Push-pull macrocycles: Donor-acceptor compounds with paired linearly conjugated or cross-conjugated pathways

Article abstract of DOI:10.1021/jo2026004

Two-dimensional π-systems are of current interest in the design of functional organic molecules, exhibiting unique behavior for applications in organic electronics, single-molecule devices, and sensing. Here we describe the synthesis and characterization of "push-pull macrocycles": electron-rich and electron-poor moieties linked by a pair of (matched) conjugated bridges. We have developed a two-component macrocyclization strategy that allows these structures to be synthesized with efficiencies comparable to acyclic donor-bridge-acceptor systems. Compounds with both cross-conjugated (m-phenylene) and linearly conjugated (2,5-thiophene) bridges have been prepared. As expected, the compounds undergo excitation to locally excited states followed by fluorescence from charge-transfer states. The m-phenylene-based systems exhibit slower charge-recombination rates presumably due to reduced electronic coupling through the cross-conjugated bridges. Interestingly, pairing the linearly conjugated 2,5-thiophene bridges also slows charge recombination. DFT calculations of frontier molecular orbitals show that the direct HOMO-LUMO transition is polarized orthogonal to the axis of charge transfer for these symmetrical macrocyclic architectures, reducing the electronic coupling. We believe the push-pull macrocycle design may be useful in engineering functional frontier molecular orbital symmetries.

Full text of DOI:10.1021/jo2026004

Article
pubs.acs.org/joc
Push−Pull Macrocycles: Donor−Acceptor Compounds with Paired
Linearly Conjugated or Cross-Conjugated Pathways
Wade C. W. Leu, Amanda E. Fritz, Katherine M. Digianantonio, and C. Scott Hartley*
Department of Chemistry & Biochemistry, Miami University, Oxford, Ohio 45056, United States
S
* Supporting Information
ABSTRACT: Two-dimensional π-systems are of current
interest in the design of functional organic molecules, exhibiting
unique behavior for applications in organic electronics, single-
molecule devices, and sensing. Here we describe the synthesis
and characterization of “push−pull macrocycles”: electron-rich
and electron-poor moieties linked by a pair of (matched)
conjugated bridges. We have developed a two-component
macrocyclization strategy that allows these structures to be
synthesized with efficiencies comparable to acyclic donor−
bridge−acceptor systems. Compounds with both cross-conjugated (m-phenylene) and linearly conjugated (2,5-thiophene)
bridges have been prepared. As expected, the compounds undergo excitation to locally excited states followed by fluorescence
from charge-transfer states. The m-phenylene-based systems exhibit slower charge-recombination rates presumably due to
reduced electronic coupling through the cross-conjugated bridges. Interestingly, pairing the linearly conjugated 2,5-thiophene
bridges also slows charge recombination. DFT calculations of frontier molecular orbitals show that the direct HOMO−LUMO
transition is polarized orthogonal to the axis of charge transfer for these symmetrical macrocyclic architectures, reducing the
electronic coupling. We believe the push−pull macrocycle design may be useful in engineering functional frontier molecular
orbital symmetries.
macrocycles”: single electron-rich (donor) and electron-poor
(acceptor) moieties bridged by two conjugated paths. While
conceptually simple, this class of structures, to our knowledge,
has received little attention,25 although examples of macro-
cycles with multiple donor/acceptor groups are known.26,27
Several aspects of push−pull macrocycles are of interest,
however. First, they address fundamental questions about
conjugation in two dimensions. One-dimensional oligomeric
push−pull systems are quite well-known and can be considered
models of single-molecule wires.28 However, the effect of
additional conjugative pathways on delocalization and electron-
transfer rates has received much less attention. Second,
macrocycles, especially shape-persistent macrocycles,9 are
well-known building blocks for advanced materials based on
columnar stacking (in solution,29−31 in liquid crystals,32−34 and
in low-dimensional nanostructures35−38). Push−pull macro-
cycles could therefore eventually combine control of intra-
molecular charge-transport with self-assembly into functional
materials. Third, rigid functional molecules, especially those
which exhibit switching behavior without nuclear rearrange-
ments, are better suited to future applications requiring
interacting single-molecule devices: such compounds have
greater potential to maintain coupling to their neighbors within
a network.39 The second conjugated bridge of push−pull
macrocycles rigidifies their cores. Fourth, Tobe and De Feyter
INTRODUCTION
■
Two-dimensional conjugation is currently being exploited in
the design of functional organic molecules,1−3 enabling behav-
ior not possible with one-dimensional (wire-like) conjugated
oligomers and polymers. Molecules with two-dimensional
π-systems, such as conjugated macrocycles or more complex
architectures,4−12 have been developed which exhibit enhanced
nonlinear optical properties,13 unusual emissive properties,14,15
and self-assembly into ordered monolayers16−18 or thin films
(for electronics).19 In this context, the interaction between
electron-rich and electron-poor substituents through two-
dimensional π-systems represents an important fundamental
area of study. For example, Haley has explored in detail the
effect of multiple conjugation paths in push−pull (donor−
acceptor) systems.20−22 Likewise, Bunz has used cruciform-
type structures to develop fluorescent compounds with spatially
separated frontier molecular orbitals, creating highly respon-
sive emissive materials for sensing.23,24 In principle, delocaliza-
tion in two dimensions is of great importance to the deve-
lopment of active components for single-molecule electronics,
which must ultimately exploit molecules with multiple (>2)
connections to inputs and outputs interacting in ordered
networks. Nevertheless, many fundamental configurations of
two-dimensional π-systems have yet to be synthesized and
characterized.
As part of a program on the design and synthesis of
molecular architectures with unusual electronic structures, we
became interested in a class of compounds we call “push−pull
Received: December 17, 2011
Published: February 2, 2012
© 2012 American Chemical Society
2285
dx.doi.org/10.1021/jo2026004 | J. Org. Chem. 2012, 77, 2285−2298

The Journal of Organic Chemistry
Article
have demonstrated that architectures with identical footprints
to our basic push−pull macrocycle structure (see below) self-
assemble into a large variety of two-dimensional lattices on
surfaces,17,18 highlighting their potential for the creation of
functional molecular networks.
developed a straightforward three-step synthesis of these
macrocycles. Charge-transfer excited states in these compounds
have been investigated using UV−vis and fluorescence
spectroscopy and fluorescence lifetime measurements. The
geometries and electronic structures of the compounds have
also been studied using DFT computational modeling. As
expected, the cross-conjugated donor−acceptor systems exhibit
reduced rates of radiative charge recombination. However, we
have found that the inclusion of the second matched linearly
conjugated bridge in D(Th)2A also reduces the rate of charge
recombination. We believe this effect can be rationalized on the
basis of frontier orbital symmetries. The push−pull macrocycle
design may therefore represent a useful approach to structures
with controlled intramolecular charge-transport.
Our target structures are also geometrically well-suited to the
incorporation of cross-conjugated moieties,40 specifically m-
phenylenes. Cross-conjugated π-systems are ubiquitous in
organic chemistry but are rarely exploited in materials
applications (compared to linearly conjugated π-systems). In
general, the extension of cross-conjugation does not lead to a
substantial narrowing of HOMO−LUMO gaps or increased
delocalization of charge. Consequently, cross-conjugated
moieties are not well-suited to the design of materials for
current applications in visible light emission or charge
transport. However, it has been suggested that cross-
conjugation may be exploited to control electron transport in
single molecules.41−47 Solomon and Andrews, for example,
have shown (computationally) that electron transmission
through cross-conjugated moieties may be manipulated over
many orders of magnitude and may therefore be used to create
functional devices such as transistors.43−46 Many efforts to
synthesize functional molecules based on cross-conjugated
systems have been made,48 particularly by Tykwinski,49−52
Diederich,53−56 Sherburn,57,58 and others.59−64 Of particular
RESULTS
■
Synthesis. One of the obstacles to the study of push−pull
macrocycles is that they are, in principle, more challenging to
prepare than their acyclic counterparts. In general, there are
two approaches to the synthesis of shape-persistent macro-
cycles.69 Kinetically controlled strategies make use of
irreversible bond formation and typically assemble the
macrocycle via the intramolecular cyclization of an oligomeric
precursor after a relatively long series of high-yielding steps.
Alternatively, kinetically controlled reactions may also be used
in low-yielding multicomponent intermolecular macrocycliza-
tions. Thermodynamically controlled strategies allow the
macrocycle to be assembled from simple monomers in one
high-yielding step but require that the target compound be the
most stable possible product under the chosen conditions.
While very powerful, thermodynamic control is typically used
to prepare only highly symmetrical macrocycles.
́
relevance here, Bardeen, Martinez, and Thayumanavan have
shown that cross-conjugated m-phenylene bridges allow fast
excited-state charge separation, but relatively slow charge
recombination, a property which may be useful in organic
photovoltaics.65 Van Walree has also demonstrated that push−
pull cross-conjugated systems undergo efficient formation of
charge-separated states.66,67 Very recently, Ratner and
Wasielewski have provided direct experimental evidence
demonstrating that charge transport through cross-conjugated
bridges is sensitive to substituent effects, suggesting that
electron transmission rates should be externally controllable.68
Our first targets were m-phenylene-based D(mP)2A and its
symmetrical analogues D(mP)2D and A(mP)2A. The tert-butyl
groups were incorporated into our design to promote solubility.
Compounds based on the parent macrocyclic tetrabenzo[18]-
cyclyne (TBC) core have been synthesized but have received
little attention otherwise.70−74 However, examples of related
structures are well-known, including work by Tobe17,18,75 and
Haley4 on TBCs with intraannular alkynyl groups. The
photophysical properties of some of these graphyne fragments
have been studied, and as discussed above, this structure
represents a useful platform for the control of two-dimensional
self-assembly.
Interestingly, TBC macrocycles have been used as examples
of both kinetically controlled and thermodynamically con-
trolled synthetic strategies. For example, Moore prepared one
TBC derivative by constructing a bifunctional oligomeric m-
phenylene ethynylene tetramer and carrying out an intra-
molecular macrocyclization (kinetic control).70 Alternatively,
Vollhardt efficiently assembled the parent TBC from simple
dipropynylbenzene monomers through alkyne metathesis
(thermodynamic control).71 Unfortunately, however, neither
of these approaches is ideal for the synthesis of our target
push−pull macrocycles. The stepwise synthesis of single bifunc-
tional oligomeric precursors would be too time-consuming
for the preparation of a large series of these compounds, and
alkyne metathesis would likely not be compatible with our
desired push−pull substitution pattern.
In this paper, we present the synthesis and characterization of
push−pull macrocycles D(mP)2A and D(Th)2A, along with
both symmetrical and acyclic analogues. These structures
feature electron-rich (veratrole) and electron-deficient (phtha-
limide) moieties bridged by a pair of either cross- or linearly
conjugated linkers (m-phenylene or 2,5-thiophene). We have
The best possible synthesis would allow the macrocycle to be
prepared as easily as an acyclic donor−bridge−acceptor
compound. Consequently, we focused on the simple three-
step method shown in Scheme 1 (top). To begin, standard
2286
dx.doi.org/10.1021/jo2026004 | J. Org. Chem. 2012, 77, 2285−2298

The Journal of Organic Chemistry
Article
a
Scheme 1
a
Reagents and conditions: (a) Pd(OAc)2, PPh3, CuI, HNiPr2; (b) TBAF, THF; (c) Pd(PtBu3)2, NEt3, toluene (D(mP)2A and A(mP)2A); (d)
Pd(PtBu3)2, DABCO, toluene (D(mP)2D, all thiophenes).
Sonogashira reaction conditions are effective for the coupling of
an o-diiodoarene with known TIPS-protected m-phenylene
ethynylene monomer,76,77 giving 1a(mP)2 and 1b(mP)2.
Deprotection with TBAF to give 2a(mP)2 and 2b(mP)2 then
proceeds in moderate isolated yields, which are lower than one
would usually expect for this reaction: purification of these
compounds by flash chromatography requires a difficult
separation from nearly coeluting impurities, and we were
forced to sacrifice yield for purity. The key step in our synthesis
is the final two-component intermolecular macrocyclization.
Intermolecular kinetically controlled macrocyclizations tend to
be low yielding. However, the two-component reaction is a
special case: in principle, it should give high yields if the second
(intramolecular) coupling is significantly faster than the first
(intermolecular). We have found that Sonogashira coupling
catalyzed by bis(tri-tert-butylphosphine)palladium(0) under
copper-free conditions78 gives good yields of our TBC targets,
proceeding well at room temperature and at reasonably high
concentrations (10 mM). In test reactions, we observed
unacceptable quantities of oxidative coupling byproducts
when less-active catalyst systems were used in combination
with copper(I) cocatalysts. The highest yields were obtained for
the push−pull-substituted D(mP)2A and symmetrical
A(mP)2A, which are synthesized from the activated diiodo-
phthalimide coupling partner. For these reactions, we obtained
the best results using triethylamine as the base (in toluene). For
the more demanding synthesis of D(mP)2D, which requires
coupling to the deactivated (electron-rich) 4,5-diiodoveratrole,
we obtained better results using DABCO as the base.
Importantly, in the synthesis of our key push−pull macrocycle
D(mP)2A, the yields (71% per step on average, 36% overall)
and length (three steps) are comparable to the synthesis of a
simple acyclic compound.
We then turned to the more demanding synthesis of the 2,5-
thiophene-containing D(Th)2A, D(Th)2D, and A(Th)2A
(Scheme 1, bottom). To our knowledge, compounds based
on this macrocyclic core have not been reported.79 While the
120° angle between the substituents of a m-phenylene is ideal
for our macrocyclic structures, the 140° angle of a 2,5-
thiophene is too shallow, implying that the macrocycles
D(Th)2A, D(Th)2D, and A(Th)2A should be strained (see
optimized geometries below). Coupling of the o-diiodoarenes
to known TIPS-protected 2,5-thiophene ethynylene80,81 mono-
mer (to give 1a(Th)2 and 1b(Th)2) and TBAF deprotection
(2a(Th)2 and 2b(Th)2) proceed as well as for the m-
phenylenes. Lowering the concentration to 2 mM for the key
macrocyclization step leads to significant improvements in
yields for this series, although the optimized yields are still
modest (30−40%).
For comparison of photophysical properties, we also
prepared the acyclic analogues D(mP)A and D(Th)A of our
push−pull macrocycles. The syntheses are directly analogous to
those for the macrocycles (see the Experimental Section).
UV−Vis and Fluorescence Spectroscopy. Representative
UV−vis spectra for all of the push−pull compounds in
cyclohexane are shown in Figure 1; spectra in other solvents
and of the symmetrical compounds (D(mP)2D, A(mP)2A,
D(Th)2D, and A(Th)2A) are given in the Supporting
Information (Figure S1). In all cases, the shapes of the spectra
are unchanged over more than an order of magnitude in con-
centration (ca. 10−7−10−5 M), suggesting that the compounds
do not aggregate in solution at the concentrations used.82 The
UV
max
UV−vis spectra exhibit no solvatochromism (Δλ < 10 nm)
in solvents of varying polarity (cyclohexane, toluene, dioxane,
chloroform, and dichloromethane). Further, the spectra of the
push−pull compounds are not bathochromically shifted relative
2287
dx.doi.org/10.1021/jo2026004 | J. Org. Chem. 2012, 77, 2285−2298

The Journal of Organic Chemistry
Article
D(mP)A, D(Th)2A, D(Th)A) are highly solvatochromic, with
systematic bathochromic shifts with increasing solvent polarity,
FL
max
as shown in Figure 1 (Δλ ≈ 200 nm from cyclohexane to
dichloromethane). Thus, as expected, the push−pull com-
pounds fluoresce from charge-transfer states. Note that the
acyclic cross-conjugated compound D(mP)A exhibits two
fluorescence bands when dissolved in more polar solvents
(chloroform and dichloromethane), despite numerous attempts
at purification by multiple techniques.83 We assign the band
at ∼600 nm to charge-transfer fluorescence, with compet-
ing emission from a locally excited state (band at roughly
400 nm).84
A more quantitative analysis of the fluorescence solvato-
chromism can be carried out using Weller’s modification of the
Lippert−Mataga equation, which allows the excited state dipole
moment (μe) to be estimated85
to their symmetrical analogues. Thus, excitation from the
equilibrium ground-state geometry initially generates a non-
polar locally excited state, as opposed to direct charge-transfer
excitation. For both the 2,5-thiophene- and m-phenylene-based
systems, the macrocycles exhibit absorption bands at similar
wavelengths to their acyclic counterparts, with the main dif-
ference being a relatively weak (ε < 104 M−1 cm−1) band to the
red that we assign to weakly allowed π−π* transitions on the
basis of TD-DFT calculations (see below).
All of the synthesized compounds are fluorescent. The
shapes of the fluorescence spectra are again independent of
concentration (ca. 10−5−10−7 M), indicating that there is no
excimer emission. As expected, the symmetrically substituted
macrocycles (D(mP)2D, A(mP)2A, D(Th)2D, A(Th)2A) do
not exhibit significant fluorescence solvatochromism (Figure
S2). However, in contrast to the UV−vis spectra, the
fluorescence spectra of all push−pull compounds (D(mP)2A,
2
2μ
e
ν = −
Δf′ + ν
̅
0
̅
f
3
(1)
hca
where νf is the fluorescence emission maximum (in cm−1), ν0 is
the emission maximum in vacuum, a is the radius of the
Onsager cavity, and Δf′ is a solvent-dependent parameter
calculated from its dielectric constant (ε) and refractive index
(n):
̅
̅
2
ε − 1
2ε + 1
n − 1
Δf′ =
−
2
(2)
4n + 2
Lippert−Mataga plots for the push−pull compounds are
given in Figure 2. Although this analysis may not be applicable
to the dual fluorescence compound D(mP)A, examples of
Figure 1. UV−vis (left) and fluorescence spectra (right) of macrocycles D(mP)2A and D(Th)2A (top) and acyclic analogues D(mP)A and D(Th)A
(bottom). UV−vis spectra are in cyclohexane, with a small amount of dichloromethane (<2%).
2288
dx.doi.org/10.1021/jo2026004 | J. Org. Chem. 2012, 77, 2285−2298

The Journal of Organic Chemistry
Article
Table 1. Excited-State Dipole Moments, Quantum Yields,
Fluorescence Lifetimes, and Radiative and Nonradiative
Rate Constants
a
μe
kr
knr
b
(D) solvent
Φf
τf (ns) (×108 s−1
)
(×108 s−1
)
D(mP)2A
D(mP)A
D(Th)2A
D(Th)A
48
48
25
33
Cy
0.48
0.31
6.7
22.2
5.0
0.71
0.14
0.04
2.7
0.77
0.31
2.0
Diox
DCM
Cy
0.021
0.40
1.5
4.1
d
e
Diox
DCM
Cy
0.26
16.7
nd
0.16
0.44
<0.005
0.017
0.02
c
2.3
0.08
0.09
0.05
5.5
4.3
4.3
10
Diox
DCM
Cy
2.3
Figure 2. Lippert−Mataga plots for D(mP)2A, D(mP)A, D(Th)2A,
and D(Th)A. Inset: expanded view of the plot for D(Th)2A in binary
mixtures of dichloromethane and cyclohexane.
0.005
0.29
1.0
0.52
0.56
0.62
3.3
13
Diox
DCM
Cy
0.19
3.4
15
0.11
1.8
14
Lippert−Mataga plots of dual fluorescence compounds exist in
the literature;86,87 the data for its charge-transfer emission band
is included for comparison. Reasonably good linear correlations
between μe and Δf′ are observed, given that eq 1 is an
approximation that neglects second-order effects and specific
solvent−solute interactions (R2 ≥ 0.88).88 The fluorescence
spectra of D(Th)2A are less solvent-dependent than those of
the other compounds. To examine this effect more closely, we
measured fluorescence spectra in binary mixtures of cyclo-
hexane and dichloromethane in order to reduce solvent-specific
effects (Figure 2, inset).89 In this solvent system, the Lippert−
Mataga plot has two distinct linear regions. Plots in binary
mixtures of cyclohexane/dichloromethane for the other
compounds are given in the Supporting Information (Figure
S3) and are in very good agreement with Figure 2.
The Lippert−Mataga plots (Figure 2 and Figure S3,
Supporting Information) and comparison with the spectra for
the symmetrical compounds indicate that three of the push−
pull compounds, D(mP)2A, D(mP)A, and D(Th)A, fluo-
rescence from a charge-transfer state in solvents more polar
than cyclohexane. However, the solvent-dependence of the
emission of D(Th)2A is more complex. The two linear regions
in the Lippert−Mataga plot for this compound suggest that its
fluorescence in cyclohexane, toluene, and dioxane arises from a
nonpolar locally excited state and shifts to a charge-transfer
state in the more polar solvents. This behavior is also consistent
with comparison of its spectra to those of its symmetrical
analogues A(Th)2A and D(Th)2D.90
D(mP)2D
A(mP)2A
D(Th)2D
A(Th)2A
0.63
1.9
1.1
0.93
9
Cy
0.49
5.5
0.90
0.07
0.07
Cy
0.008
0.014
1.1
f
Tol
2.1
4.7
a
Calculated from Lippert−Mataga plots for binary mixtures of
cyclohexane and dichloromethane. Cy = cyclohexane, DCM =
dichloromethane, Tol = toluene. Based on the linear portion at the
high Δf′ region of the Lippert−Mataga plot. Double exponential fit:
τ1 = 21.5 ns (73%) and τ2 = 3.7 ns (27%). Double exponential fit:
τ1 = 0.85 ns (87%) and τ2 = 2.3 ns (13%). Double exponential fit: τ1 =
1.8 ns (86%) and τ2 = 3.8 ns (14%).
b
c
d
e
f
(1.2 nm across the long axis of the macrocycle, measured
between the centroids of the opposing aromatic rings).71
Conversely, the estimated μe for the linearly conjugated
thiophene-based compound D(Th)2A (calculated from the
data in cyclohexane/dichloromethane at the high Δf′ limit) is
lower at 25 D.
Quantum yields (Φf) for all compounds were determined
relative to quinine bisulfate (emission from 350−600 nm) or
rhodamine 101 (emission above 550 nm) and are listed in
Table 1.93,94 In cyclohexane and dioxane, most of the com-
pounds are highly fluorescent, with the exception of the
thiophene-based macrocycles (D(Th)2A, D(Th)2D, and
A(Th)2A). The quantum yields for the push−pull compounds
were also determined in dichloromethane and decrease
significantly in all cases (Φf ≤ 0.02), with the exception of
D(Th)A (Φf = 0.11).
In order to better understand the effect of different
conjugation pathways on charge recombination rates, we also
measured fluorescence lifetimes (τf) for all compounds by time-
correlated single photon counting (excitation at 350 nm), with
the lifetimes compiled in Table 1. Fluorescence decays for the
compounds are given in Figure S4, Supporting Information.
Most fluorescence decays are adequately fit by monoexpo-
nential functions. When biexponential fits are needed (see
Table 1), one component dominates; average lifetimes were
used in these cases for comparison with the other compounds.
In general, the compounds featuring linearly conjugated (2,5-
thiophene) bridges exhibit shorter lifetimes than those with
cross-conjugated (m-phenylene) bridges, and the macrocyclic
compounds exhibit longer lifetimes than their acyclic analogues.
To aid in interpretation of the experimental data, radiative
and nonradiative rate constants (kr and knr) were calculated
The slopes extracted from the Lippert−Mataga plots were
used to estimate the excited-state dipole moments for the
compounds. The volume of the Onsager cavity (a3) of each
molecule was approximated by the total volume of a space-
filling model at a PM3-minimized geometry.91 This is a crude
approximation, particularly as these structures should have rod-
like or elliptical shapes (as opposed to spherical). However, the
estimate should allow consistent comparisons to be made
between the compounds, and is similar to methods used in
other studies.65,92 Calculated approximate μe values are
compiled in Table 1. Use of either form of the Lippert−
Mataga plot (pure solvents or binary cyclohexane/dichloro-
methane mixtures) leads to qualitatively the same results.
The estimated μe for the cross-conjugated macrocycle
D(mP)2A is approximately 48 D. Interestingly, this value
corresponds to the separation of an electron−hole pair by
1.0 nm, which is slightly less than the diameter of the parent
tetrabenzo[18]cyclyne core as determined by X-ray crystallography
from the quantum yield and lifetime data (kr = Φf τf−1, knr
(1 − Φf)τf−1), and are also compiled in Table 1.
=
2289
dx.doi.org/10.1021/jo2026004 | J. Org. Chem. 2012, 77, 2285−2298

The Journal of Organic Chemistry
Article
Figure 3. Left: PCM(cyclohexane)/B3LYP/6-31+G(d,p)-minimized geometries for D(mP)2A′, D(mP)A′, D(Th)2A′, and D(Th)A′. Right:
PCM(cyclohexane)/CAM-B3LYP/6-311+G(2d,2p)//PCM/B3LYP/6-31+G(d,p) frontier molecular orbitals.
Ab Initio Calculations. In order to better understand the
electronic structures of the synthesized compounds, we
explored their properties using DFT.95 To reduce the
computational time, the structures were simplified by replacing
the tert-butyl and hexyl groups with hydrogen atoms (D(mP)2A′,
D(Th)2A′, etc.). A brief screening of different functionals
(B3LYP,96 PBE0,97 and CAM-B3LYP98) suggested that they
have little effect on the obtained optimized geometries. We
therefore used the popular B3LYP functional with the 6-
31+G(d,p) basis set for geometry optimization. Solvent effects
(cyclohexane) were included using the PCM model.99 Geo-
metries for the push−pull compounds are given in Figure 3
(left), with the rest in the Supporting Information (Figure S5).
For the acyclic structures (D(mP)A′ and D(Th)A′), there are
three additional possible conformers differing in the relative
orientations of the veratrole, bridge, and phthalimide groups.
The energetic differences between these various conformers are
very small at the B3LYP/6-31+G(d,p) level (<0.1 kcal/mol).
The calculated properties (e.g., frontier molecular orbitals) are
similar for all conformations, thus we have chosen to focus on
the conformers most closely corresponding to the macrocyclic
analogues for the discussion below.
not as well-accommodated within the macrocyclic framework.
Taking D(Th)2A′ as a representative example, the veratrole,
phthalimide, and ethynyl carbons are coplanar, with the
thiophene rings twisted at an angle of 42° to this plane. Also,
the ethynyl units are distorted from linearity, with bond angles
of 171−177°.
For comparison with our experimental results, we calculated
the UV−vis spectra for these minimized ground-state geo-
metries using TD-DFT.101 Since standard functionals, such as
B3LYP, are known to poorly reproduce the energies of charge-
transfer states,102 we decided to adopt the long-range-corrected
CAM-B3LYP functional.103−105 Excited state properties were
therefore calculated at the TD/PCM(Cyclohexane)/CAM-
B3LYP/6-311+G(2d,2p) level at our PCM/B3LYP/6-31G
+(d,p) geometries. The HOMOs and LUMOs calculated by
CAM-B3LYP for the push−pull compounds are shown in
Figure 3 (right). As expected for donor−acceptor compounds,
the HOMOs are localized on the electron-rich (veratrole)
halves of the structures, whereas the LUMOs are localized on
the electron-poor (phthalimide) halves.
UV−vis transitions and the corresponding simulated spectra
for D(mP)2A′ and D(Th)2A′ are shown in Figure 4, with the
remainder given in the Supporting Information (Figure S6). In
general, the TD-DFT calculations do an excellent job of
(qualitatively) reproducing the experimental spectra (note that
vibronic structure is not included in the calculations). For
D(Th)2A′, the strength of the lowest energy transition is
underestimated, which is reasonable considering that this
transition has a very low oscillator strength (calcd f = 0.002),
and the calculations do not account for symmetry-breaking
All of the m-phenylene-based macrocycles are predicted to
adopt planar geometries, consistent with the reported crystal
structure of the parent compound.71,100 The TBC macrocycle is
essentially the perfect size to accommodate the intraannular
hydrogen atoms: in D(mP)2A′, they are calculated to be 2.14 Å
apart, a close match to the sum of their van der Waals radii
(2.18 Å). Bond angles for the ethynyl groups are close to
linearity (178−179°). In contrast, the 2,5-thiophene moiety is
2290
dx.doi.org/10.1021/jo2026004 | J. Org. Chem. 2012, 77, 2285−2298

The Journal of Organic Chemistry
Article
DISCUSSION
■
Inspection of the calculated orbitals in Figure 3 highlights a key
difference between our cross-conjugated and linearly con-
jugated push−pull structures. The HOMOs and LUMOs of the
linearly conjugated compounds (D(Th)2A′ and D(Th)A′) are
delocalized through the bridges, whereas those of the cross-
conjugated compounds (D(mP)2A′ and D(mP)A′) are
distinctly localized on the two halves of the compounds: direct
interaction between the donor and acceptor groups is largely
blocked by the m-phenylene moieties. This reflects the reduced
ground-state resonance interaction between the substituents
through the cross-conjugated π-systems. Interestingly, com-
pounds with spatially separated HOMOs and LUMOs are
currently being investigated as highly responsive detectors of
various analytes.23 We are not aware of systems that have
exploited this feature of donor−acceptor-substituted cross-
conjugated π-systems.106
Although the ground-state interaction between the sub-
stituents is small in the cross-conjugated π-systems, clearly this
does not interfere with the efficiency of photoinduced charge
separation,107 as reflected in the similar charge-transfer
fluorescence quantum yields for D(mP)2A and D(mP)A
versus D(Th)A (in dioxane). This is consistent with the
reported behavior of other push−pull-substituted cross-
conjugated molecules.65−67 The excited state dipole moment
(Table 1) is substantially larger for D(mP)2A (and D(mP)A)
compared to the linearly conjugated systems, implying more
complete charge-transfer for the cross-conjugated compound
despite the similar donor−acceptor separations in all of the
compounds.108 This further emphasizes the potential of cross-
conjugated systems for applications in, for example, photo-
voltaics. Presumably, the origin of this effect is the increased
orbital localization apparent in Figure 3. It could be argued that
2,5-thiophene bridges are not electronically passive compared
to m-phenylenes; however, both sets of compounds have
similar HOMO−LUMO gaps (Figure 3), and the thiophene
bridges do not appear to shift either the HOMOs and LUMOs
away from the electron-rich or electron-poor sides of D(Th)2A
or D(Th)A.
Compounds with donor and acceptor groups bridged by m-
phenylenes,65 or other cross-conjugated bridges,68 should
undergo slower charge recombination after photoinduced
electron transfer compared to linearly conjugated analogues.
The behavior of our acyclic compounds, D(mP)A and
D(Th)A, is consistent with these observations. Comparing
the fluorescence data in dioxane, which should correspond to
radiative charge-recombination for D(mP)A, D(Th)A, and
D(mP)2A, the linearly conjugated D(Th)A exhibits a roughly
30-fold shorter fluorescence lifetime compared to D(mP)A,
corresponding to a 21-fold increase in its radiative rate constant
kr (and a 34-fold increase in knr). The reduced efficiency of
charge recombination across the cross-conjugated bridges is
also apparent when comparing D(mP)2A to D(Th)A. These
results are in good agreement with previous comparisons of
m-phenylene to (linearly conjugated) p-phenylene bridges.65
However, the distinguishing feature of our compounds is the
addition of the second bridge in the push−pull macrocycles.
According to our DFT calculations (Figure 3), the additional
bridges narrow the HOMO−LUMO gaps of the macrocycles
compared to their acyclic analogues. Overall, however, there is
relatively little effect on the overall shapes of the frontier
orbitals (i.e., in terms of the relative contributions from
Figure 4. TD-DFT calculations of UV−vis transitions for D(mP)2A′
(top) and D(Th)2A′ (bottom) at the TD/PCM(Cyclohexane)/CAM-
B3LYP/6-311+G(2d,2p)//PCM/B3LYP/6-31+G(d,p) level. The
points represent the oscillator strengths for the 24 lowest energy
transitions. The solid lines are the simulated spectra based on these
transitions with a 0.333 eV half-width at half-height. The dotted lines
are the experimental spectra of D(mP)2A and D(Th)2A in
cyclohexane (au).
structural fluctuations. A summary of key data for the first
singlet excited states (S1) of the push−pull compounds is given
in Table 2. For the m-phenylene-based push−pull compounds
Table 2. Calculated First Excited-State Energies (E),
Oscillator Strengths (f), Compositions, and Ground- and
calc
Excited-State Dipoles (μg and μecalc) at the Ground-State
Geometries
a
compd
E (eV)
f
μgcalc
μecalc
composition
D(mP)2A′
D(mP)A′
D(Th)2A′
D(Th)A′
3.28
3.74
2.61
3.05
0.112
1.049
0.002
1.857
6.3
5.1
6.8
5.9
13.0
14.7
14.9
16.6
H−1 → L (71%)
H−1 → L (71%)
H → L (74%)
H → L (68%)
a
H = HOMO, L = LUMO.
(D(mP)2A′ and D(mP)A′), the first excited state (at the
ground-state geometry) corresponds to a HOMO−1 →
LUMO transition localized on the phthalimide side of the
macrocycle. This agrees well with the experimental match
between the longest wavelength bands of D(mP)2A and
A(mP)2A. For the thiophene-based compounds (D(Th)2A′
and D(Th)A′), the first excited state corresponds to a direct
HOMO → LUMO transition. In all cases, the predicted
excited-state dipole moments (μecalc) at the ground-state geo-
metries are substantially lower than those calculated by
Lippert−Mataga analysis of the fluorescence spectra, consistent
with the limited solvatochromism of the UV−vis spectra.
2291
dx.doi.org/10.1021/jo2026004 | J. Org. Chem. 2012, 77, 2285−2298

The Journal of Organic Chemistry
Article
corresponding atoms). The appearance of the UV−vis spectra
does change dramatically, due principally to the very low
oscillator strengths of the lowest-energy transitions for the
macrocycles.
symmetry-breaking vibrational modes). For the push−pull
macrocycles, the addition of a second conjugated bridge
therefore has the somewhat counterintuitive consequence of
reducing the interaction between the charge-transfer and ground
states. In the case of the cross-conjugated macrocycle
D(mP)2A, this effect appears to be small, given the already
weak electronic coupling due to cross-conjugation. However,
the effect is much more pronounced in the case of the linearly
conjugated systems resulting in a large decrease in kr. We note,
however, that we cannot at this point separate the symmetry
effect from any reduction in V caused by the out-of-plane
twisting of the thiophenes.
Of course, the electronic coupling is only one part of the
generation of long-lived charge-transfer states, which also
requires control of nonradiative decay rates. Nevertheless, the
push−pull macrocycle structure may be useful for the design
of molecules with functional molecular orbital symmetries.
The generation of long-lived charge-transfer states is of key
importance to organic photovoltaics and artificial photosyn-
thesis.116,121,122 Similarly, since the reduced electronic coupling
is dependent on molecular symmetry, it could be used as the
basis for the creation of molecular switches.41 This idea should
be complementary to the use of m-phenylenes (or other cross-
conjugated structures) as a way to control charge-transfer. The
reduced conductance in cross-conjugated π-systems has been
predicted to attenuate when non-nearest neighbor and vibra-
tional effects are taken into account.47 It may therefore be
possible to use paired bridges to reinforce these effects and
improve switching performance.
Intuitively, one might expect that the addition of a second
conjugated bridge in a push−pull system either would have no
effect on the properties of the charge-transfer state or would
increase the rate of charge recombination. This is reflected in
the cross-conjugated push−pull compounds D(mP)2A and
D(mP)A, which have similar quantum yields and lifetimes for
radiative charge-recombination. However, very different behav-
ior is observed for D(Th)2A and D(Th)A. For these linearly
conjugated systems, the macrocyclic D(Th)2A exhibits a
roughly 40-fold smaller radiative decay constant (kr) in
dichloromethane, in which both compounds emit from
charge-transfer states, as well as a reduced nonradiative decay
constant (knr).
The slower nonradiative excited-state decay of D(Th)2A
compared to D(Th)A likely results from reduced vibrational
deactivation due to the added conformational rigidity from the
second bridge. The difference in radiative rates can be partly
attributed to reduced π-overlap in the macrocycle due to the
twisting of the thiophene bridges. However, we believe this
effect also results from the symmetry of the macrocyclic
compounds. We consider here these push−pull compounds in
the context of a simple two-state charge-recombination model.
This is an oversimplification, as it neglects the role of coupling
of the charge-transfer state to locally excited states.109
Nevertheless, it does provide a reasonable qualitative frame-
work for discussion of the role of the frontier molecular
orbitals.110
As usual, kr depends on the magnitude of the transition
dipole moment (Mν) between the charge-transfer and ground
states. For a simple charge-transfer transition, Mν depends on
the change in dipole moment for the transition (Δμ), its energy
(hν), and the electronic coupling matrix element between the
two states (V) as Mν = (VΔμ)/(hν). The transition moment
should be directed along the axis of charge transfer.111,112
Accordingly, for the acyclic compounds D(mP)A′ and
D(Th)A′ the calculated (TD-DFT) transition moments with
significant HOMO−LUMO contributions lie along their long
molecular axes. However, inspection of the molecular orbitals
for both push−pull macrocycles (Figure 3) reveals that the
HOMO−LUMO transition moment must be polarized
orthogonal to the axis of charge transfer, the orbitals being
symmetric (HOMO) and antisymmetric (LUMO) with respect
to the molecular mirror plane (e.g., for D(mP)2A′, belonging to
the C2v point group, the HOMO has B1 symmetry and the
LUMO has A2 symmetry). Assuming that the symmetries do
not change on geometric relaxation of the excited state, the
electronic coupling (V) between the charge-transfer and
ground states should therefore be weak in the macrocycles, as
a direct consequence of the symmetry of the matched
conjugated bridges.
CONCLUSIONS
■
We have designed, synthesized, and characterized a series of
push−pull macrocycles: donor and acceptor groups connected
by a pair of identical (cross-) conjugated bridges. With
optimization, the key macrocyclization synthetic step can be
carried out with comparable efficiency to the final bond-
forming step in the syntheses of acyclic donor−bridge−
acceptor compounds. Examining both m-phenylene- and 2,5-
thiophene-bridged macrocycles, charge-separation occurs with
similar efficiency compared to their acyclic push−pull counter-
parts. Cross-conjugated bridges appear to promote more
complete charge-transfer as judged by the experimental
excited-state dipole moments. The addition of the second,
matched conjugated bridge has the effect of reducing the
electronic coupling between the charge-transfer state and the
ground state, particularly for the linearly conjugated systems.
EXPERIMENTAL SECTION
■
General Methods. Unless otherwise noted, all starting materials,
reagents, and solvents were purchased from commercial sources and
used without further purification. Melting points were determined
using a differential scanning calorimeter at a heating rate of 10 °C/min
or by optical microscopy with a variable temperature stage. NMR
spectra were measured in CDCl3 solutions using 200, 300, or 500
MHz NMR spectrometers. Chemical shifts are reported in δ (ppm)
relative to TMS, with the residual solvent protons used as internal
In general, the electronic coupling is a critical parameter for
both radiative and nonradiative charge-transfer,113 and, for
example, relates to charge-transfer in donor−bridge−acceptor
compounds as molecular wires.114 The effect of orbital
symmetry on V has been previously investigated, particularly
in model donor−bridge−acceptor systems with (single) rigid,
saturated bridges (typically belonging to the Cs point
group).115−120 Symmetry effects on orbital overlap are indeed
found to reduce V (although the effect tends to be limited by
1
standards (7.26 for H, 77.16 for 13C). MALDI mass spectra were
recorded in reflectron mode using dithranol, HABA, or TCNQ as the
matrix. In many cases, the MALDI samples were doped with CuI to
promote ionization of Cu+ adducts.
Synthesis of Macrocycles (Scheme 1). Trimer 1a(mP)2.
A
Schlenk tube containing Pd(OAc)2 (20.1 mg, 0.090 mmol),
CuI (12.1 mg, 0.064 mmol), PPh3 (118 mg, 0.46 mmol),
and 4,5-diiodoveratrole123 (422 mg, 1.08 mmol) was evacuated
2292
dx.doi.org/10.1021/jo2026004 | J. Org. Chem. 2012, 77, 2285−2298

The Journal of Organic Chemistry
Article
and backfilled with argon (3×). To this solid mixture was
added a solution of 1-tert-butyl-3-ethynyl-5-[(triisopropylsilyl)-
ethynyl]benzene76,77 (800 mg, 2.36 mmol) in HNiPr2 (5 mL).
The suspension was degassed by three freeze−pump−
thaw cycles and heated with stirring at 80−85 °C for 18 h. The
resulting dark brown suspension was diluted with EtOAc
(50 mL), and the solids were removed by filtration. The filtrate
was washed with satd NH4Cl (2 × 25 mL) and water (2 ×
25 mL) and then dried (MgSO4), filtered, and concentrated.
Purification by flash chromatography (19:1 hexanes/CH2Cl2
then 7:3 hexanes/EtOAc) gave 1a(mP)2 as a brown oil
(870 mg, 99%): 1H NMR (300 MHz, CDCl3) δ 1.13 (s, 42H),
1.25 (s, 18H), 3.93 (s, 6H), 7.04 (s, 2H), 7.42 (s, 2H), 7.50
(s, 4H); 13C NMR (75 MHz, CDCl3) δ 11.3, 18.7, 31.0, 34.5,
56.0, 88.1, 90.4, 92.0, 106.8, 114.0, 118.8, 123.1, 123.5, 128.95,
129.00, 131.9, 149.1, 151.5; MALDI-TOF-MS (dithranol) calcd
for C54H74CuO2Si2 (M + Cu+) 873.45, found 873.42.
Trimer 2a(mP)2. To a solution of 1a(mP)2 (699 mg, 0.862 mmol)
in THF (6.5 mL) was added dropwise a 1.0 M solution of TBAF in
THF (550 μL, 0.55 mmol).124 The reaction mixture was stirred at
room temperature for 2 h under an argon atmosphere. The result-
ing solution was diluted with EtOAc (30 mL), washed with brine
(30 mL) and water (2 × 15 mL), dried (MgSO4), filtered, and
concentrated. Purification by flash chromatography (9:1 hexanes/
EtOAc) gave 2a(mP)2 as a yellow solid (226 mg, 52%): mp 166−
167 °C; 1H NMR (300 MHz, CDCl3) δ 1.28 (s, 18H), 3.04
(s, 2H), 3.93 (s, 6H), 7.04 (s, 2H), 7.47 (s, 2H), 7.52 (s, 2H), 7.55
(s, 2H); 13C NMR (75 MHz, CDCl3) δ 31.1, 34.6, 56.0, 77.1, 83.4,
88.3, 91.9, 114.0, 118.8, 122.2, 123.3, 129.2, 129.2, 132.0, 149.2,
151.9; MALDI-TOF-MS (dithranol) calcd for C36H34O2 498.26,
found 498.14.
Trimer 2b(mP)2. To a solution of 1b(mP)2 (555 mg, 0.613 mmol)
in THF (5 mL) was added dropwise a 1.0 M solution of TBAF in
THF (390 μL, 0.39 mmol). The solution was stirred at room
temperature for 2 h under an argon atmosphere. The resulting solution
was diluted with EtOAc (30 mL), washed with water (2 × 15 mL),
dried (MgSO4), filtered, and concentrated. Purification by flash
chromatography (1:1 hexanes:CH2Cl2) gave 2b(mP)2 as a pale yellow
1
solid (171 mg, 48%): mp 162.5 °C; H NMR (300 MHz, CDCl3) δ
0.87 (t, J = 7.0 Hz, 3H), 1.29 (br s, 24H), 1.67 (t, J = 6.1 Hz, 2H), 3.07
(s, 2H), 3.66 (t, J = 7.2 Hz, 2H), 7.53 (s, 4H), 7.56 (s, 2H), 7.97 (s,
2H); 13C NMR (75 MHz, CDCl3) δ 14.0, 22.5, 26.5, 28.5, 31.0, 31.3,
34.7, 38.4, 77.6, 83.1, 87.2, 97.5, 122.2, 122.5, 126.2, 129.4, 130.3,
130.8, 131.1, 132.4, 152.0, 167.2; MALDI-TOF-MS (HABA) calcd for
C42H41CuNO2 (M + Cu+) 654.24, found 654.12.
Macrocycle D(mP)2A. A Schlenk tube containing 2a(mP)2 (40 mg,
0.08 mmol), N-hexyl-4,5-diiodophthalimide (46.3 mg, 0.096 mmol),
and Pd(PtBu3)2 (4.0 mg, 8.0 μmol) was evacuated and backfilled with
argon (3×). To this solid mixture was added a mixture of NEt3
(2.4 mL) and toluene (5.6 mL). The reaction mixture was degassed
by three freeze−pump−thaw cycles and then stirred at room tem-
perature for 18 h. The resulting solution was diluted with toluene
(20 mL), washed with water (15 mL), dried (MgSO4), filtered, and
concentrated. Purification by flash chromatography (toluene) gave
D(mP)2A as a yellow solid (40.4 mg, 70%): mp >300 °C
1
(chloroform/hexanes); H NMR (300 MHz, CDCl3) δ 0.87 (t, J =
6.9 Hz, 3H), 1.28−1.30 (m, 6H), 1.37 (s, 18H), 1.50−1.65 (m, 2H),
3.57 (t, J = 7.3 Hz, 2H), 3.95 (s, 6H), 7.02 (s, 2H), 7.50 (t, J = 1.7 Hz,
2H), 7.58 (t, J = 1.7 Hz, 2H), 7.78 (t, J = 1.6 Hz, 2H), 7.88 (s, 2H);
13C NMR (75 MHz, CDCl3) δ 14.0, 22.5, 26.5, 28.5, 31.1, 31.3, 34.8,
38.2, 56.0, 87.6, 88.9, 91.8, 97.5, 114.0, 118.8, 122.4, 123.6, 126.2,
128.6, 129.2, 130.5, 131.0, 132.9, 149.2, 151.9, 167.2; MALDI-TOF-
MS (dithranol) calcd for C50H47NO4 (M+) 725.35, found 725.33;
HRMS (ESI) calcd for C50H47NNaO4 (M + Na+) 748.3403, found
748.3416.
N-Hexyl-4,5-diiodophthalimide. To a round-bottom flask contain-
ing 4,5-diiodophthalic acid125 (3.99 g, 9.54 mmol) was added SOCl2
(5.9 mL, 0.0805 mol) at 60 °C and then the suspension was heated at
90 °C for 16 h. The excess SOCl2 was removed under reduced
pressure, and the crude yellow solid was dried under vacuum to give
the 4,5-diiodophthalic anhydride which was used without further
purification: 1H NMR (300 MHz, DMSO-d6) δ 8.55 (s, 2H). A
suspension of 4,5-diiodophthalic anhydride (3.81 g) and hexylamine
(1.25 mL, 9.53 mmol) in toluene (80 mL) was heated at reflux for 18
h and then concentrated. Purification by flash chromatography (4:1
CHCl3/hexanes, loading the crude material as a solution in a minimum
volume of hot CHCl3) gave N-hexyl-4,5-diiodophthalimide as a pale
yellow solid (3.88 g, 8.03 mmol, 84% over two steps): mp 133−134
Macrocycle D(mP)2D. A Schlenk tube containing 2a(mP)2 (40 mg,
0.08 mmol), 4,5-diiodoveratrole (37.4 mg, 0.096 mmol), and
Pd(PtBu3)2 (4.0 mg, 8.0 μmol) was evacuated and backfilled with
argon (3×). To this solid mixture was added a solution of DABCO
(1.88 g) in toluene (8 mL). The reaction mixture was degassed by
three freeze−pump−thaw cycles and stirred at room temperature for
18 h. The resulting solution was diluted with toluene (20 mL), washed
with water (15 mL), dried (MgSO4), filtered, and concentrated.
Purification by flash chromatography (toluene) gave D(mP)2D as a
1
yellow solid (29.0 mg, 58%): mp >300 °C (chloroform/hexanes); H
NMR (300 MHz, CDCl3) δ 1.37 (s, 18H), 3.96 (s, 12H), 7.08 (s, 4H),
7.56 (d, J = 1.5 Hz, 4H), 7.85 (t, J = 1.6 Hz, 2H); 13C NMR (75 MHz,
CDCl3) δ 31.1, 34.7, 56.1, 88.6, 92.2, 114.7, 119.2, 123.6, 127.9, 132.6,
149.6, 151.7; MALDI-TOF-MS (dithranol) calcd for C44H40O4 (M+)
632.29, found 632.29; HRMS (ESI) calcd for C44H40NaO4 (M + Na+)
655.2824, found 655.2817.
1
°C; H NMR (300 MHz, CDCl3) δ 0.86 (t, J = 6.4 Hz, 3H), 1.29 (s,
6H), 1.63 (m, 2H), 3.64 (t, J = 7.4 Hz, 2H), 8.30 (s, 2H); 13C NMR
(75 MHz, CDCl3) δ 14.0, 22.5, 26.5, 28.4, 31.3, 38.4, 114.9, 132.2,
133.6, 166.6.
Trimer 1b(mP)2. A Schlenk tube containing Pd(OAc)2 (20.1 mg,
0.0895 mmol), CuI (12.3 mg, 0.0637 mmol), PPh3 (119 mg, 0.455
mmol), and N-hexyl-4,5-diiodophthalimide (521 mg, 1.08 mmol) was
evacuated and backfilled with argon (3×). To this solid mixture was
added a solution of 1-tert-butyl-3-ethynyl-5-[(triisopropylsilyl)-
ethynyl]benzene (799 mg, 2.36 mmol) in HNiPr2 (5 mL). The
suspension was degassed by three freeze−pump−thaw cycles and
heated with stirring at 75 °C for 18 h. The resulting suspension was
diluted with EtOAc (50 mL), and the solids were removed by
filtration. The filtrate was washed with satd NH4Cl (2 × 25 mL) and
water (2 × 25 mL), dried (MgSO4), filtered, and concentrated.
Purification by flash chromatography (4:1 hexanes/CH2Cl2 then 1:1
hexanes/CH2Cl2) gave 1b(mP)2 as a pale yellow solid (775 mg, 79%):
mp 63.3 °C; 1H NMR (300 MHz, CDCl3) δ 0.89 (t, J = 6.6 Hz, 3H),
1.13 (s, 42H), 1.26 (s, 18H), 1.32 (br s, 6H), 1.68 (m, 2H), 3.68 (t, J =
7.2 Hz, 2H), 7.47 (t, J = 1.8 Hz, 2H), 7.50 (t, J = 1.8 Hz, 2H), 7.51 (t,
J = 1.5 Hz, 2H), 7.99 (s, 2H); 13C NMR (75 MHz, CDCl3) δ 11.3,
14.0, 18.7, 22.5, 26.5, 28.5, 31.0, 31.3, 34.6, 38.4, 87.0, 91.0, 97.8,
106.4, 122.0, 123.8, 126.2, 129.1, 130.0, 130.8, 131.3, 132.2, 151.8,
167.2; MALDI-TOF-MS (TCNQ) calcd for C60H81CuNO2Si2 (M +
Cu+) 966.51, found 966.54.
Macrocycle A(mP)2A. A Schlenk tube containing 2b(mP)2 (40 mg,
0.068 mmol), N-hexyl-4,5-diiodophthalimide (39.1 mg, 0.081 mmol),
and Pd(PtBu3)2 (3.5 mg, 6.7 μmol) was evacuated and backfilled with
argon (3×). To this solid mixture was added a mixture of NEt3
(2.4 mL) and toluene (5.6 mL). The reaction mixture was degassed by
three freeze−pump−thaw cycles and then stirred at room temperature
for 18 h. The resulting suspension was diluted with toluene (20 mL),
washed with water (15 mL), dried (MgSO4), filtered, and
concentrated. Purification by flash chromatography (toluene) gave
A(mP)2A as a yellow solid (39.6 mg, 72%): mp >300 °C (chloroform/
1
hexanes); H NMR (300 MHz, CDCl3) δ 0.88 (t, J = 6.9 Hz, 6H),
1.31 (s, 12H), 1.39 (s, 18H), 1.65 (m, 4H), 3.60 (t, J = 6.5 Hz, 4H),
7.58 (s, 4H), 7.73 (s, 2H), 7.87 (s, 4H); 13C NMR (75 MHz, CDCl3)
δ 14.0, 22.5, 26.6, 28.5, 31.1, 31.3, 34.9, 38.3, 87.8, 97.1, 122.5, 126.2,
129.9, 130.7, 130.8, 133.2, 152.3, 167.0; MALDI-TOF-MS (dithranol)
calcd for C56H54N2O4 (M + H+) 818.42, found 819.39; HRMS (ESI)
calcd for C56H54N2NaO4 (M + Na+) 841.3981, found 841.3978.
Trimer 1a(Th)2. A Schlenk tube containing Pd(OAc)2 (14.6 mg,
0.0652 mmol), CuI (8.8 mg, 0.0463 mmol), PPh3 (86.5 mg, 0.330
mmol), and 4,5-diiodoveratrole (306 mg, 0.785 mmol) was evacuated
2293
dx.doi.org/10.1021/jo2026004 | J. Org. Chem. 2012, 77, 2285−2298

The Journal of Organic Chemistry
Article
and backfilled with argon (3×). To this solid mixture was added a clear
orange solution of 2-ethynyl-5-[2-(triisopropylsilyl)ethynyl]-
thiophene80,81 (498 mg, 1.73 mmol) in HNiPr2 (5 mL). The sus-
pension was degassed by three freeze−pump−thaw cycles and heated
with stirring at 75 °C for 18 h. The resulting mixture was diluted with
EtOAc (50 mL), and the solids were removed by filtration. The filtrate
was washed with satd NH4Cl (2 × 25 mL) and water (2 × 25 mL),
dried (MgSO4), filtered, and concentrated. Purification by flash
chromatography (19:1 hexanes/Et2O) gave 1a(Th)2 as a viscous pale
(300 MHz, CDCl3) δ 0.88 (t, J = 6.4 Hz, 3H), 1.31 (s, 6H), 1.66 (m,
2H), 3.67 (t, J = 7.3 Hz, 2H), 3.90 (s, 6H), 6.87 (s, 2H), 7.04−7.14
(m, 4H), 7.79 (s, 2H); 13C NMR (75 MHz, CDCl3) δ 14.0, 22.5, 26.5,
28.5, 31.3, 38.4, 56.1, 85.8, 91.3, 94.2, 96.5, 112.9, 119.7, 124.0, 124.9,
127.2, 131.0, 131.2, 132.2, 132.4, 149.6, 167.1; MALDI-TOF-MS
(dithranol) calcd for C38H27NO4S2 (M+) 625.14, found 625.16;
HRMS (ESI) calcd for C38H27NNaO4S2 (M + Na+) 648.1279, found
648.1289.
Macrocycle D(Th)2D. A Schlenk tube containing 2a(Th)2 (88 mg,
0.22 mmol), 4,5-diiodoveratrole (103 mg, 0.264 mmol), and
Pd(PtBu3)2 (11.2 mg, 0.022 mmol) was evacuated and backfilled
with argon (3×). To this solid mixture was added a solution of
DABCO (24.1 g) in toluene (100 mL). The suspension was degassed
by three freeze−pump−thaw cycles and then stirred at room
temperature for 18 h. The resulting solution was diluted with
CH2Cl2 (50 mL), washed with water (3 × 100 mL) and brine (3 × 50 mL),
and then concentrated. Purification by flash chromatography
(7:3 hexanes/EtOAc then 1:1 hexanes/CH2Cl2) gave D(Th)2D as a
pale brown solid (48 mg, 41%): mp 261.7 °C dec; 1H NMR
(300 MHz, CDCl3) δ 3.90 (s, 12H), 6.87 (s, 4H), 7.02 (s, 4H); 13C
NMR (125 MHz, CDCl3) δ 56.1, 86.1, 95.6, 112.9, 119.9, 125.6, 130.9,
149.5; MALDI-TOF-MS (dithranol) calcd for C32H20O4S2 (M+)
532.08, found 531.95; HRMS (ESI) calcd for C32H20NaO4S2 (M + Na+)
555.0701, found 555.0703. Prior to UV−vis and fluorescence spectros-
copy, the compound was further purified by flash chromatography.
Macrocycle A(Th)2A. A Schlenk tube containing 2b(Th)2 (40 mg,
0.0814 mmol), N-hexyl-4,5-diiodophthalimide (47.0 mg, 0.0976 mmol),
and Pd(PtBu3)2 (4.2 mg, 8.14 μmol) was evacuated and backfilled
with argon (3×). To this solid mixture was added a solution of
DABCO (1.91 g) in toluene (40 mL). The suspension was degassed
by three freeze−pump−thaw cycles and then stirred at room
temperature for 18 h. The resulting solution was diluted with
CH2Cl2 (30 mL), washed with water (3 × 10 mL) and brine (3 ×
10 mL), and then concentrated. Purification by flash chromatography
(2:3 hexanes/CH2Cl2 then CH2Cl2) gave A(Th)2A as a pale brown
solid (17.1 mg, 29%): mp 235.4 °C dec; 1H NMR (200 MHz, CDCl3)
δ 0.88 (m, 6H), 1.30 (br s, 12H), 1.65 (m, 4H), 3.68 (m, 4H), 7.18 (s,
4H), 7.82 (s, 4H); 13C NMR (125 MHz, CDCl3) δ 14.2, 22.6, 26.7,
28.6, 31.5, 38.6, 91.1, 95.0, 125.2, 125.6, 131.6, 132.2, 132.7, 167.1;
MALDI-TOF-MS (dithranol) calcd for C44H34N2O4S2 (M + H+)
719.20, found 719.02.126 Prior to UV−vis and fluorescence spectros-
copy, the compound was further purified by flash chromatography.
Synthesis of D(mP)A and D(Th)A (Scheme 2). Dimer 1a(mP).
A Schlenk tube containing Pd(PtBu3)2 (9.1 mg, 0.046 mmol)
and CuI (3.4 mg, 0.018 mmol) was evacuated and backfilled
with argon (3×). To this solid mixture was added a clear
solution of 4-bromoveratrole (66 μL, 0.46 mmol) and 1-tert-
butyl-3-ethynyl-5-[(triisopropylsilyl)ethynyl]benzene (327 mg,
0.966 mmol) in a mixture of NEt3 (2.4 mL) and toluene
(5.6 mL). The reaction mixture was degassed by three freeze−
pump−thaw cycles and stirred at room temperature for 18 h.
The mixture was diluted with EtOAc (10 mL), washed with
water (2 × 15 mL) and brine (2 × 15 mL), dried (MgSO4),
filtered, and concentrated. Purification by flash chromatography
(99:1 hexanes/EtOAc) gave a mixture of 1a(mP) with some
residual 4-bromoveratrole (210 mg) that was used without further
purification.
1
orange oil (497 mg, 89%): H NMR (300 MHz, CDCl3) δ 1.13 (s,
42H), 3.92 (s, 6H), 6.98 (s, 2H), 7.10−7.14 (m, 4H); 13C NMR (75
MHz, CDCl3) δ 11.3, 18.6, 56.0, 85.4, 92.6, 97.1, 98.9, 113.7, 118.2,
124.4, 124.9, 131.5, 132.4, 149.4; MALDI-TOF-MS (HABA) calcd for
C42H54CuO2S2Si2 (M + Cu+) 773.24, found 773.22.
Trimer 2a(Th)2. To a solution of 1a(Th)2 (497 mg, 0.699 mmol) in
THF (5 mL) was added dropwise a 1.0 M solution of TBAF in THF
(440 μL, 0.44 mmol). The solution was stirred at room temperature
for 2 h under an argon atmosphere. The resulting solution was diluted
with CH2Cl2 (30 mL), washed with water (2 × 15 mL), dried
(MgSO4), filtered, and concentrated. Purification by flash chromato-
graphy (1:1 toluene:CH2Cl2) gave 2a(Th)2 as a yellow solid (220 mg,
79%): mp 114.2 °C dec; 1H NMR (300 MHz, CDCl3) δ 3.40 (s, 2H),
3.92 (s, 6H), 6.98 (s, 2H), 7.13−7.19 (m, 4H); 13C NMR (75 MHz,
CDCl3) δ 56.1, 82.3, 85.2, 92.7, 113.7, 118.1, 123.3, 125.0, 131.4,
133.2, 149.5; MALDI-TOF-MS (TCNQ) calcd for C24H14O2S2 (M+)
398.04, found 397.93.
Trimer 1b(Th)2. A Schlenk tube containing Pd(OAc)2 (28 mg,
0.125 mmol), CuI (17 mg, 0.089 mmol), PPh3 (166 mg, 0.634
mmol), and N-hexylamine-4,5-diiodophthalimide (732 mg,
1.51 mmol) was evacuated and backfilled with argon (3×). To this solid
mixture was added a solution of 2-ethynyl-5-[2-(triisopropylsilyl)-
ethynyl]thiophene (958 mg, 3.33 mmol) in HNiPr2 (9.0 mL). The
suspension was degassed by three freeze−pump−thaw cycles and
heated with stirring at 75 °C for 18 h. The resulting mixture was
diluted with CH2Cl2 (50 mL), and the solids were removed by
filtration. The filtrate was washed with satd NH4Cl (2 × 25 mL) and
water (2 × 25 mL), dried (MgSO4), filtered, and concentrated.
Purification by flash chromatography (7:3 hexanes/CH2Cl2) gave
1
1b(Th)2 as bright yellow solid (1.14 g, 94%): H NMR (300 MHz,
CDCl3) δ 0.88 (t, J = 6.2 Hz, 3H), 1.13 (s, 42H), 1.31 (br s, 6H), 1.65
(m, 2H), 3.65 (t, J = 7.2 Hz, 2H), 7.13−7.24 (m, 4H), 7.89 (s, 2H);
13C NMR (75 MHz, CDCl3) δ 11.3, 14.0, 18.6, 22.5, 26.5, 28.5, 31.3,
38.4, 91.0, 91.5, 98.4, 98.5, 122.8, 125.9, 126.8, 130.3, 130.8, 132.5,
133.2, 167.0; MALDI-TOF-MS (HABA) calcd for C48H61CuNO2S2Si2
(M + Cu+) 866.30, found 866.30.
Trimer 2b(Th)2. To a solution of 1b(Th)2 (1.04 g, 1.3 mmol) in
THF (10 mL) was added dropwise a 1.0 M solution of TBAF in THF
(900 μL, 0.90 mmol). The solution was stirred at room temperature
for 2 h under an argon atmosphere. The resulting solution was diluted
with CH2Cl2 (40 mL) and washed with water (2 × 20 mL), then dried
(MgSO4), filtered, and concentrated. Purification by flash chromatog-
raphy (19:1 hexanes/EtOAc then 1:1 hexanes/EtOAc) then trituration
with hexanes gave 2b(Th)2 as a tan solid (306 mg, 48%): mp 118.3 °C
dec; 1H NMR (300 MHz, CDCl3) δ 0.88 (t, J = 6.8 Hz, 3H), 1.31 (br
s, 6H), 1.67 (m, 2H), 3.45 (s, 2H), 3.68 (t, J = 7.3 Hz, 2H), 7.21−7.26
(m, 4H), 7.95 (s, 2H); 13C NMR (75 MHz, CDCl3) δ 14.0, 22.5, 26.5,
28.5, 31.3, 38.4, 76.2, 83.3, 90.6, 91.7, 123.5, 125.1, 125.9, 130.1, 130.9,
133.0, 133.3, 167.0; MALDI-TOF-MS (dithranol) calcd for
C30H22NO2S2 (M + H+) 492.11, found 492.03.
Dimer 2a(mP). To a solution of 1a(mP) (210 mg) in THF (5 mL)
was added dropwise a solution of 1.0 M TBAF in THF (500 μL,
0.50 mmol). The solution was stirred at room temperature for 2 h.
The resulting solution was diluted with CH2Cl2 (20 mL), washed with
water (2 × 10 mL), dried (MgSO4), filtered, and concentrated.
Purification by flash chromatography (9:1 hexanes/EtOAc) gave a
mixture of 2a(mP) and 4-bromoveratrole (109 mg) that was used
without further purification.
N-Hexyl-4-bromophthalimide. A clear solution of 4-bromophthalic
anhydride (2.00 g, 8.81 mmol) and hexylamine (1.16 mL, 0.891 g, 8.81
mmol) in toluene (50 mL) was heated at reflux for 18 h. The reaction
mixture was concentrated under reduced pressure to give N-hexyl-4-
bromophthalimide as a white solid (2.71 g, 8.73 mmol, 99%): mp
Macrocycle D(Th)2A. A Schlenk tube containing 2a(Th)2 (40 mg,
0.1 mmol), N-hexyl-4,5-diiodophthalimide (53.1 mg, 0.11 mmol), and
Pd(PtBu3)2 (5.1 mg, 0.01 mmol) was evacuated and backfilled with
argon (3×). To this solid mixture was added a solution of DABCO
(12.10 g) in toluene (50 mL). The suspension was degassed by three
freeze−pump−thaw cycles and then stirred at room temperature for
18 h. The resulting mixture was diluted with toluene (30 mL), washed
with water (3 × 50 mL), then concentrated. Purification by flash
chromatography (1:1 toluene/CH2Cl2 then 1:9 toluene/CH2Cl2)
followed by trituration with Et2O gave D(Th)2A as a pale orange solid
1
(21 mg, 34%): mp 212.1 °C dec (chloroform/hexanes); H NMR
2294
dx.doi.org/10.1021/jo2026004 | J. Org. Chem. 2012, 77, 2285−2298

The Journal of Organic Chemistry
Article
a
Scheme 2
a
Reagents and conditions: (a) Pd(PtBu3)2, CuI, NEt3, toluene; (b) TBAF, THF; (c) Pd(PtBu3)2, NEt3, toluene; (d) Pd(OAc)2, CuI, PPh3, HNiPr2.
59−61 °C; 1H NMR (500 MHz, CDCl3) δ 0.88 (t, J = 7.1, 3H), 1.30
(br s, 6H), 1.65 (m, 2H), 3.68 (t, J = 7.3, 2H), 7.70 (d, J = 7.8 Hz,
1H), 7.85 (d, J = 7.3 Hz, 1H), 7.97 (s, 1H); 13C NMR (125 MHz,
CDCl3) δ 14.0, 22.5, 26.5, 28.5, 31.3, 38.3, 124.5, 126.6, 128.7, 130.7,
133.8, 136.8, 167.1, 167.6.
with 4-bromoveratrole (82 mg) that was used without further
purification.
Trimer D(Th)A. A Schlenk tube containing 2a(Th) (82.4 mg), N-
hexyl-4-bromophthalimide (124 mg, 0.4 mmol), and Pd(PtBu3)2 (6.0
mg, 0.012 mmol) was evacuated and backfilled with argon (3×). To
this solid mixture was added a mixture of NEt3 (2.4 mL) and toluene
(5.6 mL). The reaction mixture was degassed by three freeze−pump−
thaw cycles and stirred at room temperature for 18 h. The resulting
suspension was diluted with CH2Cl2 (30 mL), washed with water (2 ×
15 mL), dried (MgSO4), filtered, and concentrated. Purification by
flash chromatography (1:1 hexanes/CH2Cl2) gave D(Th)A as a bright
Trimer D(mP)A. A Schlenk tube containing N-hexyl-4-bromoph-
thalimide (138 mg, 0.45 mmol) and Pd(PtBu3)2 (8.7 mg, 0.017 mmol)
was evacuated and backfilled with argon (3×). To this solid mixture
was added a solution of 2a(mP) (109 mg) in NEt3 (2.4 mL) and
toluene (5.6 mL). The reaction mixture was degassed by three freeze−
pump−thaw cycles and stirred at room temperature for 18 h. The
resulting suspension was diluted with CH2Cl2, washed with water (2 ×
10 mL), dried (MgSO4), filtered, and concentrated. Purification by
flash chromatography (9:1 hexanes/EtOAc) gave D(mP)A as a yellow
solid (115 mg, 0.21 mmol, 46% over three steps): mp 99−101 °C
1
yellow solid (75 mg, 16% over three steps): mp 136−137 °C; H
NMR (500 MHz, CDCl3) δ 0.88 (t, J = 7.1 Hz, 3H), 1.31 (m, 6H),
1.67 (m, 2H), 3.67 (t, J = 7.3 Hz, 2H), 3.91 (s, 6H), 6.85 (d, J = 8.7
Hz, 1H), 7.02 (d, J = 1.8 Hz, 1H), 7.13−7.15 (m, 2H), 7.22 (d, J = 4.1
Hz, 1H), 7.80 (m, 2H), 7.93 (s, 1H); 13C NMR (125 MHz, CDCl3) δ
14.0, 22.5, 26.5, 28.5, 31.4, 38.3, 56.0, 80.7, 86.8, 92.4, 95.1, 111.1,
114.1, 114.5, 122.8, 123.2, 125.1, 125.8, 126.6, 128.8, 131.0, 131.6,
132.5, 133.1, 136.4, 148.8, 150.1, 167.7, 167.8; MALDI-TOF-MS
(dithranol) calcd for C30H27NO4S (M+) 497.16, found m/z = 496.79;
HRMS (ESI) calcd for C30H27NNaO4S (M + Na+) 520.1558, found
520.1569.
1
(hexanes): H NMR (300 MHz, CDCl3) δ 0.87 (t, J = 6.4 Hz, 3H),
1.31 (s, 6H), 1.35 (s, 9H), 1.67 (m, 2H), 3.67 (t, J = 7.3 Hz, 2H), 3.90
(d, J = 1.7 Hz, 6H), 6.85 (d, J = 8.3 Hz, 1H), 7.05 (d, J = 1.7 Hz, 1H),
7.13−7.17 (dd, J = 8.2, 1.8 Hz, 1H), 7.52 (t, J = 1.5 Hz, 1H), 7.54 (t,
J = 1.3 Hz, 1H), 7.56 (t, J = 1.6 Hz, 1H), 7.81 (s, 2H), 7.95 (t, J = 1.0
Hz, 1H); 13C NMR (75 MHz, CDCl3) δ 14.0, 22.5, 26.5, 28.5, 31.1,
31.4, 34.8, 38.2, 55.9, 87.4, 87.7, 93.5, 110.7, 111.1, 114.3, 122.1, 123.2,
123.7, 125.0, 126.0, 128.6, 129.3, 129.5, 131.0, 131.8, 136.7, 151.8,
167.8, 167.8; MALDI-TOF-MS (dithranol) calcd for C36H37NO4
(M+) 547.27, found 546.98; HRMS (ESI) calcd for C36H37NaNO4
(M + Na+) 570.2620, found 570.2623.
Dimer 1a(Th). A Schlenk tube containing Pd(OAc)2 (17.6 mg,
0.0784 mmol), CuI (10.5 mg, 0.055 mmol), and PPh3 (104 mg, 0.39
mmol) was evacuated and backfilled with argon (3×). To this solid
mixture was added a clear yellow solution of 4-bromoveratrole
(136 μL, 0.945 mmol) and 2-ethynyl-5-[2-(triisopropylsilyl)ethynyl]-
thiophene (300 mg, 1.04 mmol) in HNiPr2 (5 mL). The reaction
mixture was degassed by three freeze−pump−thaw cycles and heated
with stirring at 75 °C for 18 h. The resulting suspension was diluted
with CH2Cl2 (20 mL), washed with water (2 × 15 mL), dried
(MgSO4), filtered, and concentrated. Purification by flash chromatog-
raphy (1:1 hexanes/CH2Cl2) gave 1a(Th) as a mixture with 4-
bromoveratrole (253 mg) that was used without further purification.
Dimer 2a(Th). To a solution of 1a(Th) (126 mg) in THF (5 mL)
was added dropwise a 1.0 M solution of TBAF in THF (85.7 μL, 0.09
mmol). The solution was stirred at room temperature for 2 h under an
argon atmosphere. The resulting solution was diluted with CH2Cl2
(20 mL), washed with water (2 × 15 mL), dried (MgSO4), filtered,
and concentrated. Purification by flash chromatography (4:1
hexanes:CH2Cl2) gave a yellow solid that was further purified by
preparative thin-layer chromatography to give 2a(Th) as a mixture
UV−Vis and Fluorescence Spectroscopy. UV−vis and
fluorescence spectra were determined in spectrophotometric-grade
solvents used without further purification. UV−vis spectra were
recorded using 10 mm quartz cuvettes (except in quantum yield
determinations, for which 100 mm cuvettes were used). Emission
spectra are corrected.127 Quantum yields were determined according
to the established procedure for nitrogen-sparged solutions relative to
either quinine bisulfate in 0.5 M H2SO4 (Φf = 0.54) for emission
between 350 and 600 nm or rhodamine 101 in ethanol (Φf = 0.91) for
emission above 550 nm.93,94 The absorbance of the sample solutions
was kept below 0.10 (10 mm cuvette) to avoid the inner filter effect.
Measurements were performed at room temperature, with both
sample and reference solutions excited at the same wavelength
(350 nm) to avoid possible errors caused by neglecting the difference
between the light intensities at different excitation wavelengths. The
quinine bisulfate standard was cross-checked with 9,10-diphenylan-
thracene in cyclohexane (Φf = 0.91) and the rhodamine 101 with
cresyl violet in methanol (Φf = 0.54); in both cases the measured
quantum yields agree with literature values to within 10%.
Fluorescence lifetimes were measured on nitrogen-sparged solutions.
2295
dx.doi.org/10.1021/jo2026004 | J. Org. Chem. 2012, 77, 2285−2298

The Journal of Organic Chemistry
Article
(20) Pak, J. J.; Weakley, T. J. R.; Haley, M. M. J. Am. Chem. Soc. 1999,
121, 8182−8192.
ASSOCIATED CONTENT
■
S
* Supporting Information
(21) Marsden, J. A.; Miller, J. J.; Shirtcliff, L. D.; Haley, M. M. J. Am.
Chem. Soc. 2005, 127, 2464−2476.
Solvatochromism of UV−vis spectra; fluorescence spectra for
symmetrical compounds; Lippert−Mataga plots in cyclo-
hexane/CH2Cl2 mixtures; TCSPC measurements; optimized
geometries of symmetrical compounds; TD-DFT calculated
UV−vis spectra; energies, zero-point vibrational energy
corrections, and Cartesian coordinates of optimized geo-
metries; complete ref 95; NMR spectra of all isolated
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
(22) Spitler, E. L.; Monson, J. M.; Haley, M. M. J. Org. Chem. 2008,
73, 2211−2223.
(23) Zucchero, A. J.; McGrier, P. L.; Bunz, U. H. F. Acc. Chem. Res.
2010, 43, 397−408.
(24) McGrier, P. L.; Solntsev, K. M.; Zucchero, A. J.; Miranda, O. R.;
Rotello, V. M.; Tolbert, L. M.; Bunz, U. H. F. Chem.Eur. J. 2011, 17,
3112−3119.
(25) For another example of a push−pull macrocycle, see: Wu, Y.-L.;
Bures, F.; Jarowski, P. D.; Schweizer, W. B.; Boudon, C.; Gisselbrecht,
̌
J.-P.; Diederich, F. Chem.Eur. J. 2010, 16, 9592−9605.
(26) Baxter, P. N. W. J. Org. Chem. 2004, 69, 1813−1821.
(27) Spitler, E. L.; McClintock, S. P.; Haley, M. M. J. Org. Chem.
2007, 72, 6692−6699.
(28) Meier, H. Angew. Chem., Int. Ed. 2005, 44, 2482−2506.
(29) Shetty, A. S.; Zhang, J.; Moore, J. S. J. Am. Chem. Soc. 1996, 118,
1019−1027.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: scott.hartley@muohio.edu.
Notes
The authors declare no competing financial interest.
(30) Tobe, Y.; Utsumi, N.; Kawabata, K.; Nagano, A.; Adachi, K.;
Araki, S.; Sonoda, M.; Hirose, K.; Naemura, K. J. Am. Chem. Soc. 2002,
124, 5350−5364.
ACKNOWLEDGMENTS
■
This work was supported by the Air Force Office of Scientific
Research (FA9550-10-1-0377) and Miami University. The
MALDI mass spectrometer was purchased with funds from the
National Science Foundation (CHE-0839233). We thank Profs.
David Tierney and Michael Novak for critical readings of the
manuscript and Dr. Jian He for experimental assistance.
(31) Gallant, A. J.; MacLachlan, M. J. Angew. Chem., Int. Ed. 2003, 42,
5307−5310.
(32) Zhang, J.; Moore, J. S. J. Am. Chem. Soc. 1994, 116, 2655−2656.
(33) Fischer, M.; Lieser, G.; Rapp, A.; Schnell, I.; Mamdouh, W.; De
Feyter, S.; De Schryver, F. C.; Hoger, S. J. Am. Chem. Soc. 2004, 126,
̈
214−222.
(34) Seo, S. H.; Jones, T. V.; Seyler, H.; Peters, J. O.; Kim, T. H.;
Chang, J. Y.; Tew, G. N. J. Am. Chem. Soc. 2006, 128, 9264−9265.
(35) Hasegawa, M.; Enozawa, H.; Kawabata, Y.; Iyoda, M. J. Am.
Chem. Soc. 2007, 129, 3072−3073.
REFERENCES
■
(1) Spitler, E. L.; Johnson, C. A.; Haley, M. M. Chem. Rev. 2006, 106,
5344−5386.
(36) Zang, L.; Che, Y.; Moore, J. S. Acc. Chem. Res. 2008, 41, 1596−
1608.
(37) Jin, Y.; Zhang, A.; Huang, Y.; Zhang, W. Chem. Commun. 2010,
(2) Opsitnick, E.; Lee, D. Chem.Eur. J. 2007, 13, 7040−7049.
(3) Diederich, F.; Kivala, M. Adv. Mater. 2010, 22, 803−812.
(4) Kehoe, J. M.; Kiley, J. H.; English, J. J.; Johnson, C. A.; Petersen,
R. C.; Haley, M. M. Org. Lett. 2000, 2, 969−972.
(5) Marsden, J. A.; Haley, M. M. J. Org. Chem. 2005, 70, 10213−
10226.
46, 8258−8260.
(38) Luo, J.; Yan, Q.; Zhou, Y.; Li, T.; Zhu, N.; Bai, C.; Cao, Y.;
Wang, J.; Pei, J.; Zhao, D. Chem. Commun. 2010, 46, 5725−5727.
(39) Liljeroth, P.; Repp, J.; Meyer, G. Science 2007, 317, 1203−1206.
(40) A cross-conjugated π-system consists of “three unsaturated
groups, two of which although conjugated to a third unsaturated
center are not conjugated to each other”. See: Phelan, N. F.; Orchin, M.
J. Chem. Educ. 1968, 45, 633−637.
(6) Takeda, T.; Fix, A. G.; Haley, M. M. Org. Lett. 2010, 12, 3824−
3827.
(7) Hilger, A.; Gisselbrecht, J.-P.; Tykwinski, R. R.; Boudon, C.;
Schreiber, M.; Martin, R. E.; Luthi, H. P.; Gross, M.; Diederich, F.
̈
J. Am. Chem. Soc. 1997, 119, 2069−2078.
(41) Patoux, C.; Coudret, C.; Launay, J.-P.; Joachim, C.; Gourdon, A.
Inorg. Chem. 1997, 36, 5037−5049.
(8) Mossinger, D.; Hornung, J.; Lei, S.; De Feyter, S.; Hoger, S.
̈
̈
Angew. Chem., Int. Ed. 2007, 46, 6802−6806.
(42) Cardamone, D. M.; Stafford, C. A.; Mazumdar, S. Nano Lett.
2006, 6, 2422−2426.
(9) Hoger, S. Pure Appl. Chem. 2010, 82, 821−830.
̈
(10) Schmaltz, B.; Weil, T.; Mullen, K. Adv. Mater. 2009, 21, 1067−
̈
(43) Solomon, G. C.; Andrews, D. Q.; Van Duyne, R. P.; Ratner, M. A.
J. Am. Chem. Soc. 2008, 130, 7788−7789.
1078.
(11) Huang, H.-H.; Prabhakar, C.; Tang, K.-C.; Chou, P.-T.; Huang,
G.-J.; Yang, J.-S. J. Am. Chem. Soc. 2011, 133, 8028−8039.
(44) Andrews, D. Q.; Solomon, G. C.; Van Duyne, R. P.; Ratner, M. A.
J. Am. Chem. Soc. 2008, 130, 17309−17319.
(12) Wettach, H.; Hoger, S.; Chaudhuri, D.; Lupton, J. M.; Liu, F.;
̈
(45) Solomon, G. C.; Andrews, D. Q.; Goldsmith, R. H.; Hansen, T.;
Wasielewski, M. R.; Van Duyne, R. P.; Ratner, M. A. J. Am. Chem. Soc.
2008, 130, 17301−17308.
Lupton, E. M.; Tretiak, S.; Wang, G.; Li, M.; De Feyter, S.; Fischer, S.;
Forster, S. J. Mater. Chem. 2011, 21, 1404−1415.
(13) Bhaskar, A.; Guda, R.; Haley, M. M.; Goodson, T. III. J. Am.
Chem. Soc. 2006, 128, 13972−13973.
(14) Kobayashi, S.; Yamaguchi, Y.; Wakamiya, T.; Matsubara, Y.;
Sugimoto, K.; Yoshida, Z.-i. Tetrahedron Lett. 2003, 44, 1469−1472.
(15) Jiang, X.; Bollinger, J. C.; Lee, D. J. Am. Chem. Soc. 2006, 128,
11732−11733.
(16) Pan, G.-B.; Cheng, X.-H.; Hoger, S.; Freyland, W. J. Am. Chem.
Soc. 2006, 128, 4218−4219.
(17) Tahara, K.; Okuhata, S.; Adisoejoso, J.; Lei, S.; Fujita, T.; De
Feyter, S.; Tobe, Y. J. Am. Chem. Soc. 2009, 131, 17583−17590.
(18) Tahara, K.; Lei, S.; Adisoejoso, J.; De Feyter, S.; Tobe, Y. Chem.
Commun. 2010, 46, 8507.
̈
(46) Andrews, D. Q.; Solomon, G. C.; Goldsmith, R. H.; Hansen, T.;
Wasielewski, M. R.; Van Duyne, R. P.; Ratner, M. A. J. Phys. Chem. C
2008, 112, 16991−16998.
(47) Kocherzhenko, A. A.; Grozema, F. C.; Siebbeles, L. D. A. J. Phys.
Chem. C 2010, 114, 7973−7979.
(48) Gholami, M.; Tykwinski, R. R. Chem. Rev. 2006, 106, 4997−
5027.
̈
(49) Zhao, Y.; Tykwinski, R. R. J. Am. Chem. Soc. 1999, 121, 458−
459.
(50) Zhao, Y.; McDonald, R.; Tykwinski, R. R. J. Org. Chem. 2002,
67, 2805−2812.
(51) Zhao, Y.; Campbell, K.; Tykwinski, R. R. J. Org. Chem. 2002, 67,
336−344.
(19) Klare, J. E.; Tulevski, G. S.; Sugo, K.; de Picciotto, A.; White, K.
A.; Nuckolls, C. J. Am. Chem. Soc. 2003, 125, 6030−6031.
2296
dx.doi.org/10.1021/jo2026004 | J. Org. Chem. 2012, 77, 2285−2298

The Journal of Organic Chemistry
Article
(52) Gholami, M.; Melin, F.; McDonald, R.; Ferguson, M. J.;
Echegoyen, L.; Tykwinski, R. R. Angew. Chem., Int. Ed. 2007, 46,
9081−9085.
(83) Purification of D(mP)A was carried out by preparative thin-
layer chromatography, flash chromatography, recrystallization, and
semipreparative gel permeation chromatography.
(84) A reviewer suggested the possibility of a TICT (twisted
intramolecular charge transfer) state for D(mP)A. While we cannot
exclude the possibility, we believe that the similarity between its
charge-transfer fluorescence and that of D(mP)2A, which presumably
cannot form such a state, suggests that it is unlikely.
(53) Boldi, A. M.; Diederich, F. Angew. Chem., Int. Ed. Engl. 1994, 33,
468−471.
(54) Tykwinski, R. R.; Schreiber, M.; Gramlich, V.; Seiler, P.;
Diederich, F. Adv. Mater. 1996, 8, 226−231.
(55) Bosshard, C.; Spreiter, R.; Gunter, P.; Tykwinski, R. R.;
̈
Schreiber, M.; Diederich, F. Adv. Mater. 1996, 8, 231−234.
(85) Beens, H.; Knibbe, H.; Weller, A. J. Chem. Phys. 1967, 47,
(56) Nielsen, M. B.; Schreiber, M.; Baek, Y. G.; Seiler, P.; Lecomte,
S.; Boudon, C.; Tykwinski, R. R.; Gisselbrecht, J.-P.; Gramlich, V.;
1183−1184.
(86) Lippert, E.; Luder, W.; Boos, H. In Advances in Molecular
̈
Skinner, P. J.; Bosshard, C.; Gunter, P.; Gross, M.; Diederich, F.
̈
Spectroscopy; Mangini, A., Ed.; Pergamon Press: Oxford, 1962; p 443−
457.
Chem.Eur. J. 2001, 7, 3263−3280.
(57) Fielder, S.; Rowan, D. D.; Sherburn, M. S. Angew. Chem., Int. Ed.
2000, 39, 4331−4333.
(87) Sakurai, H.; Sugiyama, H.; Kira, M. J. Phys. Chem. 1990, 94,
1837−1843.
(88) Lakowicz, J. R. Principles of Fluorescence Spectroscopy, 3rd ed.;
Springer: New York, 2006.
(89) Dielectric constants and refractive indices for the mixtures were
estimated as the averages of those for the pure solvents, weighted
according to their volume fraction: Hirata, Y.; Mataga, N. J. Phys.
Chem. 1984, 88, 3091−3095.
(90) While we do not have a concrete explanation for the distinct
behavior of D(Th)2A, we speculate that the combination of linear
conjugation, a longer conjugation length, and structural rigidity lowers
the energy of its locally excited state compared to the other systems.
(91) Spartan ’08, v. 1.1.1; Wavefunction, Inc.: Irvine, CA, 2009.
(92) Estimation of a as, for example, 40% of the length of the long
molecular axes does not significantly affect the results.
(93) Eaton, D. F. Pure Appl. Chem. 1988, 60, 1107−1114.
(94) Rurack, K.; Spieles, M. Anal. Chem. 2011, 83, 1232−1242.
(95) Frisch, M. J. et al. Gaussian 09, Rev. B.01 ; Gaussian, Inc.:
Wallingford, CT, 2010.
(96) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5642.
(97) Adamo, C.; Barone, V. J. Chem. Phys. 1999, 110, 6158−6170.
(98) Yanai, T.; Tew, D. P.; Handy, N. C. Chem. Phys. Lett. 2004, 393,
51−57.
(99) Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105,
2999−3093.
(100) We have not yet grown high-quality single crystals of our
synthesized macrocycles.
(58) Payne, A. D.; Bojase, G.; Paddon-Row, M. N.; Sherburn, M. S.
Angew. Chem., Int. Ed. 2009, 48, 4836−4839.
(59) Swager, T. M.; Grubbs, R. H. J. Am. Chem. Soc. 1987, 109, 894−
896.
(60) Londergan, T. M.; You, Y.; Thompson, M. E.; Weber, W. P.
Macromolecules 1998, 31, 2784−2788.
(61) Lepetit, C.; Nielsen, M. B.; Diederich, F.; Chauvin, R. Chem.
Eur. J. 2003, 9, 5056−5066.
(62) Klokkenburg, M.; Lutz, M.; Spek, A. L.; van der Maas, J. H.; van
Walree, C. A. Chem.Eur. J. 2003, 9, 3544−3554.
(63) Tobe, Y.; Umeda, R.; Iwasa, N.; Sonoda, M. Chem.Eur. J.
2003, 9, 5549−5559.
(64) van Walree, C. A.; van der Wiel, B. C.; Jenneskens, L. W.; Lutz,
M.; Spek, A. L.; Havenith, R. W. A.; van Lenthe, J. H. Eur. J. Org.
Chem. 2007, 4746−4751.
(65) Thompson, A. L.; Ahn, T.-S.; Thomas, K. R. J.; Thayumanavan,
S.; Martínez, T. J.; Bardeen, C. J. J. Am. Chem. Soc. 2005, 127, 16348−
16349.
(66) van Walree, C. A.; Kaats-Richters, V. E. M.; Veen, S. J.;
Wieczorek, B.; van der Wiel, J. H.; van der Wiel, B. C. Eur. J. Org.
Chem. 2004, 3046−3056.
(67) van der Wiel, B. C.; Williams, R. M.; van Walree, C. A. Org.
Biomol. Chem. 2004, 2, 3432−3433.
(68) Ricks, A. B.; Solomon, G. C.; Colvin, M. T.; Scott, A. M.; Chen,
K.; Ratner, M. A.; Wasielewski, M. R. J. Am. Chem. Soc. 2010, 132,
15427−15434.
(101) We have been unable to obtain satisfactory electrochemical
data for comparison with the calculated orbital energies.
(102) Magyar, R. J.; Tretiak, S. J. Chem. Theory. Comput. 2007, 3,
976−987.
(69) Zhang, W.; Moore, J. S. Angew. Chem., Int. Ed. 2006, 45, 4416−
4439.
(70) Zhang, J.; Pesak, D. J.; Ludwick, J. L.; Moore, J. S. J. Am. Chem.
̀
(103) Jacquemin, D.; Perpete, E. A.; Scalmani, G.; Frisch, M. J.;
Soc. 1994, 116, 4227−4239.
̌
Kobayashi, R.; Adamo, C. J. Chem. Phys. 2007, 126, 144105.
(104) Peach, M. J. G.; Benfield, P.; Helgaker, T.; Tozer, D. J. J. Chem.
Phys. 2008, 128, 044118.
(71) Miljanic,
29−34.
́
O. S.; Vollhardt, K. P. C; Whitener, G. D. Synlett 2003,
(72) Zhang, W.; Brombosz, S. M.; Mendoza, J. L.; Moore, J. S. J. Org.
(105) In any event, other functionals, including B3LYP, PBE0, and
BH&HLYP, give similar results.
Chem. 2005, 70, 10198−10201.
̌
(73) Miljanic,
4001−4004.
́
O. S.; Holmes, D.; Vollhardt, K. P. C Org. Lett. 2005, 7,
(106) We stress that we are using “cross-conjugated” here exclusively
in the sense of quantum interference effects. The term is also used to
refer to linearly conjugated π-systems that are spatially crossed, as in
cruciform-type systems.
(107) As mentioned in the Results, we exclude the possibility of
excimer emission based on the negligible concentration dependence of
the UV−vis and fluorescence spectra of all studied compounds in all
solvents.
(108) 1.18 nm vs 1.22 Å for D(mP)2A′ and D(Th)2A′, respectively,
measured between the centroids of the opposing aromatic rings.
(109) Bixon, M.; Jortner, J.; Verhoeven, J. W. J. Am. Chem. Soc. 1994,
116, 7349−7355.
(110) Williams, R. M.; Koeberg, M.; Lawson, J. M.; An, Y.-Z.; Rubin,
Y.; Paddon-Row, M. N.; Verhoeven, J. W. J. Org. Chem. 1996, 61,
5055−5062.
(111) Mulliken, R. S. J. Am. Chem. Soc. 1952, 74, 811−824.
(112) Oevering, H.; Verhoeven, J. W.; Paddon-Row, M. N.; Warman,
J. M. Tetrahedron 1989, 45, 4751−4766.
(74) Kaneko, T.; Horie, T.; Matsumoto, S.; Terguchi, M.; Aoki, T.
Macromol. Chem. Phys. 2009, 210, 22−36.
(75) Tahara, K.; Yoshimura, T.; Ohno, M.; Sonoda, M.; Tobe, Y.
Chem. Lett. 2007, 36, 838−839.
(76) Hoger, S.; Meckenstock, A.-D.; Pellen, H. J. Org. Chem. 1997,
̈
62, 4556−4557.
(77) Tobe, Y.; Utsumi, N.; Nagano, A.; Sonoda, M.; Naemura, K.
Tetrahedron 2001, 57, 8075−8083.
(78) Soheili, A.; Albaneze-Walker, J.; Murry, J. A.; Dormer, P. G.;
Hughes, D. L. Org. Lett. 2003, 5, 4191−4194.
(79) For an example of a similar macrocyclic structure, see: Iyoda,
M.; Enozawa, H.; Miyake, Y. Chem. Lett. 2004, 33, 1098−1099.
(80) Nakano, Y.; Ishizuka, K.; Muraoka, K.; Ohtani, H.; Takayama,
Y.; Sato, F. Org. Lett. 2004, 6, 2373−2376.
(81) Roue,
́
S.; Lapinte, C. J. Organomet. Chem. 2005, 690, 594−604.
(82) We also did not observe any evidence for aggregation at higher
concentrations by NMR spectroscopy (in chloroform-d).
2297
dx.doi.org/10.1021/jo2026004 | J. Org. Chem. 2012, 77, 2285−2298

The Journal of Organic Chemistry
Article
(113) Gould, I. R.; Noukakis, D.; Gomez-Jahn, L.; Young, R. H.;
Goodman, J. L.; Farid, S. Chem. Phys. 1993, 176, 439−456.
(114) Ratner, M. A. J. Phys. Chem. 1990, 94, 4877−4883.
(115) Oliver, A. M.; Paddon-Row, M. N.; Kroon, J.; Verhoeven, J. W.
Chem. Phys. Lett. 1992, 191, 371−377.
(116) Verhoeven, J. W.; van Ramesdonk, H. J.; Groeneveld, M. M.;
Benniston, A. C.; Harriman, A. ChemPhysChem 2005, 6, 2251−2260.
(117) Paddon-Row, M. N. Aust. J. Chem. 2003, 56, 729−748.
(118) Haselbach, E.; Pilloud, D.; Suppan, P. Helv. Chim. Acta 1998,
81, 670−675.
(119) Zeng, Y.; Zimmt, M. B. J. Am. Chem. Soc. 1991, 113, 5107−
5109.
(120) Cooley, L. F.; Han, H.; Zimmt, M. B. J. Phys. Chem. A 2002,
106, 884−892.
(121) Gust, D.; Moore, T. A.; Moore, A. L. Acc. Chem. Res. 2001, 34,
40−48.
(122) Wasielewski, M. R. J. Org. Chem. 2006, 71, 5051−5066.
(123) Kovalenko, S. V.; Peabody, S.; Manoharan, M.; Clark, R. J.;
Alabugin, I. V. Org. Lett. 2004, 6, 2457−2460.
(124) We have found that the use of sub-stoichiometric quantities of
TBAF affords complete conversion of the starting material, as judged
by TLC and NMR. Use of 2 equiv did not improve the yield. For an
example of a TBAF-catalyzed desilylation, see: Wong, F. F.; Chuang, S.
H.; Yang, S.-c.; Lin, Y.-H.; Tseng, W.-C.; Lin, S.-K.; Huang, J.-J.
Tetrahedron 2010, 66, 4068−4072.
(125) Jiang, J. Y.; Kaafarani, B. R.; Neckers, D. C. J. Org. Chem. 2006,
71, 2155−2158.
(126) Unfortunately, we were unable to obtain an ESI mass spectrum
of this compound.
(127) Pfeifer, D.; Hoffmann, K.; Hoffmann, A.; Monte, C.; Resch-
Genger, U. J. Fluoresc. 2006, 16, 581−587.
2298
dx.doi.org/10.1021/jo2026004 | J. Org. Chem. 2012, 77, 2285−2298