Ligand-to-ligand charge transfer in heteroleptic Ir-complexes: Comprehensive investigations of its fast dynamics and mechanism

Article abstract of DOI:10.1039/c6cp02087a

To gain new insights into ligand-to-ligand charge-transfer (LLCT) dynamics, we synthesised two heteroleptic Ir³⁺ complexes: (Ir(dfppy)₂(tpphz)) and (Ir(dfppy)₂(dpq)), where dfppy, tpphz, and dpq are 2-(4,6-difluorophenyl)p

Full text of DOI:10.1039/c6cp02087a

PCCP
View Article Online
View Journal | View Issue
PAPER
Ligand-to-ligand charge transfer in heteroleptic
Ir-complexes: comprehensive investigations of its
fast dynamics and mechanism†
Cite this: Phys.Chem.Chem.Phys.,
2016, 18, 15162
Yang-Jin Cho,a So-Yoen Kim,a Minji Cho,a Kyung-Ryang Wee,b Ho-Jin Son,a
Won-Sik Han,c Dae Won Cho*ad and Sang Ook Kang*a
To gain new insights into ligand-to-ligand charge-transfer (LLCT) dynamics, we synthesised two
heteroleptic Ir3+ complexes: (Ir(dfppy)2(tpphz)) and (Ir(dfppy)2(dpq)), where dfppy, tpphz, and dpq
are 2-(4,6-difluorophenyl)pyridine, tetrapyrido[3,2-a:20,30-c:300,200-h:20 0 0,30 0 0-j]phenazine, and 2,3-bis-(2-
pyridyl)-quinoxaline, respectively. The tpphz and dpq ligands have longer p-conjugation than dfppy.
Therefore, the excited ligand-centred (LC) state and the metal-to-ligand charge transfer (MLCT) state of
dfppy are higher than those of tpphz and dpq. The LLCT dynamics from dfppy to tpphz (or dpq) was
probed using femtoscond transient absorption (TA) spectroscopy after the selective excitation of dfppy.
The TA band for the LC/MLCT state of dfppy is observed at 480 nm. Because of the LLCT process, the TA
bands related to the MLCT states of tpphz and dpq ligands increased, whereas those of dfppy decreased. The
time constants for the LLCT process were 17 ps for Ir(dfppy)2(tpphz) and 5 ps for Ir(dfppy)2(dpq). The MLCT
emission of Ir(dfppy)2(tpphz) showed strong temperature dependence, indicating that the LLCT process has a
significant energy barrier. In comparison, the temperature weakly influenced the emission of Ir(dfppy)2(dpq),
and thus, its LLCT process may have a smaller barrier. The anomalous rigidochromism and photodynamic
behaviours can be explained in terms of the barrier between cyclometalating and ancillary ligands.
Received 30th March 2016,
Accepted 6th May 2016
DOI: 10.1039/c6cp02087a
www.rsc.org/pccp
One of the most effective methods for tuning the energy
of the emissive excited state involves changing the degree of
Introduction
The photophysical properties of heteroleptic Ir(III) complexes p-conjugation in the structures of the participating ligands.3–5
with cyclometalating and ancillary ligands have been extensively First, the p* orbitals of cyclometalating ligands contribute to
investigated for applications in organic light-emitting diodes the excited-state emissive transitions. Therefore, low-energy
(OLEDs) because varying the ligand combination can produce Ir emission can be readily achieved in metal complexes by using
complexes that efficiently emit phosphorescence throughout the cyclometalating ligands with extended p-conjugation.6,7
visible region.1 The phosphorescent states of metal complexes are
The ancillary ligand (e.g., picolinate [pic] or 2,4-pentadione
usually attributed to their metal-to-ligand charge-transfer state [acac]) plays a relatively passive role in determining the nature
(MLCT, d–p*) mixing with the strongly spin-orbit-coupled triplet of the emissive state.7 However, these ligands have been used to
manifold of the cyclometalating ligand.2 The upper excited adjust the energy of the MLCT state and the degree of mixing
states, such as the ligand-centred (LC, p–p*) state and the between the excited states (LC and MLCT). Indeed, Park and
singlet MLCT (1MLCT) state, are assumed to eventually reach co-workers8 observed emission throughout the visible spectrum
the emissive 3MLCT state through an efficient intersystem using a series of complexes with ancillary ligands with different
crossing aided by the heavy metal effect.
structures. Prior to emission, ligand-to-ligand charge transfer
(LLCT) may occur from one ligand to another, i.e., from the
cyclometalating ligand to the ancillary ligand or vice versa. In
this case, the net properties of the excited metal-complexes can
be controlled almost exclusively by the ancillary ligand.
In this regard, Park et al. reported that slow LLCT (B10 ns)
is envisaged among orbitals localized on the cyclometalating
and ancillary ligands before the observed phosphorescence is
generated.9,10 However, Chi and co-workers insisted that the
a Department of Advanced Materials Chemistry, Korea University (Sejong), Sejong,
30019, South Korea. E-mail: hjson@korea.ac.kr, sangok@korea.ac.kr
b Department of Chemistry, Daegu University, Gyeongsan, 38453, South Korea.
E-mail: krwee@daegu.ac.kr
c Department of Chemistry, Seoul Women’s University, Seoul, 01797, South Korea.
E-mail: wshan@swu.ac.kr
d Center for Photovoltaic Materials, Korea University (Sejong), Sejong, 30019,
South Korea. E-mail: dwcho@korea.ac.kr
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6cp02087a S1–Tn intersystem crossing and the Tn–T1 internal conversion (IC)
15162 | Phys. Chem. Chem. Phys., 2016, 18, 15162--15169
This journal is ©the Owner Societies 2016

View Article Online
Paper
PCCP
Scheme 1 Molecular structures.
should be very fast (less than the picosecond range) and that the
LLCT mechanism is not needed to describe the excitation
properties of the Ir complexes studied by Park’s group.11,12
These issues seem to be of general significance regarding the
excited-state behaviour of related transition-metal complexes
and of particular interest for the molecular design of Ir complexes
associated with emission-colour tuning.
To disentangle the longstanding arguments regarding the
LLCT event, we designed and synthesized heteroleptic Ir3+
complexes 1 and 2 using 2-(4,6-difluorophenyl)pyridine (dfppy)
as the cyclometalating ligand and tetrapyrido[3,2-a:20,30-c:300,200-h:
20 0 0,300 0-j]phenazine (tpphz) and 2,3-bis(2-pyridyl)quinoxaline (dpq)
as ancillary ligands, as shown in Scheme 1.
In general, the electronic states should vary with the relative
p* levels of the ligands, such that the electronic coupling
between the two ligands might be weak for widely separated
p*s. The p* levels of the cyclometalating (dfppy) and ancillary
ligands (tpphz or dpq) are considerably different. Therefore, we
expect that the dynamics and behaviours of LLCT processes can
be readily recognized in these Ir complexes, even under ambient
conditions. Herein, we observed the LLCT dynamics directly using
time-resolved transient absorption (TA) spectroscopic techniques.
Fig. 1 Steady-state absorption spectra of free ligands (dpq, dfppy, and
The anomalous hypsochromic behaviours were investigated in a tpphz) and complexes 1 and 2 (10 mM) and the emission spectra of these
complexes in CH2Cl2 at 298 K. Low-temperature emission spectra of
rigid environment, and the orbital characters were analysed via
theoretical calculations using density functional theory (DFT).
complexes 1 and 2 were measured in 2-methyltetrahydrofuran (2-MTHF)
at 77 K (lex = 309 nm). The bottom photographs show the luminescence of
complexes 1 and 2 under ultraviolet (UV) irradiation (365 nm) in 2-MTHF
measured at 77 and 298 K, respectively.
Results and discussion
Steady-state absorption and emission properties
The absorption spectrum of the dfppy free ligand shows the MLCT (Ir - tpphz) (3MLCTtpphz) bands (Fig. 1a). The weak
p–p* transition bands at 273 nm with an intense band occurring 3MLCT absorption band of complex 2 is observed clearly in the
at approximately 250 nm, as shown in Fig. 1.13 In contrast, the lower-energy region (B500 nm), as shown in Fig. 1b. The 3MLCT
tpphz free ligand shows a strong p–p* transition band at band of Ir(dfppy)2(pic) was detected at a wavelength shorter
280 nm. The two sharp bands in the range of 350–390 nm than 450 nm.16 Therefore, for complexes 1 and 2, the absorption
correspond to the characteristic n–p* transition of tpphz bands at long wavelengths (4450 nm) can be assigned to the
(Fig. 1a).14 The dpq free ligand exhibits a moderately intense 3MLCT transition bands of Ir - dpq or Ir - tpphz. However,
absorption at 334 nm and relatively strong absorptions at 275 the moderately intense absorption at approximately 380 nm is
and 250 nm related to p–p* transitions (Fig. 1b).15 Thus the excited mainly contributed by the 1MLCT transitions. The spectroscopic
singlet (1p*) state of dfppy is higher than those of other ligands.
parameters are listed in Table 1.
In contrast, the absorption spectra of the Ir(III) complexes
As shown by the bottom images in Fig. 1, dramatic emission
(1 and 2) showed two distinguishable bands: the p–p* transition colour changes of complexes 1 and 2 occurred at 77 and 298 K.
band at short wavelengths and the MLCT bands at longer The orange-red emission of complex 1 at 298 K changed
wavelengths. The weak absorption band of complex 1 in the drastically to green emission at 77 K, whereas the red emission
visible region of 420–500 nm can be assigned to spin-forbidden of complex 2 at 298 K changed to orange-red at 77 K.
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys., 2016, 18, 15162--15169 | 15163

View Article Online
PCCP
Paper
Table 1 Absorption and emission spectroscopic parameters for com-
plexes 1 and 2 and free ligands
Compounds labs (nm), log ea (Mꢀ1 cmꢀ1
)
lem (nm) lem (nm)
b
c
1
2
288 (4.8), 394 (3.9), B460 (ꢀ)
615
635
474
575
261 (4.7), 296 (4.6), 365 (4.1),
500 (2.8)
dfppy
dpq
tpphz
242 (4.9), 273 (4.7)
247 (4.8), 270 (4.5), 334 (4.2)
280 (4.3), B310, 371, 381, 391
—
—
—
474
500
540
a
The values in parentheses indicate the molar extinction coefficient
b
c
values on a logarithmic scale. In CH2Cl2 at 298 K. In 2-MTHF at
77 K. lex = 309 nm.
At 298 K, complexes 1 and 2 showed strong red emissions at
approximately 615 and 635 nm, respectively, in CH2Cl2. At
room temperature, the red-emissions showed similar maxima
in both solvents; the reasons are not only the similar polarity of
CH2Cl2 (dielectric constant (e) = 6.97) and 2-MTHF (e = 9.08),
but the intrinsic emission behaviour of both complexes. The red
emissions of [Ir(dpq)2Cl2]17 and [Ru(tpphz)2Cl2]18 at approximately
Fig. 2 Absorption (blue lines) and excitation (orange lines) spectra of
complexes 1 and 2 (10 mM) in CH2Cl2 at 298 K. The monitoring wave-
lengths for 1 and 2 were 600 and 650 nm, respectively. Asterisks indicate
the second order of the monitoring wavelength.
3
630 nm have been reported to be the MLCT emissions of the
main ligands. Therefore, the red emissions observed here can be
ascribed to the 3MLCT emissions of tpphz or dpq.
transition bands at approximately 480 and 500 nm, respectively.
Therefore, the red emission can be obtained by direct excitation to
the triplet MLCT states of dpq and tpphz.
In a glassy matrix at 77 K, complex 1 showed the green
emission with a well-developed vibronic structure. The emission
spectral features and wavelength are similar to those found in
the emission spectrum of Ir(dfppy)2(pic), in which the dfppy and
picolinate ligands act as cyclometalating and ancillary ligands,
respectively.19 The p* state of the picolinate ligand is higher
than that of dfppy, and thus, the green emission (approximately
The extinction coefficient of dfppy is higher than those of
tpphz and dpq. Moreover, complexes 1 and 2 contain two dfppy
ligands. Upon photoexcitation at 309 nm, as shown in Fig. 1
and Table 1, dfppy efficiently absorbs the excitation light.
However, no blue-green emission related to 3MLCTdfppy was
observed at 298 K. This lack of emission from dfppy implies
3
474 nm) of Ir(dfppy)2(pic) is attributed to the MLCTdfppy state.
Moreover, the emission wavelength of complex 1 is also similar
3
that MLCTdfppy is the result of an energy-sinking process and
to that of the dfppy free ligand measured at 77 K, as listed in
that efficient energy transfer occurs from dfppy to other ligands via
3
Table 1. Therefore, the emissive MLCT state of complex 1 is
3
3
LLCT. Thus, the MLCTtpphz and MLCTdpq states of complexes 1
and 2 are the final emissive states because of the LLCT at 298 K.
close to the 3p* state of the dfppy free ligand at 77 K. In contrast,
the phosphorescence emission of the tpphz free ligand was
observed at approximately 540 nm at 77 K (Table 1). Therefore,
the emissive triplet state of an excited tpphz free ligand is lower
Theoretical DFT calculations
3
than the MLCTdfppy state of complex 1. Because only the high
To obtain further insights into the photophysical properties of
complexes 1 and 2, we performed theoretical DFT calculations
employing the Gaussian-09 program.20 The frontier orbital
contour plots of three lowest unoccupied molecular orbitals
(LUMOs [L], L+1 and L+2) and three highest occupied molecular
orbitals (HOMO [H], Hꢀ1 and Hꢀ2) of complex 1 involved in
the lower-energy transitions are presented in Fig. 3. The orbitals
in the H, Hꢀ1, and Hꢀ2 levels are localized on the dfppy
ligands and Ir ions. In contrast, those in the L, L+1, and L+2
levels are predominantly localized on the tpphz ligand. The
frontier orbital changes of complex 2 have been previously
reported;19 the H, Hꢀ1, and Hꢀ2 are mainly localized on the
dfppy ligands, whereas the L and L+1 are localized on the dpq
ligand.
3MLCTdfppy state contributes to the emission spectrum at 77 K,
we can suggest that a significant energy barrier exists in the
energy flow between the 3MLCTdfppy and 3MLCTtpphz states.
Therefore, in summary, complex 1 showed two types of emissions—
greenish-blue emission (474 nm) at 77 K and red emission (615 nm)
at 298 K—originating from the 3MLCTdfppy and 3MLCTtpphz
states, respectively.
For complex 2, the emission at 77 K occurred at a shorter
wavelength (575 nm) than that at 298 K (635 nm). The emissions
3
at 77 and 298 K are both attributed to the MLCTdpq state. We
will discuss the effect of temperature in greater detail later.
Excitation spectroscopic properties
As shown in Fig. 2, the excitation spectra of the red emissions of
Based on time-dependent (TD)-DFT calculations, the transitions
complexes 1 and 2 are identical to the steady-state absorption at longer wavelengths have very low oscillation strengths, as listed in
spectra throughout the examined range of wavelengths. The Table S1 in the ESI,† and correspond to forbidden transitions
excitation spectra clearly showed weak 3MLCTtpphz and 3MLCTdpq according to the selection rule related to the spin multiplicity.
15164 | Phys. Chem. Chem. Phys., 2016, 18, 15162--15169
This journal is ©the Owner Societies 2016

View Article Online
Paper
PCCP
Fig. 3 Selected frontier orbitals and energy levels of complex 2.
These transitions are attributed to combinations of various
HOMOs and, primarily, L and L+1 or L+2. Therefore, these
theoretical orbital characteristics imply that low-energy transitions
(S1 and T1) are mainly contributed by the LLCT from dfppy to
dpq or tpphz. The simulated absorption spectrum of complex 1
precisely described the measured absorption spectrum, as
shown in Fig. S1 of the ESI.† Based on the TD-DFT calculations,
we were able to easily interpret the transitions in the absorption
spectra of complex 1. We have previously reported the results of
the TD-DFT calculations of complex 2 elsewhere.19 Notably, the
absorption band at 4450 nm can be assigned to the 3MLCT
transition of dpq or tpphz ligands.
Time-resolved TA spectroscopic studies
Fig. 4 TA spectra of (a and b) complex 1 and (b and c) complex 2
in CH2Cl2 measured after selected time delays. All transient spectra
were corrected for the chirp. The figures on the right are the temporal
profiles monitored at different wavelengths. The excitation wave-
lengths were 290 nm (a and c) and 480 nm (b and d), respectively.
The negative peaks (#) at 480 nm in (b) and (d) are attributed to the
excitation pulses.
Fig. 4 shows the TA spectra of complexes 1 and 2 in CH2Cl2
probed at various time delays following the 100 fs pulse
excitations at 290 and 480 nm. Singlet manifolds (such as the
1LC (p*) or 1MLCT states) could be formed by excitation at
290 nm. As shown in Fig. 4a, the TA bands of complex 1 occur
over a broad range of 450–750 nm. The TA band at approximately
500 nm was observed after a short time delay and can be assigned
as the TA band of the 3MLCTdfppy state. This TA band decreases the MLCTdfppy state was smaller than that of the MLCTtpphz
as that at approximately 620 nm increases. The isosbestic point state at 620 nm, as shown in Fig. S2 in the ESI,† unlike the
occurs at approximately 550 nm, indicating that two excited results obtained after excitation at 290 nm. All temporal decay
species in equilibrium change over time. The TA band at or rise profiles started with significant DA values at time delays
approximately 620 nm can be attributed to 3MLCTtpphz. Therefore, of near zero, possibly because of the fast intersystem crossing
this observation confirms the occurrence of the LLCT from (1MLCT - 3MLCT) process of each ligand. Then, the LLCT
3MLCTdfppy to 3MLCTtpphz. The TA band at 500 nm decays with a process occurs in the triplet manifold. Identical decay and rise
time constant of 17 ps (decay profile in Fig. 4a). The TA trace was time constants of 17 ps were also observed upon excitation at
collected at 615 nm to observe the formation dynamics of 350 nm.
3
3
3MLCTtpphz via LLCT. This temporal rise profile exhibited a
qualitatively identical time constant of 17 ps as the decay trace achieved directly by excitation at 480 nm. Thus, the TA band
collected at 500 nm. at approximately 620 nm dominates, as shown in Fig. 4b, and
In contrast, the 3MLCTtpphz state of complex 1 can be
3
1
The MLCTdfppy and MLCTtpphz transitions of both ligands the TA band at 500 nm is insignificant. Moreover, no rise or
can occur upon photoexcitation at 350 nm. The molar extinction decay time-profiles were observed, even after short time delays.
1
coefficient of the MLCTtpphz transition can be larger than that The 3MLCTtpphz state is the final lowest state because no
of 3MLCTdfppy because this transition is allowed by the selection evolution was observed as the time delay increased, as shown
rule for spin multiplicity. Therefore, the 1MLCTtpphz state is in Fig. 4b.
generated readily upon excitation at 350 nm. After short time
Upon excitation at 290 nm, the TA spectra of complex 2 showed
delays, the intensity of the TA band at 500 nm corresponding to the characteristic 3MLCTdfppy band at 500 nm, as discussed above,
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys., 2016, 18, 15162--15169 | 15165

View Article Online
PCCP
Paper
with a decay time of 5 ps, as shown in Fig. 4c. We attribute this
time constant to the LLCT process from dffpy to dpq in the excited
state. However, no significant spectral changes were observed after
a delay of approximately 10 ps. Therefore, the TA spectral features
3
of 3MLCTdpq are similar to those of MLCTdfppy. As shown in
3
Fig. 4d, when complex 2 was excited directly to the MLCTdpq
state by 480 nm light, the TA spectra showed the 3MLCTdpq
band at 500 nm, and no evolution was observed as the time
delay increased; thus, the lowest excited state of complex 2 is
3
also suggested to be the MLCTdpq state.
The LLCT is a forbidden process in octahedral metal-complexes
because the ligands have an orthogonal structure. The tpphz
ligand has a rigid p-conjugated planar structure, whereas the
dpq ligand is more flexible. The LLCT process may occur
through the structural distortion of the perfect octahedral
geometry. Therefore, the LLCT process of complex 2 is expected
to be efficient and fast and to have a small energy barrier.
Fig. 6 Emission spectra of 2 in 2-MTHF recorded as a function of
temperature. The excitation wavelength was 509 nm. The inset figure shows
the compilation of the emission intensity as a function of temperature.
from charge transfer exhibits broad spectral features without a
vibrational structure. The broad emission at long wavelengths
is assigned to the 3MLCT emission of tpphz in fluid solutions at
high temperature. At 77 K, when complex 1 was excited to the
3MLCT state of tpphz with 509 nm, broad emission at 572 nm
was observed, as shown by the red line in Fig. 5. Thus, the final
Temperature dependence of the emission spectra
Complex 1 exhibited emission with a vibronic structure at
B474 nm at 77 K. The emission intensity decreased, and it
gradually shifted to longer wavelengths as the temperature
increased, as shown in Fig. 5. At temperatures below 134 K,
the emission positions and vibronic structures were similar to
the phosphorescence spectrum of Ir(dfppy)2(pic). Therefore,
the low-temperature emission of complex 1 can be assigned
to the 3MLCT state of dfppy. At temperatures over 134 K,
however, the vibronic structure vanished, and the emission
maximum shifted drastically from 474 nm to B570 nm. A
considerable change in the emission intensity was observed
in the range of 130–145 K, as shown in the inset of Fig. 5.
Because the melting point of 2-MTHF is 137 K, the change in
the intensity and shift of the emission peak depends strongly
on the rigidity of the solvent. In a glassy matrix at low-
temperature, the main lowest emissive state may be attributed
to the 3MLCT state of dfppy. Usually, the emission resulting
3
emissive state of complex 1 is the MLCTtpphz state.
Fig. 6 shows the emission spectra of complex 2 in 2-MTHF
upon excitation at 509 nm as a function of temperature. The
excitation wavelength of 509 nm corresponds to the 3MLCT
transition of Ir - dpq; thus, the emission can be attributed to
the 3MLCTdpq state. At 77 K, complex 2 exhibits emission with a
moderate vibronic structure at approximately 575 nm. When
complex 2 was excited at 355 nm (or 309 nm), overall, the
observed spectral changes in its emission were identical to
those resulting from excitation at 509 nm. Thus, unlike complex
1, the emission behaviours of complex 2 do not depend on the
excitation wavelength. As mentioned above, the LLCT process
of complex 2 is not affected by an activation barrier, even at
77 K. Therefore, observing the emission originating from the
MLCTdfppy state is difficult.
As the temperature increased, the 3MLCTdpq emission
shifted to longer wavelengths and broadened. A drastic change
in the emission intensity was also observed in the range of
130–140 K, as shown in the inset of Fig. 6. Complexes 1 and 2
exhibited rigidochromism, in which the emission properties
depend on the viscosity of the system. Usually, the excited state
is stabilized by solvent reorganization. In a rigid environment,
however, solvent reorganization is extremely slow or may
not occur. Therefore, the excited state cannot be stabilized
completely, resulting in a blue shift of the emission wavelength.
For complex 2, the hypsochromic shift of approximately 1600 cmꢀ1
occurred at somewhat a lower energy state than the LC state of
free ligands (see emission wavelengths in Table 1). Therefore,
the hypsochromic change is attributed to emission from the
3MLCTdpq state.
Fig. 5 Emission spectra of 1 in 2-MTHF recorded as a function of
temperature. The excitation wavelength was 355 nm. The red spectrum
was measured at 77 K upon excitation at 509 nm. The inset shows the
compilation of the emission intensity monitored at 470 nm as a function of
the reciprocal temperature.
However, complex 1 showed a very large hypsochromic shift
of B4800 cmꢀ1. The emission wavelength at 77 K is similar to the
15166 | Phys. Chem. Chem. Phys., 2016, 18, 15162--15169
This journal is ©the Owner Societies 2016

View Article Online
Paper
PCCP
Fig. 8 (a) The proposed LLCT mechanisms for complexes 1 and 2. (b) The
LLCT dynamics at 300 K and the phosphorescence lifetimes measured at
77 K and 300 K.
Fig. 7 Compilation of the emission lifetimes as a function of temperature
for complex 2. The inset figure shows the decay profiles of emission at
550 nm for complex 2 in 2-MTHF at different temperatures.
state occurs because of strong spin–orbit coupling with the Ir
ion. The 3MLCT state of dfppy has different fates depending on
the energy level of the ancillary ligand (dpq or tpphz). For
complexes 1 and 2, because the 3MLCT states of dpq and tpphz
are lower than that of dfppy, the LLCT from the 3MLCT state of
dfppy to the 3MLCT state of dpq or tpphz is exothermic. In
contrast, the strong temperature effects observed for complex 1
constitute clear evidence that a significant activation barrier
exists in the LLCT process. Therefore, the decay and corres-
ponding rise components can be clearly observed in the TA
profiles. Furthermore, when molecules are directly excited to
3MLCTtpphz, no temperature effects or evolution in the TA decay
profiles are observed.
For complex 2, the activation barrier in the LLCT process can
be ignored. As a result, we were unable to observe any emission
from the 3MLCTdfppy state, even at 77 K. Even when the
molecule was excited at a high-energy wavelength, the observed
emission originated from the 3MLCTdpq state through an
efficient LLCT process.
3MLCTdfppy emission, which is much higher than the phos-
phorescence wavelength (or the LC state) of the tpphz free ligand.
This unusual emission behaviour is explained as follows: very
3
intense emission from the MLCTdfppy state occurred at 77 K,
and thus, a significant energy barrier exists in the LLCT between
3
3
the MLCTdfppy and MLCTtpphz states.
Temperature dependence of the emission lifetimes
The temperature dependence of the excited-state lifetimes of
complex 1 is shown for comparison in Fig. 7. At 77 K, the
intrinsic emission lifetimes of complexes 1 and 2 are 6.8 and
6.2 ms, respectively, in 2-MTHF. The emissions of complexes 1
and 2 exhibited constant lifetimes at temperature oꢀ130 K. In
contrast, near the melting point of 2-MTHF, the emission
lifetime drastically decreased as the temperature increased.
Additionally, at temperatures over 140 K, the emission lifetimes
remained nearly constant as the temperature increased. At
298 K, the emission lifetimes for complexes 1 and 2 are 1.2
and 0.8 ms, respectively. However, when the solvent was changed
to CH2Cl2, the temperature range in which the emission changed
drastically shifted to approximately 174–177 K. Because the
melting point of CH2Cl2 is 176 K, we can conclude that the
lifetimes of the 3MLCT states of complexes 1 and 2 are affected
by the solvent rigidity. In particular, for complex 1, the influences
of temperature variation and solvent rigidity on the LLCT process
between the dfppy and tpphz ligands are strong.
Conclusions
LLCT processes can be directly observed in the presence of
energetically different ligands in metal complexes. To obtain
insights into the LLCT process, we prepared a series of hetero-
leptic Ir(III) metal complexes chelated by two cyclometalated
dfppy ligands and an ancillary ligand (dpq or tpphz) to ensure a
larger pp* energy gap than that achieved with cyclometalated
ligands. Based on the temperature dependency results, the
rigidity of the environment is a very important factor affecting
Mechanism
Based on the experimental results of the temperature dependency the LLCT process. Indeed, the LLCT appears to require some
and fs-TA, we propose a rational mechanism for the LLCT process, structural distortion because this transition is forbidden when
as shown in Fig. 8. Strictly speaking, it is difficult for the LLCT the ligands are arranged in an orthogonal geometry. Complex 1,
transition to occur in a perfect octahedral structure. Complexes 1 which contained a flat and rigid ancillary ligand, showed slower
and 2 have orthogonal geometries with different types of bidentate LLCT dynamics than complex 2. Complexes 1 and 2 constitute
ligands that are electronically non-conjugated. In a fluid solution, good models for understanding the LLCT dynamics based on
however, the octahedral structure changes constantly, and this TA spectroscopy with excitation at different wavelengths. The
distortion allows the LLCT between orthogonal ligands.
LLCT properties were affected by several important factors,
1
After exciting dfppy in the singlet manifold (1LC or MLCT such as the energy differences between the main and ancillary
states), a highly efficient inter-system crossing to the 3MLCT ligands, the temperature, and the rigidity of the environment.
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys., 2016, 18, 15162--15169 | 15167

View Article Online
PCCP
Paper
Experimental
General procedures
127.6, 125.9, 124.7, 124.5, 124.3, 123.5, 123.4, 114.2, 113.6,
99.6, 99.0. MALDI-TOF MS: m/z = 857.174 ([M–PF6]+; C40H24F4IrN6 ;
+
calcd 857.163); anal. calcd for C40H24F10IrN6P: C, 47.95; H, 2.41; N,
8.39. Found: C, 47.86; H, 2.41; N, 8.42.
Starting materials for the synthesis were purchased from
Aldrich Chemical Co. and TCI Co., and were used without
further purification. Reactions were carried out under a dried
N2 atmosphere. CH2Cl2 and MeOH were freshly distilled over
calcium hydride before use. Glassware, syringes, magnetic
stirring bars, and needles were dried in a convection oven over
4 h. In order to monitor the progress of the reaction, commercial
thin layer chromatography (TLC) plates (Merck Co.) were developed,
and the spots were seen under UV light at 254 and 365 nm.
Column chromatography was done with silica gel 60G (particle
size 5–40 mm, Merck Co.).
Spectroscopic measurements
Steady-state absorption spectra were recorded using a UV-vis
spectrophotometer (Agilent Technology, Cary 5000) at room
temperature. Fluorescence spectra were measured using a
fluorescence spectrophotometer (Varian, Cary Eclipse). Elemental
analyses were carried out using a Carlo Erba Instruments CHNS-O
EA 1108 analyzer and MALDI-TOF mass spectroscopy (ABI
Voyager STR) was performed at the Ochang Branch of the
Korean Basic Science Institute.
1H and 13C NMR spectra were recorded on a Varian Mercury
300 spectrometer operating at 300.1 and 75.4 MHz, respectively.
The chemical shifts are given in ppm relative to the residual signal
of the solvents: CDCl3 7.26 ppm (1H) and CDCl3 77.0 ppm (13C).
The phosphorescence spectra at various low temperatures
were measured using an intensified charge-coupled device
(ICCD, Andor, iStar). The samples were excited by nanosecond
laser pulses of the third harmonic generation (355 nm, FWHM
of 4.5 ns) from a Q-switched Nd:YAG laser (Continuum, Surelite
II-10). Phosphorescence lifetimes were measured using a digital
oscilloscope (Tektronix, TDS-784D) equipped with a fast photo-
multiplier tube (Zolix Instruments Co., CR 131).
The sub-picosecond time-resolved absorption spectra were
collected using a pump–probe TA spectroscopy system (Ultra-
fast Systems, Helios). The pump light was generated by using a
regenerative amplified titanium sapphire laser system (Spectra
Physics, Spitfire Ace, 1 kHz) pumped by a diode-pumped
Q-switched laser (Spectra Physics, Empower). The seed pulse
was generated using a titanium sapphire laser (Spectra Physics,
MaiTai SP). The pulses (290, 350 and 480 nm) generated from
an optical parametric amplifier (Spectra Physics, TOPAS prime)
were used as the excitation pulse. And a white light continuum
pulse, which was generated by focusing the residual of the
fundamental light to a thin Sapphire crystal after the controlled
optical delay, was used as a probe beam and directed to the
sample cell with an optical path of 2.0 mm and detected with a
CCD detector installed in the absorption spectroscopy. The
pump pulse was chopped by the mechanical chopper synchronized
to one-half of the laser repetition rate, resulting in a pair of the
spectra with and without the pump, from which absorption change
induced by the pump pulse was estimated. To obtain the transient
spectrum at a chosen delay time, the raw data were corrected for the
chirp, which is carried out after the measurement.
Synthesis
The cyclometalating ligand (dfppy) and ancillary ligands (dpq21
and tpphz22) were synthesized as described by a previously
developed synthesis protocol. The synthetic route towards
dfppy, dpq and tpphz is shown in Scheme 2. The synthesis
details of [Ir(dfppy)2(dpq)](PF6) (2) are described earlier.19
[Ir(dfppy)2(tpphz)](PF6) (2): a solution of [Ir(dfppy)2(m-Cl)]2
(1.0 g, 0.8 mmol) and 2.2 equivalent ancillary ligand (tpphz)
(0.7 g, 1.8 mmol) in 2 : 1 CH2Cl2/MeOH (60 mL) was heated to
reflux under an inert atmosphere of nitrogen in the dark for
14 h. The red solution was cooled to room temperature, and
then a 10-fold excess of KPF6 (1.5 g, 8.2 mmol) was added to the
solution. The suspension was stirred for 2 h and then filtered to
remove insoluble inorganic salts. The organic solvent was
removed under reduced pressure and the residue was purified
by silica gel column chromatography using acetone/CH2Cl2
(v/v = 1 : 15) as the eluent to afford pure complex 2. Yield:
1
0.59 g (52%). H NMR (300.1 MHz, DMSO-d6) d 8.70 (d, 2H),
8.58 (d, 1H), 8.53 (d, 1H), 8.42–8.44 (m, 2H), 8.15–8.26 (m, 6H),
7.93–8.01 (m, 3H), 7.62–7.92 (m, 4H), 7.36 (t, 1H), 7.23 (t, 1H),
6.84 (t, 1H), 6.74 (t, 1H), 6.05 (d, 1H). 13C NMR (75.4 MHz,
DMSO-d6) d 158.1, 156.2, 153.8, 153.7, 152.3, 150.3, 149.9,
142.6, 140.3, 140.0, 138.7, 133.5, 133.0, 131.0, 130.2, 128.6,
Theoretical calculations
Computations on the electronic ground state of complex 1 were
performed using Becke’s three-parameter density functional
combined with the nonlocal functional of Lee, Yang, and Parr
(B3LYP).23 6-31G(d) basis sets were employed for the ligand and
the relativistic effective core potential of Los Alamos and
double-zeta basis set (LANL2DZ) were employed for Iridium.24
The ground-state geometry was fully optimized and time-dependent
Scheme 2 Synthetic route towards the cyclometalating ligand (dfppy)
and ancillary ligands (dpq and tpphz).
15168 | Phys. Chem. Chem. Phys., 2016, 18, 15162--15169
This journal is ©the Owner Societies 2016

View Article Online
Paper
PCCP
(c) Y.-Y. Lyu, Y. Byun, O. Kwon, E. Han, W. S. Jeon, R. R. Das
and K. Char, J. Phys. Chem. B, 2006, 110, 10303–10314.
14 J. Bolger, A. Gourdon, E. Ishow and J.-P. Launay, Inorg.
Chem., 1996, 35, 2937–2944.
15 X. Gu, T. Fei, H. Zhang, H. Xu, B. Yang, Y. Ma and X. Liu,
J. Phys. Chem. A, 2008, 112, 8387–8397.
density functional theory (TD-DFT) calculations were performed
to obtain the vertical singlet and triplet excitation energies. All
calculations were performed using the Gaussian-09 package.20
On the other hand, theoretical calculation results of complex 2
have been reported earlier.19
16 Y. Zhou, W. Li, Y. Liu, L. Zeng, W. Su and M. Zhou, Dalton
Trans., 2012, 41, 9373–9381.
17 S. C. Rasmussen, M. M. Richter, E. Yi. H. Place and
K. J. Brewer, Inorg. Chem., 1990, 29, 3926–3932.
18 E. Ishow, A. Gourdon, J. P. Launay, P. Lecante, M. Verelst,
C. Chiorboli, F. Scandola and C.-A. Bignozzi, Inorg. Chem.,
1998, 37, 3603–3609.
19 A.-R. Hwang, W.-S. Han, K.-R. Wee, H. Y. Kim, D. W. Cho,
B. K. Min, S. W. Nam, C. Pac and S. O. Kang, J. Phys. Chem.
C, 2012, 116, 1973–1986.
Acknowledgements
This study was supported by the Basic Science Research Program
through the National Research Foundation of Korea (NRF) funded
by the Ministry of Education (NRF-2014R1A6A1030732 and
NRF-2014R1A1A-2004199). This research was supported by the
Functional Districts of the Science Belt support program,
Ministry of Science, ICT and Future Planning (2015K000287).
20 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta,
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin,
V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,
A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi,
N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross,
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski,
R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth,
P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels,
Notes and references
1 M. S. Lowry and S. Bernhard, Chem. – Eur. J., 2006, 12,
7970–7977.
2 M. S. Lowey, W. R. Hudson, R. A. Pascal, Jr. and S. Bernhard,
J. Am. Chem. Soc., 2004, 126, 14129–14135.
3 A. B. Tamayo, S. Garon, T. Sajoto, P. I. Djurovich, I. M. Tsyba,
R. Bau and M. E. Thompson, Inorg. Chem., 2005, 44, 8723–8732.
4 S. Lamansky, P. Djurovich, D. Murphy, F. Abdel-Razzaq,
H. E. Lee, C. Adachi, P. E. Burrows, S. R. Forrest and
M. E. Thompson, J. Am. Chem. Soc., 2001, 123, 4304–4312.
5 A. J. Wilkinson, A. E. Goeta, C. E. Foster and J. A. G.
Williams, Inorg. Chem., 2004, 43, 6513–6515.
6 M. Polson, S. Fracasso, V. Bertolasi, M. Ravaglia and F. Scandola,
Inorg. Chem., 2004, 43, 1950–1956.
¨
7 J. Li, P. I. Djurovich, B. D. Alleyne, M. Yousufuddin, N. N.
Ho, J. C. Thomas, J. C. Peters, R. Bau and M. E. Thompson,
Inorg. Chem., 2005, 44, 1713–1727.
O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski and
D. J. Fox, Gaussian 09, Revision D.01, Gaussian, Inc., Wallingford
CT, 2009.
8 Y. You and S. Y. Park, J. Am. Chem. Soc., 2005, 127, 12438–12439. 21 A. Ohta, S. Masano, S. Iwakura, A. Tamura, H. Watahiki,
9 Y. You, K. S. Kim, T. K. Ahn, D. Kim and S. Y. Park, J. Phys.
Chem. C, 2007, 111, 4052–4060.
M. Tsutsui, Y. Akita and T. Watanabe, J. Heterocycl. Chem.,
1982, 19, 465–472.
10 Y. You, J. Seo, S. H. Kim, K. S. Kim, T. K. Ahn, D. Kim and 22 (a) J. Bolger, A. Gourdon, E. Ishow and J. P. Launay, Inorg.
S. Y. Park, Inorg. Chem., 2008, 47, 1476–1487.
11 H.-S. Duan, P.-T. Chou, C.-C. Hsu, J.-Y. Hung and Y. Chi,
Inorg. Chem., 2009, 48, 6501–6508.
Chem., 1996, 35, 2937–2944; (b) E. Ishow, A. Gourdon, J. P.
Launay, P. Lecante, M. Verelst, C. Chiorboli and F. Scandola,
Inorg. Chem., 1998, 37, 3603–3609.
12 Y. Chi and P.-T. Chou, Chem. Soc. Rev., 2010, 39, 638–655.
13 (a) X. Gu, T. Fei, H. Zhang, H. Xu, B. Yang, Y. Ma and X. Liu,
J. Phys. Chem. A, 2008, 112, 8387–9393; (b) F. O. Garces,
23 (a) A. D. Becke, J. Chem. Phys., 1993, 98, 5648–5652;
(b) C. Lee, W. Yang and R. G. Parr, Phys. Rev., 1988, 37,
785–789.
K. A. King and R. J. Watts, Inorg. Chem., 1988, 27, 3464–3471; 24 P. J. Hay, J. Phys. Chem. A, 2002, 106, 1634–1641.
This journal is ©the Owner Societies 2016
Phys. Chem. Chem. Phys., 2016, 18, 15162--15169 | 15169