Parent o-phenylene oligomers: Synthesis, conformational behavior, and characterization

Article abstract of DOI:10.1021/ma201866p

The o-phenylenes are an unusual class of conjugated polymer, defined largely by substantial steric twisting along their backbones. Consequently, they exhibit limited conjugation but also interesting conformational behavior: they have been shown to adopt well-defined helical secondary structures, both in the solid state and in solution. While several examples of functionalized o-phenylene oligomers have been reported, most of the basic properties of the parent compounds are unknown. Here we report the synthesis and characterization of the series of unsubstituted o-phenylene oligomers up to the octamer. Through a combination of NMR spectroscopy, including dynamic NMR (EXSY), and computational chemistry, we have found that these compounds adopt compact helical conformations in solution with three repeat units per turn. Although formally conjugated, the oligomers have a very short effective conjugation length of n_ecl ≈ 4 (based on UV-vis spectra), significantly shorter than most other conjugated systems. Also, unlike other (substituted) o-phenylenes, no hypochromicity is observed in their UV-vis spectra. The fluorescence spectra of the series exhibit a systematic blue shift with increasing length. We believe this unusual property results from increased steric congestion in the longer oligomers, which are therefore less able to accommodate structural relaxation in the excited state.

Full text of DOI:10.1021/ma201866p

ARTICLE
pubs.acs.org/Macromolecules
Parent o-Phenylene Oligomers: Synthesis, Conformational
Behavior, and Characterization
Sanyo M. Mathew and C. Scott Hartley*
Department of Chemistry & Biochemistry, Miami University, Oxford, Ohio 45056, United States
S Supporting Information
b
ABSTRACT: The o-phenylenes are an unusual class of conju-
gated polymer, defined largely by substantial steric twisting along
their backbones. Consequently, they exhibit limited conjugation
but also interesting conformational behavior: they have been
shown to adopt well-defined helical secondary structures, both in
the solid state and in solution. While several examples of
functionalized o-phenylene oligomers have been reported, most
of the basic properties of the parent compounds are unknown.
Here we report the synthesis and characterization of the series of
unsubstituted o-phenylene oligomers up to the octamer. Through
a combination of NMR spectroscopy, including dynamic NMR (EXSY), and computational chemistry, we have found that these
compounds adopt compact helical conformations in solution with three repeat units per turn. Although formally conjugated, the
oligomers have a very short effective conjugation length of necl ≈ 4 (based on UVꢀvis spectra), significantly shorter than most other
conjugated systems. Also, unlike other (substituted) o-phenylenes, no hypochromicity is observed in their UVꢀvis spectra. The
fluorescence spectra of the series exhibit a systematic blue shift with increasing length. We believe this unusual property results from
increased steric congestion in the longer oligomers, which are therefore less able to accommodate structural relaxation in the
excited state.
’ INTRODUCTION
structural defects (triphenylene moieties). More recently, elec-
tropolymerization of catechol derivatives was reported to yield
poly(o-phenylenes),17ꢀ19 but these materials were subsequently
found to be small-molecule triphenylene derivatives.20 Before
2010, there were three reports of series of o-phenylene oligomers
of which we are aware:21 by Wittig,22 Ibuki and Ozasa,23ꢀ26 and
Simpkins.27 However, these compounds were not studied in
detail, and most of their basic properties were not reported.
Recently, we decided to investigate the series of methoxy-
substituted o-phenylene oligomers oP(OMe)n (n = 4ꢀ12).28,29
We found that these compounds exhibit weak but surprisingly
long-range conjugation: as n increases, a relatively small overall
bathochromic shift of the UVꢀvis spectra is observed; however,
significant changes are observed even for long oligomers, giving
an effective conjugation length of necl = 8.28 Unusually, the
oP(OMe)n series also exhibits a systematic hypsochromic shift in
their fluorescence spectra with increasing n. Using a combination
of NMR spectroscopy and ab initio calculations, we found that
these compounds adopt helical conformations in solution, with
some disorder at their ends.29 Simultaneously, Fukushima and
Aida reported a series of o-phenylene oligomers up to the [48]-
mer.30 Remarkably, an octamer exhibits spontaneous deracemi-
zation in the solid state. The helical compound racemizes rapidly
o-Phenylenes represent a simple, fundamental class of con-
jugated polymers.1 They are isomers of the p-phenylenes, now
well-established as conducting films (when doped),2 blue-emit-
ting materials,3,4 and single-molecule wires.5,6 Similarly, the
isomeric m-phenylenes have been studied as helical polymers
and oligomers.7ꢀ11 In contrast, the o-phenylenes have received
little attention, and their properties are only now beginning to be
investigated. Their defining characteristic is the substantial steric
congestion along their backbones, which necessitates highly
twisted conformational states. Presumably, this steric hindrance
is why they were largely ignored: they are relatively difficult to
synthesize by conventional arylꢀaryl bond-forming methods,
and π-overlap between repeat units is limited.1,12 Consequently,
their properties are not particularly desirable in the context
of conjugated polymers for thin-film electronics. However, the
o-phenylenes do have some interesting potential applications. In
principle, they could be oxidatively planarized to give very narrow
graphene nanoribbons, although they tend to rearrange under
the standard reaction conditions for oxidative cyclodehydro-
genation.13 Perhaps more interestingly, the steric interactions
along the backbone can produce well-defined conformational
behavior, leading to applications as three-dimensional scaffolds
and promoting through-space overlap between repeat units.
As early as the 1960s, Kovacic reported the syntheses of some
o-phenylene polymers,14ꢀ16 but the resulting materials received
little subsequent characterization and were thought to contain
Received: August 15, 2011
Revised:
September 21, 2011
Published: October 11, 2011
r
2011 American Chemical Society
8425
dx.doi.org/10.1021/ma201866p Macromolecules 2011, 44, 8425–8432
|

Macromolecules
ARTICLE
Scheme 1a
a Reagents and conditions: (a) iodobenzene, Pd(OAc)2, SPhos, K3PO4, THF/H2O (4/1), Δ; (b) Tf2O, pyridine, DCM; (c) Mg, NH4Cl, Pd/C, MeOH;
(d) phenylboronic acid, Pd(OAc)2, SPhos, K3PO4, THF/H2O (4/1), Δ; (e) 1, Pd(OAc)2, SPhos, K3PO4, THF/H2O (4/1), Δ.
in solution, but this process slows considerably on oxidation,
which causes the springlike oligomer to compress.
are low yielding and not well-suited to the preparation of more
complex o-phenylene-based compounds. Reexamination of this
series is therefore needed to develop improved synthetic meth-
ods and establish baseline properties for the o-phenylenes as a
class of conjugated oligomers and polymers.
’ RESULTS AND DISCUSSION
Synthesis. We desired a unified synthetic approach that
allows the stepwise synthesis of oligomers of arbitrary length,
that takes advantage of modern developments in arylꢀaryl cross-
coupling reactions, and that would be amenable to the future
synthesis of oligomer heterosequences. Thus, we adapted our
previous synthetic strategy for the oP(OMe)n series.28 Our
approach, shown in Scheme 1, is based on Manabe’s strategy
for oligophenylene synthesis,31,32 which uses phenols as masked
triflates to control an iterative sequence of SuzukiꢀMiyaura
coupling reactions.
The key monomer for this series is boroxarene 1, which is
readily obtained by treatment of commercially available 2-phe-
nylphenol with BCl3.33 SuzukiꢀMiyaura coupling of 1 with
iodobenzene using Pd(OAc)2 and SPhos34 gives hydroxy-func-
tionalized o-phenylene trimer oP(H)3-OH, which is then con-
verted to the corresponding triflate oP(H)3-OTf. At this point,
the oligomer can be terminated by reductive deoxygenation35
With the emerging interest in o-phenylenes, we realized that
little was known about the parent (unsubstituted) series oP(H)n.
Unlike their para isomers, unsubstituted o-phenylenes should
exhibit workable solubility even for the higher oligomers,1,12
enabling convenient spectroscopic characterization. Many of
these compounds are known in the literature.22ꢀ26 However,
to our knowledge, most of the basic properties of the oP(H)n
series, such as their NMR and fluorescence spectra, have not been
reported, nor have modern techniques been used to determine
their conformational behavior. Moreover, the reported syntheses
8426
dx.doi.org/10.1021/ma201866p |Macromolecules 2011, 44, 8425–8432

Macromolecules
ARTICLE
Figure 1. Labeling of dihedral angles (ϕi) in an o-phenylene.
(giving oP(H)3) or by coupling to phenylboronic acid (giving
oP(H)4). Extension of the oligomer is accomplished by coupling
to another equivalent of 1, affording oP(H)5-OH. These steps
can then be repeated to prepare the higher oligomers. The
isolated yields are uniformly high, and even the longest oligomers
(oP(H)7 and oP(H)8) were obtained in only eight steps from
commercially available starting materials. We also note that this
approach offers a straightforward method for the preparation of
end-functionalized o-phenylenes by varying the coupling part-
ners for the first and last steps.
All of the compounds in Scheme 1 exhibit good solubility in
organic solvents. However, we were unable to synthesize oligo-
mers longer than the octamer due to the poor solubility of the
hydroxyl- or triflate-substituted nonamer intermediates. Thus, as
we were unable to prepare any unfunctionalized oligomer longer
than oP(H)8, we cannot say at this point whether insolubility
beyond n = 8 is an intrinsic property of the target compounds
themselves. Nevertheless, the unsubstituted o-phenylenes are
substantially more soluble than their p-phenylene analogues,
which exhibit poor solubility beyond the heptamer even when
functionalized with large branched alkyl solubilizing groups.6
The increased solubility reflects the twisting of the o-phenylene
backbone, which should prevent extensive πꢀπ contacts in the
solid state. Melting points for the series were measured by
differential scanning calorimetry (Table S1 and Figure S1) and
show a slight oddꢀeven effect.
Figure 2. Closed helical (AAA, left) and open helical (BBB, right)
conformers of oP(H)6. Geometries optimized at the BH&HLYP/
6-31+G(d,p) level.
Conformational Analysis. One of the most interesting as-
pects of the o-phenylenes is their conformational behavior.
Previously, we reported a complete conformational analysis
of the oP(OMe)n series,29 which should be applicable to the
o-phenylenes in general. Briefly, the overall conformation of an
o-phenylene oligomer is dictated by the internal torsional angles
ϕ2ꢀϕnꢀ2, shown in Figure 1. (The terminal torsional angles, ϕ1
and ϕnꢀ1, do not need to be considered as rotation about these
bonds is degenerate.) On the basis of computational modeling,
we found that each of these dihedrals can assume one of four
possible values: ϕi ≈ (70° or (130°. However, because of
coupling along the backbone, the total conformational pool can
be divided into two separate enantiomeric subsets: within a single
molecule, each ϕi can assume one of only two possible values,
either ϕi ≈ +70°/ꢀ130° or ϕi ≈ ꢀ70°/+130°. Here we discuss
and visualize the oligomers in terms of the ꢀ70°/+130° set, but
we stress that we have not resolved any of these compounds;
thus, all structures presented represent racemates. To distinguish
different conformers, we label bonds in the ꢀ70° state “A” and
those in the +130° state “B” (e.g., the “AB” conformer of oP(H)5
has ϕ2 ≈ ꢀ70° and ϕ3 ≈ +130°).
Figure 3. 1H NMR of oP(H)6 (500 MHz, CDCl3, ꢀ5 °C). Selected
signals corresponding to conformers I2, II1, and III2 are labeled. The
remaining (overlapping) signals can be separated and assigned based on
2D spectra; assignments are tabulated in the Supporting Information.
Conversely, if all ϕi ≈ +130° (the Bnꢀ3 conformer), then the
oligomers are in an extended, open helical conformation without
intramolecular stacking. These two conformers are illustrated
in Figure 2 for oP(H)6. The specific conformational state of the
o-phenylenes is important. Aromatic repeat units in the open
conformer are closer to coplanarity and thus should exhibit more
significant π-system delocalization. The closed conformer has
more interesting spatial correlations between repeat units as
every third aromatic ring is closely stacked.
Several reported crystal structures of o-phenylene oligomers
have suggested that they are predisposed to closed helical
conformations (Anꢀ3) in the solid state.27,28,30 However, given
the complexity of the conformational pool, it was not clear if they
adopted these well-defined three-dimensional structures in solu-
tion (as opposed to a large number of random coil con-
formations). Using a combination of NMR spectroscopy and
computational chemistry (see below), we demonstrated29 that
our previous series of o-phenylenes (oP(OMe)n) exist primarily
Even with the constraint that each ϕi can assume only two
possible states within a single molecule, there is, in principle,
a large number of available conformers for an o-phenylene
oligomer.36 However, it is illustrative to consider the two possible
limiting cases. If all ϕi ≈ ꢀ70° (the Anꢀ3 conformer), then the
oligomer’s conformation can be described as a compact, closed
helix, with offset, coplanar stacking of every third monomer.
8427
dx.doi.org/10.1021/ma201866p |Macromolecules 2011, 44, 8425–8432

Macromolecules
ARTICLE
1H NMR of well-separated signals (and accounting for their
relative symmetries).
In order to associate these experimentally observed confor-
mers with specific molecular geometries, we undertook to com-
pletely assign their 1H chemical shifts. Our strategy29 was to first
obtain a complete set of assignments for the major conformer
(I2) using a standard set of two-dimensional NMR experiments
(DQFCOSY, HMQC, and HMBC). Because the signals corre-
sponding to I2 are significantly more intense than the others (due
to both its greater population and higher symmetry), they are
readily identified in the contour plots of the 2D spectra. With the
assignments for I2 in hand, we then used the EXSY spectrum (in
combination with weak COSY signals) to map them onto the
minor conformers. In this way we obtained a nearly complete set
of 1H chemical shifts for III2.37 Because II1 has lower symmetry
than I2, it is impossible to unambiguously assign the correspond-
ing rings on either side of the oligomer (i.e., rings 1 vs 6, 2 vs 5,
and 3 vs 4). Instead, we were able to obtain (nearly) complete
sets of chemical shifts for each of the six individual rings, but their
connectivity was determined according to the best match to the
spectra computed by DFT (see below).
Because each of ϕ2, ϕ3, and ϕ4 of oP(H)6 can assume either the
A (ꢀ70°) or B (+130°) states, there are a total of six possible
conformers: four that are C2-symmetric (AAA, BAB, ABA, BBB)
and two that are unsymmetric (AAB and ABB). For comparison
with the experimental chemical shifts, we then carried out DFT
geometry optimizations of each of these conformers (in the gas
phase). We chose the BH&HLYP/6-31+G(d,p) level since this
functional should better estimate the strength of πꢀπ stacking
interactions;38 in any event, we did not observe any significant
differences in the performance of BH&HLYP compared to more
common methods (e.g., B3LYP). After geometries were ob-
tained for each of the six conformers, shown in Figure 5 and
Figure 4. EXSY spectrum of oP(H)6 (500 MHz, CDCl3,ꢀ5°C, tm =0.5s).
in the stacked helical conformations (Anꢀ3) in solution, with
some minor conformers identified that had defects (B states)
only at the very ends. These results suggest that o-phenylenes
may have an important role to play as scaffolds for the positioning
of substituents in space. However, considering the limited
number of reported examples of o-phenylene oligomers, most
of which have been functionalized with electron-donating
groups, the conformational behavior of the class as a whole,
independent of substituent effects, was not clear.
1
Figure S2, H chemical shifts were calculated using the GIAO
method39,40 at the PCM/WP04/6-31G(d) level,41 which has
been shown to be an efficient and economical approach to the ab
initio prediction of 1H NMR spectra in chloroform.42 We then
compared the sets of calculated chemical shifts for our six
possible conformers with the experimental data for I2, II1, and
III2. As a figure of merit, we chose the root-mean-squared (rms)
error of the calculated chemical shift (δcalc) against the experi-
mental chemical shift (δexpt), scaled according to a linear fit
between the two data sets.43
Similar to oP(OMe)n, the unsubstituted oP(H)n exhibit
complex NMR spectra at room temperature (and below) for
n g 5. As a representative example, here we discuss oP(H)6 in
1
detail, as it gives the most complete data set. Its H NMR
spectrum at ꢀ5 °C is shown in Figure 3 (the spectrum is similar
at room temperature, but with slightly broadened signals).
Remarkably, although this compound consists only of unsub-
1
stituted benzene rings, the H chemical shifts span >2 ppm,
presumably due to the high density of aromatic π-systems and
their associated ring currents.
All three experimentally observed conformers were thus read-
ily assigned to one specific three-dimensional geometry. The
good matches all had rms errors less than 0.16 ppm, and bad
matches had rms errors greater than 0.44 ppm. These thresholds
are in good agreement with the established errors for the method
based on a set of test molecules (0.12 ppm).42 Based on this
analysis, I2 is AAA, II1 is AAB, and III2 is BAB; the plots of the
calculated against experimental NMR data are given in Figure 5.
The three identified conformers are directly analogous to those
observed for oP(OMe)6.29 The conformer distribution is similar
but slightly different: 49:42:10 for oP(H)6 compared to 55:36:10
for oP(OMe)6. Notably, of the six possible backbone conforma-
tions, only those with the central bond in the “A” state are
observed (ϕ3 ≈ ꢀ77.5°, ꢀ73.9°, and ꢀ82.0° for AAA, AAB, and
BAB, respectively).
At first inspection, it would appear that the sample of oP(H)6
is contaminated with significant amounts of impurities given the
large number of small signals with nonintegral relative areas.
However, as we previously demonstrated for the methoxy-
substituted series, these smaller signals are due to minor con-
formational states in slow exchange on the NMR time scale. This
is demonstrated in the EXSY (i.e., NOESY) NMR spectrum,
shown in Figure 4. All of the minor signals exhibit cross-peaks
with corresponding major signals. These cross-peaks are of the
same phase as the diagonal; since this is a small molecule, they
must therefore arise from chemical exchange (and not through-
space NOESY effects). Using the EXSY spectrum, the NMR
signals can be assigned to three exchanging conformers. The
major conformer, I2, is 2-fold symmetric. There are then two
minor conformers: one, II1, which is nonsymmetric and one, III2,
which is 2-fold symmetric. Conformers I2, II1, and III2 have
relative populations of 49:42:10 based on the integration
We then sought to extend our NMR analysis to the other
members of the oP(H)n series. oP(H)4, as expected, exists in
rapid conformational exchange near room temperature and thus
exhibits a standard first-order NMR spectrum. For oP(H)5, while
8428
dx.doi.org/10.1021/ma201866p |Macromolecules 2011, 44, 8425–8432

Macromolecules
ARTICLE
Figure 5. Geometry-minimized conformers for oP(H)6 and plots of calculated vs experimental 1H chemical shifts.
Figure 6. Closed helical (Anꢀ3) conformers for the oP(H)n series, minimized at the BH&HLYP/6-31+G(d,p) level.
there is some evidence for slow exchange at 0 °C (a few small
For oP(H)7 and oP(H)8, the experimental chemical shift data
for the major conformer are in good agreement with that
calculated for the closed helical Anꢀ3 conformers (rms errors
e0.11 ppm, Figure S3). We therefore conclude that the oP(H)n
series adopts stacked helical conformations in solution, shown in
Figure 6. On the basis of the results for oP(H)6, and by analogy
with our results for the oP(OMe)n series, it seems likely that
conformational disorder occurs primarily at the ends of the
oligomers. Consequently, we conclude that the preference for
stacked helical conformations is a basic property of simple
o-phenylenes. Further studies will be needed to determine the
role of substituent effects in controlling the conformational state
of the oligomers. However, considering the similar behavior of
oP(H)n compared to oP(OMe)n, it appears at this point that
substituents do not have a dramatic effect on the conformational
distribution.
1
signals in the H NMR spectrum and associated EXSY cross-
peaks), the data cannot be completely assigned to multiple
independent conformers, suggesting that there are still significant
backbone rearrangements that are rapid on the NMR time scale.
Like oP(H)6, the longer oligomers oP(H)7 and oP(H)8 exist in a
slow exchange regime. For these oligomers, we were able to
obtain chemical shift assignments for the major conformer in
solution, which is 2-fold symmetric in both cases. Unfortunately,
although EXSY spectroscopy indicates that the minor signals in
the spectra of oP(H)7 and oP(H)8 result from less-populated
conformational states, we were unable to map assignments for
the major conformer onto the minor conformers due to the
considerable signal overlap. This is perhaps not surprising, as the
oP(H)n series represents the worst case for this NMR analysis
strategy, with the largest possible number of aromatic signals and
no signal separation from substituent effects.
In contrast to our o-phenylenes, Ito has investigated a series of
structurally related poly(2,3-quinoxalines), many of which adopt
8429
dx.doi.org/10.1021/ma201866p |Macromolecules 2011, 44, 8425–8432

Macromolecules
ARTICLE
Figure 8. Plots of the extrapolated UVꢀvis onset vs (a) n or (b) 1/n.
The solid line in (a) is a fit to eq 1 (χ2 = 5.39).
Figure 7. UVꢀvis spectra (cyclohexane) of oP(H)n (n = 2ꢀ8).
conformations analogous to the open Bnꢀ3 conformers.44ꢀ46
However, the 2,3-quinoxaline repeat units in these oligomers lack
the hydrogen atoms ortho to the biaryl bonds and thus should be
better able to accommodate a more planar conformation; further,
these polymers are typically functionalized with very bulky substit-
uents (e.g., tolyl), which likely bias the compounds toward the
open helix through steric interactions. Interestingly, oligo(2,3-
naphthalenes) lacking the bulky substituents exhibit closed helical
conformations similar to those observed for the o-phenylenes.47
UVꢀvis Spectroscopy. Although the UVꢀvis spectra of
several members of the oP(H)n series have been reported,24
they have not been analyzed in detail. The UVꢀvis spectra of the
complete oP(H)n series (n = 2ꢀ8) are shown in Figure 7; the
spectra for the known members of the series are in agreement
with those previously reported. The spectra show no solvato-
chromism, as indicated by comparisons of spectra of cyclohexane
and dichloromethane solutions (Figure S4), nor are their shapes
affected by concentration. Curiously, while the oP(OMe)n series
exhibits a substantial hypochromic effect in its UVꢀvis spectra
(i.e., when normalized for n, ε decreases with increasing n), we
observe no significant hypochromicity for the oP(H)n series
(Figure S4). As hypochromicity is associated with well-ordered
arrangements of chromophore transition dipole moments,48 its
absence either suggests less rigid conformational ordering for
oP(H)n or reflects the different polarizations of the transitions.
Inspection of the UVꢀvis spectra clearly indicates a significant
red shift as the oligomer is extended from oP(H)2 (biphenyl) to
oP(H)3; however, the change in spectra attenuates sharply for n
> 3. More quantitatively, the degree of conjugation in a con-
jugated oligomer (or polymer) may be evaluated in terms of the
effective conjugation length necl. The effective conjugation length
can be determined from a plot of the UVꢀvis transitions against
n using an empirical method developed by Meier:49
Accordingly, we have adopted the same approach we used to
analyze the oP(OMe)n series, evaluating the change in the
50
extrapolated UVꢀvis absorption onset λext,UV
.
Meier has
shown that this provides functionally the same results,49 and
we have confirmed that it is equivalent to the use of absorption
maxima for other conjugated oligomers (e.g., p-phenylenes),6
allowing meaningful comparison with necl values for other classes
of compounds.
A plot of λext,UV against n is shown in Figure 8a. A fit to eq 1
yields ΔλUV = 142 ( 47 nm, λ∞,UV = 298.88 ( 0.57 nm, and b =
1.79 ( 0.33. From these parameters, necl ≈ 4. The small effective
conjugation length for oP(H)n is also obvious when λext,UV is
plotted against 1/n, as shown in Figure 8b. For most conjugated
oligomers, a good linear relationship is observed between λext,UV
and 1/n at low n, with a break from linearity around necl.
However, for the oP(H)n series, the data are not adequately fit
by a linear function, even for n = 2ꢀ4, reflecting the very short
necl for the parent o-phenylenes and consistent with the value of
necl ≈ 4 obtained using eqs 1 and 2.
Compared to other classes of conjugated oligomers, the unsub-
stituted o-phenylenes, perhaps not surprisingly, have a short effective
conjugation length, roughly comparable to certain phenylene
ethynylenes or sterically hindered polypyroles.49,51 For comparison,
3,49
for unsubstituted p-phenylenes
n
ecl
≈ 10 (if planarity is forced
along the backbone necl g 12).52ꢀ54 These values may in fact
represent significant underestimates of necl for p-phenylenes, given
recent results which question its determination based on the
extrapolation of UVꢀvis data (particularly the 1/n dependence).55
Nevertheless, the value of necl ≈ 4 for the oP(H)n series should be
accurate as we have used Meier’s approach (i.e., eqs 1 and 2 rather
than plots of UV data vs 1/n) and especially considering that we
have synthesized oligomers longer than the saturation point of the
UVꢀvis data. The short necl of the o-phenylenes is reasonable in
view of the large twisting of the o-phenylene backbone, which would
be expected to attenuate conjugation. However, the effective
conjugation length of the oP(H)n series is significantly less than
that of the oP(OMe)n series (necl ≈ 8). In our original paper on the
oP(OMe)n series, we attributed their low Δλ (= 39 nm) to the
twisting of their π-systems and their long necl to conformational
rigidity. For our current series of compounds, Δλ is much larger
than for the methoxy-substituted case, although the large experi-
mental uncertainty for this parameter makes it difficult to draw firm
conclusions as to the cause. However, while the global conforma-
tional distribution is essentially the same for the two sets of
oligomers (reflected in the similar populations of I2, II1, and III2),
it is possible that the conformational rigidity is reduced for the
λn, UV ¼ λ∞, UV ꢀ ΔλUV eꢀbðn ꢀ 1Þ
ð1Þ
where λn,UV is the low-energy absorption maximum for an
oligomer with n monomer units. The parameter λ∞,UV is simply
λn,UV extrapolated to the polymer limit, ΔλUV = λ∞,UV ꢀ λ1,UV is
the overall effect of conjugation, and b is the extent of conjuga-
tion. If we define the effective conjugation length as the value of
n for which λ∞,UV ꢀ λn,UV e 1 nm, it follows that
ln ΔλUV
ð2Þ
necl
¼
þ 1
b
For the oP(H)n oligomers (n > 2), we unfortunately do not have
well-defined peaks in the low-energy regions of the spectra.
8430
dx.doi.org/10.1021/ma201866p |Macromolecules 2011, 44, 8425–8432

Macromolecules
ARTICLE
behavior for the oP(OMe)n series, the blue shift in fluorescence
spectra with increasing length appears to be an unusual but
general feature of the o-phenylenes. This behavior is not ex-
hibited by most conjugated oligomers, with the exception of
certain donorꢀacceptor-substituted compounds for which ex-
tension of the chain length attenuates a charge-transfer interac-
tion. The oP(H)n series clearly should not exhibit charge-transfer
states (nor do we observe any solvatochromism in either the
UVꢀvis or fluorescence spectra). Since analogous hypsochromic
shifts are not observed in the UVꢀvis spectra of the oligomers,
this effect must originate from varying degrees of structural
rearrangements in the excited states, leading to reduced Stokes
shifts. This phenomenon therefore must result from the unusual
conformational behavior of the o-phenylenes compared to most
other conjugated oligomers. It is well-known that the p-pheny-
lenes undergo planarization in the excited state.59,60 Clearly, for a
stacked helical o-phenylene this is not possible. We speculate at
this point that the shorter oligomers must be better able to
accommodate reductions in ϕi in the excited state due to reduced
steric hindrance. We are currently undertaking time-dependent
DFT investigations of the excited-state properties of the oP(H)n
series in order to test this hypothesis.
Figure 9. Steady-state fluorescence spectra of oP(H)n (cyclohexane).
Table 1. Fluorescence Properties of oP(H)n in Cyclohexane
λmax,FL (nm) Φf τf (ns) kr (ꢁ107 sꢀ1
)
knr (ꢁ107 sꢀ1
)
oP(H)4
oP(H)5
oP(H)6
oP(H)7
oP(H)8
396.5
390.5
383.0
380.0
376.0
0.01
0.13
0.11
0.17
0.18
0.4a
3.3
3.8
5.8
6.0
3.9
2.9
2.9
3.0
26
23
14
14
’ CONCLUSIONS
a Biexponential fit: 0.4 ns (97.7%), 6.8 ns (2.3%).
We have developed an improved method for the synthesis of
the parent o-phenylene oligomers oP(H)n up to the octamer.
Using NMR spectroscopy, we have shown that unsubstituted
o-phenylenes predominantly adopt stacked helical conformations
in solution. A more detailed analysis was possible for oP(H)6,
which adopts only three (of six) conformers, differing only in the
orientation of the ends of the chain. The conformational
distribution favors the stacked helical conformer and is similar
to that of our previously reported methoxy-substituted oP-
(OMe)6. Changes in the UVꢀvis spectra with length saturate
quickly for the oP(H)n series, which has an effective conjugation
length of necl ≈ 4 (compared to necl ≈ 8 for oP(OMe)n). Further,
the UVꢀvis spectra do not exhibit increasing hypochromicity
with increasing n. These behaviors may result from decreased
conformational rigidity for oP(H)n. The fluorescence spectra
exhibit a hypsochromic shift with increasing length, an unusual
feature for conjugated oligomers that we believe derives from
inhibited excited-state structural relaxation for the longer
oligomers.
oP(H)n series, possibly due to reduced πꢀπ-type interactions
between the monomers. A looser, less well-defined conformational
state for the oP(H)n series could explain the reduced necl. Greater
conformational flexibility is also consistent with the lack of hypo-
chromicity in the UVꢀvis spectra of oP(H)n, which was quite
pronounced for oP(OMe)n (up to 60%).
Fluorescence Spectroscopy. Beyond n = 3, the unsubstituted
o-phenylenes are fluorescent, as shown in Figure 9. Despite the
expected attenuated conjugation for the o-phenylenes, the emis-
sion maximum for oP(H)4 is actually red-shifted compared to the
analogous p-quaterphenyl (396 vs 365 nm).56 Like the UVꢀvis
spectra, the shapes of the fluorescence spectra are found to be
independent of concentration and solvent (Figure S5). Excita-
tion spectra are in good agreement with the UVꢀvis spectra
(Figure S6). Quantum yields were measured relative to 9,10-
diphenylanthracene (Φf = 0.91)57 and are tabulated in Table 1.
While oP(H)3 is essentially nonfluorescent (Φf could not be
determined on our instruments), the quantum yields increase
with increasing length and plateau around 0.18, which is identical
to the reported quantum yield of biphenyl (0.18).56,58 This
behavior is very similar to the behavior of the oP(OMe)n series
(e.g., for oP(OMe)8, Φf = 0.19). Fluorescence lifetimes (τf) were
determined by time-correlated single photon counting in
degassed cyclohexane. The lifetimes increase with increasing
chain length, although in all cases are substantially less than that
of biphenyl (16 ns).58 Radiative rate constants (kr = Φfτfꢀ1) fall
to a constant limiting value of 3 ꢁ 107 sꢀ1 for n g 6, whereas
nonradiative rate constants (knr = (1 ꢀ Φf)τfꢀ1) appear to fall
to a limiting value of 14 ꢁ 107 sꢀ1 for n g 7.
’ ASSOCIATED CONTENT
S
Supporting Information. Supplementary figures referred
b
to in the text; NMR spectra (1D and 2D); TCSPC data and fits;
MALDI spectra; experimental procedures; and Cartesian coordi-
nates of optimized geometries. This material is available free of
charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: scott.hartley@muohio.edu.
The fluorescence spectra exhibit a small but systematic
hypsochromic shift with increasing length, with a overall shift of
about 20 nm from oP(H)4 to oP(H)8. While changes in the
UVꢀvis spectra reach saturation quite quickly (necl ≈ 4), even
for oP(H)7 and oP(H)8 there is a significant difference in the
fluorescence maxima. As we had previously observed similar
’ ACKNOWLEDGMENT
Primary support from the National Science Foundation
(CHE-0910477) is gratefully acknowledged. Gaussian 09 was
purchased with the assistance of the Air Force Office of Scientific
8431
dx.doi.org/10.1021/ma201866p |Macromolecules 2011, 44, 8425–8432

Macromolecules
ARTICLE
Research (FA9550-10-1-0377), and the MALDI mass spectrom-
eter was purchased with the assistance of the NSF (CHE-
0839233).
(35) Sajiki, H.; Mori, A.; Mizusaki, T.; Ikawa, T.; Maegawa, T.;
Hirota, K. Org. Lett. 2006, 8, 987–990.
(36) For an oligomer of length n, the number of possible conforma-
tions is N = 2nꢀ4 + 2(1/2)nꢀ2
.
(37) Unfortunately, because of spectral overlap, 1 of 11 protons of
III2 and 2 of 22 protons of II1 could not be assigned. See the Supporting
Information for details.
(38) Waller, M. P.; Robertazzi, A.; Platts, J. A.; Hibbs, D. E.;
Williams, P. A. J. Comput. Chem. 2006, 27, 491–504.
(39) Ditchfield, R. Mol. Phys. 1974, 27, 789–807.
(40) Wolinski, K.; Hinton, J. F.; Pulay, P. J. Am. Chem. Soc. 1990,
112, 8251–8260.
’ REFERENCES
(1) Berresheim, A. J.; M€uller, M.; M€ullen, K. Chem. Rev. 1999, 99,
1747–1785.
(2) Ivory, D. M.; Miller, G. G.; Sowa, J. M.; Shacklette, L. W.;
Chance, R. R.; Baughman, R. H. J. Chem. Phys. 1979, 71, 1506–1507.
(3) Grimsdale, A. C.; M€ullen, K. Adv. Polym. Sci. 2006, 199, 1–82.
(4) Grimsdale, A. C.; Chan, K. L.; Martin, R. E.; Jokisz, P. G.;
Holmes, A. B. Chem. Rev. 2009, 109, 897–1091.
(5) Weiss, E. A.; Ahrens, M. J.; Sinks, L. E.; Gusev, A. V.; Ratner,
M. A.; Wasielewski, M. R. J. Am. Chem. Soc. 2004, 126, 5577–5584.
(6) Banerjee, M.; Shukla, R.; Rathore, R. J. Am. Chem. Soc. 2009,
131, 1780–1786.
(41) Wiitala, K. W.; Hoye, T. R.; Cramer, C. J. J. Chem. Theory
Comput. 2006, 2, 1085–1092.
(42) Jain, R.; Bally, T.; Rablen, P. R. J. Org. Chem. 2009, 74, 4017–
4023.
(43) Following the approach of Rablen and co-workers,42 before
0
calculating the error, we scale the δcalc values according to δcalc
=
(7) Goto, H.; Katagiri, H.; Furusho, Y.; Yashima, E. J. Am. Chem. Soc.
(δcalc ꢀ b)/m, where m is the slope and b is the intercept from the plots
in Figure 5. This is similar to the approach we took in our previous
paper,29 where we considered the standard error of the regression,
except that it favors fits with slopes close to unity. Either method leads to
the same conclusions.
(44) Ito, Y.; Ihara, E.; Murakami, M.; Sisido, M. Macromolecules
1992, 25, 6810–6813.
(45) Ito, Y.; Ohara, T.; Shima, R.; Suginome, M. J. Am. Chem. Soc.
1996, 118, 9188–9189.
(46) Ito, Y.; Miyake, T.; Hatano, S.; Shima, R.; Ohara, T.; Suginome,
2006, 128, 7176–7178.
(8) Goto, H.; Furusho, Y.; Yashima, E. J. Am. Chem. Soc. 2007, 129,
9168–9174.
(9) Goto, H.; Furusho, Y.; Yashima, E. J. Am. Chem. Soc. 2007, 129,
109–112.
(10) Ben, T.; Goto, H.; Miwa, K.; Goto, H.; Morino, K.; Furusho, Y.;
Yashima, E. Macromolecules 2008, 41, 4506–4509.
(11) Miwa, K.;Furusho, Y.;Yashima, E. Nature Chem. 2010, 2, 444–449.
(12) Tour, J. M. Adv. Mater. 1994, 6, 190–198.
(13) Ormsby, J. L.; Black, T. D.; Hilton, C. L.; Bharat; King, B. T.
Tetrahedron 2008, 64, 11370–11378.
M. J. Am. Chem. Soc. 1998, 120, 11880–11893.
(47) Motomura, T.; Nakamura, H.; Suginome, M.; Murakami, M.;
Ito, Y. Bull. Chem. Soc. Jpn. 2005, 78, 142–146.
(14) Kovacic, P.; Uchic, J. T.; Hsu, L.-C. J. Polym. Sci., Part A: Polym.
Chem. 1967, 5, 945–964.
(48) Bloomfield, V. A.; Crothers, D. M.; Tinoco, I., Jr. Physical
Chemistry of Nucleic Acids; Harper & Row: New York, 1974.
(49) Meier, H.; Stalmach, U.; Kolshorn, H. Acta Polym. 1997, 48,
379–384.
(50) Zhao, Y.; Campbell, K.; Tykwinski, R. R. J. Org. Chem. 2002, 67,
336–344.
(15) Kovacic, P.; Ramsey, J. S. J. Polym. Sci., Part A: Polym. Chem.
1969, 7, 111–125.
(16) Hsing, C.-F.; Jones, M. B.; Kovacic, P. J. Polym. Sci., Part A:
Polym. Chem. 1981, 19, 973–984.
(17) Xu, J.; Liu, H.; Pu, S.; Li, F.; Luo, M. Macromolecules 2006,
39, 5611–5616.
(51) Schumm, J. S.; Pearson, D. L.; Tour, J. M. Angew. Chem., Int. Ed.
Engl. 1994, 33, 1360–1363.
(18) Ma, M.; Liu, H.; Xu, J.; Li, Y.; Wan, Y. J. Phys. Chem. C 2007,
111, 6889–6896.
(52) Grimsdale, A. C.; M€ullen, K. Adv. Polym. Sci. 2008, 212, 1–48.
(53) Grimme, J.; Kreyenschmidt, M.; Uckert, F.; M€ullen, K.; Scherf,
U. Adv. Mater. 1995, 7, 292–295.
(54) Where applicable, the literature data have been reevaluated
using eqs 1 and 2 for consistency.
(55) Wang, Q.; Qu, Y.; Tian, H.; Geng, Y.; Wang, F. Macromolecules
2011, 44, 1256–1260.
(56) Berlman, I. B. Handbook of Fluorescence Spectra of Aromatic
Molecules, 2nd ed.; Academic Press: New York, 1971.
(57) Cross-checked against quinine bisulfate (Φf = 0.54). By this
method, our measured quantum yield of oP(H)2 (biphenyl) was in good
agreement with the literature (0.18).
(19) Dong, B.; Zheng, L.; Xu, J.; Liu, H.; Pu, S. Polymer 2007,
48, 5548–5555.
(20) Voisin, E.; Williams, V. E. Macromolecules 2008, 41, 2994–2997.
(21) Here, we consider only cases of oligomers in excess of six
repeat units.
(22) Wittig, G.; Lehmann, G. Chem. Ber. 1957, 90, 875–892.
(23) Ibuki, E.; Ozasa, S.; Murai, K. Bull. Chem. Soc. Jpn. 1975, 48,
1868–1874.
(24) Ozasa, S.; Fujioka, Y.; Fujiwara, M.; Ibuki, E. Chem. Pharm. Bull.
1980, 28, 3210–3222.
(25) Ibuki, E.; Ozasa, S.; Fujioka, Y.; Kitamura, H. Chem. Pharm.
Bull. 1980, 28, 1468–1476.
(58) Berlman, I. B. J. Chem. Phys. 1970, 52, 5616–5621.
(59) Momicchioli, F.; Bruni, M. C.; Baraldi, I. J. Phys. Chem. 1972,
76, 3983–3990.
(60) Heimel, G.; Daghofer, M.; Gierschner, J.; List, E. J. W.; Grimsdale,
A. C.; M€ullen, K.; Beljonne, D.; Brꢀedas, J.-L.; Zojer, E. J. Chem. Phys. 2005,
122, 054501.
(26) Ibuki, E.; Ozasa, S.; Fujioka, Y.; Okada, M.; Yanagihara, Y.
Chem. Pharm. Bull. 1982, 30, 2369–2379.
(27) Blake, A. J.; Cooke, P. A.; Doyle, K. J.; Gair, S.; Simpkins, N. S.
Tetrahedron Lett. 1998, 39, 9093–9096.
(28) He, J.; Crase, J. L.; Wadumethrige, S. H.; Thakur, K.; Dai, L.;
Zou, S.; Rathore, R.; Hartley, C. S. J. Am. Chem. Soc. 2010, 132,
13848–13857.
(29) Hartley, C. S.; He, J. J. Org. Chem. 2010, 75, 8627–8636.
(30) Ohta, E.; Sato, H.; Ando, S.; Kosaka, A.; Fukushima, T.;
Hashizume, D.; Yamasaki, M.; Hasegawa, K.; Muraoka, A.; Ushiyama,
H.; Yamashita, K.; Aida, T. Nature Chem. 2011, 3, 68–73.
(31) Ishikawa, S.; Manabe, K. Chem. Commun. 2006, 2589–2591.
(32) Ishikawa, S.; Manabe, K. Chem. Lett. 2006, 35, 164–165.
(33) Zhou, Q. J.; Worm, K.; Dolle, R. E. J. Org. Chem. 2004, 69,
5147–5149.
(34) Walker, S. D.; Barder, T. E.; Martinelli, J. R.; Buchwald, S. L.
Angew. Chem., Int. Ed. 2004, 43, 1871–1876.
8432
dx.doi.org/10.1021/ma201866p |Macromolecules 2011, 44, 8425–8432