Synthesis, excited state dynamics, and optical characteristics of oligophenyl-based swivel cruciforms in solution and solid state

Article abstract of DOI:10.1021/jp1028883

Oligophenyl-based swivel cruciforms are an amorphous class of materials for potential use in display applications. In this work, we describe the design, synthesis, and structural and optical properties of a group of chromophores with varying degrees of π-conjugation that produce blue emission and high photoluminescence quantum yields (PLQYs). The swivel cruciforms are branched complexes consisting of two arms, and the relative rotation of these arms is an important factor that determines the optical properties of these systems. The studies reveal that the physical size of the system, that is, the arms, and the presence and position of functional groups results in different degrees of sterical effects. This has a dramatic impact on the interplay between nonradiative and radiative decay of the excited state. For the least π-extended system, control of the excited state decay can be achieved by modifying the properties of the medium through an increase in the viscosity, which is demonstrated accordingly. The data shows that an intramolecular excited state is responsible for the swivel cruciform emission in solution. The character of this "aggregate" and its emission is not dissimilar to an excimer, and we therefore attribute it to an intramolecular excimer. In the solid state, a combination of intra- and intermolecular excimers are the most likely source of the emission. The data also shows that intra molecular excimers can produce a surprisingly high PLQY, when either properties of the medium facilitate this or when functional groups introduce steric hindrance, which subsequently prevents the nonraditive decay through conformational change.

Full text of DOI:10.1021/jp1028883

J. Phys. Chem. B 2010, 114, 12765–12776
12765
Synthesis, Excited State Dynamics, and Optical Characteristics of Oligophenyl-Based Swivel
Cruciforms in Solution and Solid State
Lars-Olof Pålsson,*,†,§ Benjamin S. Nehls,‡,| Frank Galbrecht,‡ Benjamin A. Coombs,†
Fernando B. Dias,§ Tony Farrell,‡,⊥ Ullrich Scherf,‡ and Andrew P. Monkman§
Department of Chemistry, Durham UniVersity, South Road, DH1 3LE Durham, United Kingdom,
Macromolecular Chemistry Group and Institute for Polymer Technology, Bergische UniVersita¨t Wuppertal,
Gauss-Strasse 20, D-42097 Wuppertal, Germany, and Department of Physics, Durham UniVersity,
South Road, DH1 3LE Durham, United Kingdom
ReceiVed: March 31, 2010; ReVised Manuscript ReceiVed: June 14, 2010
Oligophenyl-based swivel cruciforms are an amorphous class of materials for potential use in display
applications. In this work, we describe the design, synthesis, and structural and optical properties of a group
of chromophores with varying degrees of π-conjugation that produce blue emission and high photoluminescence
quantum yields (PLQYs). The swivel cruciforms are branched complexes consisting of two arms, and the
relative rotation of these arms is an important factor that determines the optical properties of these systems.
The studies reveal that the physical size of the system, that is, the arms, and the presence and position of
functional groups results in different degrees of sterical effects. This has a dramatic impact on the interplay
between nonradiative and radiative decay of the excited state. For the least π-extended system, control of the
excited state decay can be achieved by modifying the properties of the medium through an increase in the
viscosity, which is demonstrated accordingly. The data shows that an intramolecular excited state is responsible
for the swivel cruciform emission in solution. The character of this “aggregate” and its emission is not dissimilar
to an excimer, and we therefore attribute it to an intramolecular excimer. In the solid state, a combination of
intra- and intermolecular excimers are the most likely source of the emission. The data also shows that intra
molecular excimers can produce a surprisingly high PLQY, when either properties of the medium facilitate
this or when functional groups introduce steric hindrance, which subsequently prevents the nonraditive decay
through conformational change.
Introduction
bandgap of the conjugated polymer in these LEDs. In addition,
the color and hue of some LEDs is less well-defined due to the
large extent of inhomogeneous broadening of the electronic
transition, which can result in a relatively broad PL and EL
spectrum. Another issue relates to the processing of these
materials in the actual LED fabrication process. Conjugated
polymers with a high molecular weight and high polydispersity
are demanding to work with due to their low solubility. The
crystallinity can be extensive in some cases. One prominent
example in this context is PPV,1 and hence, spin-coating of thin
films in the LED fabrication procedure is more problematic.
The strong tendency to form aggregates is also a serious problem
with the spectroscopic properties of these, being less ideal for
the LED operation because the nonradiative decay mechanisms
are relatively strong upon aggregate formation.7-17 What is
beneficial, though, in this context is the transport properties,
which are greatly enhanced with strong intermolecular interac-
tions, and this, in turn, is a key requirement in organic electronic
devices such as LEDs.
It is therefore of great interest to be able to control the spatial
packing and orientation of the active material while at the same
time achieve maximum π-π overlap between separate chro-
mophore units to maximize charge transport and device operation.
This is, of course, a difficult challenge in synthetic chemistry, but
some recent advances using amorphous organic materials suggest
that this way forward may be rewarding. Some structures that have
been considered in this context include dendrimeric forms,18-20
starbust,21,22 oligomers, spiro-compounds,19,20,23-25 planar and swivel
type cruciforms.26-32
Organic materials are rapidly becoming more common in
various micro- and optoelectronic applications, such as organic
electronics, display applications, photovoltaic devices, etc.1,2 For
many of these applications, the key material is an organic
conjugated polymer system, especially in the case of active
display applications of the light-emitting diodes (LEDs). Elec-
troluminescence (EL) from a conjugated polymer was first
realized using the poly(p-phenylene vinylene) (PPV) system,
and this work generated an enormous momentum in the entire
field of organic electronics.3-6
Although this work clearly was an unprecedented success,
there still remained important problems to address and chal-
lenges to deal with. One such issue concerns the color and hue
of the photoluminescence (PL) of the LED based upon organic
conjugated polymers. There are a range of materials available
for the yellow and red regions, but less so for the UV-blue parts
of the optical spectrum. In the proceeding efforts, it was apparent
that the development of stable conjugated polymer materials
for the UV-blue was challenging due to the relatively large
* Corresponding author. Phone: 0044-191-3342135. Fax: 0044-191-
3342051. E-mail: lars-olof.palsson@durham.ac.uk.
† Department of Chemistry, Durham University.
‡ Bergische Universita¨t Wuppertal.
§ Department of Physics, Durham University.
| Current address: BASF SE, CA/C, E100, 67056 Ludwigshafen,
Germany.
⊥ Current address: Sabic Innovative Plastics, Plasticslaan 1, PO BOX
117, Bergen & Zoom 4600 AC, The Netherlands.
10.1021/jp1028883  2010 American Chemical Society
Published on Web 09/20/2010

12766 J. Phys. Chem. B, Vol. 114, No. 40, 2010
Pålsson et al.
SCHEME 1: Synthesis of SC3a and SC3b
In this present work, we explore the photophysical and
structural relationships of a series of oligophenyl-based swivel
cruciforms, designed using a versatile synthetic strategy. This
group of materials show unusually high PL efficiency in the
UV-blue part of the spectrum, and these are attractive material
properties often sought in materials for display applications. The
PL properties in the steady state and time domain of solutions
and films are described in this work. In complement to the
spectroscopy, we have also performed theoretical calculations
on a subset of the compounds. The work shows how the
structure of the system can affect the optoelectronic properties
and, thereby, the PL efficiency (PLQY). The work also shows
the impact the properties of the medium may have on the
photophysics, in some cases with dramatic effects.
For thin films, the PLQYs were measured as outlined by de
Mellow et al. and Pålsson et al.40 In the low-temperature
measurements, a liquid nitrogen cryostat (JANIS) was used. The
time-resolved PL was measured using the time-correlated,
single-photon counting technique (TCSPC) described recently,41
except that here, the excitation source consisted of a mode-
locked cavity dumped (APE Pulse switch) Ti:Saphire laser
(MIRA, Coherent). Near-infrared optical pulses with a temporal
width of ∼100 fs (fwhm) were used to generate the third
harmonic generation UV pulses at a wavelength of ∼300 nm
with an Inrad (model 5-050) harmonic generator. The PL was
detected using a single-photon counting module based upon an
avalanche photo diode (Id Quantique model 100-50) linked to
a time-to-amplitude converter (Ortec 567) and multichannel
analyzer (E.G. & G. Trump Card and Maestro for Windows v.
5.10). The instrument response function (IRF) of the apparatus
was measured from the Rayleigh scattered light, giving an IRF
with a duration of ∼125 ps fwhm. All PL decays were recorded
to a minimum of 10 000 counts in the peak channel of the pulse
height analyzer. The data was analyzed using the standard
method of iterative reconvolution and nonlinear least-squares
fitting in a Microsoft Excel spreadsheet.41 The quality of the
calculated fits was judged using statistical parameters, including
the Durbin-Watson parameter, reduced ꢀ2, random residuals,
and autocorrelated residuals. The experimental data were
subsequently fitted to a sum of exponentials,
Materials and Methods
Density functional theory calculations. The DFT calculation
were performed as previously outlined.33-38 Molecular geom-
etries of the model complexes were optimized without con-
straints via DFT calculations using the Becke3LYP (B3LYP)33
functional. The basis set used was the 6-31G(d) Pople basis
set.34 All of the DFT calculations were performed with the
Gaussian 03 package.38
Optical Spectroscopy. All compounds where dissolved in
methyl cyclohexane (MCH) unless otherwise stated. Thin films
were obtained on quartz disks by spin-coating solutions of ∼15
mg/mL solvent spun at 1500 rpm for 30 s. Absorption spectra
were measured on a Perkin-Elmer Lambda 19 spectrophotom-
eter, and the steady state PL was recorded using a Jobin Yvon
Horiba Fluoromax 2. The PLQYs in solution were calculated
using the comparative method39 with anthracene/ethanol as the
standard, both sample and standard with absorbencies ∼0.1 OD.
i
F(λ , t) )
A (λ )·e-k ·t
(1)
∑
em
i
em
i
Results
Synthesis and Structural Characterization. Synthesis of
swivel-type cruciforms has been an active topic during recent

Oligophenyl-Based Swivel Cruciforms
J. Phys. Chem. B, Vol. 114, No. 40, 2010 12767
SCHEME 2: Synthesis of the Extended Swivel Cruciforms SC5 and SC7.
years.26-32 These oligomers are based on tetrasubstituted biaryl
cores, such as binaphthyl-,31 biphenyl-,42-45 or bithienyl46,47
cores.
Despite its twisted structure, SC3a shows a tendency to
crystallize. Optimization of the thermal properties would seem
to be necessary with respect to potential use in optoelectronic
applications such as LEDs. The easiest way to obtain an
increased amorphous character is to enlarge the sterical hin-
drance of the arms so that internal folding and crystallization
are prevented. SC3b as an analogue to SC3a exhibits an
increased steric hindrance due to additional tert-butyl groups
on the outer phenyl rings.
Compared to Bunz’s26-29 and Nuckolls’s30 planar cruciforms,
these swivel type cruciforms allow, at least in principle, rotation
between the biparyl moieties and within the homologous parts
of the molecule, which we call the “arms” of the cruciforms.
This should, in principle, lead to a greater degree of confor-
mational freedom. However, in the swivel cruciform architec-
tures designed by Farrell and Scherf, it was found that the three-
dimensional arrangement of these structures could be biased
by introducing the possibility for π-π interactions to occur.42
For example, it was found that the cruciform SC3a shown in
Scheme 1 adopts a folded configuration in the crystal structure
and that this type of foldamer is also retained in solution.
All syntheses of swivel cruciform dimers follow a similar
straightforward method utilizing the different reactivity of
aromatic bromo/iodo and chloro groups in a Suzuki-type,
aryl-aryl, cross-coupling reaction (Scheme 1).48-50 The syn-
thetic strategy toward the sterically hindered terphenyl based
cruciform SC3b follows the strategy developed by Farrell and
Scherf with the isolation of the chloroaryl precursor 3 (Scheme
1) from the reaction of 1-chloro-2,5-dibromobenzene with 3,5-
bis-tert-butyl-phenylboronic acid under microwave or conven-
tional conditions, or both.51,52
Synthesis of extended swivel cruciforms requires additional
steps. To increase the arm length, we have prepared a versatile
central building block 7 that can be reacted with a broad variety
of substituents and is still of current synthetic interest. To prepare
building block 7, 1,4-dibromo-2-chlorobenzene (2) was coupled
via a Suzuki type cross coupling reaction with 4-trimethylsilyl-
phenylene-boronic acid (5) giving 1-chloro-2,5-(trimethylsilyl-
phenyl)benzene (6) in more than 75% yield after recrystallization
from heptane/dichloromethane. No further purification is needed
at that step.
The TMS protected terphenyl 6 was subsequently converted
to the diiodo derivative 7 by treating 6 with ICl in yields close
to 90%. Subsequent Suzuki type cross-coupling with the
corresponding boronic acid/ester allows access to a broad variety
of swivel cruciform arms. Within this publication, 3,5-bis-tert-
butylphenyl-pinacolatoboronate (1) and 2-(9,9-di-n-dioctylfluo-
rene)-4,4′,5,5′-tetramethyl-1,3,2,dioxaborolane (8) have been
coupled to building block 7, leading to the cruciform precursors
9 and 10. The synthetic protocol for the preparation of the
corresponding cruciforms SC5 and SC7 is similar to that used
in the preparation of SC3a and SC3b by utilizing a nickel-
mediated, Yamamoto-type, aryl-aryl coupling reaction under
microwave conditions.53
Within the initial step, a chloroaryl precursor (here 3; see
Scheme 1) is formed by reacting 1-chloro-2,5-dibromobenzene
with 3,5-bis-tert-butyl-phenylboronic acid under microwave or
conventional conditions51,52 in good yields, which is subse-
quently homocoupled in a nickel-mediated Yamamoto-type
reaction53 toward the desired cruciform. The syntheses of SC3a
and the corresponding monomeric terphenyl derivative (SC1)
have been reported in an earlier paper and are also depicted in
Scheme 2 for better understanding.42,49 Although SC1 is, strictly
speaking, not a cruciform, we will use this notation throughout
the text for reasons of consistency.
All swivel cruciform materials were obtained in reasonable
yields and were purified via column chromatography. Structural
integrity was proved via NMR spectroscopy and MS, respectively.

12768 J. Phys. Chem. B, Vol. 114, No. 40, 2010
Pålsson et al.
1
Figure 1. Solution H NMR spectra in the aromatic region of SC3a (bottom) and SC3b (top) in C2D2Cl4.
All swivel type cruciforms and intermediate products show
good solubility in common solvents, which allowed a detailed
characterization of all materials by NMR spectroscopy. This
method also allows judging of the strength of interaction
between the cruciform arms. As mentioned previously,42 SC3a
shows a strong π-π stacking of two phenyl-rings, which is
proved by the high-field position of two directly coupled
doublets at δ ) 6.51 and 6.93 ppm with a coupling constant of
8.5 Hz. As expected, the upfield shift of the signals cannot be
found in the related SC3b, as shown in Figure 1, displaying
that there is no significant π-π interaction of the arms in
solution.
The crystal structure of SC3b was obtained from solutions
in dichlormethane or THF, respectively, and investigated by
X-ray diffraction at room temperature. Analysis of SC3b shows
that there is no folded helical conformation, as in SC3b in the
solid state. The large distances between the outer phenyl rings
can be clearly seen in n the top view of SC3b shown in Figure
2.
Density Functional Theory Calculations. To learn more
about the structure and function relationship, density functional
theory (DFT) calculations were performed on SC3a and the
related system SC3b. These procedures resulted in energy-
minimized conformations of the two systems, and the calcula-
tions reveal that there is a nonzero dihedral angle between the
phenyl rings structures within the two terphenyl arms in both
SC3a and SC3b. The relative distortion of the two terphenyl
arms in SC3a is at an angle of ∼30°. Equally, for SC3b, the
arms containing the phenyl rings are crossed, but the relative
distortion angle in this case is ∼40°. Figures 3 and 4 show the
optimized configurations for the SC3a and SC3b, respectively.
Also shown in both figures are the frontier electronic orbitals
for the HOMO and LUMO states. These show, for both systems,
that although the HOMO appears to be confined to one arm
exclusively, the LUMO is delocalized over both arms in the
cruciform structure. In parallel, we also calculated the gas phase
HOMO-LUMO energy gap for both systems. For SC3a, the
energy gap was 5.39 eV (43478 cm-1), and for SC3b, the energy
gap was 4.55 eV (36630 cm-1).
We note that the optimized structures as obtained in the DFT
calculations are close to conformations of SC3b as seen in the
crystal structure (Figures 2 and 4). Regarding the HOMO-LUMO
energy gap, the experimentally measured S0 f S1 transitions
show a slight difference for SC3b and larger difference for
SC3a. We attribute this difference to the calculations’ essentially
being performed in a vacuum, whereas the optical measurements
(excitations) have been performed in a dielectric medium with
distinctively different properties.
Optical Spectroscopy. Figure 5 shows the absorption and
PL of the different swivel cruciform molecules in solution and
solid state (films) at ambient temperatures. The data shows that
there is a steady progression of the absorption maximum toward
longer wavelengths with an increasing size of the π-electron
system. For SC3a, the long wavelength absorption maximum
is located at ∼275 nm (35714 cm-1), but for SC5, it is located
at ∼295 nm (33898 cm-1), and for SC7, the position is ∼330
nm (30303 cm-1). The PL, on the other hand, is concentrated
to a spectrally narrower region and centered in the 380-390
nm range for all swivel cruciforms SC3a, SC5, and SC7.
The Stokes shift of the swivel cruciforms also varies
accordingly with the extent of the π-electron system, as can be
concluded from the data in Table 1, with SC3a showing the
largest Stokes shift, followed by smaller shifts for SC5 and SC7.
In addition, the PLQY depends strongly on the size of the
π-electron system as well as the temperature. Table 1 displays
the data of PLQYs for the various swivel cruciforms at ambient
temperatures, and Figure 6 shows the PLQY, φf(T), as function
of temperature in the temperature range 300-80 K. The data

Oligophenyl-Based Swivel Cruciforms
J. Phys. Chem. B, Vol. 114, No. 40, 2010 12769
Figure 2. Two perspective views of the molecular structure of compound SC3b.
2
show the presence of a temperature-dependent nonradiative
decay channel, in particular for the form SC3a.
εr - 1
2εr + 1
nD - 1
∆f )
-
(2)
2
(
)
2nD + 1
To better understand the photophysics of the swivel type
, where εr is the dielectric constant and nD is the refractive index.
For MCH, ∆f ≈ 0, whereas for ethanol, ∆f ) 0.29, which is a
relatively high value, indicating a fairly polar dielectric medium.
Figure 8a shows that changing the dielectric constant of the
medium has little impact on the absorption maxima or the fine
structure of the PL spectrum. The PLQY of these two
compounds was found to be remarkably high, as Table 1 shows,
with values of 0.82 and ∼1.0 for SC3b and SC1, respectively.
Accordingly, there was also a moderate to insignificant tem-
perature dependence on the PLQY of these two materials (data
not shown). Interestingly, we also found that the PLQY of the
form SC1 is completely media-independent. Measurements in
both polar and nonpolar media, as described previously, returned
the same value (see table 1).
cruciforms SC3a, SC5, and SC7, the related compounds SC3b
and SC1 where tested in a similar fashion. In Figures 7a and
8a, the absorption and PL of the solutions are shown. The
absorption spectrum of SC3b shows, apart from the main band,
a long wavelength absorption band indicative of more than one
absorbing species. However the PL of SC3b is, on the other
hand, broad and devoid of structure, as in the case of SC3a.
For the compound SC1, on the other hand, we observe a broad,
nonstructured absorption, and the PL spectrum exhibits a well-
resolved vibronic progression with peaks at 332, 347.5, and 362
nm. This corresponds to an energy separation of 1343.5 cm-1
between the higher and the middle peak; the energy separation
between the middle peak and the lowest is 1153 cm-1. SC1
was also investigated in a more polar environment. A measure
of solvent polarity can be given by the solvent density parameter,
∆f,
Figure 5b shows the absorption and PL of the swivel
cruciforms in the solid state (films). As expected for low
crystallinity materials, there is moderate impact on the absorp-

12770 J. Phys. Chem. B, Vol. 114, No. 40, 2010
Pålsson et al.
Figure 3. Frontier molecular orbitals calculated for SC3a showing
the HOMO configuration (below) and corresponding LUMO config-
uration (above). See text for details.
tion when going from solution to solid state. The absorption
maximum and the shape of the absorption band remain
essentially unchanged for SC3a. However, for SC5 and SC7,
there is a bathochrome shift of ∼1000 cm-1, which is the only
observed difference. The PL spectra are also affected by the
change of state and, in particular, the PL of SC5 and SC7. The
PL maxima of both these swivel cruciforms are red-shifted
relative to the corresponding solution spectra (table 1). A
significant broadening of the PL for all three systems, SC3a,
SC5 and SC7, is observed for the films as compared with the
corresponding solution spectra. In addition, the PLQY is affected
by the change of state; we observe a reduction for all swivel
cruciforms. The related compound SC1 shows a similar behavior
when going from the solution to the solid state, with only a
small impact on the absorption, whereas the PL is bathochrome
shifted. For SC3b, there is no impact on the position of the
absorption and PL maxima, although an extra red-shifted feature
appears in the film, thus reflecting the fully amorphous character
of the films made from this highly substituted swivel cruciform
(see the Discussion and Table 1 for details).
The time-domain measurements of the PL decay provide more
information about the excited state decay mechanisms. Table 2
lists the data obtained from an analysis of the PL decay of all
swivel cruciforms in solution and the solid state for selected
detection wavelengths. The PL decays and the nonexponential
fits are also shown in Figures 7-10. The data are fitted to a
sum of exponentials according to eq 1, and we found that in
general, two or three exponential terms are needed to get an
acceptable fit of the data. In some cases, a fast component of
∼60 ps was observed. We remark that due to the time resolution
of our detection system, this decay phase (when observed) is
very likely not well-defined by the analysis. Hence, we also
recognize that even faster decay phases in the PL dynamics
could be present but that these accordingly may have escaped
detection.
Figure 4. Frontier molecular orbitals calculated for SC3b showing
the HOMO configuration (below) and corresponding LUMO config-
uration (above). See text for details.
Not included in Table 2 is the temperature dependence of
the time-resolved PL. The impact of the temperature is
particularly dramatic for SC3a in solution, as can be seen in
Figure 9a. Already at 200 K, the long-lived decay component
for this system has increased to 5.7 ns, whereas for SC5 and
SC7 in solution, there is hardly any impact at all. For SC3b,
the PL lifetime increases from 1.3 to 4.0 ns as the temperature
is lowered from 300 to 100 K. For SC1, on the other hand,
there is no impact of the temperature on the PL dynamics.
Discussion
When comparing the optical data of SC3a, SC5, and SC7,
one can derive a correlation between size and extent of the
π-electron systems, the degree of substitution, and the optical
characteristics. This correlation is clearly observed in the
apparent Stokes shift of the emission and the number of phenyl
rings in the cruciform structures. We observe that SC3a has
the largest Stoke shift; SC7 shows the smallest, and SC5 is
intermediate. The implication is that the least extended system,
SC3a, has a large degree of structural relaxation in the excited
state, which successively is reduced as the extent or the size of
the π-electron system is increased and, thus, introduces sterical

Oligophenyl-Based Swivel Cruciforms
J. Phys. Chem. B, Vol. 114, No. 40, 2010 12771
Figure 5. Absorption and luminescence of the swivel cruciforms in solution and film. See text for details.
TABLE 1: Data of the Optical Transitions and PLQYs in
size of the π-conjugated system (number of phenyl units in each
Solution and Solid Statea
branch of the molecule). It is interesting to note that the
absorption maximum of SC3a is very similar to the monomeric
form SC1. The PL of SC3a is, however, bathochrome-shifted,
and the spectrum is broad and devoid of any fine structure,
which was observed for SC1. This suggests that the PL of SC1
is essentially due to S1 f S0 fluorescence.
compound
ν˜abs (cm-1
)
ν˜em (cm-1
)
∆ν˜ (cm-1
)
φf (300 K)
SC3a/MCH 35 714 (280) 25 316 (395)
10 398
7 923
4 987
10 398
7 781
5 625
9 047
9 740
5 306
5 306
7 932
0.10
0.75
0.82
0.04
0.15
0.30
0.82
0.18
0.98
1.00
0.40
SC5/MCH
SC7/MCH
SC3a film
SC5 film
SC7 film
33 898 (295) 25 975 (385)
30 303 (330) 25 316 (395)
35 714 (280) 25 316 (395)
32 787 (305) 25 006 (400)
29 154 (343) 23 529 (425)
Semiempirical calculations reveal that the ground state dipole
moment for SC1 is small, µdip < 1 D. As a consequence,
intramolecular dipole-dipole ground state interactions between
the two arms in SC3a must therefore also be weak, which would
explain why the absorption maximum of SC1 and SC3a is
nearly the same. However, there are clearly interactions in the
excited state, and the data suggest that an intramolecular excited
state aggregate is formed. Support for this conclusion comes
from the broad and red-shifted PL spectrum (relative to SC1)
as observed for SC3a and from the DFT calculations in which
the LUMO was found to encompass both arms in SC3a. The
features of the emission of this excited state aggregate are similar
to what is commonly observed for a classical excimer or
exciplex, the redshift and the absence of vibronic structure.
Because there is no evidence for any ground state interactions,
we therefore suggest attributing this emission to an intra
molecular excimer. In solution, we rule out any intermolecular
excimer formation because the solution is too dilute for this to
occur (∼10-6 M). The data from the time-resolved PL experi-
ments also support this idea. We observed that the PL dynamics
is more than two times slower in SC3a, as compared with SC1,
which is consitent with the formation of intramolecular excimers
and the associated PL emission process.
SC3b/MCH 35 714 (280) 26 667 (375)
SC3b film
SC1/MCH
SC1/EtOH 35 336 (283) 30 330 (333)
SC1 film 36 101 (277) 28 169 (355)
35 714 (280) 25 974 (385)
35 336 (283) 30 030 (333)
a Numbers in parentheses are the corresponding wavelength in
nm. The Stokes shift is calculated from ∆ν˜ ) ν˜abs - ν˜em. MCH is
methylcyclohexane, and EtOH is ethanol. See text for further
details.
More information about the excited state properties can be
obtained from the temperature dependence of the PL, as shown
in Figure 6. There is a dramatic increase in the PLQY at lower
temperatures for SC3a, whereas the other two systems, SC5
and SC7, show less temperature dependence. This strong
temperature dependence suggests that the excited state undergoes
structural changes in which the relative orientation of the
branches is affected, subsequently leading to a nonradiative
decay of the excited state. The temperature dependence suggests
that there is an associated activation energy, Ea, for this
nonradiative decay mechanism. This activation energy is related
to the PLQY φf(T) according to the relation54
Figure 6. Temperature dependence of the three cruciforms in solution.
The symbols represent the measured data, and the line is a fit of data
using eq 3. See text for details.
hindrances in the relaxation process. In the same context, we
note that the related systems SC3a and SC3b have nearly the
same Stokes shift and that the absorption and PL maxima for
these systems are very similar. The implication is therefore that
the main contribution to the Stokes shift originates from the

12772 J. Phys. Chem. B, Vol. 114, No. 40, 2010
Pålsson et al.
Figure 7. Absorption and luminescence of the cruciform SC3b in solution and film. See text for details.
Figure 8. Absorption and luminescence of the cruciform SC1 in solution and film. See text for details.
TABLE 2: Data of the Time-Resolved PL for the Swivel
-Ea/kBT -1
(
)
φf(T) ) A + B·e
(3)
Cruciforms in Solution and Film, and for Excitation at 290
nma
, where A and B are arbitrary constants. The fit of the data (see
Figure 6) and subsequent analysis gives an activation energy
for SC3a of 5.6 kJ/mol; for SC5 and SC7, the activation energy
is 9.7 and 10.1 kJ/mol, respectively. For SC3a, this means that
the activation energy, Ea ) 58 meV, is close to the thermal
energy; kBT ) 26 meV at T ) 300 K, and conformational
changes can therefore be thermally activated. Interestingly, the
related structures SC3a and SC3b showed very different
temperature dependence. At ambient temperatures, SC3b has,
in contrast to SC3a, a relatively high PLQY that rises slightly
at lower temperatures, but its temperature dependence is more
moderate and more complicated as compared with SC3a (data
not shown). For this reason Ea could be determined for SC3b
only. This behavior indicates more sterical hindrance for SC3b
as compared with SC3a but similar dependence as for SC5 and
SC7.
Compound λem (nm) τ1 (ns) A1 τ2 (ns) A2 τ3 (ns) A3
SC3a/MCH
SC3a film
SC5//MCH
SC5 film
340
420
340
440
340
440
340
420
360
440
380
420
390
350
450
360
340
400
0.07 0.150 1.60 0.090
0.07 0.060 1.79 0.050
0.11 0.042 0.87 0.016 3.30 0.034
0.31 0.031 1.53 0.030 3.70 0.026
0.10 0.060 1.70 0.027
0.07 0.040 1.81 0.032
0.06 0.092 0.36 0.026 1.12 0.016
0.33 0.024 1.00 0.008 4.40 0.015
0.06 0.100 0.66 0.050
SC7//MCH
SC7 film
0.07 0.019 0.71 0.056
0.06 0.140 0.41 0.061 0.71 0.041
0.16 0.041 0.46 0.095 0.93 0.022
1.35 0.023
0.26 0.039 1.30 0.045 3.80 0.029
4.31 0.070
0.07 0.120 0.75 0.140
0.42 0.080 1.14 0.027
1.21 0.110 4.79 0.010
SC3b/MCH
SC3b film
SC1/MCH
SC1 film
Direct measurements of the excited state kinetics provided
more information about the interplay between radiative and
nonradiative decay channels. In the PL decay, a fast component
is observed in solution, and this decay time is on the order of
60-70 ps. This phase is observed for SC3a, SC5, and SC7.
a The data are fitted to a sum of exponentials according to eq 1.
All the data are recorded at ambient temperatures, 300 K. See text
for details.

Oligophenyl-Based Swivel Cruciforms
J. Phys. Chem. B, Vol. 114, No. 40, 2010 12773
Figure 9. Time-resolved PL of the swivel cruciforms in solution and film. For the solutions, the PL decay is also recorded at different temperatures.
The PL is recorded in the maximum of all the systems and for all the conditions. See the text for details.
Figure 10. Time -resolved luminescence of the SC3B and SC1 in solution and film. For the solutions, the luminescence decay is also
recorded at different temperatures. The luminescence is recorded in the PL maximum of all the systems and for all the conditions. See the text for
details.

12774 J. Phys. Chem. B, Vol. 114, No. 40, 2010
Pålsson et al.
As the data in Table 2 show, there is little detection wavelength
dependence in this phase, and we therefore attribute it to
planarization of the phenyl rings within the arms of the swivel
cruciforms.32 We note that this decay phase is also observed
for the form SC1, but for the more sterically hindered system
SC3b, such a fast decay of the PL is not present. Although the
two more extended cruciforms SC5 and SC7 had a largely
temperature-independent PL decay, SC3a undergoes dramatic
changes, in particular when the more long-lived decay phase
of the PL is monitored. The PL decay dynamics of SC3a
changes by a factor of 3.6 as the temperature is lowered from
ambient to cryogenic temperature regions (100 K), and the decay
is completely dominated by the long-lived component. For
SC3b, the same scenario is observed, but with a slightly less
dramatic change in the longest lifetime by a factor of ∼2. For
SC1, there is, on the other hand, no temperature dependence
on the excited state decay. In view of the high PLQY for this
system, this is an expected result. The PLQY, φf, and the PL
decay are related and connected properties through the relation
reflected in the absorption spectra because SC3a has a broader
band devoid of any fine structure, which is indicative of a larger
number of subconformations with their inherent variations of
the S0 f S1 transition. Because the LUMO is delocalized over
both arms in SC3a in its the frozen configuration, the rotations
will therefore have a dramatic impact on the excited state, with
strong nonradiative decay, which is implied by the low PLQY
at 300 K. Lowering the temperature will then have the effect
of increasing the viscosity of the medium, which will hinder
rotations, thereby locking the conformation of SC3a closer to
its optimized geometry. As a consequence, the PLQY increases
in parallel to an increased PL lifetime, as can be seen in Figures
3 and 6, which was concluded earlier.
In this context, it should be mentioned that no DFT calcula-
tions were performed on the two more extended analogues SC5
and SC7. The motivation for this is the less pronounced
temperature dependence in these two systems that indicates to
us that there is significantly less relative rotation of the two
arms, primarily due to the increased friction in the condensed
phase. Hence, there is less nonradiative decay caused by
conformational change in those two systems already at ambient
temperatures. In this context, it is relevant to consider the
hydrodynamic impact rotations in viscous medium may have
on these processes. The relative rotation of the arms as in these
systems is, as far as we know, experimentally difficult to assess
because the impact of steric hindrance cannot be correctly
quantified. However, the rotational diffusion can give us some
guidience of the impact due to the physical size of these systems,
assuming spherical shapes. Using the Stokes-Einstein-Debye
relation,
kF
φf )
(4)
k +
k
nr
∑
F
, where kF is the rate of the radiative PL decay and Σknr denotes
the nonradiative decay routes. Because we in this case can
assume that the nonradiative decay is temperature-dependent
(i.e. Σknr ) knr(T)), we can rationalize the increase in the PLQY,
φf, at lower temperatures with a strong decrease in the
nonradiative decay rate knr(T), even though kF also decreases.
In passing, we note that intersystem crossing (kisc) to a triplet
state can also affect the Σknr decay routes. However, we have
observed that the amount of phosphorescence is insignificant
when compared to the fluorescence or excimer PL, and it is
hence safe to assume that kisc ≈ 0. Therefore, it is fair to
conclude that the PL is completely dominated by fluorescence
or excimer PL, at least under the experimental conditions used
in this study.32 Because the PLQY of SC1 was larger than that
of SC3b, this suggests that there is still some degree of
nonraditive decay of the excited state in SC3b, which must then
be due to conformational relaxation of the branches in the
excited state of SC3b. It is therefore also interesting to note
that the tert-butyl groups attached to the phenyl units of SC3b
have a dramatic impact on the extent of excited state confor-
mational relaxation, clearly in contrast to the case of SC3a.
The DFT calculations provide further insight into the relation-
ship between structure and function regarding the excited state
dynamics of SC3a and SC3b, as previously discussed. For both
these systems, the LUMO appears to be delocalized over both
arms, and this is significant when we consider the results of
the optical spectroscopy. The difference in structure between
these two systems is one extra tert-butyl groups on each terminal
phenyl rings on SC3b, and this has the effect of introducing
steric hindrance. There is accordingly limited room for relative
rotations between and within the arms in SC3b, which is also
evident when one considers the optimized conformation as seen
in Figure 4.
kBT
Ds )
(5)
6ηV
the size-dependent rotaional diffusion coefficient can be calcu-
lated. For the arm that is SC3a, we find that the rotational
correlation time will be on the order of 3-4 ns, which is close
to the PL lifetime. For the more extended systems SC5 and
SC7, on the other hand, the rotational diffusion time will be on
the order of tens of nanoseconds, which is well beyond the PL
lifetime in solution of these two systems. On this basis, we can
therefore rationalize the difference in nonradiative decay through
arm rotations for SC3a, as compared with SC5 and SC7.
Interestingly, we observe that going from solution to solid
state (films) did not result in any significant changes in the
Stokes shift of the structures SC3a or SC5. The form SC7,
though, did show a slightly larger Stokes shift, caused mainly
by a bathochrome shift of the PL maximum relative to the
corresponding solution spectrum. Similar observations were
made in studies on related oligothiophene cruciforms based on
that performed by Pina et al.55 In that study, solid state films
where fabricated from oligothiophene chromophores dispersed
in a Zeonex matrix, and although intermolecular interactions
(dipolar) were considered probable, no shift of the absorption
maxima was observed.55 However, bathochrome shifts were
observed in drop-cast films of these oligothiophene cruciforms.55
These are significant observations in view of the results obtained
here, particulary because in this work, the systems where not
dispersed in a matrix before the films were fabricated. Instead,
the films were prepared by allowing the solvent to evaporate,
which likely would result in even closer packing of chromophores.
Accordingly, the data obtained from the films, in particular
the absorption data, suggest that there is no formation of ground
state aggregates, such as physical dimers or higher aggregates.
These aggregates would reveal themselves through a significant
In contrast, SC3a has a less compact structure, and hence,
there is more room for relative motions of the two arms to occur.
This was evident when the calculations where performed
because it was considerably more demanding to find the
optimized structure of SC3a. On the basis of the DFT calcula-
tions, we can therefore conclude that at ambient temperatures,
the two arms in SC3a can rotate more freely relative to each
other, as compared with the related systems SC3b. This is also

Oligophenyl-Based Swivel Cruciforms
J. Phys. Chem. B, Vol. 114, No. 40, 2010 12775
bathochrome shift of the absorption maxima or, more likely, a
new band (Davidov splitting). The slight bathochrome shift that
was observed for SC5 and SC7 could instead be due an
environmental effect. The same data therefore also imply that
intermolecular dipole-dipole interactions in the films are weak
and insufficient to form physical aggregates. The fact that the
permanent dipole moment of SC1 is small (µdip < 1 D), which
in turn implies small permanent dipole moments in SC3a and
its extended analogues SC5 and SC7, supports this assumption.
There is accordingly not the prerequisite in the shape of a large
permanent dipole moment for the formation of physical ag-
gregates in these systems, either. For SC3a, there are, however,
clearer indications for the formation of an intermolecular excited
state aggregate in the films. The absence of any new low-energy
bands, spectral shifts, or broadening in the absorption and the
observed dramatic broadening of the PL (relative to the solution
spectra of SC3a) points to the formation of excited state species
such as excimers. This is also in agreement with the conclusions
made in ref 55. On reflection, the observations made on SC1
and SC3a in solution, and for SC3a films, show a progression
from fluorescence of the monomer (SC1) to the intramolecular
excimer emission from SC3a in solution. Finally, in the solid
state, there is a mixture of inter- and intramolecular excimers
being responsible for the PL of SC3a films.
The presence of EET within the inhomogenous distribution
function will also explain the reduction in the PLQY observed
when going from solution to the solid state (films). Solid state
quenching is observed in a number of cases and not clearly
understood. However, we lean toward an explanation of the EET
to trapped or defect states that will quench the PL as an
explanation for the reduction of the PLQYs in the solid state,
relative to the solution cases. The tighter chromophore packing
of the film will promote EET, and the likelihood of the migrating
exciton reaching such a state is therefore high in the solid state.
However, for the emissive systems, the very likely source of
the emission is excited state aggregates such as excimers, mainly
of intermoloecular nature but possibly with some contributions
from intramolecular excimers, as previously stated. This is
supported by the fact that the PL film spectra are very broad
and structureless and that the PL decay is relatively long-lived.
Acknowledgment. A.P.M. and L.O.P. acknowledge financial
support from CENAMPS, One North East (ONE). We also
thank Professor David J. Tozer (Durham) for useful input to
the DFT calculations.
Supporting Information Available: A detailed description
of the synthetic protocol of all swivel cruciforms. This material
is available free of charge via the Internet at http://pubs.acs.org.
The excited state decay of the films monitored through the
time-resolved PL is more complicated than the corresponding
solution measurements. This difference is due to the packing
of chromophores in the films, which stimulates additional excited
state decay mechanisms. The solid state therefore makes it
unlikely that the decays reflect conformational relaxation
because the free volume is not available to facilitate this. In
addition, we also refer to the very high solution PLQYs,
particularly for SC5 and SC7, which indicates a rigid structure,
and this should also prevail for these two systems in the solid
state. Also for SC3a, we rule out conformational relaxation as
contributing factor to the reduction of that already low PLQY
in going from solution to films. Because this system will also
be tightly packed in the solid state, the free volume to facilitate
rotation of the arms is not available. Therefore, the reduction
of the PLQY in films as compared with the low temperature
solution phase is due to a different effect, as will be discussed
in the following paragraph.
References and Notes
(1) Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Burn, P. L.;
Marks, R. N.; Mackey, K.; Friend, R. H.; Holmes, A. B. Nature 1990, 347,
539.
(2) Gra¨tzel, M. Nature 2001, 414, 338.
(3) Kraft, A.; Grimsdale, A. C.; Holmes, A. B. Angew. Chem., Int.
Ed. 1998, 37, 402.
(4) Shirota, Y. J. Mater. Chem. 2000, 10, 1.
(5) Mitschke, U.; Bauerle, P. J. Mater. Chem. 2000, 10, 1471.
(6) Friend, R. H.; Gymer, R. W.; Holmes, A. B.; Burroughes, J. H.;
Marks, R. N.; Taliani, C.; Bradley, D. D. C.; Dos Santos, D. A.; Bredas,
J. L.; Logdlund, M.; Salaneck, W. R. Nature 1999, 397, 121.
(7) Samuel, I. D. W.; Rumbles, G.; Collison, C. J. Phys. ReV. B 1995,
52, 11573.
(8) Yan, M.; Rothberg, L. J.; Kwock, E. W.; Miller, T. M. Phys. ReV.
Lett. 1995, 75, 1992.
(9) Samuel, I. D. W.; Rumbles, G.; Collison, C. J.; Moratti, S. C.;
Holmes, A. B. Chem. Phys. 1998, 227, 75.
(10) Conwell, E. M.; Pearlstein, J.; Shaik, S. Phys. ReV. B 1996, 54,
R2308.
The complex behavior of the excited state decay for SC3a,
SC5, and SC7 is very likely due to electronic energy transfer
(EET) through a weak Fo¨rster coupling mechanism between
adjacent chromophores. In the solid state, the amorphous
character of these materials will result in an inhomogeneous
distribution function of the energy in the S0 f S1 transition,
and within this inhomogeneous distribution function, there will
be EET from blue to red, which will explain the wavelength-
dependent PL dynamics. Inspection of the data in Table 2 shows
that also in the solid state, a fast decay of 60-160 ps can be
detected; however, this is for blue detection wavelengths only.
For red detection, the fast component is clearly longer when
the individual system is considered. This is in contrast to the
solution case, and we therefore attribute this fast component
observed in films as due to EET. The intermediate component
observed in the films is very hard to assign because it varies
rather unsystematically with system and sometimes with detec-
tion wavelength. In addition, it has a very random intensity
variation which further complicates the interpretation. However,
a plausible source of this intermediate decay phase could be
intramolecular excimer emission, as was observed in the solution
meaurements.
(11) Grell, M.; Bradley, D. D. C.; Long, X.; Chamberlain, T.; Inbasekaran,
M.; Woo, E. P; Soliman, M. Acta Polym. 1998, 49, 439.
(12) Grell, M.; Bradley, D. D. C.; Ungar, G.; Hill, J.; Whitehead, K. S.
Macromolecules 1999, 32, 5810.
(13) Lo, S.-C.; Pålsson, L.-O.; Kilitziraki, M.; Burn, P. L.; Samuel,
I. D. W. J. Mater. Chem. 2001, 11, 2228.
(14) Sinclair, M. B.; McBranch, D.; Hagler, T. W.; Heeger, A. J. Synth.
Met. 1992, 49-50, 593.
(15) Dias, F. B.; Morgado, J.; Macu¨anita, A. L.; Da Costa, F. P.;
Burrows, H. D.; Monkman, A. P. Macromolecules 2006, 39, 5854.
(16) Pålsson, L.-O.; Wang, C.; Russel, D. L.; Monkman, A. P.; Bryce,
M. R.; Rumbles, G.; Samuel, I. D. W. Chem. Phys. 2002, 279, 229.
(17) Pålsson, L.-O.; Vaughan, H. L.; Monkman, A. P. J. Chem. Phys.
2006, 125, 164701.
(18) Wang, P. W.; Liu, Y. J.; Devadoss, C.; Bharathi, P.; Moore, J. S.
AdV. Mater. 199, 8, 237.
(19) Halim, M.; Pillow, J. N. G.; Samuel, I. D. W.; Burn, P. L. AdV.
Mater. 1999, 11, 371.
(20) Pålsson, L.-O.; Beavington, R.; Frampton, M. J.; Lupton, J. M.;
Magennis, S. W.; Markham, J. P. J.; Pillow, J. N. G.; Burn, P. L.; Samuel,
I. D. W. Macromolecules 2002, 35, 7891.
(21) Okumoto, K.; Shirota, Y. Chem. Mater. 2003, 15, 699.
(22) Blochwitz, J.; Pfeiffer, M.; Leo, K. Synth. Met. 2002, 127, 169.
(23) Johansson, N.; Salbeck, J.; Bauer, J.; Weisso¨rtel, F.; Bro¨ms, P.;
Andersson, A.; Salaneck, W. R. AdV. Mater. 1998, 10, 1136.
(24) Spehr, T.; Pudzich, R.; Fuhrmann, T.; Salbeck, J. Org. Electron.
2003, 4, 61.

12776 J. Phys. Chem. B, Vol. 114, No. 40, 2010
Pålsson et al.
(25) Dias, F. B.; Pollock, S.; Hedley, G.; Pålsson, L. O.; Monkman,
A. P.; Tavasli, M.; Perepichka, I.; Bryce, M. R. J. Phys. Chem. 2006, 110,
19329.
(26) Wilson, J. N.; Hardcastle, K. I.; Josowicz, M.; Bunz, U. H. F.
Tetrahedron 2004, 60, 7157.
(27) Wilson, J. N.; Smith, M. D.; Enkelmann, V.; Bunz, U. H. F. Chem.
Commun. 2004, 1700.
(28) Wilson, J. N.; Josowicz, M.; Wang, Y.; Bunz, U. H. F. Chem.Comm.
2003, 2962.
(29) Gerhardt, W. W.; Zucchero, A. J.; South, C. R.; Bunz, U. H. F.;
Weck, M. Chem.sEur. J. 2007, 13, 4467.
(30) Klare, J. E.; Tulevski, G. S.; Sugo, K.; de Picciotto, A.; White,
K. A.; Nuckolls, C. J. Am. Chem. Soc. 2003, 125, 6030.
(31) Nehls, B. S.; Galbrecht, F.; Bilge, A.; Scherf, U.; Farrell, T.
Macromol. Symp. 2006, 239, 21.
(32) Pina, J.; Seixas de Melo, J.; Burrows, H. D.; Galbrecht, F.; Bilge,
A.; Kudla, C. J.; Scherf, U. J. Phys. Chem. B 2008, 112, 1104.
(33) (a) Fukui, K. J. Phys. Chem. 1970, 74, 4161–33. (b) Fukui, K. Acc.
Chem. Res. 1981, 14, 363.
(34) Krishnan, R.; Binkley, J. S.; Seeger, R.; Pople, J. A. J. Chem. Phys.
1980, 72, 650.
(40) (a) Pålsson, L.-O.; Monkman, A. P. AdV. Mater. 2002, 14, 757.
(b) de Mello, J. C.; Wittmann, H. F.; Friend, R. H. AdV. Mater. 1997, 9,
230.
(41) Pålsson, L.-O.; Vaughan, H. L.; Smith, A.; Szablewski, M.; Cross,
G. H.; Roberts, A.; Masutani, A.; Yasuda, A.; Beeby, A.; Bloor, D. J. Lumin.
2006, 117, 113.
(42) Nehls, B. S.; Galbrecht, F.; Lehmann, C. W.; Brauer, D. J.; Farrell,
T.; Scherf, U. Org. Biomol. Chem. 2005, 3, 3213.
(43) Zen, A.; Pingel, P.; Jaiser, F.; Neher, D.; Grenzer, J.; Zhuang, W.;
Rabe, J. P.; Bilge, A.; Galbrecht, F.; Nehls, B. S.; Farrell, T.; Scherf, U.;
Abellon, R. D.; Grozema, F. C.; Siebbeles, L. D. A. Chem. Mater. 2007,
19–1267.
(44) He, F; Tian, L. L.; Tian, X. Y.; et al. AdV. Funct. Mater. 2007, 17,
1551.
(45) Tian, L.; He, F.; Zhang, H.; Hai, X.; Bing, Y.; Wang, C.; Ping, L.;
Muddasir, H.; Fei, L.; Ma, Y.; Jiacong, S. Angew. Chem., Int. Ed. 2007,
46, 3245.
(46) Bilge, A.; Zen, M.; Forster, F.; Galbrecht, B.; Nehls, S.; Farrell,
T.; Scherf, U.; Neher, D. J. Mater. Chem. 2006, 16, 3177.
(47) Zen, A. ; Bilge, A.; Galbrecht, F.; Alle, R.; Meerholz, K.; Grenzer,
J.; Neher, D.; Scherf, U.; Farrell, T. J. Am. Chem. Soc. 2006, 5, 3914.
(48) Suzuki, A. J. Organomet. Chem. 1999, 576, 147.
(49) Suzuki, A. J. Organomet. Chem. 2002, 653, 83.
(50) Sakamoto, J.; Rehahn, M.; Wegner, G.; Schluter, A. D. Macromol.
Rapid Commun. 2009, 30 (9-10), 653 Sp. Iss. SI.
(51) Nehls, B. S.; Galbrecht, F.; Brauer, D. J.; Lehmann, C. W.; Scherf,
U.; Farrel, T. J. Polym. Sci., Part A: Polym. Chem. 2006, 44, 5533.
(52) Nehls, B. S.; Fuldner, S.; Preis, E.; Farrel, T.; Scherf, U.
Macromolecules 2005, 38, 687.
(53) Galbrecht, F.; Yang, X. H.; Nehls, B. S.; Neher, D.; Farrell, T.;
Scherf, U. Chem. Commun. 2005, 2378.
(54) Ferrante, C.; Kensy, U.; Dick, B. J. Phys. Chem. 1993, 97, 13457.
(55) Pina, J.; Seixas de Melo, J.; Burrows, H. D.; Bilge, A.; Farrel, T.;
Forster, M.; Scherf, U. J. Phys. Chem. 2006, 110, 15100.
(35) (a) Wachters, A. J. H. J. Chem. Phys. 1970, 52, 1033. (b) Hay,
P. J. J. Chem. Phys. 1977, 66, 4377.
(36) (a) Gordon, M. S. Chem. Phys. Lett. 1980, 76, 163. (b) Hariharan,
P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213. (c) Binning, R. C., Jr.;
Curtiss, L. A. J. Comput. Chem. 1990, 11, 1206.
(37) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Miehlich, B.;
Savin, A.; Stoll, H.; Preuss, H. Chem. Phys. Lett. 1989, 157, 200. (c) Lee,
C.; Yang, W.; Parr, G. Phys. ReV. B 1988, 37, 785. (d) Stephens, P. J.;
Devlin, F. J.; Chabalowski, C. F. J. Phys. Chem. 1994, 98, 11623.
(38) Frisch, M. J. Gaussian 03, reVision B.05; Gaussian, Inc.: Pittsburgh,
PA, 2003.
(39) Williams, A. T. R.; Winefield, S. A.; Miller, J. N. Analyst 1983,
108, 1067.
JP1028883