Dicyano-substituted poly(phenylenevinylene) (DiCN-PPV) and the effect of cyano substitution on photochemical stability

Article abstract of DOI:10.1021/ma400661f

A didecyloxy-substituted poly(phenylenedicyanovinylene), DiCN-PPV, has been synthesized. The dicyano-substituted vinylene units exist in both trans and cis (～65:35) conformations as determined by ¹H NMR analysis, and cannot be converted to all trans due to the presence of a thermodynamic equilibrium of the two conformations, in contrast to the vinylene units in regular PPVs. The unusually high cis content makes this polymer highly amorphous, very soluble in organic solvent, and highly fluorescent in the solid state with an estimated quantum yield up to 0.34, four times more fluorescent than its chloroform solution. The LUMO and HOMO energies of the new polymer were measured by cyclovoltammetry. The cyano groups in DiCN-PPV brings a decrease in LUMO energy by 0.79 eV, and makes the polymer more stable to intense white light (>20 times as strong as the sunlight) than poly(2,5-didecyloxy-1,4- phenylenevinylene), C₁₀O-PPV, by more than 2 orders of magnitude. The excellent photochemical stability and high fluorescence quantum yield in the solid state make DiCN-PPV a good candidate for outdoor fluorescent applications such as remote optical sensing.

Full text of DOI:10.1021/ma400661f

Article
pubs.acs.org/Macromolecules
Dicyano-Substituted Poly(phenylenevinylene) (DiCN−PPV) and the
Effect of Cyano Substitution on Photochemical Stability
Jianyuan Sun,† Logan P. Sanow,† Sam-Shajing Sun,‡ and Cheng Zhang*,†
†Department of Chemistry and Biochemistry, South Dakota State University, Brookings, South Dakota 57007, United States
‡Center for Materials Research, Norfolk State University, Norfolk, Virginia 23504, United States
ABSTRACT: A didecyloxy-substituted poly(phenylenedicyanovinylene),
DiCN−PPV, has been synthesized. The dicyano-substituted vinylene units
1
exist in both trans and cis (∼65:35) conformations as determined by H
NMR analysis, and cannot be converted to all trans due to the presence of a
thermodynamic equilibrium of the two conformations, in contrast to the
vinylene units in regular PPVs. The unusually high cis content makes this
polymer highly amorphous, very soluble in organic solvent, and highly
fluorescent in the solid state with an estimated quantum yield up to 0.34,
four times more fluorescent than its chloroform solution. The LUMO and
HOMO energies of the new polymer were measured by cyclovoltammetry.
The cyano groups in DiCN−PPV brings a decrease in LUMO energy by
0.79 eV, and makes the polymer more stable to intense white light (>20 times as strong as the sunlight) than poly(2,5-
didecyloxy-1,4-phenylenevinylene), C10O−PPV, by more than 2 orders of magnitude. The excellent photochemical stability and
high fluorescence quantum yield in the solid state make DiCN−PPV a good candidate for outdoor fluorescent applications such
as remote optical sensing.
Poly(alkoxyphenylenevinylene)s and poly(2,5-dialkoxycya-
noterephthalylidene) have been investigated for light emission
application.20,21 Phenylenevinylene oligomers22 and poly-
mers17,23 are known to undergo fast photochemical oxidation
by singlet oxygen.22 To systematically study the effect of
LUMO energy on photostability of conjugated polymers
without the complication by different side chains,17 we
synthesized three polymers (Chart 1) with different number
1. INTRODUCTION
Fluorescent conjugated polymers have many applications such
as light emitting,1−3 solid-state dye lasing,4,5 biological imaging
and sensing,6−10 and chemical sensing.11−15 Fluorescent
materials can also be incorporated into microparticle taggants
for military applications such as remote and covert detection of
friend or enemy targets, supporting new capabilities in covert
search and rescue (downed pilots), intelligence gathering
(tracking documents, packages), and security measures
(detection of breaches and intruders).16 However, organic
materials usually have relatively low photochemical stability,17
which limit their applications and lifetime. There has been
significant effort on developing air-stable organic functional
materials for n-type transistor application,18 but little research
on photostable organic fluorescent polymers has been
reported.19
Chart 1Chemical structures of C10O−PPV, CN−PPV, and
DiCN−PPV polymers synthesized and used in photostability
study. R = n-decyl.
It is well established that oxidation stability of conjugated
organic compounds and polymers can be enhanced by lowering
HOMO or LUMO energy level of molecules or polymers via
modification of conjugated system with electron-withdrawing
groups. This strategy has been applied to air-stable n-type
organic field-effect transistor (FET) materials.18 Air stability
was observed for electron-deficient oligothiophene derivatives
and fused arene imides with LUMO energy lower than a certain
level (ranging from −4.1 to −4.8 eV, depending on the
molecular structure). When molecular materials are exposed to
light, the LUMOs are also populated with electrons (similar to
n-type materials after electron injection) and thus their LUMO
energy levels should also largely determine their photochemical
stability.
of cyano groups on the exocyclic double bond in PPVs:
poly(2,5-didecyloxy-1,4-phenylenevinylene) or C10O−PPV,
poly(2,5-didecyloxycyanoterephthalylidene) or CN−PPV, and
poly(2,5-didecyloxy-7,8-dicyano-1,4-phenylenevinylene),
DiCN−PPV. Introduction of one cyano group on every
vinylene unit lowers the LUMO energy by 0.56 eV. An
Received: April 1, 2013
Revised: May 13, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ma400661f | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
HOAc (110 mL) was stirred at 60 °C for 3 h. The resulting mixture
was diluted with water (300 mL), and then kept in a freezer for 1 h.
The solid products were collected by filtration, and then dissolved in
hexanes. The hexane solution was washed with water (50 mL),
aqueous NaHCO3 (10 mL), dried over MgSO4, and condensed by
rotary evaporation. The crude product had a composition of ∼15% of
1,4-bisdecyloxybenzene, ∼20% of 1, and ∼65% of 4 by 1H NMR
analysis. Separation of 1 and 4 was not practical using silica gel
chromatography. The crude product was used in the next step without
purification. 1H NMR (CDCl3) of 4: δ (ppm) 0.88 (t, J = 6.8 Hz, 6H),
1.1−1.6 (m, 28H), 1.77 (m, 4H), 3.88 (t, J = 6.6 Hz, 2H), 3.95 (t, J =
6.4 hz, 2H), 4.52 (s, 2H), 6.78 (m, 2H), 6.89 (d, J = 2.4 Hz, 1H). The
chemical shift of CH2Br is same as the literature value for CH2Br in a
similar compound: 2,5-bisbromomethyl-1-methoxy-4-(2-
ethylhexyloxy)benzene.26
additional cyano group further reduces LUMO energy by 0.41
eV without significantly changing the absorption peak wave-
length. Via a comparative study of photostability of films of
C10O-, CN- and DiCN−PPVs under white light from a xenon
lamp, a dramatic effect of CN substitution on decay rate
constant has been observed. It is also found that the presence of
a high content of cis CC in the polymer structure has a
dramatic effect on the photoluminescence (PL) efficiency in
both solution and solid state, and renders the new polymer the
only polymer known to us that is much more fluorescent in the
solid state than in the solution.
2. EXPERIMENTAL SECTION
2-Cyanomethyl-1,4-bis(decyloxy)benzene (5). A mixture of
the crude product of 4 (27.42 g), sodium cyanide (5.15 g) and DMSO
(130 mL) was stirred at 53 °C for 23 min. The mixture was diluted
with water (250 mL), extracted with hexanes (100 mL × 3), dried over
MgSO4 and condensed on a rotary evaporator. The crude product was
purified twice by silica gel (1L) column chromatography using 1:40
EtOAc/hexanes as the eluent. The product was further purified by
recrystallization from hexanes at −7 °C. Yield: 16.9 g, 59% for the two
steps combined. 1H NMR (CDCl3) of 5: δ (ppm) 0.88 (t, J = 6.8 Hz,
6H), 1.1−1.6 (m, 28H), 1.76 (m, 4H), 3.67 (s, 2H), 3.92 (m, 4H),
6.78 (s, 2H), 6.94 (s, 1H). 13C NMR (CDCl3) of 5: δ (ppm) 153.07,
150.30, 119.53, 118.01, 115.97, 114.57, 112.26, 68.74 (two OCH2
overlapped), 31.93, 29.61, 29.44, 29.36, 26.14, 26.07, 22.71, 18.26
(CH2CN), 14.13. Anal. Calcd: C, 78.27; H, 11.03. Found: C, 78.25; H,
11.04.
2-Bromo-5-(bromocyanomethyl)-1,4-bis(decyloxy)benzene
(6). A mixture of 5 (1.09 g, 2.537 mmol), azobis(isobutyronitrile)
(0.15 mol equiv, 62.9 mg), tetrachlorocarbon (11 g), and bromine
(0.812 g) was heated in a closed 16 mL-vial in a 100 °C oil bath (with
1/3 of vial above the oil) for 30 min. The reaction was worked up and
the crude product was purified in a similar way used for the
purification of 6 to afford 556 mg of white solid product. Yield: 35%.
1H NMR (CDCl3) of 6: δ (ppm) 0.88 (t, 4H, J = 6 Hz), 1.1−1.7 (m,
NMR data were obtained from a Bruker Avance 300 MHz NMR
Spectrometer. Elemental analyses were performed by Atlantic
Microlab Inc. Molecular weights of polymers were determined using
a Viscotek TriSec GPC system with universal calibration. Polystyrenes
were used as the standards and tetrahydrofuran as the solvent. The
polymer films were spin-coated from the polymer solutions in o-
dichlorobenzene on glass substrates. UV−vis spectra were obtained
from an Agilent 8453 photodiode array UV−vis absorption photo-
spectrometer. Thermal analyses (DSC and TGA) were performed
using Perkin-Elmer TGA-6 and DSC-6 systems.
Electrochemical studies were performed on a Bioanalytical (BAS)
Epsilon-100w trielectrode cell system with a Pt working electrode, an
ancillary Pt electrode, and a silver reference electrode in a CH3CN
solution of 0.01 M AgNO3 and 0.1 M tetrabutylammonium
hexafluorophosphonate. The measurements were performed in a N2-
purged 0.1 M TBA−HFP/acetonitrile solution at a scan rate of 100
mV/s. Ferrocene (2 mM in 0.10 M TBA−HFP/CH3CN solution) was
used as an internal reference standard and its HOMO level of −4.80
eV was used in calculations.
Photoluminescence (PL) was measured on an Edinburgh FS920
fluorescence spectrometer. All measurements were conducted in air
without using an inert gas to purge solution or protect films.
Correction of PL spectra was carried out using the emission correction
file of the spectrometer. The emission correction was performed in the
range of 200−900 nm using deuterium-tungsten light source with
known spectral distribution. A high power Newport light source
(model 66903) equipped with a 300 W xenon light bulb and a
Newport power supply (Model 69911) was used for the photostability
study. The output from the light source was used without filtering.
Light intensity was adjusted through a focus lens and measured by a
Newport 70260 power meter with a 3A-P-SH-V1 thermal head. The
aperture diameter of the power meter is 1.2 cm and the unit intensity
is calculated to be 2200−2600 mM/cm2, about 22−26 times of
standard one sun intensity.
28H), 1.83 (m, 4H), 4.07 (m, 4H), 5.82 (s, 1H), 7.10 (s, 1H), 7.15 (s,
1H). 13C NMR (CDCl3) of 6: δ (ppm) 150.04, 149.66, 121.24,
117.63, 116.37, 115.79, 113.27, 70.25(OCH2), 69.45 (OCH2), 31.92,
29.55, 29.34, 29.29, 29.13, 29.08, 26.00, 25.97, 22.70, 22.31
(CHBrCN), 14.13 (only OCH2 in the two decyl groups are well
separated, other pairs are completely overlapped). Anal. Calcd: C,
57.25; H, 7.72; Br, 27.20. Found: C, 57.42; H, 7.78; Br, 27.41.
2,3-Bis(4-bromo-2,5-bis(decyloxy)phenyl)-but-2-enedinitrile
(7, the Model Compound). To a solution of 6 (0.585 g, 1.00 mmol)
in 2 mL of THF, a solution of t-BuOK (0.97 equiv, 108.8 mg) in THF
1
(1 mL) was added. Then, 5 min after the addition, the H NMR
spectrum of the reaction mixture showed ∼55% cis and ∼45% trans
products. THF was removed by rotary evaporation. The residue was
worked up using hexane. The hexane solution was washed with water
and dried. Crystals formed in concentrated hexane solution and were
collected by filtration and washed with MeOH to afford 0.17 g yellow
pure trans product. The filtrate was concentrated and passed through a
silica gel column to yield 50 mg trans product. No pure cis product was
2,5-Bis(bromocyanomethyl)-1,4-bis(decyloxy)benzene (2). A
literature producedure25 for bromination of phenylacetonitrile was
followed with minor modifications. A solution of 2,5-bis-
(cyanomethyl)-1,4-bisdecyloxy-benzene (121) 0.905 g, 1.93 mmol),
benzoyl peroxide (33.0 mg, 0.135 mmol), tetrachlorocarbon (5.7 g),
and N-bromosuccinimide (NBS, 0.704 g, 3.96 mmol) was heated in a
closed bottle in a 85 °C oil bath for 3 h. After removing succinimide by
filtration, the residue was purified using a silica gel (200 mL) column
and 1:40 EtOAc/hexanes as the eluent. 320 mg of yellow solid was
obtained after crystallization from hexanes. Yield: 38.8%. The main
reason for the low yield is decomposition of the product in the column
into a yellow colored species. 1H NMR (CDCl3) of 2: δ (ppm) 0.89 (t,
4H, J = 6 Hz), 1.1−1.7 (m, 28H), 1.85 (m, J = 6.4 Hz), 4.09 (t, 4H, J =
5.1 Hz), 5.86 (s, 2H), 7.20 (s, 2H). 13C NMR (CDCl3) of 2: δ (ppm)
149.57, 124.75, 116.22, 112.40, 69.46, 29.54 (two carbon peaks
overlapped), 29.33, 29.30, 29.06, 26.02, 22.69, 21.83 (CHBrCN),
14.14. Anal. Calcd: C, 57.51; H, 7.40; Br, 25.51; Found: C, 57.70; H,
7.53; Br, 25.47.
1
obtained due to cis to trans isomerization in the column. H NMR
(CDCl3) of trans 7: δ (ppm) 0.87 (m, 12H), 1.2−1.6 (m, 56H), 1.82
1
(m, 8H), 4.02 (m, 8H), 7.01 (s, 2H), 7.22 (s, 2H). Characteristic H
NMR peaks of cis 7 can be easily identified in the spectrum of a
1
product mixture: H NMR (CDCl3) of cis 7: δ (ppm) 3.61 (t, J = 6.5
Hz, 4H), 3.87 (t, J = 6.8 Hz, 4H), 7.43 (s, 2H), 7.07 (s, 2H). The two
isomers can be easily identified by 1H NMR as the aromatic and
OCH2 protons in the cis isomer are significantly shifted to the higher
field as compared to those in the trans isomer due to the shielding
effect of the two cis benzene rings.
Poly(2,5-didecyloxy-7,8-dicyano-1,4-phenylenevinylene),
DiCN−PPV. To a solution of 2 (0.209 g, 0.333 mmol) and 6 (0.0104
mmol, 5.31 mg) in 2 mL of THF was added a solution of t-BuOK (2
equiv, 80.4 mg) in THF (2 mL). The reaction was continued for 10
2-Bromomethyl-1,4-bis(decyloxy)benzene (4). A mixture of
1,4-bis(decyloxy)benzene (3, 78.12 g, 0.2 mol), paraformaldehyde
(7.207g, 0.24 mol), glacial acetic acid (230 g), and 30 wt % HBr/
B
dx.doi.org/10.1021/ma400661f | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Scheme 1. Synthesis of DiCN−PPV and a Model Compound 7
1
Figure 1. H NMR spectrum of DiCN−PPV in CDCl3.
min. The resulting mixture was dropped into MeOH. The yellow
polymer solid was collected by filtration, and vacuum-dried. Yield:
0.157 g, 98.4%. Anal. Calcd: C, 76.96; H, 9.50; N, 5.86; Br, 0.82.
Found: C, 76.97; H, 9.54; N, 5.60; Br, 0.55.
room temperature. A small amount of 6 was added as a
terminator to control molecular weight (MW). When 1% molar
equivalent of 6 was used, the polymer still has a high MW (Mw/
Mn = 63015/17901) and was not fully soluble. When 5% molar
equivalent of 6 was used, the Mw was reduced to 14 k Da, and
the polymer became fully soluble and was used in the following
study. The amount of base was controlled to be exactly 1 molar
C10O−PPV24 and CN−PPV21 were synthesized according to the
literature procedures.
3. RESULTS AND DISCUSSION
1
equiv to avoid cross-linking of the formed polymer. The H
Synthesis. The synthetic route of DiCN−PPV is shown in
Scheme 1, and is different from a literature method used for
synthesis of an analogous polymer (with MeO and EHO side
chains).27 Compound 1 was brominated to give monomer 2
using a literature procedure with minor modifications.25 6 was
synthesized from 1,4-bisdecyloxybenzene (3) in three steps.
The monobromomethylation of 3 was effected using the same
procedure as that used for dibromomethylation of 324 except
that only 1.2 mol equiv of paraformaldehyde was used.
Separation of 4 from the dibromomethylation side product
was not performed due to their similar Rf values. Bromination
of 5 also happened to the benzene ring and led to the formation
of compound 6.
Compound 2 has not been reported in the literature. But its
monofunctional analogues, iodophenylacetonitrile compounds,
have been reported and can be converted into 2,3-diphenyl-but-
2-enedinitrile compounds in the presence of MeONa.28
Applying similar condition to 2 should polymerize it into
DiCN−PPV as shown in Scheme 1. Potassium tert-butoxide
was used instead of MeONa because NaOMe can lead to the
further conversion of the formed dicyanovinylene unit into
maleimide.28 The polymerization reaction was instantaneous at
NMR spectrum is shown in Figure 1. Several peaks are present
in the region of 4 ppm for O−CH2 protons and in the aromatic
region for phenyl protons. The peak at 4.15 ppm is assigned to
O−CH2 in trans units while the minor peaks in the higher field
are assigned to O−CH2 in cis units, similar to situation in
C10O−PPV.24 The intensity of this peak accounts for ∼65% of
the overall intensity of all peaks in the region of 3.5−4.2 ppm.
Accordingly, the peak at 7.2 ppm, which accounts for ∼65% of
the integration of the aromatic peaks, can be assigned to the
phenyl units next to trans-dicyanovinylene units. It is noticed
that this chemical shift (7.2 ppm) is significantly smaller than
that (7.78 ppm) of trans-bis(4-methoxyphenyl)fumaronitrile in
the literature.28 This difference is due to the presence of 2,5-
decyloxy groups which forces the benzene ring and the vinyl
unit out of plane even in the trans conformation and thus
reducing the influence of the electron withdrawing cyano
groups.
The presence of the cis structures introduce kinks in polymer
chains and reduce its conjugation length since the benzene
rings cannot be coplanar with the vinylene unit. Cis-to-trans
conversion under acidic condition, which was successful for
C
dx.doi.org/10.1021/ma400661f | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 2. Left: DSC trace of DiCN−PPV. Middle: TGA analysis of DiCN−PPV. Right: wide-angle XRD results of DiCN−PPV.
C10O−PPV24 and SF−PPV,29 did not produce a significant
1
difference in H NMR spectrum of DiCN−PPV. There could
be two different reasons for the failed conversion: (1) The
electron-withdrawing cyano groups reduce the electron density
of the vinyl to such a low level that the proton can no longer
add to the vinylene to form a cationic intermediate, or (2) the
two conformations are similar in energy and thus can
thermodynamically coexist. To find the true reason, a model
compound (7) was made from dimerization of 6 (Scheme 1) to
study the cis/trans behavior of the dicyanovinylene unit. The
product 7 was a mixture of cis/trans dimers. Pure trans isomer
was obtained from silica gel column chromatography of the
mixture as the cis compound moves slightly slower. Pure cis
isomer was not obtained due to its faster isomerization in the
column. It was found that trans−cis isomerization happened to
trans isomer of 7 in CDCl3 solution (usually contains a trace
amount of acid) at ambient temperature after for a few days.
After 40 days in the NMR tube, the trans−cis ratio reached an
equilibrium value of 2.9:1 (or 74% trans). So, the existence of
cis and trans thermodynamic equilibrium prevents the polymer
from being converted to all trans.
Solubility, Thermal Properties, and XRD Study.
Because of the presence of cis structure, which is highly
nonplanar, DiCN−PPV is highly soluble in common polar
organic solvents such as chloroform, tetrahydrofuran, etc. DSC
thermograph (Figure 2, left) of the polymer does not show any
significant phase transition or melting peak. The decomposition
onset temperature is about 250 °C by DSC. However, the
polymer does not lose weight until 354 °C as shown in the
TGA result (Figure 2, middle). XRD experiments (wide-angle)
were performed on as-cast thick films of DiCN−PPV (Figure 2,
right). No higher order peaks were observed other than a
zeroth order peak at 3.7°.
Optical Absorption and Electrochemical Properties.
The normalized UV−vis absorption spectra of C10O−PPV,
CN−PPV, and DiCN−PPV in chloroform solutions and films
are shown in Figure 3. The peak wavelengths, cutoff
wavelengths and optical bandgaps are summarized in Table 1,
together with molecular weights of the three polymers. The first
cyano substitution causes absorption peak wavelength to blue
shift significantly from 483 nm for C10O−PPV to 435 nm for
CN−PPV in solutions. The second cyano group only produces
a blue shift of 3 nm. It is also noticed that DiCN−PPV and
C10O−PPV have smaller red shift in optical absorption peak
wavelength from chloroform solution to film than that of CN−
PPV. In general, polar polymers have stronger interactions in
the solid state and have larger red shifts in absorption from
solution to solid. For examples, RSO2-substitued PPVs have
larger red shift than C10O−PPV,29 and S,S-dioxothienylenevi-
Figure 3. UV−vis absorption (top), PL spectra (middle) of C10O−
PPV, CN−PPV and DiCN−PPV in solution and film, and PL spectra
of blend films (bottom) of MEH−PPV:CN−PPV and MEH−
PPV:DiCN−PPV. Neat polymer films were excited at their peak
absorption wavelengths. Blend films were excited at 487 nm.
D
dx.doi.org/10.1021/ma400661f | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
a
Table 1. GPC, Optical, and Electrochemical Characterization Results of the PPV Polymers
Mw/Mn (kDa) λabsmax/nm (eV) sol./film Egopt/eV sol./film λPLmax/nm sol./film Ered/LUMO (V/eV) Eox/HOMO (V/eV) Egel (eV)
C10O−PPV
CN−PPV
DiCN−PPV
23/12
14/7.5
14/6.1
483/481 (2.57/2.58)
435/556 (2.85/2.23)
432/444 (2.87/2.79)
2.25/2.14
2.39/2.14
2.43/2.36
550/582
559/639
597/571
- /-2.91
- /-5.15
2.24
2.50
2.47
−1.5/-3.29
−1.09/-3.70
1.00/-5.79
1.38/-6.17
a
Chloroform was used as the solvent for the polymers.
nylene-based conjugated polymers30 have larger red shift than
nonpolar C12-polythienylenevinylene.31 In DiCN−PPV, the cis
unit is polar; however, it is highly nonplanar and does not
permit strong interchain interactions in the solid state.
Table 2. Photoluminescence Quantum Yields of PPVs with
Different CN Substitution
C10O−PPV
CN−PPV
DiCN−PPV
a
solution
0.57−0.6
1
0.35−0.4
1.87
0.08
1.71
Cyclovoltammic measurements were performed on CN−
PPV and DiCN−PPV films coated on a Pt working electrode
and the results are given in Table 1. Relative to HOMO and
LUMO levels of CN−PPV, the second cyano group lowers the
HOMO energy by 0.38 eV and the LUMO energy by 0.41 eV.
As a result, DiCN−PPV is more difficult to oxidize, but is easier
to reduce. The calculated electrochemical bandgaps of the two
polymers are almost same, with Egel of DiCN−PPV smaller by
only 0.03 eV. Therefore, the second cyano group does not
bring significant change to the bandgap, but, does lower the
energies of the frontier orbitals.
Photoluminescence (PL) Study. PL spectra of the PPV
polymer films are shown in Figure 3. The PL peak wavelengths
are 564 and 640 nm for DiCN−PPV and CN−PPV,
respectively. The peak position of CN−PPV is similar to that
reported by Chen et al.21 Given the fact that the UV−vis
absorption λmaxs are essentially same, it is surprising to find that
the PL λmax of CN−PPV is red-shifted by 76 nm from that of
DiCN−PPV. The strong interchain interactions in CN−PPV is
believed to be the major reason for the red shift of PL and the
extended long wavelength tail in the UV−vis absorption
spectrum.21 The presence of high content of cis dicyanoviny-
lene in DiCN−PPV is apparently very effective in preventing π-
stacking of DiCN−PPV.
PL quantum yields (PLQYs) of polymer chloroform
solutions were measured by the optically dilute method.32
Rhodamine 101 in ethanol, which is known to have a quantum
yield close to 1,33 was used as the reference standard.
Concentrations of all solutions were controlled low so that
absorbances of the solutions in a 1-cm cell were smaller than
0.1. Excitation wavelength of 500 nm was chosen for all
solutions. Obtained PL intensities were corrected for detector
sensitivity. Integration of corrected PL spectrum of each
polymer was compared with that of the standard solution to
yield the PLQY. The absolute PLQYs of the polymer films were
not measured due to the lack of a film standard and an
integration sphere. The relative PLQYs of the polymer films
were obtained by comparing the integrations of PL spectra of
polymer films after correction for source intensity and film
absorbance. For each polymer film, the sample holder was
adjusted to maximize PL intensity at emission peak wavelength.
The results are listed in Table 2. It is noticed that PL of DiCN−
PPV solution is very weak, as evident from its noisy PL
spectrum in Figure 3, top. The calculated PLQY of DiCN−PPV
solution is only 0.08, or 14% of that of C10O−PPV solution.
Surprisingly, it became more fluorescent than C10O−PPV in
film. Assuming that the PLQY of C10O−PPV film is one-third
of its QY in solution as reported for MEH−PPV,34 the PLQY
of DiCN−PPV film is as high as 0.34, more than four times as
high as QY of its solution. This phenomenon has been known
for molecular dyes such as auramine O and malachite green.35
b
film
a
Chloroform was the solvent for three polymers. QYs are relative to
b
that of Rhodamine 101 ethanol solution. QYs are relative to C10O−
PPV.
and a trans-stibene dye with two cyano groups on the CC
moiety.36 Report of such behavior for polymers has not found
in the literature. In all these dye molecules, twisted phenyls
groups are present and their free waggling motion in solutions
provides fast nonradiative decay for the excited states. DiCN−
PPV is a likely the first vinylene-based conjugated polymers in
which cis CC is present in high percentage (∼35%) and
phenyl groups attached to the cis CC bonds behave similarly
to the phenyls in those dyes.
DiCN−PPV as an Electron Acceptor or PL Quencher.
CN−PPV has been shown to be a good electron acceptor for
donor polymers such as MEH−PPVs.37 We are interested in
finding out how an additional CN group affects ability of
DiCN−PPV as an electron acceptor. PL spectra of the blend
films of 1:1 MEH−PPV and CN−PPV or DiCN−PPV were
recorded with excitation at 487 nm and are shown in Figure 3.
PL intensity of the MEH−PPV/CN−PPV film is quite weak
and the peak wavelength (647 nm) matches the PL peak
wavelength of pure CN−PPV. The PL of MEH−PPV is not
completely quenched and appears as a shoulder peak at 580
nm. In the case of MEH−PPV:DiCN−PPV film, the PL of
MEH−PPV is completely quenched and the residual DiCN−
PPV emission is negligible (only 5.56% of that of MEH−PPV/
CN−PPV film). The PL quenching efficiency in MEH−
PPV:DiCN−PPV film is greater than 99.5%, much more
complete than that in MEH−PPV:CN−PPV film for which
∼93% quenching was reported.37 Since MEH−PPV cannot lose
its excitation energy by energy transfer to DiCN−PPV, the
quantitative PL quenching suggests that a nearly quantitative
charge transfer from MEH−PPV to DiCN−PPV happened in
the film when illuminated.
Photochemical Stability of PPVs. We are most interested
in the photostability of DiCN−PPV (in comparison with CN−
PPV and C10O−PPV) in the sunlight for outdoor applications.
Xenon light was used as the light source since its spectrum is
close to that of sunlight. A very high intensity (2200−2600
mW/cm2, more than 20 times the normal one sun intensity
-100 mW/cm2) was used so that the photostability test could
be completed in a relatively short period of time. The
thicknesses of all films were controlled to have peak absorbance
between 0.6 and 0.7. Such optical thickness was chosen to
minimize the effect of thickness on decay constant and to allow
sufficient range of photoinduced absorbance change to be
measured with accuracy. Absorption and PL spectra were
recorded after each illumination and are shown in Figure 4.
E
dx.doi.org/10.1021/ma400661f | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
Figure 4. Evolutions of UV−vis and PL spectra of C10O−, CN−, and DiCN−PPV with increasing (accumulated) illumination time. The excitation
wavelengths are 477, 463, and 441 nm for C10O−PPV, CN−PPV, and DiCN−PPV, respectively.
Figure 5. Left: Absorbance and PL intensity as functions of (accumulated) illumination time of DiCN−PPV. Right: Photodegradation of C10O−
PPV, CN−PPV, and DiCN−PPV films shown as semilog plots of peak PL intensity v.s. illumination time. The initial intensities were normalized.
The PL intensity of DiCN−PPV film at 20 min (in the left plot) was used as the starting point for the semilog plot. The emission wavelengths are
581, 624, and 562 nm for C10O−PPV, CN−PPV, and DiCN−PPV, respectively.
Both the PL and absorbance of C10O−PPV disappeared in
about 5 min, along with the disappearance of the film color in
the illuminated area. The absorption peak gradually shifted to
shorter wavelength as the illumination continued, presumably
F
dx.doi.org/10.1021/ma400661f | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
due to shortening of conjugation length with photodegradation.
Blue shift of PL peak is not obvious since the emission from the
sample would be dominated by remaining long polymer chains
white light (>20 times as strong as the sunlight) than C10O−
PPV by more than 2 orders of magnitude. The excellent
photochemical stability and decently high fluorescence
quantum yield in the solid state make DiCN−PPV a good
candidate for outdoor fluorescent applications such as remote
optical sensing.
in the film due to the Forster resonance energy transfer38 from
̈
shorter conjugated polymer chains to longer ones. CN−PPV
film was more stable, but the original peak became
unobservable in about 10 min. Its PL spectra were only
shown up to 8 min of accumulated illumination time since
absorption spectra after illumination for longer than 8 min were
dominated by a new peak at 392 nm.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: (C.Z.) cheng.zhang@sdstate.edu.
DiCN−PPV showed 17% increase in PL intensity in initial
20 min of illumination (Figure 5 left). The absorbance of the
film (Figure 5 left) did not show any increase during the same
period of time. Instead, it experienced faster decay in the first 3
min, as compared to its trend over the whole period of
illumination. This fast decay period overlapped with the fast
rise period of photoluminescence. The coincidence of the two
observations can be explained by light-induced trans to cis
isomerization which not only shortened conjugation length of
the polymer and reduced polymer absorption, but also further
disrupted packing of π-conjugated backbones and reduced
interchain PL quenching. Semilog plots of peak PL intensity vs
illumination time are shown in Figure 5 right for C10O−PPV,
CN−PPV and DiCN−PPV films. The initial intensities were
normalized. The PL intensity of DiCN−PPV film at 20 min (in
the left plot) was used as the starting point for the semilog plot
since the changes in PL intensity during the initial 20 min were
mainly caused by photoinduced trans-cis isomerization. All
three plots can be fitted very well with a linear function of time,
indicating a first-order decay for all three polymers. The slopes
of the linear fittings are 0.239, 0.0249, and 0.0019 min−1 for
C10O−PPV, CN−PPV and DiCN−PPV films, respectively. So,
with substitution of one cyano group, CN−PPV is more
photochemical stable than C10O−PPV by 1 order of magnitude
under the white light; with two cyano groups, DiCN−PPV is
more stable than C10O−PPV by over 2 orders of magnitude.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This material is based upon work supported by the U.S. Air
Force Office of Sponsored Research under Award No.
FA955010-1-0555. C.Z. is grateful to the support by the
National Science Foundation through award 0931373 and the
Department of Chemistry and Biochemistry and South Dakota
NSF EPSCoR Program (Grant No. 0903804) for the startup
funds. S.S. is grateful to the support by the Department of
Defense through Award W911NF-06-1-0488. The authors
would like to thank Ms. Liping Si and Dr. Hongshan He for
assistant on the fluorescence measurements.
REFERENCES
■
(1) Al Salhi, M. S.; Alam, J.; Dass, L. A.; Raja, M. Int. J. Mol. Sci. 2011,
12, 2036.
(2) Perepichka, D. F.; Perepichka, I. F.; Meng, H.; Wudl, F. Opt. Sci.
Eng. 2007, 111, 45.
(3) Wu, H.; Ying, L.; Yang, W.; Cao, Y. Chem. Soc. Rev. 2009, 38
(12), 3391.
(4) Duran-Sampedro, G.; Agarrabeitia, A. R.; Garcia-Moreno, I.;
Costela, A.; Banuelos, J.; Arbeloa, T.; Arbeloa, I. L.; Chiara, J. L.; Ortiz,
M. J. Eur. J. Org. Chem. 2012, 2012 (32), 6335.
(5) Navarro-Fuster, V.; Calzado, E. M.; Boj, P. G.; Quintana, J. A.;
Villalvilla, J. M.; Díaz-García, M. A.; Trabadelo, V.; Juarros, A.;
Retolaza, A.; Merino, S. Appl. Phys. Lett. 2010, 97, 171104.
(6) Kaeser, A.; Schenning, A. P. H. J. Adv. Mater. 2010, 22, 2985.
(7) Kai, L.; Bin, L. J. Mat. Chem. 2012, 22 (4), 1257.
(8) Feng, X.; Liu, L.; Wang, S.; Zhu, D. Chem. Soc. Rev. 2010, 39 (7),
2411.
(9) Li, K.; Bin, L. Poly. Chem. 2010, 1 (3), 252.
(10) Duan, X.; Liu, L.; Feng, F.; Wang, S. Acc. Chem. Res. 2010, 43
(2), 260.
(11) Kim, H. N.; Ren, W. X.; Kim, J. S.; Yoon, J.. Chem. Soc. Rev.
2012, 41 (8), 3210.
(12) Kim, Ha Na; Guo, Z.; Zhu, W.; Yoon, J.; Tian, H. Chem. Soc.
Rev. 2011, 40, 79.
(13) Cumming, C. J.; Aker, C.; Fisher, M.; Fox, M.; la Grone, M. J.;
Reust, D.; Rockley, M. G.; Swager, T. M.; Towers, E.; Williams, V.
IEEE Trans. Geosci. Remote Sens. 2001, 39 (6), 1119.
(14) Alvarez, A.; Costa-Fernandez, J. M.; Pereiro, R.; Sanz-Medel, A.;
Salinas-Castillo, A. Trends Anal. Chem. 2011, 30 (9), 1513.
(15) Hu, J.; Liu, S. Macromolecules 2010, 43, 8315.
(16) These applications have been envisioned by Air Force in Topic
17 description of 2010 AFOSR MURI proposal solicitation, and the
research is being conducted by an AFOSR MURI team of which the
corresponding author is a co-PI.
́
(17) Manceau, M.; Bundgaard, E.; Carle, J. E.; Hagemann, O.;
Helgesen, M.; Søndergaard, M.; Jørgensen, M.; Krebs, F. C. J. Mater.
Chem. 2011, 21, 4132.
(18) Usta, H.; Facchetti, A.; Marks, T. Acc. Chem. Res. 2011, 44, 501.
(19) Grabchev, I.; Betcheva, R. J. Photochem. Photobio, A: Chem.
2001, 142 (1), 73.
̃
4. CONCLUSION
A didecyloxy substituted poly(phenylenedicyanovinylene),
DiCN−PPV, has been synthesized. The dicyano-substitued
vinylene units exist in both trans and cis conformations in a
ratio of about 65:35. It was found that the cis conformation
could not be converted into trans conformation by acid
treatment that worked for C10O−PPV. Study of a model
compound for the diphenyl-substituted dicyanoethylene moiety
of the polymer shows that cis and trans conformations coexist in
equilibrium, indicating that the trans conformation is not much
lower in energy than the cis conformation, in contrast to what
have been observed for other PPV polymers.24,29 The unusually
high cis content makes this polymer highly amorphous, very
soluble in organic solvent, and highly fluorescent in the solid
state with an estimated quantum yield up to 0.34. Interestingly,
this polymer is only weakly fluorescent in solution presumably
due to unrestricted waggling motions of out-of-plane phenyl
groups bonded to the cis CC. Illumination with white light
induced certain degree of trans to cis conversion as evidenced
by a reduction in light absorption, which was faster than the
reduction caused by light-induced degradation. The light-
induced increase in cis content in turn caused a significant rise
in photoluminescence of the DiCN−PPV film as a result of
further reduction in interchain interaction. The two cyano
group in DiCN−PPV brings a decrease of LUMO energy level
by 0.79 eV, and makes the polymer more stable to intense
G
dx.doi.org/10.1021/ma400661f | Macromolecules XXXX, XXX, XXX−XXX

Macromolecules
Article
(20) Halls, J. J. M.; Walsh, C. A.; Greenham, N. C.; Marseglla, E. A.;
Friend, R. H.; Moratti, S. C.; Holmes, A. B. Nature 1995, 376, 498 and
references therein.
(21) Chen, S.-A.; Chang, E.-C. Macromolecules 1998, 31, 4899.
(22) Dam, N.; Scurlock, R. D.; Wang, B.; Ma, L.; Sundahl, M.;
Ogilby, P. R. Chem. Mater. 1999, 11, 1302.
(23) O’Neill, L.; Lynch, P. J.; McGoldrick, N.; Byrne, H. J.;
McNamara, M. J. Lumin. 2012, 132, 2217.
(24) Zhang, C.; Choi, S.; Haliburton, J.; Cleveland, T.; Li, R.; Sun, S.-
S.; Ledbetter, A.; Bonner, C. E. Macromolecules 2006, 39, 4317.
(25) Molina, P.; Lop
́
ez-Leonardo, C.; Llamas-Botía, J.; Foces-Foces,
C.; Fernandez-Castano, C. Tetrahedron 1996, 52 (28), 9629.
́
̃
(26) Neef, C. J.; Ferraris, J. P. Macromolecules 2000, 33 (7), 2311.
(27) Moslin, R. M.; Andrew, T. L.; Kooi, S. E.; Swager, T. M. J. Am.
Chem. Soc. 2009, 131, 20.
(28) Yeh, H.-C.; Wu, W.-C.; Wen, Y.-S.; Dai, D.-C.; Wang, J.-K.;
Chen, C.-T. J. Org. Chem. 2004, 69, 6455.
(29) Zhang, C.; Sun, J.; Li, R.; Black, S.; Sun, S.-S. Synth. Met. 2010,
160, 16.
(30) Zhang, C.; Nguyen, T. H.; Sun, J.; Li, R.; Black, S.; Bonner, C.
E.; Sun, S.-S. Macromolecules 2009, 42, 663.
(31) Zhang, C.; Matos, T.; Li, R.; Sun, S.-S.; Lewis, J. E.; Zhang, J.;
Jiang, X. Polym. Chem. 2010, 1, 663.
(32) Demas, J. N.; Crosby, G. A. J. Phys. Chem. 1971, 75, 991.
(33) Drexhage, K. H. In Dye Lasersl Schafer, F. P., Ed.; Springer:
Berlin, 1973.
(34) Samuel, I. D. W.; Crystall, B.; Rumbles, G.; Burn, P. L.; Holmes,
A. B.; Friend, R. H. Chem. Phys. Lett. 1993, 213, 472.
(35) Valeur, B. Molecular Fluorescence; Wiley VCH: New York, 2002,
pp 65−67.
(36) Kim, D. U.; Paik, S. H.; Kim, S.-H.; Tsutsui, T. Synth. Met. 2001,
123, 43.
(37) Yu, G.; Heeger, A. J. J. Appl. Phys. 1995, 78 (7), 4510.
(38) Forster, T. Ann. Phys. 1948, 437, 55.
̈
H
dx.doi.org/10.1021/ma400661f | Macromolecules XXXX, XXX, XXX−XXX