lable at ScienceDirect

Materials Chemistry and Physics 147 (2014) 476e482

Contents lists avai

Materials Chemistry and Physics

journal homepage: www.elsevier .com/locate/matchemphys

Optoelectronic properties of a perylene substituted (cholesteryl)benzoateethynylene co-polymer

Stefania Zappia a, b, Arxel de Le�on b, Marina Alloisio a, Eduardo Arias b,Giovanna Dellepiane a, Giovanni Petrillo a, Ivana Moggio b, *, Sergio Thea a,Carlos Gallardo-Vega b, Marlene Rodríguez b

a Dipartimento di Chimica e Chimica Industriale, Universit�a di Genova, Via Dodecaneso 31, 16146 Genoa, Italyb Centro de Investigaci�on en Química Aplicada, Boulevard Enrique Reyna 140, 25294 Saltillo, Coahuila, Mexico

h i g h l i g h t s

* Corresponding author. Tel.: þ52 8444389830; faxE-mail address: ivana.moggio@ciqa.edu.mx (I. Mo

http://dx.doi.org/10.1016/j.matchemphys.2014.05.0170254-0584/© 2014 Elsevier B.V. All rights reserved.

g r a p h i c a l a b s t r a c t

� Synthesis of a benzoateethynylenebackbone co-polymer with perylenesubstituent.

� Optical and electrochemical proper-ties consistent with photoinducedenergy transfer.

� Enhancement of photovoltaic effi-ciency when PCBM is added to theco-polymer.

� EFM studies of active layers.

a r t i c l e i n f o

Article history:Received 11 July 2013Received in revised form9 January 2014Accepted 10 May 2014Available online 6 June 2014

Keywords:PolymersLuminescenceOptical propertiesAtomic force microscopy (AFM)

a b s t r a c t

A perylene cholesteryl-benzoateethynylene co-polymer was synthesized by Sonogashira reaction andcharacterized by NMR, UVeVis, static and dynamic fluorescence spectroscopy and cyclic voltammetry.The optical and electrochemical properties in solution are consistent with photoinduced energy transferfrom the electron donor conjugated backbone to the electron acceptor perylene substituent. Photovoltaicproperties are indeed found, even if the performance of the solar cells is quite low due to the formationof aggregates. The incorporation of (6,6)-phenyl C61ebutyric acid methyl ester (PCBM), however, in-creases by an order of magnitude the efficiency of the prototype (from 10�4 to 10�3%) due to both betterphase mixing and improved electrical continuity as supported by Atomic Force Microscopy (AFM) andElectrical Force Microscopy (EFM) studies.

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Bulk heterojunction (BHJ) polymer solar cells (PSCs) based onconjugated p-type polymers and n-type fullerene derivatives havebeen intensively investigated in recent years for generation ofaffordable, clean and renewable energy [1]. Advantages of the BHJ

: þ52 8444389839.ggio).

PSCs include low-cost fabrication of large-area devices, lightweight,mechanical flexibility and easy tunability of chemical properties ofthe polymer materials. The bulk heterojunction of regioregularpoly(3-hexylthiophene) (P3HT) and (6,6)-phenyl C61ebutyric acidmethyl ester (PCBM) represents one of the most successful systems[2]. However, further improvement of its performance is intrinsi-cally hindered by the narrow absorption spectrum (300e650 nm)and the high-lying HOMO energy level (�4.9 eV) of P3HT. On theother hand, a major drawback of bulk heterojunction devices ingeneral is its dependence on morphological aspects. In fact, if thedistance between the acceptor and donor materials is higher than

S. Zappia et al. / Materials Chemistry and Physics 147 (2014) 476e482 477

the exciton diffusion length (typically 10e20 nm), the separatedcharges recombine [3].

An alternative strategy is thus to combine covalently in the samestructure a donor and an acceptor group. In this context, the easiersynthetic way is to co-polymerize an electron acceptor (A) with anelectron donor (D) monomer. Different structures can be obtaineddepending on the location of A and D groups, which could be eitherpart of the main chain or side substituents of it. The main-chaindonor and side-chain acceptor co-polymers were derived fromthe extended concept of two-dimensional conjugated polymersproposed by Li. et al. [4] Compared with the well-developed main-chain donoremain-chain acceptor co-polymers, this type of poly-mer is expected to have several interesting features, such asisotropic charge transport and inside carrier transfer (from back-bone to side chain). Since the first successful demonstration of thisconcept by Huang [5], this family of co-polymers has attracted greatinterests [6]. Zhang et al. pioneered the use of the concept of main-chain donor with pendant acceptors in photovoltaic materials andreported alternating co-polymers of fluorene and triphenylaminewith mono-cyano, di-cyano or aldehyde pendant acceptors [7].Compared to linear donoreacceptor polymers, however, this familyof polymers usually exhibits low fill factor (FF) values due to thepoor packing of their polymer chains.

For this reason, more investigations are needed to further un-derstand their structureeproperty relationship and improve theirperformances. In general, the most important roles to design DeAco-polymers for photovoltaic applications are: 1) a proper combi-nation of DeA units to achieve low band gap and deep HOMO en-ergy level; 2) high planarity of the DeA units and the wholebackbone to enhance absorption and hole mobility; 3) the properincorporation of side chains to offer good solubility without causingsteric hindrance. On these bases, in this work we report the syn-thesis and physicochemical characterization of a benzoateethyny-lene co-polymer with perylenic side group in one monomeric unit.The other unit presents a cholesteryl side chain that has beenpreviously reported to give interesting optical and structuralproperties, together with high solubility [8]. Phenyleneethynylenesare known as strong fluorescent materials and have been success-fully applied in light emitting diodes and sensors [9]. Their appli-cation in solar cells is still limited with reported efficiency in theorder of 10�4%, due substantially to the high energy absorptions,rather far from the maximum sun emission [10]. For this reason,perylene is used in this macromolecule not only as acceptor group[11] but also to increase the optical overlap with the terrestrial solarradiation. It is important to notice that perylene imides are prom-ising acceptor materials also because of their high photostabilityand chemical facility to be functionalized and their use in solardevices has been recently reviewed [12]. In particular, photoin-duced electron transfer has been proven in donoreacceptor sys-tems based on perylene and conjugated molecules such asoligo(phenylene vinylene)s [13] or thiophenes [14]. However, to thebest of our knowledge, no reports have been published on phe-nyleneethynylene polymers with lateral perylene bisimides groups.Moreover, in this paper we demonstrate that the efficiency can beincreased by an order of magnitude, reaching 10�3%,when PCBM isadded in the active layer, thanks to a better phase mixing andimproved electrical continuity.

2. Experimental section

2.1. Synthesis

Experimental procedure for the synthesis of poly[(N-((pentyl)-2,5-benzoate)-N0-(1-hexylheptyl) perylene-3,4,9,10-tetracarboxylbisimide))-Co-((cholesteryl)-2,5-(benzoate) ethynylene], hereafter

named nPE(Pery-Co-Chol), chemical and physicochemical charac-terization of the co-polymer and its intermediaries are given in thesupporting information.

2.2. Reagents and materials

All the reactants were purchased from SigmaeAldrich and AlfaAesar and used as received. Spectroscopic grade toluene was fromSigmaeAldrich. Acetonitrile from SigmaeAldrich was passedthrough a plug of silica gel and distilled from NaH. Triethylammine(Et3N) was distilled from KOH. Electrochemical grade tetrabuty-lammonium hexafluorophosphate was purchased from Fluka.Poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate) (PEDOT:PSS, Clevios P) was from Clevios while PCBM and Woods metalwere from SigmaeAldrich. Quartz and Indium Tin Oxide (ITO,8e12 U cm�2) substrates were supplied by Spi.Inc.

2.3. Equipment and techniques

1H and 13C NMR spectra were recorded on Varian Mercury at300 MHz (1H) and 75 MHz (13C) at room temperature in CDCl3 andusing tetramethylsilane (TMS) as internal standard. UVeVis spectrawere recorded on a Shimadzu 2401PC or an Agilent 8453 spectro-photometer. Fluorescence spectra were obtained on a Perkin ElmerLS50B spectrofluorimeter by exciting at 10 nm below the absorp-tion maximum. The quantum yield (f) in solution was obtainedaccording to literature method [15]. Quinine sulfate in 0.1 M H2SO4

water solution (f ¼ 0.54 at lexc ¼ 310 nm) was used as standard.Temperaturewas kept at 25.0 ± 0.3 �Cwith awater circulating bath.Four different solutions with absorbance between 0.03 and 0.09were analyzed and the value of f was averaged. The error in thereported values is the standard deviation between the four de-terminations. Fluorescence lifetime (t) was obtained on a TemproSingle Photon Counting equipment with 370 nm or 560 nmnanoLED lasers. Ludox was used for pump measurements. Volta-grams were obtained by cyclic voltammetry (CV in a Basi cell) witha potentiostat/galvanostat ACM Gill AC. The reference electrodewas an Accumet calomel electrode and the electrolyte was a 0.1 MCH3CN solution of Bu4NPF6. The experiments were carried out in anitrogen atmosphere at a scanning rate of 50 mV s�1. The workingelectrode was a film of the material deposited on ITO (from Sig-maeAldrich; 15e25 U cm�2) prepared by drop casting.

2.4. Thin films and solar cells

The active layers of the solar cells were prepared by dip coatingwith a TL0.01 MTI corporation, making 4 immersions (speed of16 cm min�1 and time of immersion of 20 min) in order to get athickness of 100 nm. From one dipping to the other, the layers werekept at room temperature for 15 min. After the fourth immersion,the films were annealed at 150 �C in an Imperial V laboratory oven(that was previously set at the temperature) for 30 min. As term ofcomparison for morphological studies, films were also prepared byspin coating at 2000 rpm with a WS-400B-6NPP-LITE spin pro-cessor from Laurell Technologies. Thicknesses were measured by aDektak 6M Stylus profilometer at a scanning rate of 50 mm s�1 andan applied force of 2 mg. AFM morphological characterization wasperformed on a Nanoscope III DimensionTM 3100 from Digital In-struments with a Pt-coated Si tip (20 nm nominal radius) in tappingand Electrical Force Microscopy (EFM) at a scanning rate (256 lines)of 0.2 Hz. For EFM studies, a voltage of 12 V was applied to the tipand the separation distance was 100 nm. The surface roughness(Rq), given by the root mean square average (RMS) of height devi-ation, is taken from the data plane. Grain size analysis was made inthe software available in the microscope.

S. Zappia et al. / Materials Chemistry and Physics 147 (2014) 476e482478

Solar cells were prepared with the structure: ITO/PEDOT:PSS/active layer/Woods metal. ITO substrates were first cleaned in ul-trasonic bath (Branson 200) in dichloromethane for 100, hexane for100 and methanol for 600, then dried in an oven. PEDOT:PSS wasfiltered and then deposited on the ITO slides by spin coating at5000 rpm to give a 20 nm layer. Two different configurations havebeen tested for the active layer: 1) Single layer. 15 mg of nPE(Pery-Co-Chol) in 1 ml of toluene or chlorobenzene, 12 h of magneticstirring at 80 and 95 �C for toluene and chlorobenzene, respectively.2) 3:1 w/w nPE(Pery-Co-Chol)/PCBM heterojunction. 5 mg of PCBMwere added to a solution of nPE(Pery-Co-Chol) in chlorobenzeneprepared as abovementioned and the mixture was kept undermagnetic stirring at 95 �C for 4 h. In both cases, the film was pre-pared as previously described. Woods metal as cathode wasdeposited at 85 �C. The IeV curves were obtained using a Keithley2400 source meter by illuminating the devices from the ITO sidewith a 100 mW cm�2 white light from a Solar Light Co. Model XPS400 solar simulator with a Xenon lamp and AM1.5 filter. All cellswere prepared and measured under ambient conditions.

3. Results and discussion

The co-polymer was obtained through the Sonogashira reactionbetween (cholesteryl) 2,5-diethynyl benzoate [8] and N-((pentyl)2,5-dibromobenzoate)-N0-(1-hexylheptyl)perylene-3,4,9,10-tetracarboxylbisimide 7 (Scheme 1). 7 was obtained from a modifiedprocedure of literature [16]. The commercial tridecan-7-one 1 un-dergoes a reductive amination to afford 2 in 76% yield. The aminefrom 2 was reacted with perylene-3,4,9,10-tetracarboxylic

Scheme 1. Reagents and conditions: i) NH4OAc, NaBH3CN, MeOH abs., 35 �C, 2 h, ii) PTCD, Zn155 �C, 4 h, v) DCC, DMAP, CH2Cl2/DMF, 30 �C, 5 days. vi) [(C6H5)3P]2PdCl2, CuI, pyridine, E

dianhydride (PTCD) to obtain the symmetrically functionalizedperylenic bisimide in 96% yield. Partial alkyl hydrolysis was carriedout under alkaline conditions; care had to be taken in order thattemperature was not higher than 155 �C and reaction time did notexceed 30 min. The perylenic monoanydride monoimide 4 wasrecovered by flash chromatography in 40% yield. The so-obtainedproduct was then condensed with 5-aminopentan-1-ol, thusyielding an asymmetrically-substituted perylenic bisimide with afree alcohol group that was finally esterified with 2,5-dibromobenzoic acid through a N,N0-Dicyclohexylcarbodiimide(DCC)/4-Dimethylaminopyridine (DMAP) assisted condensation ina CH2Cl2/DMF solvent mixture at 30 �C for 5 days. The desiredmonomer 7 was obtained in 70% yield.

For the co-polymerization, after several attempts, it was foundthat the best reaction conditions were achieved firstly addingtoluene as co-solvent and to dissolve 7 and the catalyst as well.Pyridinewas thenadded inorder to start the reaction. Following this,the (cholesteryl) 2,5-diethynyl benzoate dissolved in carefully driedand degassed Et3N was added. An insoluble, high molecular weightfraction was formed during the polymerization (50%). After theprecipitationworkup to eliminate all the catalyst, co-polymer 9wasrecovered by washing with CHCl3. Although the polymer wasscarcely soluble in CHCl3 at r.t., it was soluble in hot CHCl3, tolueneand chlorobenzene. The 1H NMR spectrum of the polymer isdifferent from those of the monomers, all the resonant signals arebroader and no signal was detected in the region of the acetylenicproton (c.a. 3.5 ppm). This shows that after the polycondensationreaction and polymer purification, all traces of unreacted monomeror short oligomers with terminal acetylenic H were removed.

(AcO)2, imidazole, 155 �C, 4 h, iii) KOH, t-BuOH, 155 �C, 30 min, iv) Zn(AcO)2, imidazole,t3N, toluene, 70e80 �C, 36 h.

Fig. 1. Absorption spectra of nPE(Pery-Co-Chol) (solid line), nPEBzChol (dashed line)Pery (dotted line) in toluene. The corresponding chemical structures are included.Inserted figure: fluorescence spectrum of nPE(Pery-Co-Chol) in toluene obtained byexciting on the PE backbone (398 nm, solid line) or on the perylene group (518 nm,dotted line).

Fig. 2. Jablonski diagram describing the energetic levels for the singlet and chargeseparated state of nPE(Pery-Co-Chol).

S. Zappia et al. / Materials Chemistry and Physics 147 (2014) 476e482 479

The absorption spectrum of the co-polymer nPE(Pery-Co-Chol)in toluene is reported in Fig. 1 together with that of the perylenemonomer 5 (hereafter named Pery) as a term of comparison. Thebands of Pery appear at 459, 490 and 526 nm, as reported forother perylene bisimides [17]. Also the nPE(Pery-Co-Chol) spec-trum presents these bands, virtually at the same wavelengths,with the exception of the highest energy peak that is blue-shiftedto 442 nm. It is likely due to overlapping with the pep* electronictransition associated with the conjugated phenyleneethynylenebackbone, that gives rise to a main peak at 408 nm. A phenyl-eneethynylene homopolymer bearing perylene groups as lateralsubstituents could not be obtained as an insoluble materialprecipitated during the polymerization reaction. Co-polymerization with the cholesterol substituted unit increasedsolubility, although the formation of a cotton-like material in so-lution could be observed after some time. In Fig. 1, the spectrum ofthe poly((cholesteryl)benzoateethynylene)) nPEBzChol [8], whoseabsorption maximum is centered at 382 nm, is included as a termof comparison.

It is worth noting that nPE(Pery-Co-Chol) exhibits a large ab-sorption tail extending to the IR region that is absent in the twomodel compounds. With dilution, the spectra (Fig. S1, supportinginfo) show that this tail decreases in intensity but it is persisting upto an optical density of 0.1. This fact indicates that in concentratedsolutions, agglomeration takes place and this should be taken intoconsideration during the preparation of films. On the other hand,solutions with absorbance lower to 0.1 were used for the obtentionof the intrinsic optical properties collected hereafter in Table 1.

The absorption spectra reveal that the first excited state of theperylene group is below that of the phenyleneethynylene moiety,similarly to what it was found for oligo(phenylene vinylene)s

Table 1Spectroscopic data in toluene.

Molecule labs (nm) ε * (nm) 104 (M�1 cm�1)*

nPE(Pery-Co-Chol) 408, 442, 492,528 7.55 (408), 5.80 (528)

Pery 459,490, 526 7.98 (526)nPEBzChol 382 1.89

*Based on the molecular weight of the repeat unit. aWhen excitation wavelength is on P

(OPV)-Pery polymers [13], and is related to singlet energy transferof the electron donor phenyleneethynylene (PE) unit to the electronacceptor perylene as observed in other perylene bearing polymers[13,18].

By exciting in the perylene absorption range (518 nm), theemission spectrum (Fig. 1, insetted) presents almost the same fea-tures as those of Pery (Fig. 1, insetted). When the excitation wave-length is set at the phenyleneethynylene moiety absorptionmaximum (398 nm), an additional, very weak emission from PEconjugated backbone with a maximum at around 465 nm can bealso observed. The intensity of this band is however very lowrelative to the Pery emission, thus the quantumyield f is practicallyidentical at both excitation wavelengths.

Interestingly, f is strongly reduced with respect to bothnPEBzChol and Pery, the fluorescence quenching being 4.9 withrespect to nPEBzChol (QPE) and 16.5 with respect to Pery (QPery).From the fluorescence quantum yields and the lifetimes obtainedby single photon counting (Table 1), the radiative (kr) and nonradiative (knr) constants can be calculated. Both are quenched innPE(Pery-Co-Chol) with respect to nPEBzChol while only the radi-ative constant is lower relative to Pery.

As perylene is a known electron acceptor group and photoin-duced electron transfer has been previously found for OPV-perylene systems, the free energy for intramolecular photoin-duced electron transfer DGCS was estimated by the Weller equation[19] and found to be �2.29 eV, and the energy of the charged-separated state GCS is �0.81 eV relative to the first excited state ofthe phenyleneethynylene (S1PE) and �0.02 eV relative to that of theperylene (S1Pery). Intramolecular photoinduced electron transfer isthus an exergonic process [20]. By considering GCS, S1PE and S1Pery,the Jablonski diagram of Fig. 2 can be constructed.

After photoexcitation on the phenyeleneethynylene moiety, theintrinsic radiative and non radiative decay of this chromophore(depicted with the solid and dotted arrows, respectively) can bedeactivated by singlet energy transfer to the perylene or via charge

lemis (nm) f (%) t (ns) kr 109 (s�1) knr 109 (s�1)

465a 4.7 ± 1a 1.73a 0.027a 0.55a

520, 560 5.2 ± 0.6b 2.50b 0.021b 0.38b

536, 578 86 ± 5 4.07 0.21 0.034408 23 ± 1 0.86 0.27 0.89

E or bon Pery.

Fig. 3. Normalized absorption and excitation spectrum of nPE(Pery-Co-Chol) intoluene.

Table 2Photovoltaic parameters of the prototypes studied in this work.

Active layer Voc (V) Jsc * 10�3

(mA cm�2)FF (%) h * 10�3 (%)

nPE(Pery-Co-Chol) 0.265 6.7 0.215 0.3nPE(Pery-Co-Chol)/PCBM 0.490 9.4 0.268 1.2

S. Zappia et al. / Materials Chemistry and Physics 147 (2014) 476e482480

separation in which an electron is transferred from the conjugatedbackbone to the perylene substituents (PEþPery�). When the per-ylene chromophore is excited, singlet energy transfer to PE isenergetically unfavored as PE singlet state is at higher energy thanthe perylene one. Charge separation is still possible giving rise toquenching of the perylene fluorescence.

In particular, fluorescence quenching can be used to estimatethe rate constants for indirect ðkiCSÞ (from Pery to the chargeseparated state) charge separation and the sum of the rates of en-ergy transfer, kET (from S1PE to S1Pery) by using the followingequations:

kET þ kdCS ¼ QPE � 1tPE

kiCS ¼ QPery � 1tPery

From the values of QPE and QPery previously reported and thelifetimes of Table 1, kdCS þ kET is equal to 4.52 ns�1 and kiCS is3.81 ns�1. The comparison between the normalized absorptionand excitation spectrum (at the Pery emission) (Fig. 3), indicatesthat the PE contribution to the perylene emission is very poor,thus suggesting that the rate for direct electron transfer kdCS from

Fig. 4. Current densityebias curve for solar cells of nPE(Pery-Co-Chol).

the S1PE state is more important than the rate of energy transferkET. In conclusion, these values suggest that the fluorescencequenching is due mainly to both direct and indirect charge stateseparation.

All the previous results and energetic considerations suggest thepossibility of using nPE(Pery-Co-Chol) in solar cell devices. Pro-totypes have been prepared by dip coating. In fact, films preparedby spin coating showed macroscopic defects and therefore solarcells were not fabricated through this technique. The films pre-pared by dip coating seemed quite homogenous at simple eye.Nevertheless, as the prototypes did not photogenerate, AFM anal-ysis was carried out, revealing the presence of big grains in spite ofthe fact that the toluene solution was heated to reduce agglomer-ation, observed in the absorption spectra of concentrated solutions.Solvent was thus changed to chlorobenzene in order to furtherincrease temperature and hence solubility. Although aggregateswere still present, their size was reduced (average diameter size of20 nm vs.140 nm of the film obtained from toluene), and theroughness (in 10 � 10 mm2) decreased from 11.62 to 3.69 nm (seeFig. S3, supporting information).

In this case, the solar cells presented a typical curve (Fig. 4) butthe efficiency hwas low (3 * 10�4%, Table 2). However, the efficiencyincreased by an order of magnitude when nPE(Pery-Co-Chol) wasmixed with PCBM in 1:3 weight percent.

In order to find an explanation for this behavior, tapping andelectrical force microscopy EFM was carried out for the two activelayers on PEDOT:PSS/ITO substrates (Fig. 5). As exciton diffusionlength is shorter than 20 nm, this analysis was carried out on atinier scanning area (1 �1 mm2) than that which Fig. S3 is referringto. In EFM images, darker parts correspond to the more conductingregions. For nPE(Pery-Co-Chol) (left image) the electrical potentialis concentrated in a few, quite isolated regions, while for nPE(Pery-Co-Chol)/PCBM (right image), despite that concentrated conduc-tive regions are still evident, an electrical continuity in the sampleis observed. This is visually evidenced with a contrast pattern thatwas overwritten on the images in correspondence of theconductive points as a guide for the eye. On this basis, we cansuggest that a better charge transport is occurring in the nPE(Pery-Co-Chol)/PCBM sample. This could be due to a better phasemixing.

Fig. 5. EFM images of nPE(Pery-Co-Chol) (left) and nPE(Pery-Co-Chol)/PCBM (right)films deposited on PEDOT:PSS/ITO substrate. The colored contrast is used as guide forthe eye to show the electrical potential distribution.

Fig. 6. Energetic diagram showing the possible photoinduced processes in the pro-totypes studied in this work.

S. Zappia et al. / Materials Chemistry and Physics 147 (2014) 476e482 481

Energetic considerations must be also taken into account. Fromthe electrochemical properties of nPE(Pery-Co-Chol), nPEBzCholand Pery in films (Table S2, supporting information) and the LUMOvalue of PCBM from literature [21] the energy diagram of Fig. 6 canbe drawn.

In general, photoinduced energy transfer can occur throughholes or electrons. In the first case, the difference between theHOMO of the acceptor and that of the electron donor shouldovercome the exciton energy binding that is usually considered tobe 0.4e1 eV range [22]. In the case of the electron photoinducedenergy transfer, the difference between the LUMO values shouldbe higher than 0.4 eV. For the solar cells of pure nPE(Pery-Co-Chol), the electron donor is assumed to be the cholesterol phe-nyleneethynylene moiety and the electron acceptor is Pery. In thiscase, the difference between the HOMO values is 0.27 eV whilethat between the LUMO values is 0.53 eV, so only photoinducedelectron transfer could be possible. For nPE(Pery-Co-Chol)/PCBMsolar cells, the electron donor is now the whole polymer while theelectron acceptor is PCBM. For this configuration, both differences(between HOMO and LUMO values) are around 0.6 eV, sophotoinduced energy transfer can potentially take place in bothways.

4. Conclusions

A perylene cholesterylbenzoateethynylene co-polymer, whoseabsorption spectrum extends from the UV to the visible region,was synthesized by Sonogashira reaction. The energetic consid-erations deduced from the optical and electrochemical resultssuggest the possible application of this co-polymer in solar cells.Though the cholesteryl substituted unit improves the solubilityof the perylenebenzoateethynylene backbone, thus allowing thepreparation of thin films, however the performance of the solarcells is quite low due to the formation of aggregates. Theincorporation of PCBM increases by an order of magnitude theefficiency of the prototype due to: 1) possibility of both electronand hole photoinduced transport, 2) reduction in the agglom-erates size and 3) electrical continuity in the films as observed byEFM. Chemical modifications of the present structure are inprogress to enhance solubility and hence the solar cellsperformance.

Acknowledgments

This workwas financially supported by PRIN 2009 “Materials fororganic and hybrid photovoltaic” prot. 2009PRAM8L_007 and SEP-CONACYT 98513R projects.

Technical support by Pablo Acu~na, Gabriela Padr�on, Silvia Torresand Guadalupe Mendez is acknowledged.

Appendix A. Supplementary data

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.matchemphys.2014.05.017.

References

[1] G. Yu, J. Gao, J.C. Hummelen, F. Wudl, A.J. Heeger, Science 270 (1995)1789e1791.

[2] (a) G. Li, V. Shrotriya, J. Huang, Y. Yao, T. Moriarty, K. Emery, Y. Yang, Nat.Mater. 4 (2005) 864e868;(b) W. Ma, C. Yang, X. Gong, K. Lee, A.J. Heeger, Adv. Funct. Mater. 15 (2005)1617e1622.

[3] J.J.M. Halls, K. Pichler, R.H. Friend, S.C. Moratti, A.B. Holmes, Appl. Phys. Lett. 68(1996) 3120e3122.

[4] (a) J. Hou, Z. Tan, Y. Yan, Y. He, C. Yang, Y. Li, J. Am. Chem. Soc. 128 (2006)4911e4916;(b) Y.F. Li, Y.P. Zou, Adv. Mater. 20 (2008) 2952e2958.

[5] (a) C. Duan, W. Cai, F. Huang, J. Zhang, M. Wang, T. Yang, C. Zhong, X. Gong,Y. Cao, Macromolecules 43 (2010) 5262;(b) C. Duan, K.-S. Chen, F. Huang, H.-L. Yip, S. Liu, J. Zhang, A.K.Y. Jen, Y. Cao,Chem. Mater. 22 (2010) 6444e6452.

[6] (a) S.-L. Hsu, C.-M. Chen, K.-H. Wei, J. Polym. Sci. Part A Polym. Chem. 48(2010) 5126e5134;(b) Y.-J. Cheng, L.-C. Hung, F.-Y. Cao, W.-S. Kao, C.-Y. Chang, C.-S. Hsu, J. Polym.Sci. Part A Polym. Chem. 49 (2011) 1791e1801;(c) D. Sahu, H. Padhy, D. Patra, J.-H. Huang, C.-W. Chu, H.-C. Lin, J. Polym. Sci.Part A Polym. Chem. 48 (2010) 5812e5823;(d) C. Duan, W. Cai, C. Zhong, Y. Li, X. Wang, F. Huang, Y. Cao, J. Polym. Sci. PartA Polym. Chem. 49 (2011) 4406e4415;(e) C. Duan, X. Hu, K.-S. Chen, H.-L. Yip, W. Li, F. Huang, A.K.Y. Jen, Y. Cao, Sol.Energy Mater. Sol. Cells 97 (2012) 50e58.

[7] Z.-G. Zhang, K.-L. Zhang, G. Liu, C.-X. Zhu, K.-G. Neoh, E.-T. Kang, Macromol-ecules 42 (2009) 3104e3111.

[8] G. Castruita, V. García, E. Arias, I. Moggio, R. Ziolo, A. Ponce, V. Gonz�alez,J.E. Haley, J.L. Flikkema, T. Cooper, J. Mater. Chem. 22 (2012) 3770e3780.

[9] See for instance the following recent reviews and references therein. (a)U.H.F. Bunz, in: T.A. Skotheim, J.R. Reynolds (Eds.), Conjugated Polymers.Theory, Synthesis, Properties and Characterization, CRC Press, Boca Raton,2007, pp. 1e51 (Chapter 6);(b) I. Moggio, E. Arias, in: A.H. Lima (Ed.), Biomimetic and SupramolecularSystems Research, Nova Science Publishers Inc., New York, 2008, pp.179e212;(c) S.W. Thomas, G.D. Joly, T.M. Swager, Chem. Rev. 107 (2007) 1339e1386.

[10] J. Mao, Q. Liu, S. Wang, X. Lv, Y. Huang, Y. Ma, Y. Chen, S. Yin, J. Nanosci.Nanotechnol. 8 (2008) 3343e3350.

[11] (a) C.W. Tang, Appl. Phys. Lett. 48 (1986) 183e185;(b) J.J.M. Halls, R.H. Friend, Synth. Met. 85 (1997) 1307e1308;(c) L. Schmidt-Mende, A. Fechtenk€otter, K. Müllen, E. Moons, R.H. Friend,J.D. MacKenzie, Science 293 (2001) 1119e1122;(d) J.J. Dittmer, E.A. Marseglia, R.H. Friend, Adv. Mater. 12 (2000) 1270e1274;(e) J.J. Dittmer, K. Petritsch, E.A. Marseglia, R.H. Friend, H. Rost, A.B. Holmes,Synth. Met. 102 (1999) 879e880.

[12] C. Li, H. Wonneberger, Adv. Mater. 24 (2012) 613e636.[13] (a) E.E. Neuteboom, S.C.J. Meskers, P.A. van Hal, J.K. van Duren, E.W. Meijer,

R.A.J. Janssen, H. Dupin, G. Pourtois, J. Cornil, R. Lazzaroni, J.-L. Br�edas,D. Beljonne, J. Am. Chem. Soc. 125 (2003) 8625e8638;(b) A. Marcos Ramos, S.C.J. Meskers, E.H.A. Beckers, R.B. Prince, L. Brunsveld,R.A.J. Janssen, J. Am. Chem. Soc. 126 (2004) 9630e9644.

[14] (a) L.J. Bu, X.Y. Guo, B. Yu, Y. Qu, Z.Y. Xie, D.H. Yan, Y.H. Geng, F.S. Wang, J. Am.Chem. Soc. 131 (2009) 13242e13243;(b) Z.A. Tan, E.J. Zhou, X.W. Zhan, X. Wang, Y.F. Li, S. Barlow, S.R. Marder, Appl.Phys. Lett. 93 (2008) 073309;(c) X.W. Zhan, Z.A. Tan, B. Domercq, Z.S. An, X. Zhang, S. Barlow, Y.F. Li,D.B. Zhu, B. Kippelen, S.R. Marder, J. Am. Chem. Soc. 129 (2007) 7246e7247;(d) X.W. Zhan, Z.A. Tan, E.J. Zhou, Y.F. Li, R. Misra, A. Grant, B. Domercq,X.H. Zhang, Z.S. An, X. Zhang, S. Barlow, B. Kippelen, S.R. Marder, J. Mater.Chem. 19 (2009) 5794e5803.

[15] A.T.R. Williams, S.A. Winfield, J.N. Miller, Analyst 108 (1983) 1067e1071.[16] (a) N. Soh, T. Ariyoshi, T. Fukaminato, H. Nakajima, K. Nakano, T. Imato, Org.

Biomol. Chem. 5 (2007) 3762e3768;

S. Zappia et al. / Materials Chemistry and Physics 147 (2014) 476e482482

(b) M.W. Holman, R. Liu, D.M. Adams, J. Am. Chem. Soc. 125 (2003)12649e12654.

[17] J.J.M. Halls, R.H. Friend, Synth. Met. 85 (1997) 1307e1308.[18] (a) L.M. Herz, C. Silva, R.H. Friend, R.T. Phillips, S. Setayesh, S. Becker,

D. Marsitsky, K. Müllen, Phys. Rev. B 64 (2001) 195203;(b) D. Beljonne, G. Pourtois, C. Silva, C.E. Hennebicq, L.M. Herz, R.H. Friend,G.D. Scholes, S. Setayesh, K. Mullen, J.L. Br�edas, Proc. Natl. Acad. Sci. 99 (2002)10982e10987.

[19] A. Weller, Z. Phys. Chem. Neue Folge 133 (1982) 93e98.[20] All the parameters used for this calculus are reported in supporting

information.[21] W. Li, T. Lee, S.J. Oh, C.R. Kagan, ACS Appl. Mater. Interfaces 3 (2011)

3874e3883.[22] X.Y. Zhu, Q. Yang, M. Muntwiler, Acc. Chem. Res. 42 (2009) 1779e1787.