Synthesis and spectroscopic properties of a soluble semiconducting porphyrin polymer

Article abstract of DOI:10.1039/c4cp02105c

A semiconducting porphyrin polymer that is solution processable and soluble in organic solvents has been synthesized, and its spectroscopic and electrochemical properties have been investigated. The polymer consists of diarylporphyrin units that are linked at meso-positions by aminophenyl groups, thus making the porphyrin rings an integral part of the polymer backbone. Hexyl chains on two of the aryl groups impart solubility. The porphyrin units interact only weakly in the ground electronic state. Excitation produces a local excited state that rapidly evolves into a state with charge-transfer character (CT) involving the amino nitrogen and the porphyrin macrocycle. Singlet excitation energy is transferred between porphyrin units in the chain with a time constant of ca. 210 ps. The final CT state has a lifetime of several nanoseconds, and the first oxidation of the polymer occurs at ca. 0.58 V vs. SCE. These properties make the polymer a suitable potential excited state electron donor to a variety of fullerenes or other acceptor species, suggesting that the polymer may find use in organic photovoltaics, sensors, and similar applications. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4cp02105c

PCCP
View Article Online
View Journal | View Issue
PAPER
Synthesis and spectroscopic properties of a
soluble semiconducting porphyrin polymer
Cite this: Phys.Chem.Chem.Phys.,
2014, 16, 17569
Robert A. Schmitz, Paul A. Liddell, Gerdenis Kodis, Michael J. Kenney,
Bradley J. Brennan, Nolan V. Oster, Thomas A. Moore,* Ana L. Moore* and
Devens Gust*
A semiconducting porphyrin polymer that is solution processable and soluble in organic solvents has been
synthesized, and its spectroscopic and electrochemical properties have been investigated. The polymer
consists of diarylporphyrin units that are linked at meso-positions by aminophenyl groups, thus making the
porphyrin rings an integral part of the polymer backbone. Hexyl chains on two of the aryl groups impart
solubility. The porphyrin units interact only weakly in the ground electronic state. Excitation produces a
local excited state that rapidly evolves into a state with charge-transfer character (CT) involving the amino
nitrogen and the porphyrin macrocycle. Singlet excitation energy is transferred between porphyrin units in
the chain with a time constant of ca. 210 ps. The final CT state has a lifetime of several nanoseconds, and
the first oxidation of the polymer occurs at ca. 0.58 V vs. SCE. These properties make the polymer a
suitable potential excited state electron donor to a variety of fullerenes or other acceptor species,
suggesting that the polymer may find use in organic photovoltaics, sensors, and similar applications.
Received 14th May 2014,
Accepted 4th July 2014
DOI: 10.1039/c4cp02105c
www.rsc.org/pccp
and polyporphyrin-fullerene dyads34 based on monomers featuring
both a free meso-position and a meso-aminophenyl group. Electro-
Introduction
Porphyrins and their tetrapyrrolic relatives are important com- polymerization generates semiconducting polymers in which por-
pounds because of their various roles in biology as light absorbers, phyrin macrocycles are joined by aminophenyl linkers reminiscent
redox centers and binding sites for small molecules, and because of of polyaniline. Although these and other electropolymerized
their applications (real and potential) in artificial photosynthesis, porphyrins are interesting and potentially useful materials, both
molecule-based (opto)electronics and data processing, sensing, their study and their applications are limited by the fact that they
medical imaging and treatment, and related areas. Many of these are bound as thin films to electrodes from which they cannot be
applications use both the light-absorbing and redox properties of removed, and are essentially insoluble in liquids.
porphyrins to convert light energy into electrochemical potential.
We now report the synthesis and spectroscopic properties
For actual device applications, the incorporation of porphyrins into of a structurally closely related porphyrin polymer which is
conducting or semiconducting media is especially convenient prepared by chemical reaction in solution; soluble in a variety
because it allows photoinduced charge separation involving a of organic solvents including tetrahydrofuran, chlorobenzene,
porphyrin followed by charge migration to an electrode and thence anisole, and benzonitrile; and solution processable via spin
into an electronic circuit. However, given the plethora of literature coating, drop casting or other methods. This polymer (P-(PN)n,
on porphyrin chemistry, relatively few conducting or semiconduc- Fig. 1) has the same polymeric backbone as the electropolymers
ting porphyrin polymers in which the porphyrin forms the polymer we reported earlier, and features 3,5-dihexyl substituted phenyl
backbone (rather than a side chain) have been reported. Much of groups at two meso-positions. These hexyl chains greatly
the research on conducting porphyrin polymers has been carried increase the solubility of the polymer relative to the electro-
out by electropolymerization of porphyrin monomers to give semi- chemically produced polymers, which had mesityl groups at
conducting films on electrodes,1–16 although some examples of these two meso-positions.
solution chemical methods for formation of long porphyrin oligo-
mers or polymers have been reported.17–31 We have reported the
electrochemical preparation and properties of polyporphyrins32,33
Results and discussion
Synthesis and characterization
Department of Chemistry and Biochemistry, Center for Bioenergy and
Zinc monomer PBr was synthesized from readily available
precursors by a series of reactions (see Experimental section),
Photosynthesis, Arizona State University, Tempe, AZ 85287-1604, USA.
E-mail: gust@asu.edu, tmoore@asu.edu, amoore@asu.edu
This journal is ©the Owner Societies 2014
Phys. Chem. Chem. Phys., 2014, 16, 17569--17579 | 17569

View Article Online
Paper
PCCP
Fraction B was less easily volatilized in the mass spectrometer
than A, and showed smaller signals for chains with 2, 3, 4 and
6 porphyrins, but significant peaks for 5-unit and longer chains.
Fraction C was dominated by high molecular weight polymer that
did not volatilize, and the signal to noise obtained was therefore
low. Chain lengths up to 13 were observed, but most of the
material likely consists of longer chains.
1
The polymer was also investigated by H-NMR spectroscopy.
Solutions of the polymer in deuterated chloroform were studied at
400 MHz. The spectra obtained were characterized by very broad
resonances due to the long chain lengths and possible aggrega-
tion of the chains. The broadness of the resonances also accen-
tuated resonances from minor impurities with sharp resonances,
and this complicated analysis. However, for samples of all three
types, resonances expected for the postulated structure were
observed, thus confirming the structure. An attempt was made
to estimate chain lengths by comparison of the integrals
of unique protons at the ends of the polymer chains (i.e. the
meso- and beta-protons of the terminal porphyrin which lacks a
meso-amino group, and the aryl and amino protons of the
terminal porphyrin at the other end of the chain) with those of
the aliphatic protons of the hexyl side chains of the internal
porphyrins. This method was only approximate, as overlapping
resonances and large differences in relaxation times are present,
Fig. 1 Structure and synthesis of porphyrin polymer P-(PN)n. Synthetic
details are given in the text.
and the polymer P-(PN)n was prepared by palladium-catalyzed but suggested that fraction A contained significant amounts of
coupling of PBr (Fig. 1). The catalyst was prepared in situ in chains with an average length of about 6 porphyrin units, whereas
tetrahydrofuran by mixing palladium(II) acetate and bis[(2- fractions B and C had average chain lengths of at least 20 units.
diphenylphosphino)phenyl] ether, and then adding an excess
Several model compounds were prepared, including a
of cesium carbonate. Polymerization ensued, leading to the model32 MP for the porphyrin P at the end of the chain, a
zinc form of the polymer. Treatment with trifluoroacetic acid in model MPN for the porphyrin at the amino end of the polymer
dichloromethane removed the zinc to yield free base P-(PN)n. chain, and a dimer MP-PN model for a two-porphyrin section of
Column chromatography was used to remove impurities and the polymer. These are shown in Fig. 2, and details of their
any remaining monomer. Subsequently, chromatography was preparation are reported in the Experimental section.
used to separate the polymer into three fractions, A, B, and C.
Spectroscopic properties
Details of a typical preparation of the polymer are given in the
Experimental section.
Monomers. The spectroscopic properties of model compounds
Based on the order of elution in column chromatography,
fractions A, B, and C were postulated to contain P-(PN)n of
increasing average molecular weight, respectively. Inductively
coupled plasma elemental analysis showed that there was no
detectable palladium in any of these fractions. Mass spectrometry
was consistent with this assumption. Analysis was performed
using MALDI-TOF instrumentation and a diphenylbutadiene
matrix. This technique was not quantitative because the higher
oligomers did not volatilize readily in the mass spectrometer.
Thus, the ratios of peak heights does not correspond to the
relative abundances of the various chains, but the technique
verified the presence of the expected polymeric material and
allowed some conclusions about average chain length to be
drawn. All three fractions showed that no monomer was present,
and chain lengths up to at least 13 units could be detected
unambiguously. Longer chains were present in these samples,
but their low volatility coupled with a tendency of the material to
associate with itself in the mass spectrometer prevented their
quantification. The mass spectrum of fraction A showed large
contributions of chains with 2–7 units plus longer chains.
were investigated in order to aid in the elucidation of the
Fig. 2 Structures of model monomeric and dimeric porphyrins.
17570 | Phys. Chem. Chem. Phys., 2014, 16, 17569--17579
This journal is © the Owner Societies 2014

View Article Online
PCCP
Paper
Fig. 3 Spectra of model compounds in 2-methyltetrahydrofuran.
Absorption spectra of MP (red), MPN (black), and MP-PN (blue) and
corrected emission spectra of MP (red dash), MPN (black dash) and MP-PN
(blue dash). All spectra have been normalized for ease in comparison.
Fig. 4 Fluorescence emission from a solution of MPN with excitation at
420 nm. The spectra are normalized at the emission maximum. Solvents
were cyclohexane (solid), toluene (dashed), 2-methyltetrahydrofuran
(dash-dot), 1-decanol (dash-dot-dot), and acetonitrile (dot).
spectroscopic properties of the polymer. The absorption spectra in only 7 nm. Clearly, more polar solvents stabilize the excited
2-methyltetrahydrofuran solution of model porphyrin MP, which state of MPN relative to the ground state. These data suggest
has only hydrogen at meso-position 15, and model MPN, which has that the excited state of MPN has significant charge-transfer
an aminotolyl group at the corresponding position, are shown in (CT) character, and the more polar solvents stabilize the charge-
Fig. 3. The spectrum of MP is typical of free base porphyrins with a transfer state. Presumably this state arises from increased
sharp Soret band at 413 nm and four Q-band maxima at 509, 544, electron donation from the amino group to the porphyrin
588 and 642 nm. The spectrum of MPN, on the other hand, has a macrocycle in the excited state, giving the amino group more
broad Soret band at 428 nm and broad Q-band absorption, with positive character.
maxima at 521, ca. 570, ca. 590, and 675 nm. Thus, the presence
Because the excited state behavior observed in MP and MPN
of the meso-amino group in MPN leads to significant distortion of is also expected to play a role in the photochemistry of the
the spectrum, relative to those of other free base tri- or tetra- polymer, we investigated the emission of the model compounds
arylporphyrins. Differences are also seen in the fluorescence as a function of time after excitation. Fluorescence decays were
emission spectra (Fig. 3). The spectrum of MP is similar to that measured using the single photon timing technique. The emis-
of other triarylporphyrins, with maxima at 649 and 713 nm. sion spectrum of MP in 2-methyltetrahydrofuran was measured
Molecule MPN, on other hand, shows an emission peak at at 8 wavelengths in the 620–760 nm region, and the spectra were
714 nm, with a shoulder at ca. 775 nm.
fitted globally to derive decay-associated spectra (DAS). The data
These results suggest that the excited state of MPN is signifi- were fitted well as a single exponential decay (w2 = 1.06) with a
cantly perturbed from that of other porphyrins due to the time constant of 9.53 ns. Such an excited state lifetime is typical
presence of the meso-amino group. We investigated the solvent for porphyrins of this general type.
dependence of the spectra in order to get more information
The decay-associated spectra from time-resolved fluorescence
about this perturbation. Five solvents were investigated: acetoni- experiments on MPN are shown in Fig. 5. The decays were
trile (e = 37.5), 1-decanol (e = 8.1), 2-methyltetrahydrofuran multiexponential in all of the solvents investigated. Two compo-
(e = 7.0), toluene (e = 2.4) and cyclohexane (e = 2.0). The nents were observed in cyclohexane, 2-methyltetrahydrofuran
absorption spectra of MP and MPN showed relatively small and acetonitrile, and four components were seen in 1-decanol.
solvent effects over this range of solvents. Porphyrin MP also In all solvents, a relatively long-lived component was observed.
showed only small shifts in emission. The shortest-wavelength The shape of this component is similar to that of the emission
emission band of MP is found at 647 nm in both cyclohexane spectra shown in Fig. 4: this is expected, as this component is
and acetonitrile. On the other hand, larger solvent effects were mainly responsible for the steady-state emission spectrum. The
observed for MPN in emission (Fig. 4). The main emission band lifetime ranges from 7.2 ns in cyclohexane, the least polar solvent,
appears at 696 nm in cyclohexane and 720 nm in acetonitrile.
It is clear from these studies that the energy of the relaxed emission to the charge-transfer excited state discussed above.
first excited singlet state of MPN has a significant dependence The shorter-lived components are assigned to local excited
to 4.9 ns in acetonitrile, the most polar solvent. We ascribe this
on solvent polarity, as illustrated by the emission results, (LE) states of MPN that evolve into the final charge-transfer
whereas that of MP shows only small effects. Upon increasing state with the indicated time constants. Consistent with this
the dielectric constant of the solvent from 2.0 to 37.5, the assignment is the fact that the emission maxima of these states
shortest wavelength emission band of MPN shifts by 24 nm are all found at slightly shorter wavelengths than those of the
to longer wavelengths, whereas there is no observed shift in CT states, indicating that these LE states are of higher energy
the corresponding emission of MP. The Stokes shift for MPN than the CT state. The results in decanol most clearly show that
in 2-methytetrahydrofuran is 38 nm, whereas that for MP is the decays of the emissions of these shorter-lived states
This journal is ©the Owner Societies 2014
Phys. Chem. Chem. Phys., 2014, 16, 17569--17579 | 17571

View Article Online
Paper
PCCP
Thus, it is a model for P-(PN)1. The absorption spectrum of
MP-PN (Fig. 3) features a Soret maximum at 410 nm with a
strong shoulder at ca. 425 nm. The Q-band region is broad, with
absorption maxima discernable at 512, ca. 550, 587, and
ca. 670 nm. The spectrum is essentially a linear combination
of the spectra of MP and MPN, with very minimal changes due
to interactions between the chromophores. The fluorescence
emission spectrum of MP-PN with excitation at 520 nm is very
similar to that of MPN (Fig. 3). There is only minimal emission
around 650 nm, where MP emits. Both porphyrins have some
absorbance at 520 nm, yet emission from PN dominates the
spectrum. This indicates that singlet excitation energy is
rapidly transferred from the excited singlet state of MP to PN.
This is expected because the linkage between the two chromo-
phores is short, and the emission of MP overlaps well with the
absorption of the longest-wavelength Q-band of PN. These
conditions favor rapid singlet-singlet energy transfer by the
35,36
¨
Forster-type (dipole–dipole) mechanism.
Thus, in dimer
MP-PN, the two chromophores appear to not interact strongly
in the ground state, and to exhibit efficient energy transfer in
the excited state.
Additional information concerning MP-PN comes from time-
resolved fluorescence studies. Fig. 6a shows the DAS for MP-PN
in 2-methyltetrahydrofuran with excitation at 520 nm. The DAS
features a long-lived component (6.0 ns) with a shape similar to
that of the CT emission from MPN which is assigned to decay of
the MP-1PN charge-transfer state. Additional decay components
of 540 ps and 36 ps have maxima at slightly shorter wavelengths
than does the CT state, and the shapes of the DAS prove
that these shorter components decay to form the final CT state.
Fig. 5 Decay-associated emission spectra in various solvents of MPN
with excitation at 520 nm. (a) Cyclohexane: 7.2 ns (circles), 190 ps
(squares); (b) 2-methyltetrahydrofuran: 5.7 ns (circles), 100 ps (squares);
(c) acetonitrile: 4.9 ns (circles), 120 ps (squares); (d) 1-decanol: 5.4 ns
(circles), 1.96 ns (squares), 380 ps (diamonds), 63 ps (triangles).
(positive amplitudes in Fig. 5d) have the same time constants as
a rise of emission amplitude in the region of CT emission
(negative amplitudes in Fig. 5d). This indicates that the LE
states decay to give the CT state. Decanol is a viscous solvent
that may retard intramolecular motions that accompany relaxa-
tion of the LE states to the CT state, and permit observation of
relaxations that are too fast to measure in the other solvents. It
is also possible that different hydrogen bonding interactions
between decanol and the meso-amino group of MPN play a role.
Dimer. We now turn to dimer MP-PN, which features a
porphyrin bearing a meso-amino group linked via this moiety
to a porphyrin with only a hydrogen atom at one meso position.
Fig. 6 Decay-associated emission spectra with excitation at 520 nm in
2-methyltetrahydrofuran solution. (a) MP-PN, 6.0 ns (circles), 540 ps
(squares), 36 ps (triangles); (b) P-(PN)n, 5.4 ns (circles), 1.1 ns (triangles),
200 ps (squares).
17572 | Phys. Chem. Chem. Phys., 2014, 16, 17569--17579
This journal is © the Owner Societies 2014

View Article Online
PCCP
Paper
The shape of the emission is similar to that of the emission of
MP-PN as well as that of MPN (Fig. 3), indicating that emission
arises entirely from the PN subunits. With identical absorbance
at the excitation wavelength, the emission of P-(PN)n is about
10 times weaker than that from MP-PN.
A time-resolved study of the emission of P-(PN)n was per-
formed in 2-methyltetrahydrofuran, with excitation at 520 nm.
As shown in Fig. 6b, the main emission component has a decay
time constant of 5.4 ns, which is only slightly shorter than that
for MP-PN. Additional decay components were observed, as was
the case with the dimer and MPN monomer. These had values of
200 ps and 1.1 ns. We assign these short components to LE
states that decay to give the longer-lived CT state, as occurs in the
monomer and dimer. We assume that the shorter-lived transient
states in P-(PN)n and MP-PN decay into the longest-lived emitting
states, as was seen in MP-PN. Without knowing the yields of the
conversion of the LE states into the CT state for each molecule,
we cannot determine whether a portion of the 10-fold quenching
in emission intensity in the polymer relative to MP-PN may be
Fig. 7 Absorption spectra of MP-PN (solid) and P-(PN)n (dash) in
2-methyltetrahydrofuran, absorption spectrum of P-(PN)n as a film on
glass (dash-dot-dot), and emission with excitation at 520 nm of MP-PN
(dash-dot) and P-(PN)n (dot). The absorption spectra have been normal-
ized at the Soret maximum. The emission spectra show emission intensity
ratios with excitation of solutions of equal absorbance at 520 nm.
We ascribe these components to LE states similar to those noted due to differences in the rates of processes that occur in both
above for MPN alone. It is tempting to ascribe one of these compounds, or to processes not present in the model dimer.
1
components instead to energy transfer from MP-PN. However,
Fluorescence anisotropy and energy transfer. In principle,
this could not be verified unambiguously. As mentioned earlier, singlet–singlet energy transfer between porphyrin units in the
MPN alone shows similar short components, especially in polymer is possible. We isolated, by careful and repeated chro-
1-decanol. In addition, a cyclic pentamer, PN5, has been matography, a fraction of polymer containing essentially pure
prepared (to be described elsewhere), and its DAS in various P-(PN)3 in order to investigate this phenomenon. Fluorescence
solvents show similar components. Energy transfer from a anisotropy decay measurements were carried out on MP-PN and
porphyrin having only a hydrogen atom on a meso-position is P-(PN)3 in 2-methyltetrahydrofuran solution at ambient tempera-
impossible in this pentamer due to the absence of such a tures (Fig. 8). The molecules were excited at 680 nm where the
1
porphyrin. Thus, energy transfer from MP-PN to yield MP-1PN porphyrins bearing meso-amino substituents absorb and the
definitely occurs in MP-PN, but its time scale cannot be unam- fluorescence anisotropy was measured at 780 nm, where
biguously deduced from our measurements. Ideally, excitation of
MP only with observation of emission from 1PN could allow
determination of such transfer, but as seen in Fig. 3, there is no
suitable wavelength for such an experiment.
Polymer. With the results for model compounds in hand, we
now examine the spectroscopic properties of the polymer
P-(PN)n. Experiments were performed on fraction B from the
synthesis discussed above. The absorption spectrum in
2-methyltetrahydrofuran of the polymer, which consists of a
range of chain lengths, is shown in Fig. 7, along with that of
dimer MP-PN for comparison. The Soret band appears at
415 nm in the polymer (vs. 410 for MP-PN), and is very broad,
with no sharp peak in the region where MP absorbs. This is due
to the higher ratio of PN to P in the oligomeric material.
Absorption in the Q-band region is broad, with maxima at
520, 575, and 664. In fact, the Q-band region appears very
similar to the corresponding region for MPN (Fig. 3). Thus, the
spectrum indicates limited ground-state interaction between
the porphyrin units.
The absorption spectrum of P-(PN)n when cast as a film on a
glass slide is also shown in Fig. 7. The spectrum has similar
features to those of the solution spectrum, but the peaks are
Fig. 8 Fluorescence anisotropy decays with excitation at 680 nm and
detection at 780 nm. (a) Dimer MP-PN. The white line is a best exponential
fit to the data with a time constant of 635 ps. (b) Tetramer P-(PN)3. The
broader and shifted to longer wavelengths.
Excitation of a solution of P-(PN)n in 2-methyltetrahydrofuran
at 520 nm yielded the emission spectrum shown in Fig. 7.
white line is a best exponential fit to the data with a time constants of
105 ps and 1.49 ns.
This journal is ©the Owner Societies 2014
Phys. Chem. Chem. Phys., 2014, 16, 17569--17579 | 17573

View Article Online
Paper
PCCP
emission is also from the PN components. Fluorescence aniso-
tropy will decay with time constants related to overall tumbling
of the molecule in solution and also energy transfer between
porphyrins, which results, in effect, in reorientation of the transition
dipole. The anisotropy decay for dimer MP-PN was well fitted by a
single exponential time constant of 635 ps. This is ascribed to
rotational tumbling of the molecule in solution. The decay for
tetramer P-(PN)3 required two exponential components with time
constants of 105 ps (49% of the decay) and 1.49 ns (51% of the
decay). The 1.49 ns component is due to overall tumbling of the
molecule, and is longer than the corresponding component in
MP-PN because of the larger size of the molecule. The 105 ps
component is assigned to energy transfer between adjacent meso-
aminoporphyrins, and corresponds to an approximate time
constant for singlet–singlet energy transfer of 210 ps. Although
two adjacent porphyrin moieties are relatively close together spa-
tially their excited states are isoenergetic, which results in a relatively
small spectral overlap integral in the Forster equation36 for singlet–
¨
singlet energy transfer and relatively slow energy transfer compared
to systems in which the spectral overlap integral is large (e.g., energy
transfer from zinc porphyrins to free base porphyrins).
Electrochemical properties
The cyclic voltammograms of the polymer and model com-
pounds were obtained in benzonitrile solution with tetra-n-
butylammonium hexafluorophosphate as supporting electrolyte.
As reported previously,32 monomer MP shows two irreversible
overlapping oxidation peaks in the region 0.9–1.1 V vs. SCE and a
third irreversible peak around 1.3 V. These potentials are similar
to the first and second oxidation potentials of 5,10,15,20-
tetraphenylporphyrin in benzonitrile (1.08 and 1.25 V)37 and
the redox potential of aniline in acetonitrile (0.90 V).38 The lack
of reversibility is consistent with formation of an electropoly-
merized film on the platinum electrode. Fig. 9a shows cyclic
voltammograms for dimer MP-PN in benzonitrile. The solubility
of the dimer and the polymer were low in this solvent, which
resulted in a large capacitive current contribution to the voltam-
mogram. The first potential sweep shows oxidations at 0.58 V
and 1.04 V that are essentially irreversible. The peak at 1.04 V is
Fig. 9 Cyclic voltammograms for (a) dimer MP-PN and (b) polymer
P-(PN)n in deaerated benzonitrile containing tetra-n-butylammonium
hexafluoro-phosphate. Voltage was swept at 100 mV sꢀ1
.
assigned to the first oxidation of the porphyrin with the free and 1.0 V vs. SCE are observed. The 0.6 V wave is ascribed to
meso-position by analogy to monomer MP and the second oxidation of porphyrin molecules within the polymer chain, and
oxidation of PN. The peak at 0.58 V is assigned to the first the 1.0 V wave is due to the second oxidation of these porphyrins
oxidation of porphyrin PN, which bears both a meso-amino and any contributions from the terminal porphyrins.
functionality and an aniline ring. This value is consistent with
redox potentials for other porphyrins with meso-amino substitu-
ents.39 As can be seen in Fig. 9a, the second and subsequent
voltage scans of MP-PN show irreversible behavior under these
Experimental section
Synthesis
conditions, and each sweep results in less current than the
previous scan. This is consistent with formation of a film of The preparation of MP was reported previously.32 The 1H NMR
some sort on the electrode, although MP-PN cannot form poly- spectra were recorded on a Varian Inova 400 or a Varian Inova
mer chains, as does MP.
500 spectrometer. Mass spectra were obtained on a matrix-
Turning now to polymer P-(PN)n, the voltammograms in assisted laser desorption/ionization time-of-flight spectrometer
Fig. 9b, it is clear that the behavior is generally similar to that (MALDI-TOF). Ultraviolet-visible ground state absorption spec-
observed for the dimer MP-PN, although the peaks are not well tra were measured on a Shimadzu UV2100U spectrometer.
resolved due to lack of solubility, the irreversible behavior and
3,5-Dihexylbenzaldehyde (1). To a flask containing 4.0 g
deposition of a film on the electrode. Oxidation waves at ca. 0.6 (13 mmol) of methyl-3,5-dihexylbenzoate40 and 80 mL of
17574 | Phys. Chem. Chem. Phys., 2014, 16, 17569--17579
This journal is © the Owner Societies 2014

View Article Online
PCCP
Paper
tetrahydrofuran (THF) was added 0.50 g (13 mmol) of lithium together with lesser amounts of the starting material and the
aluminum hydride (LAH) in small quantities until the ester had dibrominated porphyrin. The reaction mixture was concentrated
been reduced to the corresponding alcohol. The progress of the to a viscous oil by evaporation of the solvent at reduced pressure
reaction was followed by thin layer chromatography (THC) and this oil was chromatographed on silica gel (hexanes/dichloro-
1
(hexanes/10% ethyl acetate). The reaction mixture was cooled methane, 4 : 1 to 3 : 1) to give 1.17 g of 3 (68% yield). H NMR
in ice water and small lumps of ice were added to the reaction (400 MHz) d ꢀ2.97 (2H, s, –NH), 0.93 (12H, t, J = 7 Hz, –CH3),
flask to quench excess LAH. The resulting suspension was 1.36–1.44 (16H, m, –CH2–), 1.48–1.56 (8H, m, –CH2–), 1.83–1.91
filtered through celite and the residue was washed with dichloro- (8H, m, –CH2–), 2.89 (8H, t, J = 8 Hz, –CH2–), 7.43 (2H, s, Ar–H),
methane/10% methanol. The filtrate was evaporated to dryness, 7.85 (4H, s, Ar–H), 8.99 (2H, d, J = 4 Hz, b-H), 9.00 (2H, d, J = 4 Hz,
and the residue was redissolved in dichloromethane (100 mL) b-H), 9.28 (2H, d, J = 4 Hz, b-H), 9.73 (2H, d, J = 5 Hz, b-H), 10.16
and then washed with aqueous citric acid followed by aqueous (1H, s, meso-H); MALDI-TOF-MS m/z calcd for C56H69N4Br 876.5,
sodium bicarbonate. The organic layer was dried over anhydrous obsd 876.4; UV/vis (CH2Cl2) 416, 512, 547, 588, 645 (nm).
sodium sulfate and concentrated to a viscous oil by evaporation
5-(4-t-Butylphenylcarbamate)-10,20-bis-(3,5-dihexylphenyl)-
of the solvent at reduced pressure. This material was dissolved in porphyrin (4). To a heavy walled glass tube was added 1.00 g
dichloromethane (100 mL) and to the stirred solution was added (1.14 mmol) of 3, 3.64 g (11.4 mmol) of 4-(boc-amino)-
portions of activated manganese dioxide such that the alcohol benzeneboronic acid pinacol, 4.83 g (22.8 mmol) of tribasic
was converted to the corresponding aldehyde. The progress of potassium phosphate and 20 mL of THF. The suspension was
the reaction was followed by TLC (hexanes/10% ethyl acetate). flushed with a stream of argon gas for 10 min, 132 mg
Once the reaction was complete, the suspension was filtered (0.11 mmol) of tetrakis-(triphenylphosphine)palladium(0) was
through Celite and the residue was washed with dichloro- added and the argon flushing procedure was continued for an
methane/20% methanol solution (100 mL). The filtrate was additional 10 min. The tube was sealed with a Teflont screw
evaporated to dryness and the residue was chromatographed plug and warmed to 67 1C. After 17 h, the tube was cooled and
on silica gel (hexanes/5% ethyl acetate) to give 1 as a viscous oil TLC (hexanes/dichloromethane, 1 : 1) of the contents indicated
(2.92 g, 81% yield).1H NMR (400 MHz) d 0.88 (6H, t, J = 6 Hz, that all the starting material had been consumed. The reaction
–CH3), 1.31–1.38 (12H, m, –CH2–), 1.59–1.67 (4H, m, –CH2–), 2.65 mixture was filtered through Celite and the filtrate was concen-
(4H, t, J = 8 Hz, –CH2–), 7.26 (1H, s, Ar–H partially obscured by trated to dryness by evaporation of the solvent. The residue was
CDCl3), 7.51 (2H, s, Ar–H), 9.97 (1H, s, –CHO); MALDI-TOF-MS chromatographed on silica gel (hexanes/dichloromethane, 2 : 1
1
m/z calcd for C19H30O1 274.2, obsd 273.8.
to 1 : 1) to give 985 mg (87% yield) of 4. H NMR (400 MHz) d
5,15-Bis-(3,5-dihexylphenyl)porphyrin (2). To a flask containing ꢀ2.98 (2H, s, –NH), 0.92 (12H, t, J = 7 Hz, –CH3), 1.32–1.42 (16H,
3.30 g (22.6 mmol) of 2,20-dipyrromethane, 6.20 g (22.7 mmol) m, –CH2–), 1.46–1.56 (8H, m, –CH2–), 1.64 (9H, s, –CH3), 1.82–
of 3,5-dihexylbenzaldehyde (1) and 2.3 L of chloroform was added 1.88 (8H, m, –CH2–), 2.88 (8H, t, J = 8 Hz, –CH2–), 6.83 (1H, s,
1.72 mL of boron trifluoride diethyl etherate. Stirring under an –NH), 7.41 (2H, s, Ar–H), 7.75 (2H, d, J = 8 Hz, Ar–H), 7.82 (4H, s,
argon atmosphere was carried out in the dark for 30 min. A 5.14 g Ar–H), 8.13 (2H, d, J = 8 Hz, Ar–H), 8.89 (2H, d, J = 4 Hz, b-H), 8.93
portion of 2,3-dichloro-5,6-dicyanobenzoquinone (DDQ) was added (2H, d, J = 4 Hz, b-H), 9.04 (2H, d, J = 4 Hz, b-H), 9.32 (2H, d,
to the reaction mixture and stirring was continued for 1 h. The dark J = 4 Hz, b-H), 10.19 (1H, s, meso-H); MALDI-TOF-MS m/z calcd
solution was reduced in volume to approximately 1 L by distillation for C67H83N5O2 989.6, obsd 989.6; UV/vis (CH2Cl2) 414, 510, 545,
of the solvent at reduced pressure and then gently shaken with 584, 639 (nm).
aqueous sodium bicarbonate (1 L). Once the two layers had
5-Bromo-15-(4-tert-butylphenylcarbamate)-10,20-bis-(3,5-
separated the organic phase was washed a further three times with dihexylphenyl)porphyrin (5). To a 1 L flask containing 2.90 g
fresh bicarbonate solution. The organic layer was then concentrated (2.93 mmol) of 4 and 400 mL of chloroform was added 547 mg
to a viscous oil by distillation of the solvent at reduced pressure and (3.07 mmol) of N-bromosuccinimide. The solution was stirred
this oil was chromatographed on silica gel (hexanes/dichloro- for 20 min, after which time TLC (hexanes/dichloromethane,
methane, 5 : 1 to 3 : 1). The appropriate fractions were combined, 1 : 1) indicated that the reaction was complete. The reaction
the solvent was evaporated and the residue was recrystallized from mixture was washed with aqueous sodium bicarbonate, dried
dichloromethane/methanol to give 3.56 g of 2 (40% yield). 1H NMR over anhydrous sodium sulfate, and concentrated to dryness by
(400 MHz) d ꢀ3.08 (2H, s, –NH), 0.92–0.95 (12H, t, J = 7 Hz, –CH3), distillation of the solvent under reduced pressure. The resulting
1.37–1.45 (16H, m, –CH2–), 1.50–1.58 (8H, m, –CH2–), 1.92–1.85 (8H, material was chromatographed on silica gel (hexanes/dichloro-
1
m, –CH2–), 2.92 (8H, t, J = 8 Hz, –CH2–), 7.43 (2H, s, Ar–H), 7.91 (4H, methane, 5 : 2 to 3 : 2) to give 2.81 g (90% yield) of 5. H NMR
s, Ar–H), 9.12 (4H, d, J = 5 Hz, b-H), 9.38 (4H, d, J = 5 Hz, b-H), 10.30 (400 MHz) d –2.74 (2H, s, –NH), 0.91 (12H, t, J = 7 Hz, –CH3),
(2H, s, meso-H); MALDI-TOF-MS m/z calcd for C56H70N4 798.6, obsd 1.34–1.41 (16H, m, –CH2–), 1.46–1.52 (8H, m, –CH2–), 1.64 (9H, s,
798.5; UV/vis (CH2Cl2) 408, 503, 537, 576, 631 (nm).
–CH3), 1.81–1.88 (8H, m, –CH2–), 2.87 (8H, t, J = 8 Hz, –CH2–),
5-Bromo-10,20-bis-(3,5-dihexylphenyl)porphyrin (3). To a 6.82 (1H, s, –NH), 7.41 (2H, s, Ar–H), 7.75 (2H, d, J = 8 Hz, Ar–H),
flask containing 1.00 g (1.25 mmol) of 2 and 200 mL of chloro- 7.82 (4H, s, Ar–H), 8.10 (2H, d, J = 8 Hz, Ar–H), 8.82 (4H, s, b-H),
form was added 223 mg (1.25 mmol) of N-bromosuccinimide. 8.92 (2H, d, J = 4 Hz, b-H), 9.65 (2H, d, J = 5 Hz, b-H); MALDI-TOF-
After stirring the reaction mixture for 15 min, TLC (hexanes/ MS m/z calcd for C67H82N5O2Br 1067.6, obsd 1067.5; UV/vis
dichloromethane, 2 : 1) indicated that the product was present (CH2Cl2) 422, 520, 556, 597, 654 (nm).
This journal is ©the Owner Societies 2014
Phys. Chem. Chem. Phys., 2014, 16, 17569--17579 | 17575

View Article Online
Paper
PCCP
5-Bromo-15-(4-aminophenyl)-10,20-bis-(3,5-dihexyl-phenyl)- was concentrated by distillation of the solvent at reduced pres-
porphyrin (6). To a flask containing a solution of 1.5 g sure and dried under high vacuum. The residue was dissolved in
(1.40 mmol) of 5 dissolved in 40 mL of dichloromethane was a mixture of dichloromethane (200 mL) and trifluoroacetic acid
added 60 mL of trifluoroacetic acid. The green solution was (200 mL) and stirred under an argon atmosphere for 1 h.
stirred under an argon atmosphere for 20 min, whereupon TLC The reaction mixture was then diluted with dichloromethane
(hexanes/dichloromethane, 1 : 1) indicated that the reaction (200 mL) and washed with water followed by aqueous sodium
was complete. The reaction mixture was diluted with 200 mL of bicarbonate. The solvent was then evaporated and the residue
dichloromethane and washed with water (200 mL) several was chromatographed on silica gel (dichloromethane/2% THF,
times. The organic layer was then washed with aqueous dichloromethane/20% THF and finally dichloromethane/20%
sodium bicarbonate, dried over anhydrous sodium sulfate THF/10% methanol) to give three fractions A, B and C the weight
and then concentrated to dryness by evaporation of the of each being, 0.43 g, 0.33 g, and 0.21 g. Characterization was
solvent. The residue was chromatographed on silica gel performed as discussed in the Results and discussion section.
(dichloromethane/hexanes, 3 : 4 to 1 : 1) to give 1.24 g (91%
5-Bromo-15-(4-tert-butylphenylcarbamate)-10,20-bis(2,4,6-
yield) of 6. 1H NMR (400 MHz) d ꢀ2.71 (2H, s, –NH), 0.91 (12H, trimethylphenyl)porphyrin (7). To a flask containing a solution
t, J = 7 Hz, –CH3), 1.34–1.42 (16H, m, –CH2–), 1.46–1.52 (8H, m, of 159 mg (0.203 mmol) of 10-(4-tert-butylphenylcarbamate)-5,15-
–CH2–), 1.81–1.88 (8H, m, –CH2–), 2.86 (8H, t, J = 8 Hz, –CH2–), bis(2,4,6-trimethylphenyl)porphyrin (8)32 and 50 mL of chloro-
3.98 (2H, s, –NH), 7.02 (2H, d, J = 8 Hz, Ar–H), 7.40 (2H, s, form was added 38 mg (0.21 mmol) of N-bromosuccinimide. The
Ar–H), 7.82 (4H, s, Ar–H), 7.94 (2H, d, J = 8 Hz, Ar–H), 8.84 (2H, reaction mixture was stirred for 30 min, at which time the solvent
d, J = 5 Hz, b-H), 8.87 (2H, d, J = 5 Hz, b-H), 8.92 (2H, d, J = 5 Hz, was evaporated at reduced pressure and the residue was chroma-
b-H), 9.64 (2H, d, J = 5 Hz, b-H); MALDI-TOF-MS m/z calcd for tographed on silica gel (dichloromethane/hexanes, 1 : 1) to give
C
62H74N5Br 967.5, obsd 967.5; UV/vis (CH2Cl2) 423, 521, 559, 153 mg (92% yield) of 7. 1H NMR (300 MHz) d ꢀ2.59 (2H, s, N–H),
598, 656 (nm).
1.63 (9H, s, –CH3), 1.83 (12H, s, Ar–CH3), 2.63 (6H, s, Ar–CH3), 6.82
[5-Bromo-15-(4-aminophenyl)-10,20-bis-(3,5-dihexylphenyl)- (1H, s, N–H), 7.28 (4H, s, Ar–H), 7.74 (2H, d, J = 8 Hz, Ar–H), 8.10
porphyrino]zinc(II) (PBr). To flask containing 1.20 (2H, d, J = 8 Hz, Ar–H), 8.64 (2H, d, J = 5 Hz, b-H), 8.73 (2H, d,
a
g
(1.24 mmol) of 6 and 200 mL of dichloromethane was added J = 5 Hz, b-H), 8.78 (2H, d, J = 4 Hz, b-H), 9.59 (2H, d, J = 5 Hz, b-H);
50 mL of a saturated solution of zinc acetate dihydrate in MALDI-TOF-MS m/z calcd for C49H46N5O2Br 815.28, obsd 815.29;
methanol. After stirring for 1 h, TLC (hexanes/methylene Uv/vis (CH2Cl2) 421, 520, 553, 596, 653 (nm).
chloride, 1 : 1) indicated that all the starting material had been
Carbamate-protected form of MP-PN (9). To a glass tube was
consumed and a single product had formed. The pink reaction added 86 mg (0.14 mmol) of 10-(4-aminophenyl)-5,15-bis(2,4,6-
mixture was washed with water (200 mL) several times and trimethylphenyl)porphyrin,32 100 mg (0.122 mmol) of 7, 56 mg
then with aqueous sodium bicarbonate; it was then dried over (0.17 mmol) of cesium carbonate, 10 mg (0.018 mmol) of bis[(2-
anhydrous sodium sulfate. The solvent was evaporated at diphenylphosphino)phenyl] ether and 20 mL of THF. The sus-
reduced pressure and the remaining solid was dried under high pension was flushed with argon for 10 min, 3 mg (0.012 mmol)
1
vacuum to give 1.21 g (95% yield) of PBr. H NMR (400 MHz) of palladium acetate was added and the flushing procedure was
d 0.91 (12H, t, J = 7 Hz, –CH3), 1.32–1.41 (16H, m, –CH2–), 1.45– continued for an additional 5 min. The tube was sealed with a
1.51 (8H, m, –CH2–), 1.80–1.88 (8H, m, –CH2–), 2.86 (8H, t, Teflons screw plug and the reaction mixture was warmed to
J = 7 Hz, –CH2–), 3.82 (2H, s, –NH), 6.93 (2H, d, J = 8 Hz, Ar–H), 67 1C for 18 h. Once cool, the solvent was evaporated by
7.38 (2H, s, Ar–H), 7.82 (4H, s, Ar–H), 7.92 (2H, d, J = 8 Hz, Ar–H), distillation at reduced pressure and the residue was chromato-
8.86 (2H, d, J = 4 Hz, b-H), 8.90 (2H, d, J = 4 Hz, b-H), 8.96 (2H, d, graphed on silica gel (hexanes/15% ethyl acetate) to give 163 mg of
1
J = 5 Hz, b-H), 9.70 (2H, d, J = 5 Hz, b-H); MALDI-TOF-MS m/z 9 (97% yield). H NMR (300 MHz) d ꢀ2.88 (2H, s, N–H), ꢀ2.26
calcd for C62H72N5BrZn 1029.4, obsd 1029.4; UV/vis (CH2Cl2) (2H, s, N–H), 1.63 (9H, s, –CH3), 1.82 (12H, s, Ar–CH3), 1.89 (12H,
424, 517(sh), 553, 595 (nm).
s, Ar–CH3), 2.65 (12H, s, Ar–CH3), 6.82 (1H, s, N–H), 7.29 (4H, s,
Polymer P-(PN)n. To a 250 mL heavy walled glass flask was Ar–H), 7.30 (4H, s, Ar–H), 7.3 (2H, m, Ar–H partially obscured),
added 1.00 g (0.97 mmol) of PBr, 78 mg (0.15 mmol) of bis[(2- 7.73 (2H, d, J = 8 Hz, Ar–H), 8.02 (2H, d, J = 8 Hz, Ar–H), 8.05 (1H,
diphenylphosphino)phenyl] ether, 442 mg (1.36 mmol) of s, N–H), 8.11 (2H, d, J = 8 Hz, Ar–H), 8.65 (2H, d, J = 5 Hz, b-H),
cesium carbonate and 200 mL of THF. The suspension was 8.73–8.81 (8H, m, b-H), 9.03 (2H, d, J = 5 Hz, b-H), 9.23 (2H, d,
flushed with a stream of argon for 15 min, 21 mg (0.01 mmol) J = 5 Hz, b-H), 9.60 (2H, d, J = 5 Hz, b-H), 10.06 (1H, s, meso-H);
of palladium(II) acetate was added and the argon flushing MALDI-TOF-MS m/z calcd for C93H84N10O2 1372.68, obsd 1372.68;
process was continued for a further 10 min. The flask was sealed Uv/vis (CH2Cl2) 413, 426 (sh), 514, 551, 587, 647, 668 (nm).
with a Teflont screw plug and the reaction mixture was stirred at
Dyad MP-PN. Compound 9 (150 mg, 0.109 mmol) was added
67 1C for 42 h. A TLC on silica gel (dichloromethane/hexanes, to a flask along with 30 mL of trifluoroacetic acid. The green
1: 1) of the reaction mixture indicated that most if not all of the solution was stirred under a nitrogen atmosphere for 20 min
starting material had been consumed and that many other and then diluted with dichloromethane (150 mL). After
compounds (polymers of different chain lengths) had formed. washing with water and aqueous sodium bicarbonate, the
The reaction mixture was filtered through Celite and the residual solution was dried over sodium sulfate and then concentrated
material was thoroughly washed with THF. The combined filtrate by distillation of the solvent at reduced pressure. The residue was
17576 | Phys. Chem. Chem. Phys., 2014, 16, 17569--17579
This journal is © the Owner Societies 2014

View Article Online
PCCP
Paper
chromatographed on silica gel (dichloromethane/20–5% hexanes) a CHI 620 potentiostat (CH Instruments) using a Pt disk working
to give 125 mg of MP-PN (90% yield). 1H NMR (300 MHz) d ꢀ2.88 electrode, a Pt wire mesh counter electrode, and a silver wire quasi-
(2H, s, N–H), ꢀ2.25 (2H, s, N–H), 1.82 (12H, s, Ar–CH3), 1.89 (12H, reference electrode in a conventional three-electrode cell. Electro-
s, Ar–CH3), 2.64 (12H, s, Ar–CH3), 3.98 (2H, s, N–H), 7.03 (2H, d, chemical studies were carried out in anhydrous benzonitrile
J = 8 Hz, Ar–H), 7.29 (4H, s, Ar–H), 7.30 (4H, s, Ar–H), 7.3(2H, m, containing 0.10 M tetra-n-butylammonium hexafluorophosphate
Ar–H partially obscured), 7.96 (2H, d, J = 8 Hz, Ar–H), 8.02 (2H, d, as the supporting electrolyte. The working electrode was cleaned
J = 8 Hz, Ar–H), 8.02 (1H, s, N–H), 8.65 (2H, d, J = 4 Hz, b-H), 8.73– between experiments by polishing with diamond paste slurry,
8.84 (8H, m, b-H), 9.03 (2H, d, J = 5 Hz, b-H), 9.22 (2H, d, J = 4 Hz, followed by solvent rinses. After each voltammetric experiment,
b-H), 9.59 (2H, d, J = 5 Hz, b-H), 10.06 (1H, s, meso-H); MALDI- ferrocene was added to the solution, and the potential axis was
TOF-MS m/z calcd for C88H76N10 1272.63, obsd 1272.62; Uv/vis calibrated against the formal potential of the ferrocenium/ferrocene
(CH2Cl2) 412, 426 (sh), 514, 554, 587, 649, 662 (nm).
redox couple (taken as 0.45 V vs. SCE in dichloromethane).
Steady-state spectroscopy. Absorption spectra were measured
Carbamate protected form of MPN (10). To a heavy walled
glass tube was added 190 mg (0.23 mmol) of 7, 75 mg (0.70 mmol) on a Shimadzu UV-3101PC UV-vis-NIR spectrometer. Steady-state
of p-toluidine, 106 mg (0.33 mmol) of cesium carbonate, 19 mg fluorescence spectra were measured using a Photon Technology
(0.04 mmol) of bis[(2-diphenylphosphino)phenyl] ether and 60 mL International MP-1 spectrometer and corrected for detection
of THF. The mixture was flushed with argon gas for 15 min, 5.2 mg system response. Excitation was provided by a 75 W xenon-arc
(0.02 mmol) of palladium acetate was added and the argon gas lamp and single grating monochromator. Fluorescence was
flushing process was continued for an additional 10 min. The tube detected 901 to the excitation beam via a single grating mono-
was sealed with a Teflont screw plug and the reaction mixture was chromator and an R928 photomultiplier tube having S-20 spec-
warmed to 67 1C. After 17 h the reaction mixture was cooled and tral response and operating in the single photon counting mode.
TLC (hexanes/20% ethyl acetate) indicated that all the starting
Time-resolved fluorescence. Fluorescence decay kinetics
porphyrin had been consumed. The reaction mixture was were measured using the time-correlated single-photon counting
filtered through Celite and the residue was washed thoroughly technique. The excitation source was a fiber supercontinuum
with THF (100 mL). The filtrate was concentrated to dryness by laser based on a passive modelocked fiber laser and a high-
removal of the solvent at reduced pressure and the residue was nonlinearity photonic crystal fiber supercontinuum generator
chromatographed on silica gel (hexanes/5–7.5% ethyl acetate) to (Fianium SC450). The laser provides 6 ps pulses at a repetition
give 163 mg of 10 (83% yield). 1H NMR (400 MHz) d ꢀ2.31 (2H, s, rate variable between 0.1–40 MHz. The laser output was sent
N–H), 1.63 (9H, s, –CH3), 1.84 (12H, s, Ar–CH3), 2.26 (3H, s, through an Acousto-Optical Tunable Filter (Fianium AOTF) to
Ar–CH3), 2.62 (6H, s, Ar–CH3), 6.81 (1H, s, N–H), 6.87 (2H, d, obtain excitation pulses at the desired wavelength. Fluorescence
J = 8 Hz, Ar–H), 6.99 (2H, d, J = 8 Hz, Ar–H), 7.26 (4H, s, Ar–H emission was collected at 901 and detected using a double-
partially obscured), 7.63 (1H, s, N–H), 7.73 (2H, d, J = 8 Hz, Ar–H), grating monochromator (Jobin-Yvon, Gemini-180) and a micro-
8.09 (2H, d, J = 8 Hz, Ar–H), 8.59 (2H, d, J = 5 Hz, b-H), 8.60 (2H, channel plate photomultiplier tube (Hamamatsu R3809U-50).
d, J = 5 Hz, b-H), 8.74 (2H, d, J = 5 Hz, b-H), 9.27 (2H, d, J = 4 Hz, The polarization of the emission was 54.71 relative to that of the
b-H); MALDI-TOF-MS m/z calcd for C56H54N6O2 842.4, obsd excitation. Data acquisition was done using a single photon
842.4; UV/vis (CH2Cl2) 423, 520, 572, 590, 663 (nm).
counting card (Becker-Hickl, SPC-830). The instrument response
MPN. To a flask containing 150 mg (0.18 mmol) of 10 and function had a FWHM of 50 ps, as measured from the scattering
20 mL of dichloromethane was added 20 mL of trifluoroacetic of the sample at the excitation wavelength. The data were fitted
acid. The green solution was stirred at room temperature under with a sum of exponentials decay model at a single wavelength.
an argon atmosphere for 30 min. A TLC (hexanes/20% ethyl
acetate) indicated that the reaction was complete. The reaction
mixture was diluted with dichloromethane (100 mL) and
washed with water (2 ꢁ 100 mL) and then with aqueous sodium
Conclusions
bicarbonate (100 mL). The solution was dried over anhydrous The palladium catalyzed coupling method described here for
sodium sulfate and filtered, and the filtrate was concentrated to preparation of P-(PN)n yields a semiconducting polymer with the
a purple solid. This material was chromatographed on silica gel porphyrin moieties as part of the polymer backbone that is similar
(hexanes/25% ethyl acetate) to give 120 mg of MPN (91% yield). to the polymers previously prepared by electropolymerization.32,34
1H NMR (400 MHz) d ꢀ2.31 (2H, s, N–H), 1.84 (12H, s, Ar–CH3), Replacement of the mesityl groups used in the electropolymers with
2.26 (3H, s, Ar–CH3), 2.62 (6H, s, Ar–CH3), 4.01 (2H, s, N–H), 6.86 3,5-dihexylphenyl groups yields a soluble polymer that can be
(2H, d, J = 8 Hz, Ar–H), 6.99 (2H, d, J = 8 Hz, Ar–H), 7.04 (2H, d, studied easily in solution. It is amenable to spin-coating or other
J = 8 Hz, Ar–H), 7.24 (4H, s, Ar–H partially obscured), 7.61 (1H, s, film-forming procedures and should be useful for making polymer
N–H), 7.95 (2H, d, J = 8 Hz, Ar–H), 8.59 (2H, d, J = 5 Hz, b-H), 8.61 phases intimately mixed with fullerene phases for construction of
(2H, d, J = 5 Hz, b-H), 8.80 (2H, d, J = 5 Hz, b-H), 9.27 (2H, d, bulk heterojunction organic photoelectrochemical cells. Compari-
J = 5 Hz, b-H); MALDI-TOF-MS m/z calcd for C51H46N6 742.4, obsd son of the spectroscopic properties of the polymer with those of the
742.6; UV/vis (CH2Cl2) 427, 521, 570, 590, 665 (nm).
model compounds shows that linking the polymer rings via meso-
Electrochemical measurements. The voltammetric characteri- aminophenyl groups allows the porphyrin moieties to retain many
zation of the redox processes for the molecules was performed with of the absorption and redox properties of the monomer, rather than
This journal is ©the Owner Societies 2014
Phys. Chem. Chem. Phys., 2014, 16, 17569--17579 | 17577

View Article Online
Paper
PCCP
4 J. Basu and K. K. Rohatgimukherjee, Sol. Energy Mater.,
1991, 21, 317–325.
5 D.-S. Duanmu, Z.-P. Chen, X.-S. Yu and X. Zhou, Chin.
J. Chem., 2004, 22, 779–781.
6 D. Wohrle, R. Benters, O. Suvorova, G. Schnurpfeil,
N. Trombach and T. Bogdahn-Rai, J. Porphyrins Phthalocya-
nines, 2000, 4, 491–497.
7 G. Li, S. Bhosale, S. Tao, R. Guo, S. Bhosale, F. Li, T. Zhang,
T. Wang and J.-H. Furhop, Polymer, 2005, 46, 5299–5307.
8 C. Y. Lin, Y. C. Hung, C. M. Liu, C. F. Lo, Y. C. Lin and
C. L. Lin, Dalton Trans., 2005, 396–401.
9 K. A. Macor, Y. O. Su, L. A. Miller and T. G. Spiro, Inorg.
Chem., 1987, 26, 2594–2598.
10 A. Bettelheim, B. White, S. Raybuck and R. W. Murray, Inorg.
Chem., 1987, 26, 1009–1017.
11 A. Bettelheim, B. A. White and R. W. Murray, J. Electroanal.
Chem., 1987, 217, 271–286.
12 A. Bettelheim, D. Ozer, R. Harth and R. W. Murray,
J. Electroanal. Chem., 1989, 266, 93–108.
generating a highly delocalized chromophoric system. At the same
time, the close similarity in structure between P-(PN)n and the
electropolymers studied previously32,34 suggest that the polymer
chains form a semiconducting material through which positive
charge is expected to flow readily. The relatively long excited singlet
state lifetime observed for the polymer shows that this state is
kinetically competent to inject charge into a suitably located
electron acceptor. This in turn suggests that polymers of this type
could be useful in organic photovoltaics or light emitting diodes,
sensors, or other optoelectronic applications.
The spectroscopic studies of the polymer and model com-
pounds suggest that the presence of an amino group at the meso-
position of a porphyrin leads to formation of initial local excited
states that quickly evolve into excited states with charge-transfer
character. Presumably, the lone pair electrons of the amino
group are somewhat delocalized into the porphyrin macrocycle,
leaving a partial positive charge on the amino substituent and a
partial negative charge on the macrocycle. The formation of
charge transfer states of this general type in molecules with an
amino group attached to an aromatic residue with some electron
accepting character is widely observed and has been extensively
studied, although the exact nature of the CT state has been
greatly debated and is still not completely understood.41
13 B. A. White and R. W. Murray, J. Electroanal. Chem., 1985,
189, 345–352.
14 E. M. Bruti, M. Giannetto, G. Mori and R. Seeber, Electro-
analysis, 1999, 11, 565–572.
15 S. Griveau, V. Albin, T. Pauporte, J. H. Zagal and F. Bedioui,
J. Mater. Chem., 2002, 12, 225–232.
16 E. Mazzotta and C. Malitesta, Sens. Actuators, B, 2010, 148,
186–194.
17 J. S. Lindsey and D. F. Bocian, Acc. Chem. Res., 2011, 44,
638–650.
18 H. Zhan, S. Lamare, A. Ng, T. Kenny, H. Guernon,
W. K. Chan, A. B. Djurisic, P. D. Harvey and W. Y. Wong,
Macromolecules, 2011, 44, 5155–5167.
Energy transfer between porphyrin units along the polymer
chain is moderately rapid. The singlet state lifetimes suggest
that in a film structure, energy transfer could also occur
between adjacent polymer chains. The electrochemical experi-
ments show that the polymer chain is readily oxidized at ca.
+0.6 V vs. SCE, indicating that the excited states of the por-
phyrin moieties are thermodynamically competent to inject
charge into a wide variety of electron acceptors, including
typical fullerenes. Thus, porphyrins of this general type have
properties that could prove useful in a variety of optoelectronic,
sensing and solar energy conversion applications.
19 J. Y. Lee, H. J. Song, S. M. Lee, J. H. Lee and D. K. Moon, Eur.
Polym. J., 2011, 47, 1686–1693.
20 J.-H. Fuhrhop, Langmuir, 2014, 30, 1–12.
21 Z. M. Liu, I. Schmidt, P. Thamyongkit, R. S. Loewe,
D. Syomin, J. R. Diers, Q. Zhao, V. Misra, J. S. Lindsey and
D. F. Bocian, Chem. Mater., 2005, 17, 3728–3742.
22 S. Lamare, S. M. Aly, D. Fortin and P. D. Harvey, Chem.
Commun., 2011, 47, 10942–10944.
Acknowledgements
We thank Aurelie Marcotte for inductively coupled plasma
experiments. This work was supported by a grant from the
U.S. Department of Energy (DE-FG02-03ER15393). GK was
supported as part of the Center for Bio-Inspired Solar Fuel
Production, an Energy Frontier Research Center funded by the
U.S. Department of Energy, Office of Science, Office of Basic
Energy Sciences under Award Number DE-SC0001016. MJK
gratefully acknowledges support from a Goldwater Scholarship.
23 N. Aratani and A. Osuka, Macromol. Rapid Commun., 2001,
22, 725–740.
24 N. Aratani, A. Osuka, Y. H. Kim, D. H. Jeong and D. Kim,
Angew. Chem., Int. Ed., 2000, 39, 1458–1462.
25 H. L. Anderson, Chem. Commun., 1999, 2323–2330.
26 M. S. Choi, T. Yamazaki, I. Yamazaki and T. Aida, Angew.
Chem., Int. Ed., 2004, 43, 150–158.
27 V. S. Y. Lin, S. G. Dimagno and M. J. Therien, Science, 1994,
264, 1105–1111.
Notes and references
28 P. N. Taylor, J. Huuskonen, G. Rumbles, R. T. Aplin, E. Williams
and H. L. Anderson, Chem. Commun., 1998, 909–910.
29 H. L. Anderson, S. J. Martin and D. D. C. Bradley, Angew.
Chem., Int. Ed. Engl., 1994, 33, 655–657.
30 Y. Du, S. Zhang, X. Jiang, K. Zhu, Z. Geng, Y. Fang, P. Huo,
C. Liu, Y. Song and G. Wang, J. Polym. Sci., Part A: Polym.
Chem., 2014, 52, 1282–1290.
1 F. Bedioui, J. Devynck and C. Bied-Charreton, Acc. Chem.
Res., 1995, 28, 30–36.
2 J. R. Fish, E. Kubaszewski, A. Peat, T. Malinski, J. Kaczor,
P. Kus and L. Czuchajowski, Chem. Mater., 1992, 4, 795–803.
3 C. Poriel, Y. Ferrand, P. Le Maux, C. Paul-Roth,
G. Simonneaux and J. Rault-Berthelot, J. Electroanal. Chem.,
2005, 583, 92–103.
17578 | Phys. Chem. Chem. Phys., 2014, 16, 17569--17579
This journal is © the Owner Societies 2014

View Article Online
PCCP
Paper
¨
31 E. Maligaspe, T. Kumpulainen, N. K. Subbaiyan, 36 T. Forster, Discuss. Faraday Soc., 1959, 27, 7–17.
M. E. Zandler, H. Lemmetyinen, N. V. Tkachenko and 37 M. E. Jamin and R. T. Iwamoto, Inorg. Chim. Acta, 1978, 27,
F. D’Souza, Phys. Chem. Chem. Phys., 2010, 12, 7434–7444.
135–143.
32 P. A. Liddell, M. Gervaldo, J. W. Bridgewater, A. E. Keirstead, 38 R. L. Hand and R. F. Nelson, J. Am. Chem. Soc., 1974, 96,
S. Lin, T. A. Moore, A. L. Moore and D. Gust, Chem. Mater.,
2008, 20, 135–142.
33 B. J. Brennan, P. A. Liddell, T. A. Moore, A. L. Moore and
D. Gust, J. Phys. Chem. B, 2013, 117, 426–432.
850–860.
39 K. Kurotobi, Y. Toude, K. Kawamoto, Y. Fujimori, S. Ito,
P. Chabera, V. Sundstrom and H. Imahori, Chem. – Eur. J.,
2013, 19, 17075–17081.
34 M. Gervaldo, P. A. Liddell, G. Kodis, B. J. Brennan, 40 H. F. Chow, M. K. Ng, C. W. Leung and G. X. Wang, J. Am.
C. R. Johnson, J. W. Bridgewater, A. L. Moore, T. A. Moore
and D. Gust, Photochem. Photobiol. Sci., 2010, 9, 890–900.
Chem. Soc., 2004, 126, 12907–12915.
41 Z. R. Grabowski, K. Rotkiewicz and W. Rettig, Chem. Rev.,
2003, 103, 3899–4031.
¨
35 T. Forster, Ann. Phys., 1948, 2, 55–75.
This journal is ©the Owner Societies 2014
Phys. Chem. Chem. Phys., 2014, 16, 17569--17579 | 17579