Electronic, redox, and photophysical consequences of metal-for-carbon substitution in oligo-phenylene-ethynylenes

Article abstract of DOI:10.1021/ja411354d

The electronic structures, redox chemistry, and excited-state properties of tungsten-containing oligo-phenylene-ethynylenes (OPEs) of the form W[C(p-C ₆H₄CC)_n-1Ph](dppe)₂Cl (n = 1-5; dppe =1,2-bis(diphenylphosph

Full text of DOI:10.1021/ja411354d

Article
pubs.acs.org/JACS
Electronic, Redox, and Photophysical Consequences of Metal-for-
Carbon Substitution in Oligo-Phenylene-Ethynylenes
Daniel C. O’Hanlon,† Brian W. Cohen,† Davis B. Moravec,† Richard F. Dallinger,‡
and Michael D. Hopkins*,†
†Department of Chemistry, The University of Chicago, 929 East 57th Street, Chicago, Illinois 60637, United States
‡Department of Chemistry, Wabash College, Crawfordsville, Indiana 47933, United States
S
* Supporting Information
ABSTRACT: The electronic structures, redox chemistry, and excited-state
properties of tungsten-containing oligo-phenylene-ethynylenes (OPEs) of the
form W[C(p-C6H4CC)n−1Ph](dppe)2Cl (n = 1−5; dppe =1,2-bis-
(diphenylphosphino)ethane) are reported and compared with those of organic
analogues in order to elucidate the effects of metal-for-carbon substitution on OPE
bonding and electronic properties. Key similarities between the metallo- and
organic OPEs that bear on materials-related functions include their nearly identical
effective conjugation lengths, reduction potentials, and π* orbital energies and
delocalization. In addition to these conserved properties, the tungsten centers endow OPEs with reversible one-electron
oxidation chemistry and long-lived emissive triplet excited states that are not accessible to organic OPEs. The electronic
similarities and differences between metallo- and organic OPEs can be understood largely on the basis of π/π* orbital energy
matching between tungsten and organic PE fragments and the introduction of an orthogonal mid-π/π*-gap d orbital in metallo-
OPEs. These orbital energies can be tuned by varying the supporting ligands; this provides a means to rationally implement and
control the emergent properties of metallo-OPE materials.
centers in metallo-PEs have been incorporated into the
backbone between PE segments (A, Chart 1), such as in the
INTRODUCTION
■
The p-phenylene-ethynylene moiety (PE) and its derivatives
are primary building blocks of photo- and electro-active π-
conjugated molecular assemblies, oligomers, and polymers,
because these materials are typically endowed with strong
luminescence, controllable electron conductance, structural
rigidity, and a myriad of hybrid properties derived there-
from.1−5 For example, oligomeric PEs (OPEs) have been
widely employed as defined-length bridging units for
connecting molecules to each other and to nanostructured or
solid surfaces; applications include linkers in donor−bridge−
acceptor systems6,7 and dye-sensitized solar cells,8 chromo-
phore scaffolds for energy transfer and light harvesting,7,9 and
components for molecular electronics.4,10 Along similar lines,
polymeric PEs have been developed as sensors,11−13 organic
solar cells,14 and organic light-emitting diodes11,15,16 due to
their large exciton mobilities and fluorescence quantum
yields.11,16 The synthetic parameter space for controlling the
electronic structures and properties of organic PE compounds
and materials has been extensively explored, with hundreds of
derivatives of the parent PE building block having been
studied.1−5,11,14,15
Chart 1. Structural Motifs of Metallo-PEs
extensively studied class of PtIIL2-containing oligomers and
polymers,20 or as pendant groups on the chain (B, Chart 1), as
in materials that contain 2,2′-bipyridine units to which are
ligated luminescent chromophores (e.g., RuII(bpy)32+).17,18,21
In contrast to these motifs, in which the metal center is external
to the PE building block, in motif C (Chart 1) the PE unit is
modified through replacement of an ethynyl carbon atom by a
metal center. Because the metal d orbitals are integral to the
In parallel with synthetic-organic approaches to controlling
the properties of PE materials, considerable effort has been
devoted to developing transition-metal-containing PEs. Interest
in these analogues stems from the fact that the metal centers
provide complementary avenues for manipulating electronic
structure and for introducing properties that are not accessible
to all-organic systems.2,3,17−19 Most commonly, the metal
Received: November 7, 2013
Published: January 31, 2014
© 2014 American Chemical Society
3127
dx.doi.org/10.1021/ja411354d | J. Am. Chem. Soc. 2014, 136, 3127−3136

Journal of the American Chemical Society
Article
MC bond, the electronic coupling between the metal center
and the PE π system in C is expected to be stronger than for
metallo-PEs of types A and B, which do not possess substantial
metal−ligand multiple-bond character. Thus, these metal−
alkylidyne-derived materials afford distinctive opportunities for
controlling and enhancing the properties of PE materials.
Among the wide variety of metal−alkylidyne complexes that
might be employed as ethynyl-group replacements in type C
metallo-PEs,22 d2 W(CR)L4X complexes stand out as
particularly function-rich building blocks. Characteristic fea-
tures of these compounds relevant to PE properties include
their long-lived phosphorescent excited states,23−29 widely
tunable oxidation potentials,30,31 and stable d1 metalloradical
configurations.26,32 Because of these attributes, W(CR)L4X
building blocks have been used to construct a variety of redox-
and photoactive metallo-PEs and metallo-ynes that implement
their rich properties in electronic and solar-energy-conversion
materials.23,24,27,28,33−36
In order to rationally design metallo-PE materials, it is
necessary to understand at a fundamental level how the
electronic structure−property relationships of organic PEs are
affected by the incorporation of metal centers. With this goal in
mind, we report a comparative study of the electronic structures
and redox and excited-state properties of type C tungsten-
containing OPEs of the form W[C(p-C6H4CC)n−1Ph]-
(dppe)2Cl (1−5, Chart 2; dppe =1,2-bis(diphenylphosphino)-
a means to rationally implement and control the properties
derived from the tungsten fragment in metallo-PE materials.
RESULTS AND DISCUSSION
■
Nature of the Frontier Orbitals. The electronic structures
of W[C(C6H4CC)n−1Ph](dppe)2Cl compounds 1−5 and of
their organic analogues OPE-1−OPE-5 (Chart 2) were
calculated using density functional theory. The calculated gas-
phase molecular structures of 1−5 exhibit nearly identical
geometries about their tungsten centers (Table S1 in
Supporting Information (SI)) and, like the OPE-n compounds,
essentially planar configurations within their PE units
(interphenylene dihedral angles <4.5°). Because the key
structural features of 1−5 are nearly identical, it may be
concluded that the differences among their frontier orbitals and
electronic properties are a consequence of the different lengths
of their PE alkylidyne ligands.
The frontier orbitals of 1−5 (Figure 1) are derived from the
tungsten dxz, dyz, and dxy (t2g) orbitals, as has generally been
found for d2 metal−alkylidyne complexes with compact
alkylidyne ligands.26,30,38 The HOMOs of all compounds are
nearly identical in atomic composition, being principally of
tungsten dxy character; this orbital is of π-symmetry with
respect to the dppe ligands (∼25% contribution to the
HOMO), and nonbonding relative to the axial CPh and Cl
ligands (0% contribution; Table S2 in SI). As a consequence of
its axial-nonbonding character the HOMO energy is nearly
independent of the length of the alkylidyne ligand, spanning a
range of only 0.11 eV across the series (Figure 2). The energy
and compositional invariance of the dxy-derived HOMO figures
prominently in the redox and excited-state properties of 1−5
(vide infra).
Chart 2. Tungsten-Containing and Organic Oligo-
Phenylene-Ethynylenes
The frontier π and π* orbitals of the PE backbones of 1−5
are the HOMO−1 and LUMO, respectively (Figure 1). These
orbitals possess the typical nodal relationships of a correspond-
ing π/π* orbital pair, such as for the analogous π/π* orbitals of
OPE-1−OPE-5 (Figure S1, SI), but differ markedly in the
extents to which their energies and delocalization depend on
the length of the PE chain. The properties of the π* orbitals of
1−5 are strikingly similar to those of OPE-1−OPE-5. In
particular, the π*(WCR) and π*(OPE) orbital energies both
decrease smoothly with increasing oligomer length, differing by
less than 0.3 eV for a given length n (Figure 2). Furthermore,
both the π*(WCR) (Figure 1) and π*(OPE) orbitals (Figure
S1, SI) are delocalized across the PE chain, with corresponding
[CCC6H4] subunits of 2−5 and OPE-2−OPE-5 having very
similar amplitudes. The π* LUMOs of 3 and OPE-3 (Figure 3)
are representative of these close correspondences. For this pair
of analogues, the contributions to the π* orbitals from the
terminal [WC(dppe)2Cl] and HCCC6H4 units are 30% and
32%, respectively; those from the central [CCC6H4] units are
45% and 47%; and the capping CCPh units contribute 25% and
22% (Tables S3 and S4, SI). Other pairs of analogues among
1−5 and OPE-1−OPE-5 exhibit similarly close relationships
(Figure S2, SI). Thus, replacement of the [HCC] unit with
[WC(dppe)2Cl] at the terminus of a PE chain results in only a
modest perturbation of the energies and delocalization of the
OPE π* LUMOs, despite the strong mixing between the
π*(WC) and π*[(C6H4CC)n−1Ph] fragment orbitals.
In contrast to the close relationships between the π* LUMOs
of 1−5 and OPE-1−OPE-5, the properties of their π orbitals
differ significantly from each other. For OPE-n, the π orbital
energy (Figure 2) and delocalization (Figure S1, SI) change
ethane)37 and of their organic analogues (OPE-1−OPE-5,
Chart 2). It is found that there are both striking similarities and
differences among the electronic structures and properties of
these classes of oligomers. Key similarities between metallo-
and organic PEs that bear on materials-related functions include
their nearly identical effective conjugation lengths, reduction
potentials, and π* orbital energies and delocalization. In
addition to these conserved properties, the tungsten centers
endow OPEs with reversible one-electron oxidation chemistry
and long-lived emissive triplet excited states that are not
accessible to organic OPEs. The electronic relationships
between these classes of materials can largely be understood
on the basis of π/π* energy matching between tungsten and
organic OPE fragments and the introduction in the metallo-PEs
of an orthogonal, mid-π/π*-gap d orbital. The fact that these
aspects can be tuned by varying the supporting ligands provides
3128
dx.doi.org/10.1021/ja411354d | J. Am. Chem. Soc. 2014, 136, 3127−3136

Journal of the American Chemical Society
Article
Figure 1. Top: dxy-derived HOMO of 1, viewed normal to the equatorial and alkylidyne-ligand planes; the HOMOs of 2−5 are nearly
indistinguishable from that of 1. Bottom: π(WCR) and π*(WCR) orbitals of 1−5, viewed normal to the plane of the alkylidyne ligand. Hydrogen
atoms and dppe carbon atoms are omitted for clarity. Orbitals are rendered at iso 0.98 (iso 0.97 for HOMO and LUMO of 1).
for 3 and OPE-3 are illustrated in Figure 3; the other pairs of
compounds also exhibit large differences (Figure S3, SI).
Consistent with their limited delocalization, the energies of the
π(WCR) orbitals of 1−5 are independent of the length of the
alkylidyne ligand and span a range of only 0.05 eV; by
comparison, the π(OPE) orbital energies of OPE-1−OPE-5
span a range of 1 eV (Figure 2).
The different behavior of the π(WCR) HOMO−1 and
π*(WCR) LUMO of 1−5 as a function of the length of the PE
unit is a consequence of the energy denominators that describe
the orbital interactions between the [W(CC6H4)(dppe)2Cl]
and [(CCC6H4)n−1H] fragments. The close correspondence
between the energies and delocalization of the π* LUMOs of
1−5 and OPE-1−OPE-5 is traceable to the small difference
between the π* orbital energies of parent compounds 1 and
OPE-1 (ΔE(π*) = 0.06 eV). As a result of the matched π*
energies of the [HCCC6H4] and [W(CC6H4)(dppe)2Cl]
fragments, their interactions with the π* orbitals of a given
length [(CCC6H4)n−1H] fragment are governed by similar
energy denominators. In contrast, there is a large difference
between the π orbital energies of 1 and OPE-1 (ΔE(π) = 1.53
eV). For the π(WCR) orbitals of extended W[C-
(C6H4CC)n−1Ph](dppe)2Cl complexes 2−5, the relatively
poor energy match between the π fragment orbitals of the
[W(CC6H4)(dppe)2Cl] and [(CCC6H4)n−1H] units results in a
low degree of mixing compared to that for the π HOMOs of
OPE-n compounds. Although this zero-order energy gap
decreases as the chain length increases (ΔE(π) = 0.56 eV for
5 and OPE-5), fragment-orbital mixing is still limited because
the orbitals of longer [(CCC6H4)n−1H] fragments are
increasingly localized on the central phenylene-ethynylene
units rather than near the terminal [W(CC6H4)(dppe)2Cl]
fragment. Therefore, the π(WCR) orbitals of 1−5 are primarily
of [W(CC6H4)(dppe)2Cl] character and vary little in energy
with chain length. The differences between the π(WCR) and
π*(WCR) orbitals of W[C(C6H4CC)n−1Ph](dppe)2Cl com-
plexes are manifested in a number of their electronic properties
(vide infra).
Figure 2. Orbital energies of W[C(C6H4CC)n−1Ph](dppe)2Cl (1−5)
and HCC(C6H4CC)n−1Ph (OPE-1−OPE-5).
Figure 3. Corresponding π and π* orbitals of 3 and OPE-3 (iso 0.98).
Hydrogen atoms and dppe carbon atoms are omitted for clarity.
smoothly with increasing chain length in a manner that mirrors
the properties of the π* orbitals described above. The π(WCR)
orbitals of 1−5, in contrast, do not increasingly delocalize as the
alkylidyne ligand is extended (Figure 1). Instead, the principal
contribution to these orbitals (>75%) comes from the terminal
[W(CC6H4)(dppe)2Cl] unit; for 3−5, the contributions to
π(WCR) from the [CCC6H4] units beyond the [W-
(CC6H4CCC6H4)(dppe)2Cl] fragment of 2 total only 4−6%
(Table S3, SI). The evident differences in π orbital coefficients
Electrochemical Properties. The electrochemistry of
W[C(C6H4CC)n−1Ph](dppe)2Cl complexes was investigated
to gain insight into how their redox properties differ from those
of organic OPEs. Oxidation and reduction potentials for 1−4
are set out in Table 1; cyclic voltammograms are shown in
3129
dx.doi.org/10.1021/ja411354d | J. Am. Chem. Soc. 2014, 136, 3127−3136

Journal of the American Chemical Society
Article
reduction potentials exhibit an approximate linear correlation
with the calculated π*(WCR) LUMO energy (slope =0.97,
RMSD = 0.04 V; Figure S7, SI). In view of the close parallels
between the reduction properties of 1−4 and organic PEs, it is
predicted that W[C(C6H4CC)n−1Ph](dppe)2Cl derivatives
should retain reduction-associated electronic properties of the
organic analogues such as their high electron mobilities and low
barriers to electron injection in OLEDs,2,16 while at the same
time providing the new oxidation chemistry noted above.
Electronic-Absorption Spectra and Effective Conjuga-
tion Length. The electronic-absorption spectra of W[C-
(C6H4CC)n−1Ph](dppe)2Cl complexes, shown in Figure 4,
exhibit bands that are characteristic both of d2 metal−alkylidyne
complexes25 and of organic OPEs. For 1, the lowest-energy
absorption band (λmax = 531 nm, ε = 230 M−1 cm−1; Table 2)
Table 1. Electrochemical Data for
W[C(C6H4CC)n−1Ph](dppe)2Cl Complexes
a
0/+
potential, V vs FeCp2
b
−/n−
0/+
0/−
cmpd
E1/2
E1/2
Epc
c
e
e
e
,d
b
1
2
3
4
−0.58
−0.57
−0.57
−0.57
−3.39
f
b,
g
g
−2.95
−2.54
−2.44
−3.11, −3.37
−2.86
−2.75
h
h
a
THF solution, 25 °C, ν = 0.1 V s−1 unless otherwise indicated.
b
c
d
e
Irreversible, Epc reported. Reference 30. 0.3 M [NBun ][PF6]. 0.1
4
f
g
h
M [NBun ][PF6]. ν = 0.5 V s−1. ν = 0.2 V s−1. Quasi-reversible, E1/2
4
listed.
Figures S4−S6, SI. All compounds exhibit a reversible one-
electron oxidation and one or more irreversible reductions.
Data for 1 were reported previously as part of a study of the
electrochemistry of d2 W(CR)L4X complexes in which the CR,
L, and X ligands were systematically varied.30 It was found that
the first oxidation of these complexes arises from the (dxy)2/
(dxy)1 couple; because the dxy orbital is axially nonbonding, the
oxidation potential is only weakly sensitive to the nature of the
CR ligand.30 Consistent with these general observations, and
with the invariance of the calculated energy of the dxy-derived
HOMO (Figure 2), the oxidation potentials of 1−4 are
identical within experimental error (E1/20/+ = −0.57 0.01 V vs
FeCp20/+, Table 1). Organic PEs, which lack the mid-π/π*-gap
dxy redox orbital, are oxidized at much more positive potentials
than 1−4;16 for example, the onset oxidation potential for
Table 2. Electronic-Absorption Spectroscopic Data for
a
W[C(C6H4CC)n−1Ph](dppe)2Cl.
λmax, nm (εmax, M−1 cm−1
)
1[dxy→
π*(WCR)]
1[π(WCR)→
π*(WCR)]
cmpd
1[π(R)→π*(R)]
b
,
c
b
1
2
3
4
531 (230)
604 (270)
623 (360)
628 (520)
340 (16400)
d
d
386 (32300)
404 (39100)
410 (50500)
308 (33100), 320 (33800)
326 (52600), 340 (51000)
a
Toluene solution, room temperature; spectra in THF solution are
b
c
similar (Table S5, SI). Reference 26. Reported as λmax = 525 nm and
εmax = 310 M−1 cm−1 in ref 26; values reported here were obtained
1
from deconvolution of the [dxy→π*(WCR)] band and overlapping
bands. Reference 27.
39
0/+
polymeric [C6H2(OR)2CC]n is ∼0.65 V vs FeCp2
.
Given
d
that the oxidation potentials of simple W(CR)L4X complexes
are tunable over a 2 V range via variation of the electronic
properties of the equatorial L ligands,30 it is expected that the
potentials of these midgap redox states should be subject to
similarly broad synthetic control for other W[C-
(C6H4CC)n−1Ph]L4X derivatives.
has been assigned previously to the spin- and dipole-allowed
1[dxy→π*(WCR)] (HOMO→LUMO) transition and the
strong band in the near-UV region (λmax = 340 nm, εmax
=
16400 M−1 cm−1) has been assigned to the fully allowed
1[π(WCR)→π*(WCR)] (HOMO−1→LUMO) transition on
In contrast to the significantly different oxidation properties
of metallo- and organic PEs, the reduction properties of these
analogues are strikingly similar, with the first reduction
potentials of 2−4 (Table 1) being within 0.15 V of those
reported for Ph(CCC6H4)mH compounds (Ep0/−: 2 = −2.95 V,
PhCCPh = −2.95 V; 3 = −2.54 V, Ph(CCC6H4)2H = −2.69 V;
4 = −2.44 V, Ph(CCC6H4)3H = −2.46 V; THF solution, all
potentials vs FeCp20/+).40 This observation strongly reinforces
the close correlation found between the π* LUMO energies of
1−5 and OPE-1−OPE-5 (Figure 2), given that these are the
redox orbitals for the process. Indeed, for 1−4 the measured
the basis of TD-DFT calculations.26 Corresponding [dxy→
1
1
π*(WCR)] and [π(WCR)→π*(WCR)] absorption bands are
observed in the spectra of 2−4 (Figure 4). These bands redshift
as the alkylidyne ligand is lengthened, with decreasing marginal
shifts from each PE unit. Additionally, the extinction
coefficients of the 1[dxy→π*(WCR)] and 1[π(WCR)→π*-
(WCR)] bands increase by factors of two and three,
respectively, between 1 and 4 (Table 2). Qualitatively, this
follows the trend observed for the π→π* transitions of organic
analogues of form Ph(CCC6H4)mH.41,42
Figure 4. Electronic-absorption spectra of W[C(C6H4CC)n−1Ph](dppe)2Cl compounds in toluene.
3130
dx.doi.org/10.1021/ja411354d | J. Am. Chem. Soc. 2014, 136, 3127−3136

Journal of the American Chemical Society
Article
The shift of the electronic-absorption band maxima for 1−4
as a function of the number of PE units may be used to predict
both the optical band gap for a hypothetical infinite-chain-
length W[C(C6H4CC)n−1Ph](dppe)2Cl complex and the
“effective π conjugation length” of the metal−alkylidyne ligand.
It has been observed that the dependence of the π→π* band
energies of conjugated organic oligomers on the number of
repeat units is generally well described by the following
empirical relationship (eq 1)4,43
positions and shapes of these bands overlay closely with
those of the lowest-energy 1[π→π*] bands of analogous
organic OPEs (λmax(3) = 320 nm, λmax(Ph(CCC6H4)2H) = 320
nm; λmax(4) = 326 nm, λmax(Ph(CCC6H4)3H) = 340 nm;
Figures S9−S10 in SI),42 suggesting that the orbitals between
which these transitions occur have small contributions from the
[W(C)(dppe)2 Cl] unit. For hypothetical W[C-
(C6H4CC)n−1Ph](dppe)2Cl compounds with PE alkylidyne
ligands longer than that of 4, it is expected that the these ligand-
1
localized [π(R)→π*(R)] bands will continue to lie to higher
E(n) = E∞ + (E1 − E∞)exp(−a(n − 1))
(1)
1
1
energy of the [π(WCR)→π*(WCR)] and [dxy→π*(WCR)]
bands, based on their predicted positions for hypothetical
infinite-length compounds (λmax(n = ∞): 1[π(R)→π*(R)],
∼375 nm;43,46 1[π(WCR)→π*(WCR)], 413 nm; 1[dxy→
π*(WCR)], 629 nm).
where E(n) is the band energy for the oligomer of n repeating
units, E1 is the band energy for the monomer, E∞ is the band
energy for the infinite polymer, and a is an attenuation factor.
The fits of the 1[dxy→π*(WCR)] and 1[π(WCR)→π*(WCR)]
band energies of 1−4 to eq 1 (Figure 5; see also Figure S8 in
Excited-State Properties. In contrast to the high-
1
quantum-yield [ππ*] fluorescence observed for organic PE
oligomers and polymers,2,15 the W[C(C6H4CC)n−1Ph]-
(dppe)2Cl complexes phosphoresce in fluid solution at room
temperature (Figure 6 and Table 3). The emission from 1 has
Figure 5. Energies of the 1[dxy→π*(WCR)] and 1[π(WCR)→
π*(WCR)] electronic-absorption band maxima and 3[dxy←π*(WCR)]
emission band maxima of W[C(C6H4CC)n−1Ph](dppe)2Cl com-
pounds. The curves represent the fit of the data to eq 1.
Figure 6. Emission spectra of W[C(C6H4CC)n−1Ph](dppe)2Cl
compounds in toluene at room temperature. Band areas reflect the
emission quantum yields.
SI) predict that the band maxima for a W[C(C6H4CC)n−1Ph]-
(dppe)2Cl complex with an infinitely long PE alkylidyne ligand
1
lie at E∞ = 15890 cm−1 (629 nm) for [dxy→π*(WCR)] and
E∞ = 24200 cm−1 (413 nm) for 1[π(WCR)→π*(WCR)].
These band maxima are red-shifted from those observed for 4
by only 1 nm (30 cm−1) for 1[dxy→π*(WCR)] and 3 nm (180
3
been previously assigned as originating from the [(dxy)1(π*-
(WCPh))1] excited state, based on electronic-structure
calculations and the fact that the excited-state structure
determined by X-ray transient-absorption measurements is
consistent with that anticipated for the (dxy)1(π*(WCPh))1
configuration.26 The emission bands of 1−4 redshift with
lengthening of the OPE ligand and maintain nearly constant
band widths and Stokes shifts relative to the 1[dxy→π*(WCR)]
absorption band (Table 3); this both supports assignment of
1
cm−1) for [π(WCR)→π*(WCR)]. The effective conjugation
length of an oligomer has been defined as that for which the
experimentally observed absorption maximum lies within 1 nm
of that predicted for the infinite polymer.43 Thus, the behavior
1
1
of both the [dxy→π*(WCR)] and [π(WCR)→π*(WCR)]
bands of 1−4 indicate that the effective conjugation length of
the W[C(C6H4CC)n−1Ph] unit is 4−5 repeat units. (The same
value is found from an analysis of the emission bands of the
compounds, which are described below.)44 This conjugation
length is the same as that found for organic PE analogues of the
form p-Et2N3(3-R-C6H3CC)nSiMe3 (R = alkyl, n ≤ 16), for
which a fit of π→π* absorption-band energies to eq 1 provides
a conjugation length of 5 repeat units.43
3
the emissive state as [(dxy)1(π*(WCR))1] for all compounds
in the series and is consistent with an effective conjugation
length of 4−5 units (Figure 5).44 In view of the prediction from
1
the preceding section that the [(dxy)1(π*(WCR))1] state will
1
lie lower in energy than the PE-ligand-localized [ππ*] state at
the limit of an infinitely long alkylidyne ligand, it is anticipated
that W[C(C6H4CC)n−1Ph](dppe)2Cl derivatives with longer
1
The electronic-absorption spectra of 3 and 4 also exhibit
strong bands in the 300−350 nm region whose properties are
consistent with assignment to alkylidyne-ligand-localized
1[π(R)→π*(R)] transitions (Figure 4 and Table 2). Unlike
the featureless 1[dxy→π*(WCR)] and 1[π(WCR)→π*(WCR)]
bands, these ligand-centered bands exhibit vibronic shoulders
whose spacing (∼1200−1300 cm−1) is consistent with
assignment to phenylene-based stretching modes.45 The
ligands will not exhibit PE-like [ππ*] fluorescence unless the
emitting segment can be spatially decoupled from the metal-
derived excited states.
The luminescence lifetimes and quantum yields of
compounds 1−4 depend strongly on the length of the PE
unit, with the lifetimes (τem) decreasing from 303 to 39 ns and
the quantum yield (ϕem) from 0.017 to 0.0025 from 1 to 4
(Table 3). These differences are the result of an 8-fold increase
3131
dx.doi.org/10.1021/ja411354d | J. Am. Chem. Soc. 2014, 136, 3127−3136

Journal of the American Chemical Society
Article
Table 3. Emission Spectroscopic and Photophysical Data for the 3[(dxy)1(π*(WCR))1] State of W[C(C6H4CC)n−1Ph](dppe)2Cl
a
Complexes.
b
c
d
cmpd
λmax, nm
Stokes shift, cm−1
fwhm, cm−1
τem, ns
ϕem
kr, 104 s−1
knr, 106 s−1
e
e
e
1
2
3
4
668
3860
3670
3640
2760
2640
2540
2520
303
0.017
5.61
7.64
7.14
3.24
9.36
20.3
f
f
f g
776
106
0.0081 ,
806
816
49
39
0.0035
3670
0.0025
6.41
25.6
a
b
c
d
e
Toluene solution, room temperature. Shift from the 1[dxy→π*(WCR)] absorption band (Table 3). kr = ϕem/τem. knr = 1/τem−kr. Reference 26.
f
g
Reference 27. The quantum yield reported here is larger than that reported in ref 27 by a factor of 1.5; this scaling reflects the subsequent revision
2+
to the absolute quantum yield of the Ru(bpy)3 reference compound from ϕ = 0.042 to 0.063 (ref 47).
in the nonradiative decay rate (knr) across the series; the
radiative rate remains constant. This contrasts with the
behavior of organic OPEs, for which the fluorescence lifetime
decreases with oligomer length due to an increase in the
radiative rate.48 For 2−4, the differences in nonradiative rate
appear to be a manifestation of decay in the weak-coupling
limit, as indicated by the observation of a linear (energy-gap-
law) correlation between ln(knr) and Eem (Figure 7).49 Because
thermally activated nonradiative decay channel; this is in
reasonable agreement with that predicted by extrapolation of
the energy-gap-law fit to the rates of 2−4 (3.36 × 105 s−1),
given the uncertainties inherent in extrapolating the two
relationships. We conclude, therefore, that 1 would conform to
the broader energy-gap-law relationship exhibited by 2−4 if the
higher-activation-energy nonradiative decay channel were not
present.
To further probe the properties of the emissive excited-states
of W[C(C6H4CC)n−1Ph](dppe)2Cl complexes, the transient-
absorption spectra of 1−4 were measured. The spectra acquired
∼7−10 ps following excitation are shown in Figure 8. These
3
Figure 7. Energy-gap law plot for the emissive [(dxy)1(π*(WCR))1]
excited states of 1−4. The line is a linear fit to the data for 2−4: ln(knr)
= −0.00155Eem + 36.108 (R2 = 0.9999).
3
the energy of the [(dxy)1(π*(WCR))1] excited-state energy of
4 approaches that predicted for a hypothetical complex with an
infinite-length OPE ligand (Figure 5),44 the conformation of
these compounds to the energy-gap law suggests that
chromophores with longer OPE ligands will possess emission
lifetimes of magnitude comparable to that of 4 (in the absence
of new nonradiative decay modes introduced by the extended
ligands). The nonradiative rate of 1 deviates significantly from
the linear correlation for 2−4 (knr(obs) = 3.24 × 106 s−1,
knr(pred) = 3.36 × 105 s−1), indicating the presence of an
additional nonradiative decay pathway. Comparison of the
temperature dependence of the nonradiative decay rates of 1
and 2 shows that at room temperature 1 additionally decays via
thermal population of a higher-lying excited state (see Figure
S11, Tables S6 and S7, and associated discussion in the SI).
That the energy gap between the emissive 3[(dxy)1(π*-
(WCR))1] excited state and next higher-lying excited state is
considerably smaller for 1 than for 2−4 is indicated by their
electronic-absorption spectra (Figure 4), which show that for 1
Figure 8. Transient-absorption spectra of 1−4 in toluene (λex = 400
nm) at Δt ≅ 5(τrise). Values of ΔA in the 440−760 nm and 850−1400
nm regions are not comparable because the spectra are acquired on
separate instruments at different laser powers.
time points are those at which the prominent features in the
spectra have reached their maximum absorbance, and following
which the spectra decay according to the emission lifetime but
do not evolve in shape; thus, they originate from the thermally
equilibrated 3[(dxy)1(π*(WCR))1] excited state of each
compound. Kinetic analyses of time-resolved spectra provide
time constants for the rise of the features shown in Figure 8 of
τrise ≅ 1−2 ps (Figures S12−S14, SI), which presumably
reflects vibrational cooling and/or intersystem crossing from a
1
the [dxy→π*(WCR)] band strongly overlaps with the higher-
lying absorption band whereas for 2−4 these bands are well
separated. Based on an Arrhenius analysis of emission lifetime
data (see the SI), the room-temperature nonradiative rate of 1
would lie in the range ∼5.7−7.8 × 105 s−1 in the absence of the
3
higher-lying singlet state to the emissive [(dxy)1(π*(WCR))1]
3132
dx.doi.org/10.1021/ja411354d | J. Am. Chem. Soc. 2014, 136, 3127−3136

Journal of the American Chemical Society
Article
Table 4. Ground-State and 3[(dxy)1(π*(WCR))1] Excited-State Electronic-Absorption Bands of W[C(C6H4CC)n−1Ph](dppe)2Cl
Complexes
λmax, nm
a
b
cmpd
1[π(WCR)→π*(WCR)]
3[π(WCR)→π*(WCR)]TA
3[π*(WCR)→π*n]TA
ΔE, cm−1
1
2
3
4
340
386
404
410
420
7440
7890
8200
8260
555
800−900
1130
604, 540 (sh)
620 (sh), 580, 485 (sh)
1370
a
b
Table 2. ΔE = E(1[π(WCR)→π*(WCR)]) − E(3[π(WCR)→π*(WCR)]TA), measured at the lowest-energy shoulder.
state. The spectra of 1−4 each exhibit a prominent band in the
visible region, and 2−4 also show a strong, broad band in the
near-infrared region; the positions of these bands are set out in
Table 4. The spectra of 3 and 4 exhibit the bleaching of the
ground state 1[π(WCR)→π*(WCR)] absorbance near 450 nm,
while for 1 and 2 the equivalent features lie outside the
infrared band are more strongly dependent on the length of the
3
PE unit than are the [π(WCR)→π*(WCR)]TA bands; the
3[π(WCR)→π*(WCR)]TA band red shifts only 430 cm−1
between 3 and 4, whereas 3[π*(WCR)→π*n]TA red shifts
1550 cm−1. The larger marginal red shift is likely due to the
increased density of orbitals in the π* manifold as the
tungsten−PE unit increases in length.
1
wavelength range of the experiment. Bleaching of the [dxy→
π*(WCR)] absorption band is not observed for any of the
compounds due to its relatively small extinction coefficient (ε
≤ 520 M−1 cm−1, Table 2).
An important difference between the transient-absorption
spectra of W[C(C6H4CC)n−1Ph](dppe)2Cl complexes and
those of organic analogues is in their temporal evolution. The
transient-absorption spectrum of the S1 π→π* state of
Ph(CCC6H4)2H (an analogue of 3) exhibits features that
appear within 3 ps and then red-shift over the next 50 ps.51
Corresponding evolution on these time scales of the transient-
absorption and fluorescence spectra of longer organic OPEs has
also been observed.48,52 In contrast to these observations, the
transient-absorption bands of 1−4 do not shift in position or
change shape over the course of their decay. The longer-time
spectral evolution of the organic OPEs has been interpreted as
arising from planarization of the PE units in the excited
state.48,51,52 In the ground state in solution at room
temperature, OPEs possess a broad distribution of torsional
angles due to the small barrier to rotation (0.5 kcal/mol);53 this
barrier is larger in 1[ππ*] excited states due to increased
quinoidal/cumulenic character associated with the excited-state
structural distortions.
The transient absorption bands of 1−4 in the visible region
3
are assigned to the spin- and dipole-allowed transition [(π-
(WCR))2(dxy)1(π*(WCR))1→(π(WCR))1(dxy)1(π*(WCR))2]
3
(denoted [π(WCR)→π*(WCR)]TA in Table 4), based on
1
their similarity to the ground-state [π(WCR)→π*(WCR)]
absorption bands. For 1, 1b, and 2, it was possible to derive the
excited-state electronic-absorption spectra in the visible region
from nanosecond-time-scale transient-absorption spectra (Fig-
ure 9).50 The close correspondence in shape and extinction
The logical inference from these observations is that the
tungsten−PE units of extended derivatives 2−4 do not (fully)
3
planarize in the [(dxy)1(π*(WCR))1] excited state, and that
their excited-state torsional barriers are smaller than for their
organic analogues. This can be understood on the basis of the
different orbital configurations of the 3[(dxy)1(π*(WCR))1] and
1[ππ*] excited states. Although both states are characterized by
singly occupied π* orbitals, the depopulated orbital for the
organic compounds is π bonding and for 1−4 is the axially
nonbonding dxy orbital. Thus, for an oligomer of given length,
the changes in formal π bond order associated with planarizing
quinoidal/cumulenic distortions are smaller for the
Figure 9. Electronic-absorption spectra of the ground states (solid
3
line) and [(dxy)1(π*(WCR))1] excited states (data indicated by solid
circles, connected by a smooth curve) of 1, 1b, and 2 in toluene.
coefficient between the ground-state and excited-state π→π*
bands of these compounds is evident. Further, the ground and
excited-state π→π* bands of 1−4 exhibit parallel shifts in
energy as a function of the length of the OPE ligand (ΔE =
7440−8200 cm−1, Table 4); this suggests that the effective
conjugation length within the tungsten-PE unit in the
3[(dxy)1(π*(WCR))1] excited state is 4−5 repeat units, which
is the same as that found for the ground state (vide supra). The
intense bands observed in the near-infrared region for 2−4
(Figure 8) are assigned to spin- and dipole-allowed transitions
between orbitals within the π* manifold, specifically,
3[(dxy)1(π*(WCR))1(π*n)0→(dxy)1(π*(WCR))0(π*n)1] (de-
1
3[(dxy)1(π*(WCR))1] state than for the [ππ*] state. Relevant
to this point, a recent study of the structure of 1 in the
3[(dxy)1(π*(WCR))1] state using X-ray transient-absorption
spectroscopy and DFT calculations26 found that, relative to the
ground state, the WC−Cipso bond contracts by 0.041 Å (a
cumulenic distortion) and the phenyl group undergoes a
quiniodal distortion characterized by the displacement
parameter δr = 0.038 Å.54 By comparison, the distortions of
the analogous S2 1[ππ*] state of OPE-1 are roughly twice as
large (Δd(HC−Cipso) = 0.071 Å, δr = 0.072 Å), consistent with
the 2-fold larger change in formal bond orders.55 These
observations support the hypothesis that the lack of longer-time
evolution of the transient-absorption spectra of 2−4 manifests a
3
noted [π*(WCR)→π*n]TA), where π*n is a LUMO+n WCR
or OPE-localized orbital. The transition energies of the near-
3133
dx.doi.org/10.1021/ja411354d | J. Am. Chem. Soc. 2014, 136, 3127−3136

Journal of the American Chemical Society
Article
small perturbation to their ground-state torsional distributions,
due to a small increase in excited-state torsional barriers relative
to their organic analogues.
Work is underway to test these predictions in new metallo-OPE
derivatives of type C.
ASSOCIATED CONTENT
* Supporting Information
■
CONCLUSIONS
S
■
Synthetic procedures and characterization data for 2a, 2b, 3, 3a,
3b, 4, and selected starting materials; experimental procedures
for electronic absorption and emission spectra, emission
lifetimes and quantum yields, transient-absorption spectra,
cyclic voltammograms, and density functional calculations;
tables of bond distances, Cartesian coordinates, orbital energies,
and orbital atomic parentages provided by density functional
theory calculations; tables of electronic-absorption band
maxima, temperature-dependent emission lifetime data, and
Arrhenius fitting parameters for emission lifetimes; figures
showing electrochemical data, electronic spectroscopic data,
kinetic analyses of transient-absorption spectra, and calculated
frontier orbitals. This material is available free of charge via the
Internet at http://pubs.acs.org.
The substitution of a tungsten center for the terminal ethynyl
carbon atom in an oligo-phenylene-ethynylene, yielding the
W[C(C6H4CC)n−1Ph](dppe)2Cl family of compounds (1−5),
preserves the molecular structure and key electronic properties
of the organic materials and provides a locus for introducing
new functionality. The metallo- and organic OPEs are found to
possess similar effective π-conjugation lengths (4−5 PE units),
first reduction potentials, π* orbital energies and amplitudes,
and OPE-centered π→π* electronic-absorption bands. In
addition to these conserved properties, the metallo-OPEs also
undergo reversible one-electron oxidation at potentials
approximately 1 V lower than those of organic OPEs, and
long-lived phosphorescence from the 3[(dxy)1(π*(WCR))1]
state instead of the 1[ππ*] fluorescence characteristic of organic
OPEs. The different excited-state orbital configurations of
organic (1[ππ*]) and metallo (3[(dxy)1(π*(WCR))1]) OPEs
affect the excited-state barrier to rotation in the PE backbone
and, in a point that remains under investigation, may be
important for the contrasting PE-length dependences of their
radiative rates (which are sizable for organic OPEs and
negligible for 1−4) and nonradiative rates (small for organic
OPEs, sizable for 1−4).
AUTHOR INFORMATION
Corresponding Author
mhopkins@uchicago.edu
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The fact that the conserved and new properties of metallo-
OPEs 1−4 can be understood largely on the basis of the π, π*,
and dxy orbital energies, which for tungsten−alkylidyne
complexes are known to be sensitive to the nature of the
supporting ligands, suggests that it should be possible to
control these properties through compositional variation. It has
previously been shown for W(CR)L4X compounds that the
oxidation potential is linearly correlated with the calculated dxy
orbital energy, and can be tuned over a 2 V range by varying the
equatorial ligands.30 The oxidation potentials of 1−4 conform
to this correlation; thus, these empirical relationships should be
directly applicable to the design of new redox-active metallo-
OPEs with other ligand sets. Similarly, variation of the
supporting ligands of W(CR)L4X compounds can strongly
affect π(WCR) and π*(WCR) orbital energies; this provides a
means to modulate the interactions among the fragment π/π*
orbitals of the [W(CC6H4)L4X] and [(CCC6H4)n−1H]
subunits. For example, in the series of compounds W(CPh)-
(dppe)2Cl (1), W(CPh)(dppe)(CO)2Cl, and W(CPh)-
(CO)4Cl, the π(WCPh) and π*(WCPh) orbital energies are
calculated to decrease by 0.6−0.8 eV with each substitution of a
dppe ligand by two CO ligands.30 The close correspondences
among the π* orbitals (and associated properties) of 1−5 and
OPE-1−OPE-5 are due to the nearly identical π* fragment-
orbital energies of the [W(CC6H4)(dppe)2Cl] and
[(CCC6H4)n−1H] subunits, and the dissimilarities between
the π orbitals of 1−5 and OPE-1−OPE-5 arise from the ∼1.5
eV mismatch between the corresponding π fragment orbitals.
Thus, incorporating a [W(CO)4Cl] subunit into an OPE
should bring the π(WCR) and π(OPE) fragment orbitals into
resonance and decouple the π*(WCR) and π*(OPE) fragment
orbitals, while the mixed-ligand [W(dppe)(CO)2Cl] unit would
provide intermediate mixing of the π and π* levels. The ability
to tune the delocalization of both the π and π* orbitals could
have applications for redox- and photoactive materials, such as
in selectively promoting or disfavoring hole or electron transfer.
This research was supported by the U.S. Department of Energy,
Office of Basic Energy Sciences, Solar Photochemistry Program,
under Grant DE-FG02-07-ER15910. Use of the Center for
Nanoscale Materials (CNM) was supported by the U.S.
Department of Energy, Office of Science, Office of Basic
Energy Sciences, under Contract No. DE-AC02-06CH11357.
Mass spectrometry facilities were supported in part by National
Science Foundation Grant CHE-1048528. We are grateful to
Dr. Gary P. Wiederrecht and Dr. David J. Gosztola of the
Center for Nanoscale Materials at Argonne National
Laboratory for their assistance with the picosecond transient-
absorption measurements, and Dr. Shujiang Yang for assistance
with some of the DFT calculations.
REFERENCES
■
(1) Poly(arylene ethynylene)s: From Synthesis to Application; Weder,
C., Ed.; Springer-Verlag: Berlin, 2005.
(2) Bunz, U. H. F. Chem. Rev. 2000, 100, 1605.
(3) Schwab, P. F. H.; Levin, M. D.; Michl, J. Chem. Rev. 1999, 99,
1863.
(4) Martin, R. E.; Diederich, F. Angew. Chem., Int. Ed. 1999, 38, 1350.
(5) Giesa, R. Macromol. Sci. Rev. Chem. Phys. 1996, C36, 631.
(6) (a) Wenger, O. S. Acc. Chem. Res. 2011, 44, 25. (b) Odobel, F.;
Fortage, J. C. R. Chem. 2009, 12, 437. (c) Albinsson, B.; Martensson, J.
̊
J. Photochem. Photobiol., C 2008, 9, 138.
(7) Holten, D.; Bocian, D. F.; Lindsey, J. S. Acc. Chem. Res. 2001, 35,
57.
(8) (a) Johansson, P. G.; Kopecky, A.; Galoppini, E.; Meyer, G. J. J.
Am. Chem. Soc. 2013, 135, 8331. (b) Stalder, R.; Xie, D.; Zhou, R.;
Xue, J.; Reynolds, J. R.; Schanze, K. S. Chem. Mater. 2012, 24, 3143.
(c) Jiang, H.; Taranekar, P.; Reynolds, J. R.; Schanze, K. S. Angew.
Chem., Int. Ed. 2009, 48, 4300. (d) Myahkostupov, M.; Piotrowiak, P.;
Wang, D.; Galoppini, E. J. Phys. Chem. C 2007, 111, 2827. (e) Taratula,
O.; Rochford, J.; Piotrowiak, P.; Galoppini, E.; Carlisle, R. A.; Meyer,
G. J. J. Phys. Chem. B 2006, 110, 15734. (f) Wang, D.; Mendelsohn, R.;
Galoppini, E.; Hoertz, P. G.; Carlisle, R. A.; Meyer, G. J. J. Phys. Chem.
B 2004, 108, 16642.
3134
dx.doi.org/10.1021/ja411354d | J. Am. Chem. Soc. 2014, 136, 3127−3136

Journal of the American Chemical Society
Article
(9) (a) Yin, S.; Leen, V.; Jackers, C.; Beljonne, D.; Van Averbeke, B.;
Van der Auweraer, M.; Boens, N.; Dehaen, W. Chem.Eur. J. 2011,
17, 13247. (b) Harriman, A.; Mallon, L. J.; Elliot, K. J.; Haefele, A.;
Ulrich, G.; Ziessel, R. J. Am. Chem. Soc. 2009, 131, 13375. (c) Atienza,
C.; Insuasty, B.; Seoane, C.; Martin, N.; Ramey, J.; Rahman, G. M. A.;
Guldi, D. M. J. Mater. Chem. 2005, 15, 124. (d) Zhao, Y.; Shirai, Y.;
Slepkov, A. D.; Cheng, L.; Alemany, L. B.; Sasaki, T.; Hegmann, F. A.;
Tour, J. M. Chem.Eur. J. 2005, 11, 3643.
(26) Lovaasen, B. M.; Lockard, J. V.; Cohen, B. W.; Yang, S.; Zhang,
X.; Simpson, C. K.; Chen, L. X.; Hopkins, M. D. Inorg. Chem. 2012,
51, 5660.
(27) Cohen, B. W.; Lovaasen, B. M.; Simpson, C. K.; Cummings, S.
D.; Dallinger, R. F.; Hopkins, M. D. Inorg. Chem. 2010, 49, 5777.
(28) Moravec, D. B.; Hopkins, M. D. J. Phys. Chem. A 2013, 117,
1744.
(29) (a) Jelliss, P. A.; Wampler, K. M. Organometallics 2005, 24, 707.
(b) Lee, F.-W.; Chan, M. C.-W.; Cheung, K.-K.; Che, C.-M. J.
Organomet. Chem. 1998, 552, 255. (c) Lee, F.-W.; Chi-Wang Chan,
M.; Cheung, K.-K.; Che, C.-M. J. Organomet. Chem. 1998, 563, 191.
(d) Bocarsly, A. B.; Cameron, R. E.; Mayr, A.; McDermott, G. A. In
Photochemistry and Photophysics of Coordination Compounds; Yersin, H.,
Vogler, A., Eds.; Springer-Verlag: Berlin, 1987, p 213. (e) Bocarsly, A.
B.; Cameron, R. E.; Rubin, H. D.; McDermott, G. A.; Wolff, C. R.;
Mayr, A. Inorg. Chem. 1985, 24, 3976.
(30) Haines, D. E.; O’Hanlon, D. C.; Manna, J.; Jones, M. K.; Shaner,
S. E.; Sun, J.; Hopkins, M. D. Inorg. Chem. 2013, 52, 9650.
(31) (a) Pombeiro, A. J. L. Eur. J. Inorg. Chem. 2007, 2007, 1473.
(b) Pombeiro, A. J. L. J. Organomet. Chem. 2005, 690, 6021. (c) Zhang,
L.; Guedes da Silva, M. F. C.; Kuznetsov, M. L.; Gamasa, M. P.;
Gimeno, J.; Frausto da Silva, J. J. R.; Pombeiro, A. J. L. Organometallics
2001, 20, 2782. (d) Zhang, L.; Gamasa, M. P.; Gimeno, J.; da Silva,
M.; Pombeiro, A. J. L.; Graiff, C.; Lanfranchi, M.; Tiripicchio, A. Eur. J.
Inorg. Chem. 2000, 1707. (e) Pombeiro, A. J. L. In Molecular
Electrochemistry of Inorganic, Bioinorganic, and Organometallic Com-
pounds; Pombeiro, A. J. L., McCleverty, J. A., Eds.; Kluwer Academic
Publishers: Dordrecht, 1993, p 331.
(10) (a) Kaur, I.; Zhao, X.; Bryce, M. R.; Schauer, P. A.; Low, P. J.;
Kataky, R. ChemPhysChem 2013, 14, 431. (b) Moore, A. M.;
Mantooth, B. A.; Dameron, A. A.; Donhauser, Z. J.; Lewis, P. A.;
Smith, R. K.; Fuchs, D. J.; Weiss, P. S. In Frontiers in Materials
Research; Fujikawa, Y., Nakajima, K., Sakurai, T., Eds.; Springer-Verlag:
Berlin, 2008, p 29. (c) Newton, M. D.; Smalley, J. F. Phys. Chem.
Chem. Phys. 2007, 9, 555. (d) Dube, A.; Chadeayne, A. R.; Sharma, M.;
Wolczanski, P. T.; Engstrom, J. R. J. Am. Chem. Soc. 2005, 127, 14299.
(e) James, D. K.; Tour, J. M. Top. Curr. Chem. 2005, 257, 33.
(f) Mantooth, B. A.; Weiss, P. S. Proc. IEEE 2003, 91, 1785. (g) Tour,
J. M. Acc. Chem. Res. 2000, 33, 791. (h) Dhirani, A.; Lin, P.-H.; Guyot-
Sionnest, P.; Zehner, R. W.; Sita, L. R. J. Chem. Phys. 1997, 106, 5249.
(11) Andrew, T. L.; Swager, T. M. J. Polym. Sci., Part B: Polym. Phys.
2011, 49, 476.
(12) Thomas, S. W.; Joly, G. D.; Swager, T. M. Chem. Rev. 2007, 107,
1339.
(13) Swager, T. M.; Zheng, J. In Poly(arylene ethynylene)s: From
Synthesis to Application; Weder, C., Ed.; Springer-Verlag: Berlin, 2005,
p 151.
(14) Cheng, Y.-J.; Yang, S.-H.; Hsu, C.-S. Chem. Rev. 2009, 109,
5868.
(32) van der Eide, E. F.; Piers, W. E.; Parvez, M.; McDonald, R. Inorg.
Chem. 2007, 46, 14.
(15) Grimsdale, A. C.; Leok Chan, K.; Martin, R. E.; Jokisz, P. G.;
Holmes, A. B. Chem. Rev. 2009, 109, 897.
(16) Voskerician, G.; Weder, C. In Poly(arylene ethynylene)s: From
Synthesis to Application; Weder, C., Ed.; Springer-Verlag: Berlin, 2005,
p 209.
(33) Semenov, S. N.; Blacque, O.; Fox, T.; Venkatesan, K.; Berke, H.
J. Am. Chem. Soc. 2010, 132, 3115.
(34) Semenov, S. N.; Taghipourian, S. F.; Blacque, O.; Fox, T.;
Venkatesan, K.; Berke, H. J. Am. Chem. Soc. 2010, 132, 7584.
(35) Sun, J.; Shaner, S. E.; Jones, M. K.; O’Hanlon, D. C.; Mugridge,
J. S.; Hopkins, M. D. Inorg. Chem. 2010, 49, 1687.
(36) Moravec, D. B.; Hopkins, M. D. Chem.Eur. J. 2013, 19, 17082.
(37) The tungsten-containing compounds and their precursors were
synthesized using Pd-catalyzed cross-coupling methods similar to
those employed in the preparation of organic oligo-phenylene-
ethynylenes. The compounds are slightly air sensitive in the solid
state and solution. Full details regarding their synthesis and
characterization are provided in the Supporting Information.
(38) Vyboishchikov, S. F.; Frenking, G. Chem.Eur. J. 1998, 4, 1439.
(39) Ofer, D.; Swager, T. M.; Wrighton, M. S. Chem. Mater. 1995, 7,
418.
(17) Liu, Y.; Li, Y.; Schanze, K. S. J. Photochem. Photobiol., C 2002, 3,
1.
(18) Ley, K. D.; Schanze, K. S. Coord. Chem. Rev. 1998, 171, 287.
(19) (a) Low, P. J. Coord. Chem. Rev. 2013, 257, 1507. (b) Whittell,
G. R.; Hager, M. D.; Schubert, U. S.; Manners, I. Nat. Mater. 2011, 10,
176. (c) Eloi, J.-C.; Chabanne, L.; Whittell, G. R.; Manners, I. Mater.
Today 2008, 11, 28. (d) Williams, K. A.; Boydston, A. J.; Bielawski, C.
W. Chem. Soc. Rev. 2007, 36, 729. (e) Klemm, E.; Pautzsch, T.;
Blankenburg, L. In Poly(arylene ethynylene)s: From Synthesis to
Application; Weder, C., Ed.; Springer-Verlag: Berlin, 2005, p 53.
(f) Kingsborough, R. P.; Swager, T. M. Prog. Inorg. Chem. 1999, 48,
123.
(20) (a) Liao, C.; Shelton, A. H.; Kim, K.-Y.; Schanze, K. S. ACS
Appl. Mater. Interfaces 2011, 3, 3225. (b) Muro, M. L.; Rachford, A. A.;
Wang, X.; Castellano, F. N. Top. Organomet. Chem. 2010, 29, 1.
(c) Silverman, E. E.; Cardolaccia, T.; Zhao, X.; Kim, K.-Y.; Haskins-
Glusac, K.; Schanze, K. S. Coord. Chem. Rev. 2005, 249, 1491.
(21) (a) Kokil, A.; Yao, P.; Weder, C. Macromolecules 2005, 38, 3800.
(b) Walters, K. A.; Ley, K. D.; Cavalaheiro, C. S. P.; Miller, S. E.;
Gosztola, D.; Wasielewski, M. R.; Bussandri, A. P.; van Willigen, H.;
Schanze, K. S. J. Am. Chem. Soc. 2001, 123, 8329. (c) Ley, K. D.; Li, Y.;
Johnson, J. V.; Powell, D. H.; Schanze, K. S. Chem. Commun. 1999,
1749.
(22) (a) Shi, C. A.; Jia, G. C. Coord. Chem. Rev. 2013, 257, 666.
(b) Herndon, J. W. Coord. Chem. Rev. 2012, 256, 1281. (c) Mindiola,
D. J.; Bailey, B. C.; Basuli, F. Eur. J. Inorg. Chem. 2006, 2006, 3135.
(d) Schrock, R. R. Chem. Rev. 2002, 102, 145. (e) Fischer, H.;
Hofmann, P.; Kreissl, F. R.; Schrock, R. R.; Schubert, U.; Weiss, K.
Carbyne Complexes; VCH Publishers: Weinheim, Germany, 1988.
(23) Mayr, A.; Yu, M. P. Y.; Yam, V. W.-W. J. Am. Chem. Soc. 1999,
121, 1760.
(24) Yu, M. P. Y.; Yam, V. W.-W.; Cheung, K.-K.; Mayr, A. J.
Organomet. Chem. 2006, 691, 4514.
(25) Da Re, R. E.; Hopkins, M. D. Coord. Chem. Rev. 2005, 249,
1396.
(40) The first reduction potentials of PhCCPh, Ph(CCC6H4)2H, and
Ph(CCC6H4)3H in THF have been reported relative to the Ag/Ag+
reference as −2.56 V, −2.30 V, and −2.07 V, respectively. Wiberg, J.;
Guo, L.; Pettersson, K.; Nilsson, D.; Ljungdahl, T.; Martensson, J.;
̊
Albinsson, B. J. Am. Chem. Soc. 2006, 129, 155. These were scaled to
the FeCp20/+-referenced potentials noted in the text by applying an
offset of −0.39 V, which is derived from the reported first reduction
potential of PhCCPh in THF against this standard (−2.95 V vs
FeCp20/+). Boudon, C.; Gisselbrecht, J. P.; Gross, M.; Anthony, J.;
Boldi, A. M.; Faust, R.; Lange, T.; Philp, D.; Van Loon, J. D.;
Diederich, F. J. Electroanal. Chem. 1995, 394, 187.
(41) Chu, Q.; Pang, Y. Spectrochim. Acta, A 2004, 60, 1459.
(42) Mao, G.; Orita, A.; Matsuo, D.; Hirate, T.; Iwanaga, T.; Toyota,
S.; Otera, J. Tetrahedron Lett. 2009, 50, 2860.
(43) Meier, H.; Stalmach, U.; Kolshorn, H. Acta Polym. 1997, 48,
379.
3
(44) Fitting the [dxy←π*(WCR)] emission band maxima to eq 1
predicts that for the infinite-length oligomer E∞ = 12230 cm−1 (818
nm; see Figure 5); this is only 2 nm (20 cm−1) lower than the
1
1
emission maximum of 4. Thus, the [dxy→π*(WCR)], [π(WCR)→
π*(WCR)], and 3[dxy←π*(WCR)] bands all provide effective
conjugation lengths of 4−5 units.
3135
dx.doi.org/10.1021/ja411354d | J. Am. Chem. Soc. 2014, 136, 3127−3136

Journal of the American Chemical Society
Article
́
(45) Varsanyi, G. Vibrational Spectroscopy of Benzene Derivatives;
Academic Press: New York, 1969.
(46) Schumm, J. S.; Pearson, D. L.; Tour, J. M. Angew. Chem., Int. Ed.
1994, 33, 1360.
(47) Ishida, H.; Tobita, S.; Hasegawa, Y.; Katoh, R.; Nozaki, K.
Coord. Chem. Rev. 2010, 254, 2449.
(48) Duvanel, G.; Grilj, J.; Schuwey, A.; Gossauer, A.; Vauthey, E.
Photochem. Photobiol. Sci. 2007, 6, 956.
(49) Englman, R.; Jortner, J. Mol. Phys. 1970, 18, 145.
(50) It was not possible to calculate extinction coefficients for 3 and 4
because their excited-state lifetimes were too close to the temporal
response limit of the spectrometer.
(51) Beeby, A.; Findlay, K. S.; Low, P. J.; Marder, T. B.; Matousek,
P.; Parker, A. W.; Rutter, S. R.; Towrie, M. Chem. Commun. 2003,
2406.
(52) Sluch, M. I.; Godt, A.; Bunz, U. H. F.; Berg, M. A. J. Am. Chem.
Soc. 2001, 123, 6447.
(53) Levitus, M.; Schmieder, K.; Ricks, H.; Shimizu, K. D.; Bunz, U.
H. F.; Garcia-Garibay, M. A. J. Am. Chem. Soc. 2001, 123, 4259.
(54) Defining the Cipso carbon atom of the phenyl group as C(1), δr
= 1/4[r(C(1)−C(2)) + r(C(3)−C(4)) + r(C(4)−C(5)) + r(C(6)−
C(1)) − 2r(C(2)−C(3)) − 2r(C(5)−C(6))].
(55) Pugliesi, I.; Tonge, N. M.; Hornsby, K. E.; Cockett, M. C. R.;
Watkins, M. J. Phys. Chem. Chem. Phys. 2007, 9, 5436.
3136
dx.doi.org/10.1021/ja411354d | J. Am. Chem. Soc. 2014, 136, 3127−3136