Probing triplet state properties of organic chromophores via design and synthesis of Os(II)-diketonate complexes: The triplet state intramolecular charge transfer

Article abstract of DOI:10.1021/jp0453111

We report the probe of specific triplet state properties of organic chromophores that are otherwise inaccessible in low viscous solution. The prototypical example demonstrated here is [Os(CO)₃(Cl)(NDP)] (1) ((NDP)H = 2-naphthyl-7-dimethylanilin

Full text of DOI:10.1021/jp0453111

19908
J. Phys. Chem. B 2004, 108, 19908-19911
Probing Triplet State Properties of Organic Chromophores via Design and Synthesis of
Os(II)-Diketonate Complexes: The Triplet State Intramolecular Charge Transfer
Jen-Kan Yu,† Yi-Ming Cheng,† Ya-Hui Hu,† Pi-Tai Chou,*,† Yao-Lun Chen,‡
Shin-Wun Lee,‡ and Yun Chi*,‡
Department of Chemistry, National Taiwan UniVersity, Taipei 106, Taiwan, and Department of Chemistry,
National Tsing Hua UniVersity, Hsinchu 30043, Taiwan
ReceiVed: October 14, 2004
We report the probe of specific triplet state properties of organic chromophores that are otherwise inaccessible
in low viscous solution. The prototypical example demonstrated here is [Os(CO)3(Cl)(NDP)] (1) ((NDP)H )
2-naphthyl-7-dimethylanilino-1,3-propanedione), which, upon electronic excitation, undergoes intramolecular
charge transfer in both S1 and T1 manifolds of NDP. The dipolar changes in S1 and T1 monitored via the
solvatochromism for both fluorescence and phosphorescence were deduced to be 18.0 and 11.9 D, respectively.
The appreciable difference in the dipolar change can be qualitatively rationalized by different extents of
charge-transfer character between S1 and T1 states. The results led to the probe of other reactions in triplet
manifold feasible.
Third-row heavy metal containing luminescent complexes
have increasingly gained attention due to their potential ap-
plications in electroluminescent devices.1 The strong spin-orbit
coupling effectively promotes an intersystem crossing from
singlet excited states to lower triplet emitting states, namely
3MLCT, 3ππ*, or a mixture of the two.2 The resulting materials
are suitable for use in the preparation of OLEDs with unprec-
edented phosphorescence efficiencies attained by harnessing
both triplet and singlet excitons, theoretically giving 100%
internal quantum efficiency.3
donating carbonyl ligands should further increase the d-d
energy gap,7 so that the organic chromophore of interest, i.e.,
â-diketonate, is expected to lie in the lower-lying excited states.
The â-diketonate ligand with both donor and acceptor substit-
uents was synthesized through a Claisen condensation reaction,
in which treatment of 2-acetonaphthone and 4-dimethylami-
nobenzoic acid ethyl ester in the presence of a strong base NaH
gives the desired diketone, 2-naphthyl-7-dimethylanilino-1,3-
propanedione (NDP)H. To prepare the Os(II)-metal complexes,
we selected the literature method designed for the related
dibenzoylmethanate complex [Os(CO)3(tfa)(dbm)], tfa )
CF3CO2- and (dbm)H ) dibenzoylmethane.8 In this approach,
heating the finely pulverized solid-state mixture of (NDP)H and
osmium dimer [Os2(CO)6(tfa)2] afforded [Os(CO)3(tfa)(NDP)]
(1a). Treatment of 1a with NaCl in refluxing THF solution
afforded the ligand substitution product [Os(CO)3(Cl)(NDP)]
(1) in ∼80% yield. Moreover, treatment of (NDP)H with BPh3
in THF solution at room temperature gave the respective BPh2
complex 2 in 90% yield (see Supporting Information).
Figure 1 shows absorption and emission spectra of complex
1 in various solvents. As solvent polarity increased, in contrast
to a slight bathochromic shift of the absorption spectral features,
a drastic solvent-polarity dependence in emission spectra was
observed. In aerated cyclohexane, 1 exhibited distinct dual
emission with λmax at 470 and 570 nm. While the intensity of
the 470 nm band remained unchanged, the 570 nm band revealed
drastic O2-concentration dependence and became dominant in
the degassed solution (see Figure 1). The O2 quenching rate
was calculated to be as large as 1.4 × 109 M-1 s-1 (Supporting
Information). Accordingly, the 470 and 570 nm bands in 1
unambiguously are ascribed to the fluorescence and phospho-
rescence, respectively. Table 1 lists the photophysical properties
of the studied complexes. In degassed cyclohexane the phos-
phorescence lifetime for 1 was measured to be ∼60 µs. This,
in combination with a near unity of the phosphorescence
quantum yield, renders a radiative lifetime of ∼60 µs, which is
much shorter than that of the triplet ππ* for the free NDPH
ligand (∼0.1 s measured in the 77 K methylcyclohexane glass).
The results support the enhancement of the T1 f S0 transition
We here would like to advance another angle of approach
toward probing the fundamentals of the triplet-state properties
based on the design of Os(II) complexes. Our strategy is to
functionalize ligands connected to the heavy metal, so that the
lowest triplet-state energy level can be fine tuned and thus lie
mainly in the ππ* (or nπ*) chromophore of the designated
ligand. On this basis, one can ingeniously design a special type
of triplet-state reaction, the associated properties of which can
thus be dictated by the prominent phosphorescence. A case in
point is the triplet state intramolecular charge transfer (TSICT)
reaction,4 the investigation of which is rare due to the spin
forbidden nature of the T1f S0 transition. Several methods have
been employed to circumvent the lack of phosphorescence in
solution. The micelle or polymer encapsulated organic chro-
mophore, together with the external heavy atom effect, may
resolve room-temperature phosphorescence.5 However, the
resulting heterogeneous and rigid environments make studies
of the solvation-associated properties impractical. As a result,
photophysical properties associated with TSICT, such as the
solvation relaxation dynamics as well as the steady-state
consequence, e.g., the phosphorescence solvatochromism, are
not attainable in low viscous solvents. To achieve this goal, we
have designed and synthesized an asymmetrical â-diketonate
ligand, composed of an N,N-dimethylaniline moiety that serves
as the electron donor,6 and obtained the corresponding Os(II)
carbonyl-metal complexes. The addition of back π-bond
* Corresponding author. E-mail: chop@ntu.edu.tw; ychi@mx.nthu.edu.tw.
† National Taiwan University.
‡ National Tsing Hua University.
10.1021/jp0453111 CCC: $27.50 © 2004 American Chemical Society
Published on Web 12/02/2004

Letters
J. Phys. Chem. B, Vol. 108, No. 52, 2004 19909
Figure 1. Absorption and steady-state emission spectra of 1 at 298 K
degassed cyclohexane (-9-), dichloromethane (-O-), and acetonitrile
(-4-). (-0-) denotes the emission spectrum of 1 in aerated cyclohexane.
TABLE 1: Photophysical Properties of Complexes 1, 1a,
and 2
max
em
(nm)
max
em
a
b
λ
λ
a
b
λabs
(nm) Φ b,c
τf
τp
1
C6H12
C6H6
THF
CH2Cl2 436 543
CH3CN 422 580
424 470, 570 574
∼1
15.4 (ps)d 60 (µs)
432 508
430 537
593
609
611
649
0.84
0.40
0.58
15.6 (ps)d 125 (µs)
104 (ps)
110 (ps)
74 (µs)
86 (µs)
0.005 22.0 (ps)d 70 (µs)
1a C6H12
419 497, 583 577
∼1
5.13 (ps)d 105 (µs)
113 (ps) 201 (µs)
CH2Cl2 435 532
CH3CN 432 575
C6H12 434 467
CH2Cl2 461 537
CH3CN 466 572
603
646
467
537
572
0.61
0.015 8.52 (ps)d 50 (µs)
2
0.12
0.39
0.005 0.21 (ns)
1.23 (ns)
1.07 (ns)
Figure 2. Time-gated fluorescence (upper) and phosphorescence
(lower) of 1 in degassed (a) C6H12, (b) C6H6, (c) CH2Cl2, (d) THF, and
(e) CH3CN. (inset) The plot of peak frequency versus solvent polarity
∆f.
a Data were obtained in aerated solution. b Data were obtained in
degassed solution (via three freeze-pump-thaw cycles). c Quantum
yield was obtained for fluorescence plus phosphorescence. d Due to the
system response of 30 ps, these values obtained from curve fitting are
subject to large uncertainties.
SCHEME 1: Complexes 1-2 and the Proposed
Prototypes to Study Charge (3) and Proton (4) Transfer
in the Triplet State
dipole in 1 due to the Os heavy atom effect. Conversely, the
fluorescence lifetime of 1 in cyclohexane is beyond the system
response of 30 ps (see Table 1). In comparison, the boron
complex 2 (see Scheme 1) bearing the same NDP ligand exhibits
a unique fluorescence, the lifetime of which is as long as 1.23
ns in cyclohexane. Assuming that the relaxation dynamics of
S1 in 1 is dominated by rate of S1 f Tn, the results also manifest
the enhancement of spin-orbit coupling on the S1 f Tn
intersystem crossing due to the Os heavy atom effect.
In degassed, polar solvents, fluorescence and phosphorescence
strongly overlap, giving rise to a broad emission (Figure 1).
Fortunately, as shown in Table 1, the different relaxation
dynamics of fluorescence and phosphorescence allow us to
temporally resolve each component (Supporting Information).
In the degassed solution, the spectrum taken at a delayed time
of 5 µs after the excitation ensures the acquisition of phospho-
rescence free from fluorescence interference. Conversely, gating
the time-resolved spectrum at the first ∼10 ns should obtain a
nearly phosphorescence-free fluorescence. The results shown
in Figure 2 clearly reveal remarkable solvent dependency, and
the plot of peak frequency for both fluorescence and phospho-
rescence is linearly proportional to solvent polarity (∆f)9 (see
inset of Figure 2). Accordingly, changes in dipole moment with
respect to the ground state were deduced to be 18.0 and 11.9 D
for S1 and T1, respectively, with the Lippert-Mataga equation.10,SI

19910 J. Phys. Chem. B, Vol. 108, No. 52, 2004
Letters
TABLE 2: Calculated Electronic Transitions and Associated Frontier Orbitals for Complex 1a
a A. TD-B3LYP method based on the geometry optimized S0 state (B3LYP). B. TD-B3LYP method based on the geometry optimized T1 state
(UB3LYP).
As shown in Table 1, a similar ICT reaction was observed for
1a, in which the axial ligand in 1 is replaced by CF3CO2-, and
the dipolar changes for S1 and T1 were derived to be 17.8 and
11.8 D, respectively. In comparison, only fluorescence solva-
tochromism resulting from ICT in the excited singlet state was
observed for boron complex 2, and the calculated dipolar change
of 18.4 D is consistent with that in 1 (or 1a) (Supporting
Information).
The appreciable difference in the dipolar change is intriguing,
as it may imply the intrinsic difference in the electron density
distribution between S1 and T1 for both 1 and 1a. To show such
a possibility, Table 2 depicts the composition and structures of
the crucial frontier molecular orbitals for 1 involved in the
transition of low-lying excited states using the TD-B3LYP//
B3LYP/6-31G(d′, p′) method (Supporting Information). With
this method, the Franck-Condon excitation energy from the
ground state to the S1 state was calculated to be 3.1 eV (400
nm), which is qualitatively consistent with the corresponding
experimental results of 424 nm (S1) in cyclohexane. Since the
phosphorescence originates from the geometry-relaxed T1 state,
we also performed the geometry optimization of the T1 state,
followed by a single-point calculation of the T1-S0 transition.
As shown in Table 2, differences in the frontier orbital
configuration between S1 and T1 are immediately discernible.
The HOMO and LUMO frontier orbitals in the singlet manifold
possess a π configuration in the amino phenyl ligand and a π*
pattern mainly in the 2-naphthyl ligand. One thus would expect
that the S0 f S1 transition incorporated an electron-transfer
process from the amino phenyl group to the 2-naphthyl moiety.11
In contrast, although to a large extent (90%) similar HOMO
and LUMO are involved in the S0 f T1 transition, contribution
from other frontier orbitals are also substantial. The involvement
of HOMO-1 orbital possessing a π configuration of the
2-naphthyl and MLCT is ∼10% (Supporting Information).
Different extents of charge transfer character render distinctly
different solvatochromism effects, as observed experimentally.
The success in probing TSICT in complexes 1 and 1a
warrants the design of other TSICT prototypes, among which
complex 3 bearing various degrees of donor/acceptor substit-
uents (Scheme 1) should receive considerable attention due to
their facile syntheses.12 The Os(CO)n fragment can thus be
exploited as a building block to probe other types of reactions
in the triplet manifold. For example, theoretical approaches have
predicted that, relative to the fast excited-state proton transfer
(ESPT) in the singlet state, the triplet-state proton transfer should
be much slower due to the decrease in proton acidity.13
Unfortunately, cases of triplet state proton transfer are very rare14
because of the fast, dominant proton-transfer process in the
singlet-excited state. If the rate of S1 f Tn ISC can compete
with the proton-transfer rate in a singlet manifold, studies of

Letters
J. Phys. Chem. B, Vol. 108, No. 52, 2004 19911
(3) (a) Lamansky, S.; Djurovich, P.; Murphy, D.; Abdel-Razzaq, F.;
Lee, H.-E.; Adachi, C.; Burrows, P. E.; Forrest, S. R.; Thompson, M. E. J.
Am. Chem. Soc. 2001, 123, 4304. (b) Brooks, J.; Babayan, Y.; Lamansky,
S.; Djurovich, P. I.; Tsyba, I.; Bau, R.; Thompson, M. E. Inorg. Chem.
2002, 41, 3055.
(4) (a) Tyson, D. S.; Luman, C. R.; Zhou, X.; Castellano, F. N. Inorg.
Chem. 2001, 40, 4063. (b) Song, L.-Q.; Feng, J.; Wang, X.-S.; Yu, J.-H.;
Hou, Y.-J.; Xie, P.-H.; Zhang, B.-W.; Xiang, J.-F.; Ai, X.-C.; Zhang, J.-P.
Inorg. Chem. 2003, 42, 3393. (c) Goodall, W.; Williams, J. A. G. Chem.
Commun. 2001, 2514. (e) Del Guerzo, A.; Leroy, S.; Fages, F.; Schmehl,
R. H. Inorg. Chem. 2002, 41, 359.
the triplet-state proton-transfer become feasible. Accordingly,
a proposed structure like complex 4, in which the phenol-ligand
is expected to undergo ESPT, forming a quinone isomer,15
should be intriguing to probe the proton-transfer dynamics in
the triplet state manifold. We thus believe that results presented
in this study can spark a broad range of interest in both
fundamental approach and synthetic field relevant to third-row
transition metal complexes.
(5) (a) Shtykov, S.; Melnikov, G.; Goryacheva, I. J. Mol. Struct. 1999,
482-483, 699. (b) Murillo Pulgarin, J. A.; Molina, A. A.; Alanon Pardo,
M. T. Anal. Chim. Acta 2000, 423(1), 85. (c) Perez-Ruiz, T.; Martinez-
Lozano, C.; Tomas, V.; Martin, J. J. Pharm. Biomed. Anal. 2003, 32(2),
225.
(6) Sato, Y.; Morimoto, M.; Segawa, H.; Shimidzu, T. J. Phys. Chem.
1995, 99, 35.
Acknowledgment. This work was supported by the National
Science Council, Taiwan, R.O.C. (NSC93-2119-M-002-002).
We thank the National Center for High-Performance Computing,
Taiwan, for the use of their facility.
Supporting Information Available: Syntheses, photophysi-
cal experimental details, the Lippert-Mataga plot, and crystal
data of 2. This material is available free of charge via the Internet
at http://pubs.acs.org.
(7) Zalis, S.; Farrell, I. R.; Vlcek, A. J. Am. Chem. Soc. 2003, 125,
4580.
(8) Chen, Y.-L.; Sinha, C.; Chen, I.-C.; Liu, K.-L.; Chi, Y.; Yu, J.-K.;
Chou, P.-T.; Lu, T.-H. Chem. Commun. 2003, 3046.
(9) ∆f is the solvent polarity parameter function and is generally
expressed as ∆f ) (ꢀ - 1)/(2ꢀ + 1), where ꢀ denotes the static dielectric
constant of the solvent (Supporting Information).
(10) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Plenum
Press: New York, 1983.
(11) Meyer, T. J. Pure Appl. Chem. 1986, 50, 1293.
(12) Yu, W.-S.; Cheng, C.-C.; Cheng, Y.-M.; Wu, P.-C.; Song, Y.-H.;
Chi, Y.; Chou, P.-T. J. Am. Chem. Soc. 2003, 125, 10800.
(13) For example, see Scheiner, S. J. Phys. Chem. A 2000, 104, 5898.
(14) Grellmann, K. H.; Mordzinski, A.; Heinrich, A. Chem. Phys. 1989,
136, 201.
References and Notes
(1) (a) Baldo, M. A.; O’Brien, D. F.; You, Y.; Shoustikov, A.; Sibley,
S.; Thompson, M. E.; Forrest, S. R. Nature 1998, 395, 151. (b) Welter, S.;
Brunner, K.; Hofstraat, J. W.; De Cola, L. Nature 2003, 421, 54.
(2) (a) Liu, Y.; Jiang, S.; Glusac, K.; Powell, D. H.; Anderson, D. F.;
Schanze, K. S. J. Am. Chem. Soc. 2002, 124, 12412. (b) Haskins-Glusac,
K.; Ghiviriga, I.; Abboud, K. A.; Schanze, K. S. J. Phys. Chem. B 2004,
108, 4969. (c) McMillin, D. R.; Moore, J. J. Coord. Chem. ReV. 2002,
229, 113. (d) Amini, A.; Harriman, A.; Mayeux, A. Phys. Chem. Chem.
Phys. 2004, 6, 1157.
(15) 15. Kaczmarek, L.; Borowicz, P.; Grabowska, A.; J. Photochem.
Photobiol. A: Chem. 2001, 138, 159, and references therein.