European Polymer Journal 65 (2015) 4–14

MA

CRO

MO

LECU

LAR

NA

NO

TECH

NO

LOG

Y

Contents lists available at ScienceDirect

European Polymer Journal

journal homepage: www.elsevier .com/locate /europol j

Feature Article

Polymer compatibilized self-assembling perylene derivatives

http://dx.doi.org/10.1016/j.eurpolymj.2014.11.0220014-3057/� 2015 Published by Elsevier Ltd.This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

⇑ Corresponding author.E-mail address: Sundar@Carleton.ca (P.R. Sundararajan).

Elianne Dahan, Pudupadi R. Sundararajan ⇑Department of Chemistry, Carleton University, 1125 Colonel By Drive, Ottawa, Ontario K1S 5B6, Canada

a r t i c l e i n f o

Article history:Received 20 October 2014Received in revised form 13 November 2014Accepted 16 November 2014Available online 27 November 2014

Keywords:Fluorescent gelNanostructurePerylene tetracarboxy diimidePolystyrenePoly (dimethyl siloxane)Phase separation

a b s t r a c t

Solution coating of functional small molecules in polymer films is a convenient approach tofabricating flexible optoelectronic and photo-functional devices. Uniform dispersion of thesmall molecule in the polymer and inhibition of aggregation are requirements for therobust functioning of the devices. Alternate approaches such as incorporating the func-tional molecule as part of the polymer chain have been examined, although such methodslimit the flexibility in materials choice. While perylene or perylene tetracarboxy diimide(PTCDI) or perylene imide (PI) phase separate into discrete crystals in polymer films, wefind that functionalizing the perylene imide with a polymer or oligomer segment that iscompatible with the host polymer matrix results in highly uniform dispersion of the smallmolecule, with the inherent photophysical properties of the perylene segment unaffected.We demonstrate this approach with oligostyrene–PTCDI–oligostyrene dispersed inpolystyrene, PDMS–PI and PDMS–PTCDI–PDMS in PDMS in solution cast films and intwo-component gels from organic solvents. Fluorescent gels of polystyrene and PDMS(without crosslinks or functionalization) were obtained via this route.

� 2015 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-NDlicense (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42. Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63. Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

3.1. Morphology of the composite films . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73.2. Optical properties of the composite films . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93.3. Composite gels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103.4. Morphology of the composite gels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103.5. UV–Vis and fluorescence spectra of the composite gels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

4. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13Appendix A. Supplementary material. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

1. Introduction

Dispersing functional small molecules in polymer matri-ces is a common approach to fabricating organic-based

E. Dahan, P.R. Sundararajan / European Polymer Journal 65 (2015) 4–14 5

CULA

RN

AN

OTE

CHN

OLO

GY

flexible devices, the earliest examples being the repro-graphic photoreceptors and polymer-dispersed liquidcrystal displays [1–6]. Solution coating of functional multi-layers using such compositions are preferred for flexibledevices. The relative concentration of the functionalmolecule often has to be as high as 50 wt% in the polymer.For example, Santerre et al. [7] reported the properties ofOLED devices based on N,N0-diphenyl-N,N0-bis(3-methyl-phenyl)-[1,10-biphenyl]-4,40-diamine (TPD) dispersed inpolymers with high Tg. The best performance was obtainedwith a TPD concentration of 75%. Detailed studies havebeen reported on the morphologies resulting from phaseseparation of three different charge transport moleculesin polycarbonates and polystyrene [8–11]. Molecular dis-persion of the small molecule is essential and any phaseseparation and crystallization would lead to degradationof the device performance. Smith et al. [12] have shownthat crystallization of TPD was the cause of delaminationof an OLED device. Scharfe [13] discussed the effect of suchcrystallization on charge trapping in photoreceptors. Theleaching of the transport molecule would be more preva-lent in copiers and printers using liquid developers.

As an alternate to dispersing the functional small mole-cule in polymer matrices, Limburg et al. [14] and Ong et al.[15] designed arylamine based polymers and copolymersby incorporating the photoactive molecules as part of thechain to prevent phase separation. Design and properties ofseveral such polymers with arylamine functionality in themain chain or side chain have been studied [16–18]. Conju-gated polymers for solar cell applications have also beenexamined [19,20]. All-polymer bulk hetero-junction solarcells using blends of polymers bearing donor and acceptormoieties have been reported by Jenekhe et al. [21]. Theadvantage of using a functional molecule dispersion in a hostpolymer is that either can be changed at will, whereas a spe-cific polymer with the functional segment cannot.

Fig. 1. OM of solvent cast films of (a) perylene/polystyrene (5/95 wt%), (b) PTpolycarbonate (5/95) and (e) perylene/PMMA (2/98). Tetrachloroethylene was u

Perylene and its derivatives have been investigated overthe past few decades for their applications in optoelec-tronic and photovoltaic devices [22,23]. The classic crystal-lographic investigations by Hädicke and Gracer [24]related the effect of substituents on the p-overlap andthe color. The self-assembly facilitated by the p-interactioncan be modulated by substitutions at the imide nitrogen aswell as the bay positions [25–27]. Linear and dove-tailsubstitutions have been used to create nano-fiber mor-phologies [28,29]. Supramolecular association betweenmelamine functionalized perylene tetracarboxylic diimide(PTCDI) and cyanuric acid via hydrogen bonding resultedin nano-ribbon and nano-rope morphologies [30]. Theapplications of perylene imides and diimides in solar celland organic electronics have been summarized in recentreviews [31,32].

When perylene (or PTCDI) by itself was dispersed in apolymer matrix, the small molecule formed discrete crys-tals of a few microns as shown in Fig. 1, irrespective ofthe solvent used for casting the films. We discussed theself-assembly of PTCDI substituted with oligostyrene onboth imide nitrogens (PS–PTCDI–PS, Scheme 1a) in solution[33] as well as in the gel state [34]. In another study, wesubstituted poly (dimethyl siloxane) (PDMS) on both imidenitrogens (Di-PDMS) (Scheme 1b) or on one (Mono-PDMS)(Scheme 1c) and reported on the differences in the mor-phology of the mono and di-substituted PTCDI in solution[35,36] and gels [37]. Both PS–PTCDI–PS and Di-PDMS arenon-ionic Gemini surfactants. We believed that by attach-ing an oligomeric or polymeric chain which is compatiblewith the host polymer, the mixing of such compatibilizedPTCDI would be better, and a uniformly dispersedcomposite film could be obtained. For example, with oligo-styrene attached to PTCDI, we expect that a good dispersionin polystyrene (Scheme 1d) could be achieved. In the pres-ent work, we dispersed PS–PTCDI–PS in the corresponding

CDI/polystyrene (5/95), (c) perylene/polycarbonate (2/98), (d) perylene/sed for (e) and chloroform for the others.

MA

CRO

MO

LE

Scheme 1. (a) Polystyrene–PTCDI–polystyrene Gemini surfactant, (b) PDMS–PTCDI–PDMS Gemini surfactant, (c) inverse macromolecular surfactant PDI–PDMS, (d) polystyrene and (e) PDMS.

6 E. Dahan, P.R. Sundararajan / European Polymer Journal 65 (2015) 4–14

MA

CRO

MO

LECU

LAR

NA

NO

TECH

NO

LOG

Y

polymer, polystyrene (PS), in an attempt to make compos-ite polymer films. We also prepared gels of PS–PTCDI–PSrecently, with trans-decalin [34]. Since this solvent hasbeen used in the past for gelation involving polystyrene,we prepared two-component composite gels with differentconcentration of PS–PTCDI–PS in polystyrene. In our previ-ous work, we used propylamine for gelation of Mono- andDi-PDMS [37]. This solvent was also used for gelation withPDMS without having to crosslink or functionalize thepolymer [38]. In the present work, we prepared PTCDIdispersed gels of PDMS with different concentrations ofMono- or Di-PDMS. The optical properties of the self-assembled polymer-compatibilized PTCDI remained thesame upon dispersion in the corresponding polymer matrices.

2. Experimental

The syntheses of Mono-PDMS and Di-PDMS have beendescribed in our previous publications [35,36]. PS–PTCDI–PS was synthesized as described by Islam and Sundararajan[33]. Poly (dimethylsiloxane) (Mw = 182,600; Mn = 106,000,CASRN 9016-00-6), polystyrene (Mw = 239,700 and

Mn = 119,600) and all the solvents (propylamine, chloro-benzene, THF, or chloroform of laboratory grade) were pur-chased from Aldrich Chemical Company. Solvent cast filmswere prepared by dissolving appropriate mixtures ofPS–PTCDI–PS and PS in chlorobenzene, THF, or chloroform.Relative concentrations of 1, 5 or 10 wt% of PS–PTCDI–PS inPS were used. Films were coated on a glass substrate usingan electrically driven film coater and were dried at avery low rate of the solvent evaporation at ambient condi-tions for 24 h and then under vacuum for 24 h. The finalthickness of the films was about 20 lm.

Gelation studies on PS–PTCDI–PS (a Gemini surfactant)were discussed in our previous publication [34]. In thepresent work, PS/PS–PTCDI–PS composite gels were pre-pared with 2–10 wt% of PS–PTCDI–PS in PS. A total soluteconcentration of 0.8 mM was dissolved in trans-decalinat temperatures ranging from 78 to 82 �C with constantstirring. This resulted in a red solution. Closed vials wereused to avoid the evaporation of the solvent. The gels wereprepared by slow cooling the solution and as a quick test,gelation was confirmed by tube inversion, i.e., no flowingsolvent.

E. Dahan, P.R. Sundararajan / European Polymer Journal 65 (2015) 4–14 7

LOG

Y

We reported the gelation with PDMS before, withoutany crosslinks or functionalization [38], as well as the gela-tion of Mono-PDMS and Di-PDMS [37]. In the present study,a solute concentration of 0.1 M (18.26 g/L) in propylaminewas used, with 2%, 5%, or 10 wt% Mono-PDMS and 5%,10 wt% Di-PDMS in PDMS. Solutions were prepared withconstant stirring at temperatures ranging from 65 to75 �C. Gels were then prepared by slow cooling to roomtemperature by turning off the hot plate. Gels were pre-pared in closed vials to avoid the evaporation of the solvent.Gelation was tested by tube inversion with no flowingsolvent.

The optical micrographs (OM) were recorded using aZeiss Axioplan polarized optical microscope in transmis-sion mode. Northern Eclipse (version 6.0 and 8.0) imageprocessing software was used to record the images. Scan-ning electron microscopy (SEM) images were obtainedusing a VEGAII XMU (TESCAN, Czech Republic) scanningelectron microscope. Dry samples were sputter coatedwith 80:20 Au/Pd target using a Hummer VIII SputteringSystem (Anatech Ltd., Alexandria, VA). Thermal analysiswas performed using a TA Instruments 2010 differentialscanning calorimeter at 10 �C/min heating rate. The instru-ment was calibrated for temperature and energy with

Fig. 2. OM images of PS/PS–PTCDI–PS films (top: 1 wt% PS–PTCDI–PS) and (5 wtTHF. Spin cast films with PS/PS–PTCDI–PS (95/5 wt%) in (g) chloroform and (h)

indium and tin as certified reference materials. DSC tracesfor films were recorded with about 8 mg of the samplesunder the flow of nitrogen. UV–visible absorption spectrawere recorded using a Varian CARY 3 UV–Vis spectropho-tometer. The data were processed with CARY WinUV Soft-ware version 3.00. The fluorescent emission data werecollected using a Varian CARY Fluorescence spectropho-tometer at the excitation wavelengths (kex) of 460 nmand 417 nm with a bandwidth of 5 nm for excitation and5 nm for emission. Data collection and processing weredone by Eclipse WinFLR Software (version 1.1).

3. Results and discussion

3.1. Morphology of the composite films

While perylene, without any substitution, would formcrystals, attaching short chains to perylene mono- or dii-mide leads to vesicular, nano-web or nanowire morphologywhen precipitated from solution or during gelation. Fig. 1shows the optical micrographs (OM) of films cast from chlo-roform or TCE solutions of polystyrene, polycarbonate orPMMA with low concentrations of perylene or PTCDI. Evenin the presence of the polymers, large, discrete crystals of

% PS–PTCDI–PS) cast from (a,d) chlorobenzene, (b,e) chloroform and (c, f)THF are also shown.

MA

CRO

MO

LECU

LAR

NA

NO

TECH

NO

Fig. 3. SEM images of films with 5 wt% PS–PTCDI–PS (a–c) and 10 wt% PS–PTCDI–PS (d,e) using (a) chlorobenzene, (b) chloroform and (c) THF, (d)chlorobenzene, (e) chloroform.

8 E. Dahan, P.R. Sundararajan / European Polymer Journal 65 (2015) 4–14

MA

CRO

MO

LECU

LAR

NA

NO

TECH

NO

LOG

Y

the small molecules are seen in all these cases. Fig. 2 showsthe OM of films of PS–PTCDI–PS in polystyrene, cast fromthree different solvents. Spherical domains, with almostuniform size are seen with films cast from chlorobenzeneand chloroform. Especially with a 5 wt% concentration inchloroform (Fig. 2(e)), dense spheres with uniform size areseen. Such close packed morphology ensures that percola-tion threshold has been reached and would facilitate chargehopping. The films made with THF, however, do not showdiscrete domains (see Optical Properties below). The lengthof the oligostyrene chain attached to PTCDI in this case isonly about 12 units. The molecular weight of the polysty-rene matrix is over 100,000. With the oligostyrene attach-ment, single crystals of perylene is not seen in thepolystyrene matrix. We have shown previously thatdrop-cast films of PS–PTCDI–PS show vesicular and fusedvesicular morphologies [33]. Such spherical morphology ismaintained when this Gemini surfactant is dispersed inpolystyrene. Fig. 2(g) and (h) shows spin cast films withchloroform and THF. A highly uniform spherical morphol-ogy is seen in these films. With the oligostyrene segmentdispersed in the polystyrene matrix, the domain formationis due to the self-assembly of PTCDI. Thus it is a supramolec-ular solution.

Fig. 3 shows the SEM images of films with 5 wt%PS–PTCDI–PS in polystyrene. No domains similar to thosein Fig. 2(a), (b), (d) and (e) are seen in Fig. 3(a) and (b)which would indicate that the domains seen in Fig. 2 are

sub-surface. We have reported such a phenomenon beforein the case of a functionalized phthalocyanine dispersed inpolycarbonate or PMMA [39], as well as biscarbamates dis-persed in polycarbonate [40]. In another study [41], wedescribed a polymer dispersed self-assembling small mol-ecule system, in which a homologous series of carbamates,with a hydrogen-bonding moiety and alkyl side chains,was dispersed in polycarbonate. These self-assemblingmolecules formed colloidal size domains in the polymer,which involved a hierarchy of three levels of assembly.When a film is cast from solution, the self-assembly ofthese small molecules is so rapid that they form domainsin the bulk of the polymer during solvent evaporationand do not diffuse to the surface of the polymer film. Dueto the sub-surface assembly, no domains are seen in theSEM images, although their presence is seen in the trans-mission OM. Fig. 3(c) shows that with THF, sphericaldomains of less than a micron are seen on the surface aswell. The SEM images of films with a higher concentrationof 10 wt% PS–PTCDI–PS (Fig. 3(d) and (e)) show domainson the surface as well, with those in Fig. 3(e) being of nano-meter size. The highly monodisperse size of these domainsshow that spontaneous nucleation occurs due to self-assembly in the polymer matrix. Fig. 4 shows the photo-graphs of the polystyrene films with 1, 5 and 10 wt% PTCDI.Consistent with the OM observations, the film is transpar-ent with 1 and 5 wt%, becoming opaque with higherconcentration.

Fig. 4. Transparency of the PS/PS–PTCDI–PS films. L to R: 1, 5 and 10 wt% PTCDI in polystyrene, cast from chloroform.

E. Dahan, P.R. Sundararajan / European Polymer Journal 65 (2015) 4–14 9

ROM

OLE

CULA

RN

AN

OTE

CHN

OLO

GY

The sub-surface self-assembly of the small molecules inthe polymer matrix has another advantage. Typically, thecharge transport layer (CTL) of a photoreceptor containsthe charge transport molecule (CTM) such as TBD, to a load-ing of about 50%. Such a high loading of CTM essentiallyleads to a metastable composite. Abrasion due to brush-cleaning of the photoreceptor after each copy/print cycle,leaching due to the use of liquid developer etc., result inthe loss of the charge transport molecule (CTM) whichwould change the composition of the active layer and henceits performance. Over-coating the CTL with abrasion-resis-tant layers (such as siloxane) has been adopted as a route toreduce the loss of the CTM [42]. Sundararajan et al. [43]showed that introducing cyclodextrins in the CTL to formpolycarbonate/cyclodextrin rotaxanes reduced the abra-sion significantly. The use of a self-assembling moleculesuch as PS–PTCDI–PS obviates the need for such add-oncoatings, since the active molecule is assembled sub-sur-face in the polymer matrix. An optimum concentration isneeded as well for such sub-surface morphology sincebeyond a certain concentration of the small molecule,domains are formed on the surface as well, as seen inFig. 3(d) and (e). In spite of the domains seen on the surfaceof the films, note that these are highly monodisperse in size(nanometers) in the case of films cast from chloroform(Fig. 3(e)).

To address the question of whether a compatible oligo-mer is indeed required as a side chain to form such uniformdomains described above, we dispersed Mono-PDMS and

Fig. 5. The variation of glass transition temperature of polystyrene withvarious concentrations of PS–PTCDI–PS using three different solvents.

Di-PDMS in polystyrene. As the PDMS segment and poly-styrene are not miscible, the morphology shown in Fig. S1(Supplementary data) was obtained. The dagger-like mor-phology in the OM shown in Fig. S1a is due to the fusionof the vesicles of Mono-PDMS is the polystyrene matrix,as seen in the SEM image in Fig. S1b.

As expected, the glass transition temperature (Tg) ofpolystyrene was depressed as seen in Fig. 5. Although theTg was reduced from 104 �C to about 80 �C with 10 wt% ofPS–PTCDI–PS with the three solvents, the rate is different.With the films made with chlorobenzene and chloroform,the Tg is reduced only by about 4 �C with 5 wt% PS–PTCDI–PS and the decrease is more pronounced thereafter.However, with THF the decrease is linear, with the Tg reduc-ing by 14 �C, to 90 �C with 5 wt% of PS–PTCDI–PS. The dif-ferences in the morphology and the thermal propertybetween the films made with the three solvents do notrelate to the solubility parameter. However, it should benoted that the viscosity of THF is 0.48 cP, while that of chlo-robenzene and chloroform are 0.75 and 0.56 cP, respec-tively. The effect of the viscosity of the solvent on theresulting morphology during self-assembly was discussedbefore in our work on phthalocyanine aggregation in poly-carbonate and PMMA matrices [39].

MA

C

3.2. Optical properties of the composite films

In our previous studies [37] on the self-assembly ofPS–PTCDI–PS in solution, the monomeric form of this com-pound in chloroform showed three distinct peaks at 456,489 and 526 nm in the UV–Vis spectra, corresponding tothe S0–2, S0–1 and S0–0 transitions respectively. TheUV–Vis spectra in Fig. 6 shows that PS–PTCDI–PS in thefilm with PS cast from chloroform is in aggregate form,with peaks at 498 and 532 nm, which are red-shifted1 withrespect to the solution spectrum. The films from chloroben-zene also showed a red shit to 494 nm and 528 nm. Inaddition, less intense peaks are seen in both Fig. 6(a) and(b) at 564 and 589 nm with chlorobenzene and at 568 and590 nm with chloroform, which could be assigned to theaggregated state of PS–PTCDI–PS. However, when THF wasused as the solvent, the UV–Vis spectrum (Fig. 6(c)) showedthe monomeric form of PS–PTCDI–PS. The film drop castfrom THF showed a major peak at 526 nm correspondingto the S0–0 vibronic transition, a second peak at 491 nm cor-responding to the S0–1 vibronic transition, and a hump at

1 For interpretation of color in Fig. 6, the reader is referred to the webversion of this article.

Fig. 6. UV–Vis absorption spectra of films drop cast from different solvents at different PS–PTCDI–PS concentrations.

10 E. Dahan, P.R. Sundararajan / European Polymer Journal 65 (2015) 4–14

MA

CRO

MO

LECU

LAR

NA

NO

TECH

NO

LOG

Y

461 nm corresponding to the S0–2 vibronic transition. Theabsorption spectra for the films cast with chlorobenzeneand chloroform are broader which would indicate strongeraggregation than that cast with THF. As discussed above,there is a significant difference in the morphologies of thefilms cast from the two chlorinated solvents and THF. Thelatter did not show discrete domains in Fig. 2.

3.3. Composite gels

A recent review by Ajayaghosh et al. [44] summarizedthe various p-system gels and their possible applications.As we mentioned in our recent publication, there have beena few studies on perylene diimide and phthalocyanine basedgels [26,45–49]. The poor integrity and mechanical proper-ties of these gels limit their applications. Incorporating themin polymer matrices and fabricating two-component gelsmight be a route to overcome these limitations. Such gelsin which the two components either do not have any specificinteractions or mutually complex are known [50–54]. In thepresent work, both polystyrene and PDMS are known toform gels. We recently discussed the thermo-reversiblegelation of PS–PTCDI–PS, mono-PDMS and Di-PDMS. Hence,

we prepared the composite gels, in which the PTCDI bearsthe compatible polymer segment in which it is incorporated.Note that in contrast to the work of others, the three gelatorsdo not have C@O and NAH type hydrogen bonding groups intheir structure.

The dissolution and gelation onset temperatures of thedifferent composite gels are given in Table 1. The tableshows the effect of concentration of the oligomer-function-alized perylene in the corresponding polymer on the tem-peratures of dissolution and the onset of gelation. Forexample with the PS/PS–PTCDI–PS composite gels thatwere formed with trans-decalin at 5% PS–PTCDI–PS, a redsolution was obtained at 82 �C. When the solution wasslow cooled, a deep red-colored, opaque and immobilegel began to form at 57 �C. However, when the concentra-tion of PS–PTCDI–PS increases to 10 wt% PS–PTCDI–PS thegelation onset of gelation drops to 50 �C.

3.4. Morphology of the composite gels

We previously discussed that the morphology of PDMSgels (without crosslinks or functionalization) consistedof interconnected spherical domains, the Mono-PDMS gel

Table 1The dissolution temperatures and onset of gelation of the different composite gels.

Composite gel Solvent %wt Concentration of oligomerfunctionalized PTCDI (%)

Dissolution temperature (�C) Onset of gelation (�C)

PS/PS–PTCDI–PS Trans-decalin 2 78 53PS/PS–PTCDI–PS Trans-decalin 5 82 57PS/PS–PTCDI–PS Trans-decalin 10 79 50PDMS/Mono-PDMS Propylamine 2 65 55PDMS/Mono-PDMS Propylamine 5 68 59PDMS/Mono-PDMS Propylamine 10 66 58PDMS/Di-PDMS Propylamine 5 65 55PDMS/Di-PDMS Propylamine 10 63 53

E. Dahan, P.R. Sundararajan / European Polymer Journal 65 (2015) 4–14 11

RN

AN

OTE

CHN

OLO

GY

forms vesicular morphology and the Di-PDMS gel formsnano-fibers. Fig. 7 shows the morphology of the compositegels with PDMS/Mono-PDMS and PDMS/Di-PDMS.Fig. 7(a)–(c) shows that the spherical/vesicular individualmorphologies of PDMS and Mono-PDMS are maintainedin the composite gels. Both the Mono-PDMS and the Di-PDMS self-assemble in the PDMS matrix, forming uni-formly distributed spheres for Mono-PDMS and fibers forDi-PDMS. The spheres in the background of the gel blendsare due to the PDMS which has also been known to gel inpropylamine [38]. Note that the distribution of theMono-PDMS spheres and Di-PDMS fibers are fairly uniformin the polymer matrix, considering that the relativeconcentration of the Mono- or Di-PDMS is no more than10 wt% in the polymer. Fig 7(g) and (i) shows the photo-graph of the fluorescent PDMS gel, with Mono or Di-PDMSincorporated.

The morphology in the case of PS/PS–PTCDI–PS withtrans-decalin gel is shown in Fig. 8. As we discussed before,PS–PTCDI–PS forms fibrillar morphology in trans-decalin

Fig. 7. OM images of gels of PDMS with (a) 2%, (b) 5%, (c) 10% Mono-PDMS anPDMS + 5 wt% MonoPDMS gel, (g) emission color (kEx: 470 nm), (h) PDMS + 5 wtreferences to color in this figure legend, the reader is referred to the web versio

gels, and so does polystyrene. The morphology of the com-posite gel consists of inter-meshed fibers of polystyreneand PS–PTCDI–PS, as seen in Fig. 8(c). These fibers arefew hundred microns long and also fold along their lengthto resemble an eaves trough, as seen in the higher magni-fication images in Fig. 8(d) and (e). This behavior was dis-cussed in our previous work [34]. Fig. 8(g) shows thefluorescent polystyrene gel, with PS–PTCDI–PS in it.

3.5. UV–Vis and fluorescence spectra of the composite gels

The absorption and fluorescence spectra of the PS gelswith 2%, 5% and 10% PS–PTCDI–PS are shown in Fig. 9(a)and (b). The absorption spectrum of the gel with 2%PS–PTCDI–PS shows peaks at 500 and 535 nm with intensityratio I0–0/I0–1 of 0.97. There is a slight red shift of the of theS0–0 peak, in the absorption spectrum with increasing con-centration of PS–PTCDI–PS to 5% (503 nm, 546 nm) and10% (504 nm, 548 nm) and the S0–0 peak increases in inten-sity significantly. Such a change in the relative intensity of

d (d) 5%, (e) 10% Di-PDMS. Photographs are shown for the following: (f)% DiPDMS gel, (i) emission color (kEx: 497 nm). (For interpretation of then of this article.)

MA

CRO

MO

LECU

LA

Fig. 8. OM of polystyrene gels with (a) 5% and (b) 10% PS–PTCDI–PS; (c–e) SEM images of polystyrene gels with (c) 2%, (d) 5%, (e) 10% PS–PTCDI–PS. (f) and(g) show the color of the gel and the emission color (kEx: 503 nm), respectively. (For interpretation of the references to color in this figure legend, the readeris referred to the web version of this article.)

Fig. 9. (a) UV–Vis and (b) fluorescence spectra of PS with 2%, 5%, or 10% PS–PTCDI–PS.

12 E. Dahan, P.R. Sundararajan / European Polymer Journal 65 (2015) 4–14

MA

CRO

MO

LECU

LAR

NA

NO

TECH

NO

LOG

Y

the S0–0 peak with concentration was seen with the gels ofPS–PTCDI–PS by itself [34]. Fig 9(b) shows quenching ofthe fluorescence with increasing PS–PTCDI–PS concentra-tion. This would indicate a higher order aggregation processwith higher concentration.

The absorption spectra (not shown here) of the PDMS/5% Mono-PDMS gels showed peaks at 470, 495 and ashoulder at 545 nm. This is nearly identical to the spectraof pure Mono-PDMS gels in propylamine. The quenchingof fluorescence intensity indicated stacking of the peryleneunits in the gel state.

4. Conclusions

We have shown that dispersing oligomer functionalizedperylene imides in the corresponding polymers gives riseto organized domains, without any change in the opticalproperties. While perylene or perylene diimide by itselfwill form single-crystal-like morphology when dispersedin polymers, the oligomer attached to the imide nitrogen(s)acts as a compatibilizer. Typically, concentration of 50 wt%or more of the photo-conducting molecule (such as TPD) inthe polymer matrix is required for efficient charge

E. Dahan, P.R. Sundararajan / European Polymer Journal 65 (2015) 4–14 13

NO

TECH

NO

LOG

Y

transport. However, both self-assembly and percolationthreshold were achieved with less than 10 wt% of the oli-gomer functionalized PTCDI in the corresponding poly-mers, while maintaining the transparency of thecomposite film. The concept of photoconductors with con-trolled aggregation was discussed by Dulmage et al. [55]almost 36 years ago. In this paper we have shown thatpolymer-compatibilized PTCDI enables creation of suchuniform morphology of self-assembled PTCDI segments.It is also seen that the self-assembled structures developin the bulk of the polymer matrix, and are not seen onthe surface. This would be an advantage for flexible poly-mer based devices. Although nano-wires and similar self-assembled morphologies have been reported with dove-tail or hydrophilic/hydrophobic asymmetric substitutions,the work described here involves compatibilization withcommon, commodity polymers as matrices. The compositegels discussed here show that the oligomer functionalizedphotoactive molecules in the polymer based gels retaintheir optical (aggregation) behavior and these could beeventually candidates for fluidic photoactive applications.With the gels also, we used common polymers: polysty-rene with a Tg of 100 �C and PDMS with a Tg of �100 �C.

Acknowledgements

Financial support from the Natural Sciences and Engi-neering Research Council of Canada is gratefully acknowl-edged. We thank Dr. M.R. Islam of our group for the OMshown in Fig. 1(c)–(e).

MA

CRO

MO

LECU

LAR

NA

Appendix A. Supplementary material

Supplementary data associated with this article can befound, in the online version, at http://dx.doi.org/10.1016/j.eurpolymj.2014.11.022.

References

[1] Borsenberger PM, Weiss DS. Organic photoreceptors forxerography. New York: Marcel Decker Inc.; 1998.

[2] Stolka M, Yanus JF, Pai DM. Hole transport in solid solutions of adiamine in polycarbonate. J Phys Chem 1984;88:4707–14.

[3] Borsenberger PM. Hole transport in tri-para-tolylamine-dopedbisphenol-A-polycarbonate. J Appl Phys 1990;68:6263–73.

[4] Lin LB, Jenekhe SA, Borsenberger PM. High electron mobility inbipolar composites of organic molecules. Appl Phys Lett1996;69:3495–7.

[5] Zmija J, Klosowicz SJ, Raszewski Z. Polymer-dispersed liquid-crystaldisplays: properties and applications. Opto-Electron Rev1993;4:130–4.

[6] Zmija J, Klosowicz SJ, Raszewski Z. Polymer-dispersed liquid-crystaldisplays: principles, preparation, properties and applications. ProcSPIE, Liquid Solid State Cryst: Phys Technol Appl 1993;1845:16–28.

[7] Santerre F, Bedja I, Dodelet JP, Sun Y, Lu J, Hay AS, et al. Hole transportmolecules in high Tg polymers: their effect on the performance oforganic light-emitting diodes. Chem Mater 2001;13:1739–45.

[8] Khan F, Sundararajan PR. Influence of polycarbonate flexibility onthe annealing-induced phase separation of the hole transportmolecule TPD in a model charge transport composite. J Phys ChemB 2004;108:117–26.

[9] Khan F, Hor AM, Sundararajan PR. Morphological reasoning for theenhanced charge carrier mobility of a hole transport molecule inpolystyrene. Pure Appl Chem 2004;76:1467–78.

[10] Khan F, Hor AM, Sundararajan PR. Morphological aspects andthermal behavior of tri-p-tolylamine (TTA) based model charge

transport composites: a comparison with TPD. Synth Met2005;150:199–211.

[11] Khan F, Khanna S, Hor A, Sundararajan PR. The role of molecular volumeand the shape of the hole transport molecule in the morphology ofmodel charge transport composites. Can J Chem 2010;88:247–59.

[12] Smith PF, Gerroir P, Xie S, Hor AM, Popovic Z. Degradation of organicelectroluminescent devices. Evidence for the occurrence ofspherulitic crystallization in the hole transport layer. Langmuir1998;14:5946–50.

[13] Scharfe M. Electrophotography principles and optimization. NewYork: John Wiley & Sons; 1984.

[14] Frechet JMJ, Gauthier S, Limburg WW, Loutfy RO, Murti DK, SpiewakJW. Polyarylamine compounds and systems utilizing polyarylaminecompounds. US Patent 4,801,517; 1989.

[15] Ong BS, Keoshkerian B, Baranyi G, Murti DK. Photoconductiveimaging members with diaryl biarylamine copolymer chargetransport layers. US Patent 4, 959, 288; 1990.

[16] Liu Y, Liu MS, Li XC, Jen AKY. Synthesis and characterization of anovel light-emitting polymer containing highly efficient hole-transporting aromatic diamine. Chem Mater 1998;10:3301–4.

[17] Schmitz C, Thelakkat M, Schmidt HW. A combinatorial study of thedependence of organic LED characteristics on layer thickness. AdvMater 1999;11:821–6.

[18] Liu Y, Ma H, Jen AKY. Synthesis and characterization of a bipolarlight-emitting copolymer consisting of tetraphenyldiaminobiphenyland bis-quinoline units. Chem Mater 1999;11:27–9.

[19] Chen JT, Hsu CS. Conjugated polymer nanostructures for organicsolar cell applications. Polym Chem 2011;2(12):2707–22.

[20] Cheng YJ, Yang SH, Hsu CS. Synthesis of conjugated polymers fororganic solar cell applications. Chem Rev 2009;109(11):5868–923.

[21] Earmme T, Hwang YJ, Subramaniyan S, Jenekhe SA. All-polymer bulkheterojuction solar cells with 4.8% efficiency achieved by solutionprocessing from a Co-solvent. Adv Mater 2014;26:6080–5.

[22] Law KY. Organic photoconductive materials: recent trends anddevelopments. Chem Rev 1993;93:449–86.

[23] Schmidt-Mende L, Fechtenkötter A, Müllen K, Moons E, Friend RH,MacKenzie JD. Self-organized discotic liquid crystals for high-efficiency organic photovoltaics. Science 2001;293:1119–22.

[24] (a) Graser F, Hädicke E. Crystal structure and color of perylene-3, 4:9, 10-bis (dicarboximide) pigments. Liebigs Ann Chem1980;12:1994–2011;(b) Hädicke E, Graser F. Structures of eleven perylene-3,4:9,10-bis(dicarboximide) pigments. Acta Crystallogr Sect C1986;42:189–95;(c) Klebe G, Graser F, Hädicke E, Berndt J. Crystallochromy as a solid-state effect: correlation of molecular conformation, crystal packingand colour in perylene-3,4:9,10-bis(dicarboximide) pigments. ActaCrystallogr Sect B 1989;45:69–77.

[25] Würthner F, Hanke B, Lysetska M, Lambright G, Harms GS. Gelationof a highly fluorescent urea-functionalized perylene bisimide dye.Org Lett 2005;7:967–70.

[26] Li XQ, Stepanenko V, Chen Z, Prins P, Siebbeles LD, Wurthner F.Functional organogels from highly efficient organogelator basedon perylene bisimide semiconductor. Chem Commun 2006;37:3871–3.

[27] Würthner F. Perylene bisimide dyes as versatile building blocks forfunctional supramolecular architectures. Chem Commun2004:1564–79.

[28] Balakrishnan K, Datar A, Oitker R, Chen H, Zuo J, Zang L. Nanobeltself-assembly from an organic n-type semiconductor: propoxyethyl-PTCDI. J Am Chem Soc 2005;127:10496–7.

[29] Balakrishnan K, Datar A, Naddo T, Huang J, Oitker R, Yen M, et al.Effect of side-chain substituents on self-assembly of perylenediimide molecules: morphology control. J Am Chem Soc2006;128:7390–8.

[30] Yagai S, Monma Y, Kawauchi N, Karatsu T, Kitamura A.Supramolecular nanoribbons and nanoropes generated fromhydrogen-bonded supramolecular polymers containing perylenebisimide chromophores. Org Lett 2007;9:1137–40.

[31] Chen L, Henrike W. Perylene imides for organic photovoltaics:yesterday, today, and tomorrow. Adv Mater 2012;24:613–36.

[32] Huang C, Barlow S, Marder SR. Perylene-3,4,9,10-tetracarboxylicacid diimides: synthesis, physical properties, and use in organicelectronics. J Org Chem 2011;76:2386–407.

[33] Islam M, Sundararajan PR. Nano-scale self-assembly impeded byCH� � �p interaction in block selective solvents in the case ofoligostyrene–perylene diimide–oligostyrene (Coil–Rod–Coil)molecule. Eur Polym J 2013;49:2042–51.

14 E. Dahan, P.R. Sundararajan / European Polymer Journal 65 (2015) 4–14

MA

CRO

MO

LECU

LAR

NA

NO

TECH

NO

LOG

Y

[34] Dahan E, Sundararajan PR. Solvent-dependent nanostructures of gelsof a Gemini surfactant based on perylene diimide spacer andoligostyrene tails. Eur Polym J 2014;61:113–23.

[35] Yao D, Bender TP, Gerroir PJ, Sundararajan PR. Self-assembledvesicular nanostructures of perylene end-cappedpoly(dimethylsiloxane). Macromolecules 2005;38:6972–8.

[36] Yao D, Tuteja B, Sundararajan PR. Pigment-mediated nanowebmorphology of poly(dimethylsiloxane)-substituted perylenebisimides. Macromolecules 2006;39:7786–8.

[37] Dahan E, Sundararajan PR. Thermo-reversible gelation of rod-coiland coil-rod-coil molecules based on poly (dimethyl siloxane) andperylene imides and self-sorting of the homologous pair. Soft Matter2014;10:5337–49.

[38] Dahan E, Sundararajan PR. Thermoreversible physical gels ofpoly(dimethylsiloxane) without cross-links or functionalization.Langmuir 2013;29(27):8452–8.

[39] Islam MR, Sundararajan PR. Tubular or subsurface morphology ofoctabutoxyphthalocyanine upon self-assembly in polymermatrices: effect of the casting solvent. Chem Eur J2011;17:6098–108.

[40] Islam MR, Sundararajan PR. Morphology of a hydrogen-bondmediated self-assembling small molecule in a polycarbonatematrix. Can J Chem 2008;86:600–7.

[41] Tuteja B, Moniruzzaman M, Sundararajan PR. Domains of colloidalsize, mediated by self-assembly of small molecules in a polymermatrix: a three-level hierarchy of assembly. Langmuir2007;23:4709–11.

[42] Ong BS, Keoshkerian B, Murti DK. Photoresponsive imagingmembers with electron transport overcoatings, US Patent4,835,081; 1989.

[43] Sundararajan PR, Murti DK, Odell PG, Hamer GK. Photoreceptorincluding rotaxanes. US Patent 6, 096,464; 2000.

[44] Babu SS, Praveen VK, Ajayaghosh A. Functional p-gelators and theirapplications. Chem Rev 2014;114:1973–2129.

[45] Sugiyasu K, Fujita N, Shinkai S. Visible-light-harvesting organogelcomposed of cholesterol-based perylene derivatives. Angew ChemInt Ed 2004;43:1229–33.

[46] Wurthner F, Hanke B, Lysetska M, Lambright G, Harms GS. Gelationof a highly fluorescent urea-functionalized perylene bisimide dye.Org Lett 2005;7:967–70.

[47] Wu HX, Xue L, Shi Y, Chen YL, Li XY. Organogels based on J- and H-type aggregates of amphiphilic perylenetetracarboxylic diimides.Langmuir 2011;27:3074–82.

[48] Diaz DD, Cid JJ, Vazquez P, Torres T. Strength enhancement ofnanostructured organogels through inclusion of phthalocyanine-containing complementary organogelator structures and in situcross-linking by click chemistry. Chem Eur J 2008;14:9261–73.

[49] Diaz DD, Torres T, Zentel R, Davis R, Brehmer M. Physicochemicalcharacterization of octakis(alkyloxy)-substituted Zn(II)-phthalocyanines non-covalently incorporated into an organogeland their remarkable morphological effect on the nanoscale-fibers.Chem Commun 2007:2369–71.

[50] (a) Lee HY, Nam SR, Hong JI. Microtubule formation using two-component gel system. J Am Chem Soc 2007;129:1040–1;(b) Hirst AR, Smith DK, Harrington JP. Unique nanoscalemorphologies underpinning organic gel-phase materials. Chem EurJ 2005;11:6552–9;(c) Hirst AR, Smith DK, Feiters MC, Geurts HPM, Wright AC. Two-component dendritic gels: easily tunable materials. J Am Chem Soc2003;125:9010–1;

(d) Saha A, Roy B, Garai A, Nandi AK. Two-componentthermoreversible hydrogels of melamine and gallic acid. Langmuir2009;25:8457.

[51] Dasgupta D, Srinivasan S, Rochas C, Ajayaghosh A, Guenet JM. Hybridthermoreversible gels from covalent polymers and organogels.Langmuir 2009;25:8593–8.

[52] Lopez D, Guenet JM. Encapsulation of filaments of a self-assemblingbicopper complex in polymer nanowires. Eur Phys J Part B1999;B12:405–11.

[53] Khan MK, Sundararajan PR. Encapsulation of dye molecules andnanoparticles in hollow organogel fibers of a nonchiral polyurethanemodel compound. Chem Eur J 2011;17:1184–92.

[54] Khan MK, Chan A, Sundararajan PR. Two component gels of immiscibleblends of biscarbamates (polyurethane model compounds) andpoly(e-caprolactone). Soft Matter 2012;8:10149–57.

[55] Dulmage WJ, Light WA, Marino SJ, Salzberg CD, Smith DL,Staudenmayer WJ. An aggregate organic photoconductor. I.Chemical composition, preparation, physical structure, and opticalproperties. J Appl Phys 1978;49:5543–54.

Dr. Elianne Dahan graduated from CarletonUniversity with a Doctorate of Philosophy inChemistry in January 2014. Her doctoralresearch was on The Gelation and Morphologyof Rod–Coil and Coil–Rod–Coil MoleculesBased on Perylene Diimide. E.D. was a reci-pient of the Ontario Government Scholarshipfor Science and Technology in 2011 as well aswell as a recipient of the Queen Elizabeth IIGraduate Scholarship in 2012.

Dr. Sundararajan (Sundar) graduated fromthe University of Madras in 1969. He spentpost-doctoral tenures with Professor PaulFlory at Stanford University and withProfessor Bob Marchessault at University ofMontreal. Sundar was then employed at XeroxResearch Centre of Canada in Mississauga,Ontario for 25 years, as a member of researchstaff, Manager of Materials Characterizationand as Principal Scientist. His work at Xeroxwas focused on morphology of photorecep-tors, simulation of polymer conformations

and chain-folding. He became a professor at the Department of Chemis-try, Carleton University in Ottawa, Canada in 2000 as the NSERC-XeroxIndustrial Research Chair. Sundar’s current research is on morphology of

polymer composites, self-assembly, organo- and polymer gelation. He is aFellow of the Chemical Institute of Canada and the winner of theMacromolecular Science and Engineering Award of the Chemical Instituteof Canada and the Materials Chemistry Award of the Canadian MaterialsSociety. He has published over 150 papers and holds 10 patents.