol), THF (4 mL), and iPr2NEt (2 mL) were placed in a screw-
capped tube. The tube was flushed with nitrogen. Trimethysilylacetylene
(0.048 mL, 0.337 mmol) was added, and the tube was sealed and heated at 80 °C
for 3 h. The solvent was evaporated under reduced pressure. The residue was
separated and purified by silica gel column chromatography using hexane–CH2Cl2
(6:1 v/v) as eluent, to give 18 (142 mg, quant) as a pale yellow solid. 18:
1
C33H33NSSi; MW 503.77; mp: 187–189 °C; H NMR (400 MHz, DMSO-d6) d
(ppm): 7.33 (s, 2H), 7.24 (s, 2H), 7.08 (dd, J = 7.4, 1.5 Hz, 2H), 6.99 (dt, J = 7.4,
1.6 Hz, 2H), 6.86 (dt, J = 7.4, 1.2 Hz, 2H), 6.29 (dd, J = 8.2, 1.1 Hz, 2H), 1.93 (s,
6H), 1.90 (s, 6H), 0.25 (s, 9H); IR (KBr/cm-1); 2957, 2920, 2152, 1589, 1572, 1485,
1462, 1443, 1379, 1308, 1248, 1238, 1163, 1128, 1043, 1020, 1003, 928, 843, 745,
708, 654. High-resolution MS: found m/z 503.2104; calcd for C33H33NSSi m/
z 503.2103.
Synthesis of 10-(40-ethynyl-2,20,6,60-tetramethylbiphenyl-4-yl)-10H-phenothiazine
(19)
Compound 18 (100 mg, 0.119 mmol) was dissolved in THF (6 mL). To the
solution, 1.0 M THF solution of Bu4N?ꢀF- (0.25 mL, 0.25 mmol) was added and
the mixture was stirred at room temperature for 1 h. CH2Cl2 (ca 30 mL) was added
to the mixture and washed with water. The organic layer was separated, dried over
Na2SO4, filtered, and concentrated. The mixture was separated by alumina column
chromatography using hexane–CH2Cl2 (5:1 v/v) as eluent, to give 19 (74 mg, 86 %)
1
as a pale yellow powder. 19: C30H25NS; MW 431.59; mp: 224–226 °C; H NMR
123

192
R. Sugimura et al.
(400 MHz, DMSO-d6) d (ppm): 7.35 (s, 2H), 7.24 (s, 2H), 7.07 (dd, J = 7.4,
1.5 Hz, 2H), 6.99 (dt, J = 8.2, 1.5 Hz, 2H), 6.86 (dt, J = 7.4, 1.2 Hz, 2H), 6.28 (dd,
J = 8.2, 1.2 Hz, 2H), 4.15 (s, 1H), 1.94 (s, 6H), 1.91 (s, 6H); IR (KBr/cm-1); 3312,
2916, 1591, 1576, 1489, 1464, 1441, 1375, 1308, 1279, 1246, 1234, 1049, 1003,
876, 741, 638, 606, 419. High-resolution MS: found m/z 431.1727; calcd for
C30H25NS m/z 431.1708.
Synthesis of 10PTZ–Pt–MNDI (3)
Compound 19 (100 mg, 0.232 mmol), 12 (117 mg, 0.232 mmol), 13 (124 mg,
0.232 mmol), and CuI (5.3 mg, 0.028 mmol) were placed in a two-necked flask.
CH2Cl2 (15 mL) and iPr2NH (0.75 mL) were added and the flask was flushed with
nitrogen. The mixture was stirred at room temperature for 18 h. The solvents were
evaporated under reduced pressure. The resulting mixture was separated and
purified by alumina column chromatography to give 20 (65.7 mg, 21 %, eluent:
hexane–CH2Cl2 1:2 v/v), 3 (117 mg, 36 %, eluent: CH2Cl2), and 15 (75.2 mg,
22 %, eluent: CH2Cl2–EtOAc 10:1 v/v). Analytically pure samples were obtained
by reprecipitation from CH2Cl2 with ether. 20: C78H72N4PtS2; MW 1324.65; brown
1
solid; mp: [300 °C; H NMR (400 MHz, CDCl3) d (ppm): 9.82 (d, J = 5.8 Hz,
2H), 7.97 (d, J = 1.4 Hz, 2H), 7.63 (dd, J = 5.8, 2.0 Hz, 2H), 7.40 (s, 4H), 7.14 (s,
4H), 7.00 (dd, J = 7.3, 1.4 Hz, 4H), 6.89 (dt, J = 8.2, 1.4 Hz, 4H), 6.80 (dt,
J = 7.3, 1.2 Hz, 4H), 6.31 (dd, J = 8.2, 1.2 Hz, 4H), 1.99 (s, 12H), 1.96 (s, 12H),
1.47 (s, 18H); IR (KBr/cm-1) 3404, 2964, 2914, 2106, 1618, 1589, 1547, 1462,
1443, 1418, 1377, 1308, 1236, 1159, 1126, 1086, 1042, 1001, 928, 870, 847, 745,
654, 600, 563; MS (FAB?) = 1325 [M?]. Anal. Calcd for C78H72N4PtS2: C, 70.72;
H, 5.48; N, 4.23. Found: C, 70.37; H, 5.43; N, 3.97. 10PTZ–Pt–MNDI (3):
C80H77N5O4PtS; MW 1399.65; brown solid; mp: 268 °C (dec); 1H NMR
(400 MHz, CDCl3) d (ppm): 9.81 (d, J = 6.1 Hz, 1H), 9.75 (d, J = 6.1 Hz, 1H),
8.83 (d, J = 7.6 Hz, 2H), 8.80 (d, J = 7.6 Hz, 2H), 7.97 (s, 2H), 7.63 (dt, J = 6.1,
2.0 Hz, 2H), 7.43 (s, 2H), 7.40 (s, 2H), 7.14 (s, 2H), 7.00 (dd, J = 7.6, 1.6 Hz, 2H),
6.89 (dt, J = 7.6, 1.7 Hz, 2H), 6.80 (dt, J = 7.3, 1.2 Hz, 2H), 6.31 (dd, J = 8.3,
1.2 Hz, 2H), 4.23 (t, J = 7.6 Hz, 2H), 2.10 (s, 6H), 1.99 (s, 6H), 1.96 (s, 6H), 1.76
(m, 2H), 1.47–1.26 (m, 28H), 0.88 (t, J = 6.8 Hz, 3H); IR (KBr/cm-1) 2959, 2924,
2856, 2106, 1709, 1670, 1618, 1582, 1545, 1462, 1445, 1418, 1369, 1340, 1308,
1248, 1192, 1045, 872, 824, 793, 770, 745, 706, 419. High-resolution MS: found m/
z 1398.5342; calcd for C80H77N5O4PtS m/z 1398.5344.
Measurement of cyclic voltammograms
Redox potentials were measured, by use of an ALS model 610A electrochemical
analyzer, in a conventional three-electrode cell equipped with a glassy carbon
working electrode, a platinum wire counter electrode, and a saturated calomel
electrode (SCE) as reference electrode. Measurements were performed with a sweep
rate of 100 mV/s in a suitable solvent containing 0.1 M tetra-n-butylammonium
perchlorate as supporting electrolyte. The redox potentials were finally corrected on
the basis of the ferrocene/ferrocenium (Fc/Fc?) couple.
123

Photoinduced charge separation
193
Measurements of absorption spectra for phenothiazine radical cation 9?
and 22?, and naphthalene diimide radical anion 23-
For assignment of transient species in laser photolysis, the absorption spectra of the
electrochemically generated 9?, 22?, and 23- were measured by use of an Ocean
Optics HR4000 spectrometer with a 1 mm width electrochemical cell equipped with a
fine platinum mesh working electrode, a platinum wire as a counter electrode, and a
reference SCE (Fc/Fc? = ? 0.48 V vs. SCE in CH2Cl2 and DMF) at suitable external
potentials. The radical cation and radical anion species were generated in CH2Cl2 and
in DMF, respectively, in the presence of 0.1 M tetra-n-butylammonium perchlorate.
Molecular geometry and molecular orbital calculations for model compounds
2M and 3M
Model compounds 2M and 3M, in which, for simplicity methyl groups were used
instead of the tert-butyl groups on bipyridine and the n-octyl group on MNDI in 2
and 3, were used for geometrical optimization and molecular orbital calculations for
2 and 3. There are two relative orientations (parallel and perpendicular toward the
standard Pt-plane) of the phenyl rings, A and B (formula 24), thus four basic
structures of conformers in PTZ-Pt-MNDI (2M and 3M) can be derived. More
conformers are generated from these basic conformers depending on the folding
direction and orientations of PTZs: 10-Phenylphenothiazine has a largely twisted
(almost perpendicular) phenyl ring in the 10-position. In 10PTZ–Pt–MNDI (3M),
there are two folding directions in the PTZ rings of the perpendicular orientation.
Therefore, 10PTZ–Pt–MNDI (3M) has two more conformers, six (4 ? 2)
conformers in total. In 3PTZ-Pt-MNDI (2M), there are eight conformers (two
folding directions and four possible rotational orientations) for each basic
conformer. Therefore, 3PTZ–Pt–MNDI (2M) has 32 (4 9 8) conformers in total.
The initial geometry of these conformers was generated. Geometrical optimization
was performed using MOPAC 2009 (PM6) [16]. The energies of thus obtained
conformers are as follows: 180.63019–181.25674 kcal/mol for 3PTZ–Pt–MNDI
(2M) and 156.67625–157.32413 kcal/mol for 10PTZ–Pt–MNDI (3M). The most
stable conformational isomers and their HOMO and LUMO are depicted in Fig. 2.
Laser photolysis of 2 and 3
Nanosecond time-resolved difference spectra were obtained by using the second
harmonic generator of a Q-switched Nd3?:YAG laser (Continuum Surelite I-10,
k = 532 nm) [17]. Sample solutions in a 1 cm-quartz cell were deaerated by
bubbling with argon for 5 min. White light from a Xe-arc lamp was used for
acquisition of absorption spectra. For measurements of picosecond time-resolved
difference spectra, a sample solution in a quartz cell (1 mm length) was excited
with the second harmonic pulses of a mode-locked Nd3?:YAG laser (Continuum
PY61C-10, k = 532 nm) [18]. The transient absorption spectra in the time range
-20 ps to 1 ns were acquired by using continuum pulses generated by focusing the
fundamental laser pulse into a flowing H2O/D2O mixture (1:1 by volume).
123

194
R. Sugimura et al.
Results and discussion
Synthesis of PTZ–Pt–MNDIs (2 and 3)
The synthesis of 2 is outlined in Scheme 1. Compound 4 was converted to triazene
derivative 5 followed by treatment with CH3I and I2 to give 3-iodophenothiazine 6.
Compound 6 was reacted with bis(pinacolato)diboron in the presence of Pd as catalyst
to give borate 7. The Pd-mediated cross-coupling of 7 with diiodobiphenyl 8 produced
9 in moderate yield. Then, acetylene derivative 11 was obtained from 9 via 10. Triad 2
was prepared by reaction of bipyridine–platinum–dichloride complex 13 with an
equimolar mixture of 11 and 12 using iPr2NH and CuI as a base and catalyst,
respectively, according to the synthetic procedure for 1 [12]. The mixture of 2, 14, and
15 was easily separated and purified by column chromatography.
The synthesis of 3 is shown in Scheme 2. The Ullmann coupling reaction of
phenothiazine (16) with diiodobiphenyl 8 gave 17, which was converted to 19 by
procedures similar to those from 9 to 11. Reaction of 13 with an equimolar mixture
of acetylenes (12:19 = 1:1) in the presence of excess iPr2NH and a catalytic amount
of CuI produced the desired coupling product (3) with 15 and 20.
Absorption and emission of PTZ–Pt–MNDIs (2 and 3)
Figure 1a, b shows the absorption spectra of 2 and 3, respectively, in toluene, and
those of reference compounds (Scheme 3). The absorption of these triads in the
visible region is attributed to the platinum complex moiety. The PTZ and MNDI
moieties had no strong absorption band at wavelengths longer than 400 nm. The
Me Me
R
Me
Me
I
Me
I
I
Me
H
TMS
Me Me
8
S
N
R
S
N
Me
Me
S
N
Me
Me
PdCl2(PPh3)2
CuI
Pd(OAc)2, Cs2CO3
iPr2NEt, THF
DMF
y. 76%
y. 89%
9
⋅
a) BF3 Et2O, tBuONO, CH2Cl2
4: R = NH2
5: R = N3Et2
6: R = I
Bu4N+⋅F–, THF
y. 98%
:
R = TMS
10
11
b) Et2NH, K2CO3
y. 89%
: R = H
MeI, I2, toluene
y. 76%
B2(pin)2, PdCl2(dppf), KOAc (pin: pinacolate)
DMSO
7
: R = B(pin)
y. 80%
tBu
tBu
Me Me
Me Me
S
N
Me O
O
R =
R =
tBu
tBu
N
N
H
N
N C8H17
O
Pt
Me O
Cl
Cl
N
N
14
y. 20%
12
13
CuI, iPr2NH
CH2Cl2
Pt
2
+
Me O
N
O
R
R
y. 34%
N C8H17
O
Me O
15 y. 20%
Scheme 1 Synthetic procedures for 2
123

Photoinduced charge separation
195
TMS
PdCl2(PPh3)2
CuI
H
8, Cu, K2CO3
18-crown-6-ether
MeMe
MeMe
Me Me
Me Me
S
S
N
R
S
N
I
iPr2NEt, THF
o-dichlorobenzene
N
H
y. 64%
y. quant
16
17
Bu4N+•F−, THF
y. 86%
18: R = TMS
19: R = H
tBu
tBu
MeMe
MeMe
12, 13, CuI, iPr2NH
N
N
R =
N
S
+
+
3
15
Pt
CH2Cl2
y. 36%
y. 22%
R
R
20 y. 21%
Scheme 2 Synthetic procedures for 3
absorption bands at *380 nm are assigned to the p–p* transition in the MNDI
moiety. Absorption of 3PTZ acetylene 11 (kmax = 339 nm) occurs at a longer
wavelength than that of 10PTZ acetylene 19 (kmax = 320 nm), which is indicative
of the more extended electron delocalization in 11. The PTZ absorption overlaps
with that of the MNDI moiety. The absorption of both triads is almost
superimposable with the summation of the spectra of their components (i.e., PTZ,
MNDI, and Pt), which suggests negligible electronic interaction between these
components in the ground state. Laser light at 532 nm, which selectively excites the
Pt complex moiety, was used for picosecond and nanosecond laser photolysis.
The phosphorescence spectra of 2 and 3 in toluene are shown in the insets of
Fig. 1a, b, respectively, together with that of reference compound 21. 21
phosphoresced strongly (kem = 574 nm) at room temperature in toluene under
deaerated conditions. The intensities of the phosphorescence of 2 and 3 are
significantly weaker than that of 21. The efficient quenching of the phosphorescence
in 2 and 3 indicates rapid intramolecular electron (or hole) transfer or energy
3
transfer from Pt*.
Molecular geometry and molecular orbital calculations for model compounds
2M and 3M
The molecular structure of neutral phenothiazine is not planar but rather takes a
folded butterfly form [19, 20]. In addition, rotational isomers around acetylenic
bonds [12] are possible. Therefore, several conformational isomers of 2 and 3 are
likely to exist. We carried out geometrical optimization using MOPAC 2009 (PM6)
[16]. Model compounds 2M and 3M, in which, for simplicity, methyl groups were
used instead of the tert-butyl groups on bipyridine and the n-octyl group on MNDI,
were used in the geometrical optimization and molecular orbital calculations for 2
and 3. Figure 2 depicts the optimized molecular structures and their highest
molecular orbitals (HOMOs) and lowest unoccupied molecular orbitals (LUMOs)
for the most stable conformer. The HOMOs and LUMOs are located on the PTZ and
MNDI moieties, respectively, in each case. We also compared the MO coefficients
of the CA carbon atom of the twisted biphenylene moiety (optimized structure in
123

196
R. Sugimura et al.
Fig. 1 UV–visible and phosphorescence (inset) spectra in toluene: (a) for 2 (solid line) and (b) for 3
(solid line), with the spectra of the reference compounds: platinum complex 21 (dotted line), naphthalene
diimide 12 (dashed–dotted line), in both (a) and (b), phenothiazine 11 (dashed line) in (a), and
phenothiazine 19 (dashed line) in (b); the phosphorescence spectra are compared at the same absorption
intensity at the excitation wavelength of 440 nm
Fig. 2). In 2M, the CA carbon atom makes a small but clear contribution to the
HOMO. In contrast, the CA carbon atom in 3M contributes only slightly (invisible
ꢀ
ꢁ
P
in the HOMO in Fig. 2). The sums of the squares of the HOMO coefficients
ci2
of the CA carbon are calculated to be 7.1 9 10-3 for 2M and 1.1 9 10-5 for 3M.
The smaller coefficient of the CA carbon atom in 3M is because of the large dihedral
angle between the phenothiazine ring and 10-phenyl ring [21]. This difference is
discussed further in the section ‘‘Nanosecond transient absorption spectra: lifetime
of the CS states’’.
Redox potentials and energy levels of the excited states and the ion pairs
of PTZ–Pt–MNDIs
The redox potentials of 2 and 3 were determined by cyclic voltammetry
measurements in CH2Cl2 (Table 1). Both triads furnished three reversible reduction
waves and two reversible or irreversible oxidation waves. The first and second
reduction waves are attributed to MNDI (from MNDI to MNDI– then to MNDI2-
)
whereas the third reduction wave is attributed to the platinum complex (from Pt to
Pt-). The first and second oxidation waves are assigned to PTZ (from PTZ to PTZ?,
from PTZ? to PTZ2?). These assignments were made by comparison with the cyclic
voltammograms of suitable reference compounds (9, 21, 22, and 23). The first
tBu
tBu
Me
Me
S
N
Me O
N
O
N
N
N
N
Pt
Pt
Br
N C8H17
O
Me O
A
B
tBu
21
22
23
24
Scheme 3 Chemical formula of 21, 22, 23, and 24
123

Photoinduced charge separation
197
oxidation wave of 2 was reversible whereas that of 3 was irreversible. PTZs with a
terminal acetylene (i.e., 11 and 19) furnished reversible waves for the first oxidation
step. The oxidation potential of the platinum complex was estimated to be ?0.59 V
from the oxidation peak of 21. The peaks for Pt ? Pt? in 2 and 3 appear at ?0.92–
?1.05 V with the oxidation peak for PTZ? ? PTZ2?. The higher oxidation
potential for the Pt atom in 2 and 3 is probably because of the electron-withdrawing
nature of PTZ? which destabilizes PTZ?–Pt?–MNDI.
Taking these redox potentials into account, the energies for the CS states (ECS in
eV) relative to the ground states, in toluene, were estimated by using the Weller
approximation (Eq. 1) [22, 23], in which DGS is a correction term for the polarity of
the solvent (toluene in this case) (Eq. 2). D and A in these equations represent the
donor (i.e., the PTZ or Pt moieties) and the acceptor (i.e., the Pt or MNDI moiety).
ECS ¼ e½EoxðDÞ ꢁ EredðAÞꢂ þ DGs
ð1Þ
ð2Þ
ꢀ
ꢁ
DGS ¼ e2=4pe0 ꢀ ½ð1=2rD þ 1=2rAÞ ꢀ ð1=e ꢁ 1=erÞ ꢁ 1=eRDA
ꢂ
The CS states are regarded as solvent-separated ion-pairs with a center(?)-to-
center(-) distance RDA and effective radii rD and rA. e0 is the permittivity of a
vacuum, and e and er are, respectively, the relative dielectric constants of the solvent
(2.38 for toluene) and reference solvent (8.93 for CH2Cl2) in which the redox
potentials are measured. We assume that the overall structure of the CS state will
not change significantly from that of the neutral state, although the structure of
phenothiazine is planar in the radical cation state [21]. From the molecular model
calculated at the PM6 level, the estimated distances between donors and acceptors
˚
were: RPTZ–MNDI = 21.3 A, RPTZ–Pt = 16.8 A, and RPt–MNDI = 12.4 A for 3PTZ–
˚
˚
˚
Pt–MNDI, and RPTZ–MNDI = 19.2 A, RPTZ–Pt = 14.7 A, and RPt–MNDI = 12.4 A for
˚
˚
Fig. 2 Optimized structures of 2M and 3M and their HOMOs and LUMOs, which were drawn at
isovalue = 0.02. 2M and 3M denote the model structures for 2 and 3, respectively (see text). Hydrogen
atoms are omitted for clarity
123

198
R. Sugimura et al.
Table 1 Redox potentials of 2 and 3
Compound
Eox2 (V)a
Eox1 (V)a
Ered1 (V)a
Ered2 (V)a
Ered3 (V)a
2
?0.92b
?1.05b
?0.59b
?1.01b
?1.03b
?0.21
?0.25c
–
-1.05
-1.50
-1.86
3
-1.05
-1.51
-1.88
21
11
19
9
–
–
-1.86
?0.25
?0.26
?0.23
?0.26
–
–
–
–
–
–
–
–
–
–
c
–
–
–
c
–
22
23
–
–
–
-1.05
-1.50
Measured in CH2Cl2 in the presence of 0.1 M tetra-n-butylammonium perchlorate as supporting
electrolyte
a
V vs Fc/Fc?
b
Irreversible peak potential
c
Not measured
10PTZ–Pt–MNDI. These values assume that the spin centers are located at the
center of the PTZ ring, on the Pt-atom, and at the center of the naphthalene ring. The
˚
effective radii were assumed to be rPt = rD = rA = 4.5 A, which is frequently used
as a typical value [23, 24].
The energy levels of the partial and full CS states and local excited states are shown
in Fig. 3. For both triads, transformation of PTZ–3Pt*–MNDI to the full CS states
(PTZ?–Pt–MNDI–) is exothermic by 0.3 eV. The following path for formation of
the full CS state is energetically possible: PTZ–3Pt*–MNDI ? PTZ–Pt?–MNDI-
?
PTZ?–Pt–MNDI-. The energy of PTZ–Pt–3MNDI* is slightly higher than those of
the full CS states.
Absorption spectra of electrochemically generated phenothiazine radical cation
and naphthalene diimide radical anion
To assign the transient absorption we measured the absorption spectra of 9?, 22?,
and 23– generated under electrochemical oxidation or reduction using a thin-layer
cell (Fig. 4). For radical cation 9?, which corresponds to 3PTZ?, broad absorption
was observed up to approximately 1,000 nm, with peaks at 479, 638, 820, and
920 nm (Fig. 4a). For radical cation 22?, which corresponds to 10PTZ?, absorption
at 517 nm and broad absorption up to approximately 920 nm was observed, with
peaks at 783 and 883 nm (Fig. 4b). The absorption of 9? occurred in the longer
wavelength region than that of 22?, suggesting that the 3PTZ radical cation is more
delocalized because of conjugation with the attached phenylene ring. This
consideration is compatible with the results of MO calculations, which reveal that
the CA carbon atom in 2 has a larger HOMO coefficient than that in 3. Radical anion
23– (i.e., MNDI–) has strong absorption at 473 nm and weak absorption at 606, 690,
and 768 nm (Fig. 4c). The molar absorptivity (e) values of these radical ions are
123

Photoinduced charge separation
199
3PTZ+-Pt–-MNDI
10PTZ+-Pt–-MNDI
2.56
(b)
(a)
2.60
3PTZ-3Pt*-MNDI
2.16
10PTZ-3Pt*-MNDI
2.16
3PTZ-Pt+-MNDI–
2.04
10PTZ-Pt+-MNDI–
2.04
10PTZ-Pt-3MNDI*
3PTZ-Pt-3MNDI*
~2.0
~2.0
3PTZ+-Pt-MNDI–
1.86
10PTZ+-Pt-MNDI–
1.84
Ground State
Ground State
Fig. 3 The energy levels of excited states and CS states for 2 (a) and for 3 (b)
approximate and assume clean conversion from the neutral species. The e value at
the maximum absorption of MNDI– is significantly larger than those of the PTZ?s.
Picosecond transient absorption spectra: formation of the CS states
Picosecond laser photolysis was performed with excitation at 532 nm (FWHM:
17 ps), which selectively excites the Pt-complex chromophore. Figure 5a, b shows
the time-resolved difference absorption spectra for 2 and 3, respectively. Imme-
diately after the laser pulse (0 ps), broad absorption appears in the 550–800 nm
region for both 2 and 3. According to our previous study [12], this absorption is
assigned to the 3Pt* state. The 3Pt* absorption is replaced, during 100–1,000 ps, by
sharper absorption at 470, *610, *690, and *780 nm for excitation of 2 (Fig. 5a).
Similar spectral changes were observed for excitation of 3 (Fig. 5b). In both spectra,
the sharp absorption at 470 and 610 nm is clearly assigned to MNDI- by
comparison with the spectra in the electrochemical reduction of 23 (Fig. 4c). The
absorption of 3PTZ? (from 2) and 10PTZ? (from 3) is ambiguous because it is
small and slightly broad (Fig. 4a, b) and overlaps with the larger and sharper
absorption of MNDI-. The spectra observed 1 ns after the laser pulse are indicative
Fig. 4 UV–visible spectra of the electrochemically generated radical cation 9? (a) and of 22? (b), and of
radical anion 23- (c). The gray solid, dotted, and black lines show the initial, intermediate, and final
states, respectively
123

200
R. Sugimura et al.
Fig. 5 Picosecond transient difference absorption spectra for 2 (a) and 3 (b) in toluene under 532 nm
laser excitation (FWHM: 17 ps) at room temperature using a 1-mm cell
of the formation the CS states (i.e., 3PTZ?–Pt–MNDI- and 10PTZ?–Pt–MNDI-).
3
The decay of Pt* does not differ significantly between 2 and 3 as both have time
constants of 200–400 ps. The CS time constants were determined to be 280 ps
(kCS = 3.6 9 10-9 s-1) for 2 and 230 ps (kCS = 4.4 9 10-9 s-1) for 3. The
quantum yields (U) for the formation CS states were estimated to be U [ 0.9 for
both 2 and 3.1 These rate constants and quantum yields are similar to those of
compound 1 [12]. The phosphorescence time-profile for 2 and 3 at room
temperature under nitrogen had comparable lifetimes: 280 ps for 2 and 230 ps for 3.
Nanosecond transient absorption spectra: lifetime of the CS states
Figure 6 shows the nanosecond transient absorption of 2 and 3 in toluene with laser
excitation at 532 nm (FWHM: 4 ns). Nanosecond photolysis produced sharp
absorption at 470 and 610 nm, assigned to MNDI-, which appeared 20 ns after
excitation of 2 (Fig. 6a). A similar spectrum was observed for excitation of 3
(Fig. 6b at 50 ns). The absorption in the 600–700 nm region in Fig. 6a at 20 ns is
broader than that in Fig. 6b at 50 ns. This difference reflects the overlap between the
MNDI- and 3PTZ? absorption in Fig. 6a: 3PTZ? absorbs at *638 nm, whereas
10PTZ? absorbs at *517 nm and has no strong absorption at *650 nm. The
spectra observed are thus consistent with those expected for the CS states.
The time constants for the CR process to the ground states (sCR) were estimated:
The decay curve of 3PTZ?–Pt–MNDI- (from 2) was best simulated as a
biexponential curve: sCR1 = 75 ns (95 %) and sCR2 = 285 ns (5 %), whereas the
decay for 10PTZ?–Pt–MNDI- (from 3) was simulated as a single exponential:
1
The quantum yields for the formation of CS states for the excitation of 2 and 3 were estimated by using
the relative intensities of 3Pt* (650 nm) and MNDI- (470 nm) in the ps-laser photolysis (Fig. 5).
123

Photoinduced charge separation
201
20 ns
50 ns
80 ns
120 ns
200 ns
800 ns
50 ns
400 ns
800 ns
1200 ns
2000 ns
3000 ns
(a)
(b)
+0.15
+0.1
+0.1
+0.05
0
+0.05
0
400
500
600
700
800
400
500
600
700
800
Wavelength / nm
Wavelength / nm
Fig. 6 The nanosecond transient difference absorption spectra of 2 (a) and for 3 (b) in toluene under
532 nm laser excitation (FWHM: 4 ns) at room temperature using a 10-mm cell
sCR = 830 ns. The lifetime of 10PTZ?–Pt–MNDI- was found to be substantially
longer than that of 3PTZ?–Pt–MNDI-. The longer lifetime of 10PTZ?–Pt–MNDI-
(sCR = 830 ns, single exponential in toluene) was nearly identical with that
(sCR = 839 ns, single exponential in toluene) of the CS state from 1 [12].
The origin of the biexponential decay observed for 3PTZ?–Pt–MNDI- is not
clear and may be because of conformational isomers around the acetylenic bonds or
around the C–C bond directly connected to the 3PTZ group, as discussed in the
previous section.
Mechanistic consideration of the photoinduced CS and CR processes
In principle, two mechanisms for formation of CS state are possible:
1. PTZ–3Pt*–MNDI ? PTZ–Pt?–MNDI- ? PTZ?–Pt–MNDI-; and
2. PTZ–3Pt*–MNDI ? PTZ–Pt–3MNDI* ? PTZ?–Pt–MNDI-.
In our previous study of a related system (i.e., MTA–Pt–MNDI), the radical ions
(MTA? and MNDI-) were observed in different absorption regions. Formation of
the radical ions was found to have different time constants, indicative of mechanism
1 [12]. In this case, decay of PTZ–3Pt*–MNDI was shown to be 200–400 ps in the
picosecond transient spectra for 2 and 3 (Fig. 5). The phosphorescence of
PTZ–3Pt*–MNDI decays with a similar time constant. The absorption of MNDI-
increases with a similar time constant. These results are compatible with mechanism
1. Mechanism 2 involves exothermic energy transfer from PTZ–3Pt*–MNDI to
PTZ–Pt–3MNDI* in the 1st step and long-distance electron transfer from PTZ to
3MNDI* in the 2nd step. The rate-determining step is likely to be in the 2nd step. In
such a case, PTZ–Pt–3MNDI* is expected to be observed. However, no evidence
was obtained for the formation of PTZ–Pt–3MNDI*, therefore, the CS formation for
these triads is likely to follow mechanism 1.
The driving forces for the CR process were estimated from the energy levels of
PTZ?–Pt–MNDI- (DGCR & -1.85 eV for both the 3PTZ and 10PTZ derivatives
(Table 2; Fig. 3)). However, the sCR value of 10PTZ?–Pt–MNDI- (830 ns) is much
longer than those of 3PTZ?–Pt–MNDI- (sCR1 = 75 ns (95 %) and sCR2 = 285 ns
123

202
R. Sugimura et al.
Table 2 The distances between redox moieties and the energy for the CS states
˚
RD–A (A)
˚
RPt–A (A)
˚
RD–Pt (A)
þꢁAꢁ
ED
þꢁAꢁ
EPt
þꢁPtꢁ
ED
Compound
(eV)
(eV)
(eV)
2
3
21.3
19.2
12.4
12.4
16.8
14.7
1.86
1.84
2.04
2.05
2.60
2.56
(5 %)). The difference between the decay time constants could be because of the
electronic coupling matrix element VDA, which is involved in the pre-exponential
term in the Marcus theory. 3PTZ?–Pt–MNDI- (from 2) would have a larger VDA
value than 10PTZ?–Pt–MNDI- (from 3). This consideration is supported by the
two observations:
1. the CA atom (see formulas 2 and 3 in Fig. 2) in the twisted biphenyl linker has a
larger HOMO coefficient in 2 than in 3; and
2. 9?, which is a model of 3PTZ? from 2, has an absorption maximum at a longer
wavelength than 22?, which is a model of 10PTZ? from 3.
The larger VDA value in 3PTZ?–Pt–MNDI- may affect the CS process and the
CR process. However, the CS time constants of 2 and 3 are not significantly
different in the picosecond transient absorption. A larger electronic effect on the CR
process than on the CS process has frequently been observed for several related
dyads [25, 26].
In our previous report on MTA–Pt–MNDI [12], the CS state was demonstrated to
be in a SCRP state. Although time-resolved electron paramagnetic resonance
(TREPR) studies were not carried out in this study, it is very likely that this system
also proceeds via an SCRP state. The SCRP state is an equilibrium state between
0
0
singlet–triplet mixed states, T (PTZ?–Pt–MNDI-) and S (PTZ?–Pt–MNDI)- (the
superscripts S0 and T0 denote singlet-predominant and triplet-predominant states,
respectively), at zero magnetic field. The relatively long lifetime observed for the
CS state (especially for 3) reflects not only the electronic interaction VDA but also
the spin-state differences between the SCRP and the ground states. The observed
lifetimes of the CS states also depend on the rate of intersystem crossing in these
radical pairs [27–30, and references cited therein]. Further studies to elucidate the
mechanism of the spin-sorted charge recombination are in progress.
Conclusion
We prepared two triads, 3PTZ–Pt–MNDI (2) and 10PTZ–Pt–MNDI (3). The CS
state formation and decay were studied in toluene using picosecond and nanosecond
laser photolysis with selective excitation of the Pt-moiety. Formation of the CS
states, PTZ?–Pt–MNDI- from PTZ–3Pt*–MNDI, was observed to have similar time
constants: sCS = 280 ps for 3PTZ?–Pt–MNDI- and sCS = 230 ps for 10PTZ?–Pt–
MNDI-. However, the lifetimes of the CS states were found to be very different:
sCR1 = 75 ns (95 %) and sCR2 = 285 ns (5 %) for 3PTZ?–Pt–MNDI- and
123

Photoinduced charge separation
203
sCR = 830 ns for 10PTZ?–Pt–MNDI-. The longer lifetime of the CS state of
10PTZ?–Pt–MNDI- (from 3) suggests the importance of the electronic coupling
matrix element VDA and the spin-state difference between the possible SCRP state
and the ground state.
Acknowledgments This work was supported by Grants-in-Aid: for Scientific Research (B) (no.
2350066 for K.O.), for Young Scientists (B) (no. 22750043 for S.S.), and for JSPS Fellows (no. 238419
for R.S.) from the Japan Society for Promotion of Science (JSPS). S.S. is also grateful for financial
support from the Kansai Research Foundation for Technology Promotion and the Association for the
Progress of New Chemistry.
References
1. K.E. McAuley, P.K. Fyfe, J.P. Ridge, N.W. Isaacs, R.J. Cogdell, M.R. Jones, Proc. Natl. Acad. Sci.
USA 96, 14706–14711 (1999)
2. G.J. Kavarnos, Fundamentals of Photoinduced Electron Transfer (Wiley, New York, 1993)
3. J.W. Verhoeven, J. Photochem. Photobiol. C 7, 40–60 (2006)
˚
4. M. Borgstrom, N. Shaikh, O. Johansson, M.F. Anderlund, S. Styring, B. Akermark, A. Magnuson, L.
¨
¨
Hammarstrom, J. Am. Chem. Soc. 127, 17504–17515 (2005)
5. L. Flamigni, J.-P. Collin, J.-P. Sauvage, Acc. Chem. Res. 41, 857–871 (2008)
6. L. Flamigni, E. Baranoff, J.-P. Collin, J.-P. Sauvage, Chem. Eur. J. 12, 6592–6606 (2006)
7. S. Chakraborty, T.J. Wadas, H. Hester, R. Schmehl, R. Eisenberg, Inorg. Chem. 44, 6865–6878
(2005)
8. J.E. McGarrah, Y.-J. Kim, M. Hissler, R. Eisenberg, Inorg. Chem. 40, 4510–4511 (2001)
9. M. Hissler, J.E. McGarrah, W.B. Connick, D.K. Geiger, S.D. Cummings, R. Eisenberg, Coord.
Chem. Rev. 208, 115–137 (2000)
10. S.D. Cummings, R. Eisenberg, J. Am. Chem. Soc. 118, 1949–1960 (1996)
11. W. Paw, S.D. Cummings, M.A. Mansour, W.B. Connick, D.K. Geiger, R. Eisenberg, Coord. Chem.
Rev. 171, 125–150 (1998)
12. S. Suzuki, R. Sugimura, M. Kozaki, K. Keyaki, K. Nozaki, N. Ikeda, K. Akiyama, K. Okada, J. Am.
Chem. Soc. 131, 10374–10375 (2009)
13. H. Kawauchi, S. Suzuki, M. Kozaki, K. Okada, D.-M.S. Islam, Y. Araki, O. Ito, K. Yamanaka, J.
Phys. Chem. A 112, 5878–5884 (2008)
14. Y. Nomura, Y. Takeuchi, J. Chem. Soc. B 956–960 (1970)
15. M. Hissler, W.B. Connick, D.K. Geiger, J.E. McGarrah, D. Lipa, R.J. Lachicotte, R. Eisenberg,
Inorg. Chem. 39, 447–457 (2000)
16. MOPAC2009, J.J.P. Stewart, Stewart Computational Chemistry, version 9.211W (2009). http://
OpenMOPAC.net, Accessed June 4, 2012
17. T. Ohno, K. Nozaki, M. Haga, Inorg. Chem. 31, 548–555 (1992)
18. A. Yoshimura, K. Nozaki, N. Ikeda, T. Ohno, J. Phys. Chem. 100, 1630–1637 (1996)
19. J.J.H. McDowell, Acta. Cryst. B32, 5–10 (1976)
20. C.L. Klein, J.M. Conrad III, S.A. Morris, Acta. Cryst. C41, 1202–1204 (1985)
21. T. Okamoto, M. Kuratsu, M. Kozaki, K. Hirotsu, A. Ichimura, T. Matsushita, K. Okada, Org. Lett. 6,
3493–3496 (2004)
22. A.Z. Weller, Phys. Chem. Neue Folge 133, 93–98 (1982)
23. M. Lor, L. Viaene, R. Pilot, E. Fron, S. Jordens, G. Schweitzer, T. Weil, K. Mu¨llen, J.W. Verhoeven,
M. Van der Auweraer, F.C. De Schryver, J. Phys. Chem. B 108, 10721–10731 (2004)
24. H. Oevering, M.N. Paddon-Row, M. Heppener, A.M. Oliver, E. Cotsaris, J.W. Verhoeven, N.S.
Hush, J. Am. Chem. Soc. 109, 3258–3269 (1987)
´
25. A.L. Thompson, T.-S. Ahn, K.R.J. Thomas, S. Thayumanavan, T. Martınez, C.J. Bardeen, J. Am.
Chem. Soc. 127, 16348–16349 (2005)
26. A.B. Ricks, G.C. Solomon, M.T. Colvin, A.M. Scott, K. Chen, M.A. Ratner, M.R. Wasielewski, J.
Am. Chem. Soc. 132, 15427–15434 (2010)
27. G.L. Closs, M.D.E. Forbes, J.R. Jr. Norris, J. Phys. Chem. 91, 3592–3599 (1987)
123

204
R. Sugimura et al.
28. P.J. Hore, in Advanced EPR in Biology and Biochemistry, ed. by A.J. Hoff (Elsevier, Amsterdam,
1989), pp. 405–440
29. Z.E.X. Dance, Q. Mi, D.W. McCamant, M.J. Ahrens, M.A. Ratner, M.R. Wasielewski, J. Phys.
Chem. B 110, 25163–25173 (2006)
30. T. Miura, A.M. Scott, M.R. Wasielewski, J. Phys. Chem. C 114, 20370–20379 (2010)
123