Spn

JNa

b

c

d

ARRA

KPGCN

1

ttcictisaab(

meo

0h

Synthetic Metals 176 (2013) 55– 64

Contents lists available at SciVerse ScienceDirect

Synthetic Metals

journa l h om epage: www.elsev ier .com/ locate /synmet

ynthesis and characterization of single wall carbon nanotube-graftedoly(3-hexylthiophene) and their nanocomposites with goldanoparticles

ilian N. de Freitasa, Manoko S. Maubaneb,c, George Bepeteb,c, Willem A.L. van Otterloc,d,eil J. Covilleb,c, Ana F. Nogueiraa,∗

Laboratory of Nanotechnology and Solar Energy, P.O. Box 6154, Chemistry Institute, University of Campinas (UNICAMP), 13083-970, Campinas, SP, BrazilDST/NRF Centre of Excellence in Strong Materials University of the Witwatersrand, PO Wits, 2050, Johannesburg, South AfricaMolecular Sciences Institute, School of Chemistry, University of the Witwatersrand, PO Wits, 2050, Johannesburg, South AfricaDepartment of Chemistry and Polymer Sciences, Stellenbosch University, Stellenbosch, 7600, Western Cape, South Africa

a r t i c l e i n f o

rticle history:eceived 20 February 2013eceived in revised form 6 May 2013ccepted 21 May 2013

eywords:

a b s t r a c t

Single-wall carbon nanotubes (SWCNTs) were functionalized via the Prato reaction, by using a combi-nation of N-methylglycine and 2-thiophenecarboxaldehyde (2T) or 3-thiophenecarboxaldehyde (3T).Pristine SWCNTs or functionalized SWCNTs were then reacted with 3-hexylthiophene (3HT) in thepresence of FeCl3 to produce a composite (P3HT/SWCNT) or copolymers (P3HT-2T-SWCNT and P3HT-3T-SWCNT), containing 3-hexylthiophene units as the main chain and carbon nanotubes as side groups.

oly(3-hexylthiophene)old nanoparticlesarbon nanotubesanocomposites

After that, gold nanoparticles (AuNPs) were synthesized using a two-phase, one-pot reaction involvingthe reduction of HAuCl4 by NaBH4, in the presence of P3HT, P3HT/SWCNT, P3HT-2T-SWCNT or P3HT-3T-SWCNT. This approach generated nanocomposites where AuNPs were stabilized directly with P3HT-or P3HT-containing SWCNTs. The characteristics of the ternary nanocomposites were investigated usingUV–vis and fluorescence spectroscopy, thermogravimetric analysis, X-ray diffraction and high-resolutiontransmission electron microscopy.

. Introduction

Carbon nanotubes have been used in numerous applications dueo their unique mechanical properties, thermal stability, high elec-rical conductivity and ultrahigh dimensional aspect ratios. The 1Dharacteristic of carbon nanotubes is perhaps the most promis-ng feature for their application in electronic devices, because theharges have high mobility and show ballistic transport charac-eristics along their main axis. For example, many groups haventroduced single-wall carbon nanotubes (SWCNTs) into organicolar cells to act both as an exciton dissociation center andn electron acceptor [1–5]. In these applications, the nanotubesre combined with regioregular poly(3-hexylthiophene) (P3HT),ecause of the high hole mobility (∼0.1 cm2 V−1 s−1), low band gap∼1.9 eV) and solution processability of this polymer [6,7].

Advantage can be taken of the hybridization of conjugated poly-

ers with carbon nanotubes as it can enhance the attributes of

ach component in the nanocomposite. However, the dispersionf carbon nanotubes in polymeric matrixes is poor. The van der

∗ Corresponding author. Tel.: +55 19 35213029; fax: +55 19 35213023.E-mail address: anaflavia@iqm.unicamp.br (A.F. Nogueira).

379-6779/$ – see front matter © 2013 Elsevier B.V. All rights reserved.ttp://dx.doi.org/10.1016/j.synthmet.2013.05.026

© 2013 Elsevier B.V. All rights reserved.

Waals interactions between the tubes tend to generate undesiredaggregates, thus decreasing their solubility in most solvents. Also,the presence of impurities in the sample, such as graphite, amor-phous carbon or metal nanoparticles (residues of catalysts used inthe synthesis), can alter the physical and chemical properties ofnanotubes.

Generally, dispersion can be improved through chemicalfunctionalization of carbon nanotubes [8–11]. Another elegantapproach is to use grafting methods to provide chemical bondsbetween the components [12–16]. It is reasonable to expect thatcomposites containing carbon nanotube-grafted polymers willbehave differently from carbon nantoubes/polymer physical mix-tures in which no covalent bonds exist.

Composites incorporating noble metal nanoparticles are alsoattractive materials as they combine the unique optical and cat-alytic properties of metal nanoparticles with the high surface areaof organic materials [17]. For example, the incorporation of metalnanoparticles into solar cells has been investigated [18–21]. Theaddition of 5 wt% of gold nanoparticles (AuNPs) into solar cells com-

posed of mixtures of a polymer/fullerene derivative increased theefficiency due to a combination of enhanced light absorption (lightscattering of AuNPs) and improved charge transfer (lower seriesresistance) [21].

5 thetic

ccrcnAtoeoc

2

2

TaNsl2Nt2rlmv23ac

2

FC(dst(P(fio9

a

pt3

c1r

aba

6 J.N. de Freitas et al. / Syn

Here, our interest is to explore the characteristics of three-omponent nanocomposites based on a conjugated polymer,arbon nanotubes and metal nanoparticles. We present a simpleoute to synthesize copolymers containing P3HT units as mainhain, chemically bonded to SWCNTs (side groups) and ternaryanocomposites comprising mixtures of these copolymers withuNPs. These new types of nanocomposites could provide advan-

ages over two-component systems (P3HT/SWCNTs, P3HT/AuNPsr SWCNTs/AuNPs), as they are expected to combine the uniquelectrical and mechanical properties of carbon nanotubes with theptical and catalytic features of AuNPs, as well as the solution pro-essability of P3HT.

. Experimental

.1. Functionalization of SWCNTs

SWCNTs were purchased from Aldrich and used as received.he functionalization of this material was achieved by adapting

procedure developed by Prato and co-workers [22]. SWC-Ts (30 mg) were dispersed in 1,2-dichlorobenzene (25 mL) by

onicating for 10 min at rt. The mixture was heated to 150 ◦C, fol-owed by the addition of 2-thiophenecarboxaldehyde (2 T) (300 mg,.7 mmol) or 3-thiophenecarboxaldehyde (3 T) (300 mg, 2.7 mmol).-Methylglycine (360 mg, 4.0 mmol) was then added to the reac-

ion mixture in small portions over 5 days (about 72 mg per4 h) whilst keeping at 150 ◦C. At the end of the reaction, theesulting solid was filtered using a PTFE membrane (1.0 �m, Mil-ipore) and washed with DMF (100 mL), MeOH (500 mL) and a

ixture of acetone:water (1:1 v:v, 500 mL). After drying underacuum for 48 h, the products 2T-SWCNT (functionalized with-thiophenecarboxaldehyde) and 3T-SWCNT (functionalized with-thiophenecarboxaldehyde) were isolated as black solids. Fig. S1nd Fig. S2 (Supplementary Information) show the TGA and HRTEMharacterization data for the functionalized SWCNTs.

.2. Synthesis of P3HT

The synthesis of P3HT was achieved by a simple route, usingeCl3 as catalyst [23]. FeCl3 (2.0 g, 12.3 mmol) was added to dryHCl3 (∼12 mL) and stirred for 10 min at rt. 3-Hexylthiophene3HT) (0.50 g, 3.0 mmol) in dry CHCl3 (∼12 mL) was then addedrop-wise to the FeCl3 solution and the reaction mixture wastirred overnight under Ar at rt. The polymerization reaction washen terminated by pouring the reaction mixture into excess MeOH∼50 mL). The crude polymer precipitate was filtered using aTFE membrane filter (1 �m, Millipore) and washed with ethanol100 mL), a distilled water:acetone mixture (1:1 v:v, 500 mL) andnally with acetone (500 mL). The dark brown solid productbtained was dried under vacuum for 48 h to afford P3HT (0.45 g,0%).

Synthesis of P3HT-2T-SWCNT and P3HT-3T-SWCNT copolymersnd a P3HT/SWCNT composite

The syntheses and purification of SWCNT-grafted P3HT wereerformed using a similar procedure as described in section 2.2. Inhis case, 3-hexylthiophene (200 mg, 1.2 mmol) plus 2T-SWCNT orT-SWCNT (20 mg) were added to the FeCl3 reaction medium.

The synthesis of the non-covalently bonded P3HT/SWCNTomposite was obtained by adding 3-hexylthiophene (200 mg,.2 mmol) and unfunctionalized SWCNTs (20 mg) into the FeCl3eaction medium.

The monomer:carbon nanotube ratio was kept at 10:1 wt% inll cases. At the end of the purification procedure, ∼180 mg of alack solid was obtained in each case. The synthesized materialsre listed in Table 1.

Metals 176 (2013) 55– 64

2.3. Synthesis of P3HT-capped gold nanoparticles

Synthesis of P3HT-capped AuNPs was carried out using theprocedure reported by Zhai and McCullough [24]. Tetraoctylam-monium bromide (0.27 g, 0.5 mmol) was added to toluene (8 mL)in a 25 mL round bottom flask and the solution was stirred until allthe material had dissolved. An aqueous solution of HAuCl4 (3 mL,0.1 mmol) was added and the two phase mixture was stirred vigor-ously until all the HAuCl4 was transferred into the organic layer.This was determined by observing a color change from yellowto colorless in the aqueous phase. P3HT (15 mg) was dissolvedin toluene (3 mL) and added to the mixture. The mixture wasthen stirred for 3 h at rt. A freshly prepared aqueous solution ofNaBH4 (2.5 mL, 10 mmol) was next slowly added to the mixturewith vigorous stirring. During this time, dark purple polymer-goldnanoparticles were formed and the mixture was stirred overnightat rt, before being quenched in MeOH (30 mL). Purification was per-formed by filtering the resulting mixture through a PTFE membranefilter (0.45 �m, Millipore) and washing with MeOH and water, fol-lowed by drying under vacuum at rt for 24 h.

2.4. Synthesis of P3HT, SWCNT and AuNP-based nanocomposites

Ternary nanocomposites were prepared according to the pro-cedure described in section 2.4, with the exception that P3HTwas replaced with P3HT/SWCNT composite, or P3HT-2T-SWCNT orP3HT-3T-SWCNT copolymers. AuNPs were also synthesized in thepresence of pristine P3HT and pristine SWCNTs. All the materialsthat were synthesized are listed in Table 1.

2.5. Characterization

The optical characteristics of toluene solutions (1 mg mL−1)of the synthesized materials were determined. UV–vis spectrawere recorded on a Hewlett-Packard 8453 diode array spec-trophotometer and fluorescence spectra were measured witha ISS PC1TM Photon Counting Spectrofluorometer, operating atroom temperature (excitation = 430 nm). Powder X-ray diffraction(XRD) data were obtained from a Shimadzu XRD 7000 diffrac-tometer with CuK�, 40 kV, 30 mA and 2◦ min−1 scanning rate,over the range 2–80◦. Thermogravimetric analysis (TGA) wasperformed using a Thermogravimetric Analyser 2950 from TAInstruments. All measurements were done under a continuousargon flow of 100 mL min−1, and heating from rt to 900 ◦C at a rateof 10 ◦C min−1. High-resolution transmission electron microscopy(HRTEM) images were acquired in a HRTEM-JEM 3010 URP oper-ating at 300 kV with resolution of 0.17 nm. The samples wereprepared by casting one drop of a dispersion of the material in CHCl3onto a copper grid (300 mesh) pre-covered with a thin holey carbonlayer.

3. Results and discussion

3.1. Synthesis of polymer, copolymers and nanocomposites

The functionalization of SWCNTs and the synthesis of P3HT,copolymers and a composite based on P3HT and SWCNTs wereperformed according to Scheme 1.

In the first step, the Prato reaction was used to func-tionalize SWCNTs with 2-thiophenecarboxyaldehyde (2T) or3-thiophenecarboxyaldehyde (3T), to give 2T-SWCNT and 3T-

SWCNT, respectively. After purification, it was observed thatfunctionalized SWCNT samples appeared to be cleaner, with lesstube aggregation, when compared to a pristine sample, as seen byHRTEM images (Fig. S2).

J.N. de Freitas et al. / Synthetic Metals 176 (2013) 55– 64 57

Table 1Composition of all the materials synthesized.

Sample composition Abbreviation Sample type

poly(3-hexylthiophene) P3HT Polymerpoly(3-hexylthiophene) + pristine carbon nanotubes P3HT/SWCNT Compositepoly(3-hexylthiophene) + 2T-carbon nanotubes P3HT-2T-SWCNT Copolymerpoly(3-hexylthiophene) + 3T-carbon nanotubes P3HT-3T-SWCNT Copolymerpoly(3-hexylthiophene) + gold nanoparticles P3HT-Au Nanocompositepoly(3-hexylthiophene) + pristine carbon nantoubes + gold nanoparticles P3HT/SWCNT-Au Nanocomposite

nly in

ip3bn

f2twae

Sp

poly(3-hexylthiophene) + 2T-carbon nantoubes + gold nanoparticles

poly(3-hexylthiophene) + 3T-carbon nantoubes + gold nanoparticles

poly(3-hexylthiophene) + gold nanoparticles + pristine carbon nanotubes (added o

In the second step, the in situ chemical oxidative polymer-zation of 3-hexylthiophene was carried out in the presence ofristine SWCNTs or functionalized SWCNTs. The polymerization of-hexylthiophene monomer in the presence of excess FeCl3 haseen used to synthesize both “free” polymers [23,25] and carbonanotube-grafted polymers [13,26,27].

The two functionalized SWCNT samples synthesized have dif-erent reactivity in the polymerization reaction. In the case ofT-SWCNT, there is only one free site for the polymerization reac-ion, and hence end-cap polymers (P3HT-2T-SWCNT) are expected,

hile in 3T-SWCNT there are two reactive sites at which prop-

gation can occur, so pearl type polymers (P3HT-3T-SWCNT) arexpected. The active sites are indicated by (*) in Scheme 1.

cheme 1. Synthesis of P3HT, composite P3HT/SWCNT and copolymers P3HT-2T-SWCNTolymer chain growth.

P3HT-2T-SWCNT-Au nanocompositeP3HT-3T-SWCNT-Au Nanocomposite

the final step) P3HT-Au + SWCNT Nanocomposite

Table S1 (Supplementary Information) lists the molar massesmeasured for the polymer, composite and copolymers, using the gelpermeation chromatography (GPC) technique. The data indicatesthat all the samples are composed of small chains with a hetero-geneous size distribution. The average molecular weight valuesobserved were ∼2600 g mol−1 (P3HT), ∼2000 g mol−1 (P3HT-2T-SWCNT and P3HT-3T-SWCNT) and ∼1200 g mol−1 (P3HT/SWCNT).It appears that the presence of carbon nanotubes in the reactionmedium hampers the growth of the polymer chains. However,these results must be interpreted carefully since only the fraction

of the material that could be dispersed in chloroform and filteredwas analyzed (i.e., the fraction of the material with smaller chainsand/or lower concentration of SWCNTs in the chain).

and P3HT-3T-SWCNT (R = hexyl). Note that * indicates possible sites of thiophene

5 thetic Metals 176 (2013) 55– 64

tttrc

wcmptpasPSSl

3

adagas(osrc(0t

2iPczdmnSso

rdntrptpkpmbtot

Fig. 1. XRD analysis of (a) pristine SWCNTs, P3HT and P3HT-Au; dotted lines indicateP3HT peaks, while * indicates AuNPs peaks; (b) P3HT/SWCNT, P3HT-2T-SWCNT,

8 J.N. de Freitas et al. / Syn

The composite and copolymers containing SWCNTs were foundo be more soluble in toluene and 1,2-dichlorobenzene, comparedo the material made from a simple physical mixture of pris-ine P3HT and pristine SWCNTs. Similar observations have beeneported for a poly(3-octylthiophene)/SWCNT in situ polymerizedomposite [27] and also P3HT-MWCNT-based copolymers [13].

In the third step, the synthesis of nanocomposites with AuNPsas performed, as shown in Scheme 2. The control sample

onsisted of P3HT-stabilized AuNPs, synthesized according to theethod reported by Zhai and McCullough [24], using a room tem-

erature, two-phase, one-pot reaction involving the reduction ofetrachloroauric acid by sodium borohydride in the presence of theolymer. In this approach, the AuNPs are stabilized by the inter-ction of the sulfur atoms of the polymer with the gold surface. Aimilar procedure was carried out in the presence of composite3HT/SWCNT or the copolymers P3HT-2T-SWCNT or P3HT-3T-WCNT, to synthesize ternary nanocomposites containing P3HT,WCNTs and AuNPs. All the materials synthesized in this study areisted in Table 1.

.2. Structural analysis

Fig. 1a shows the X-ray diffractograms of pristine SWCNTs, P3HTnd the P3HT-Au nanocomposite. The SWCNT powder has a broadiffraction pattern centered at 25.1◦, which merges with the halo-morphous pattern, related to the plane 002 reflection from theraphite-like structure [28]. The pristine P3HT film has a sharp peakt 5.20◦, which correspond to an interchain lamella peak with d-pacing ∼17 A and a peak in the halo-amorphous region at 24.10◦

d-spacing ∼3.8 A). These peaks are possibly related to crystallinerientations of the thiophene units (stacking) with respect to theubstrate [29,30]. In the P3HT-Au film, the reflections of AuNPs areeadily identified. The peaks at 38.42◦, 44.56◦, 64.82◦ and 77.73◦,orresponding to the face-centered cubic (fcc) bulk gold patternplanes (1 1 1), (2 0 0), (2 2 0) and (3 1 1), respectively; JCPDS 04-784) confirms the crystalline structure of the AuNPs loaded intohe polymer matrix.

Fig. 1b shows the X-ray diffractograms of P3HT/SWCNT, P3HT-T-SWCNT and P3HT-3T-SWCNT. In all three materials, the

nterchain lamella and amorphous region peaks characteristic of3HT can still be observed. This is indicative that the presence ofarbon nanotubes does not disrupt the crystallization and organi-ation of the polymeric material. In addition, the peaks positionso not change; only a slight variation of the full-width at half-aximum (FWHM) is observed. Other authors have also observed

o significant change in the peak positions after the addition ofWCNTs [27] or MWCNTs [31] to the polymer matrix, and observedimilar changes in the FWHM, attributed to the physical wrappingf the polymer onto the walls of the nanotubes.

The diffractograms of the ternary nanocomposites (Fig. 1c)eveal the presence of the peaks attributed to AuNPs, which are evi-ence of the effective incorporation of metal nanoparticles into theanocomposite during the synthesis. Another interesting feature ishat the AuNPs disrupt the crystalline phase of the polymer. Theseesults are different from what was found for films prepared fromhysical mixtures of P3HT, fullerene and AuNPs separately syn-hesized [20]. A strong interaction between the sulfur-containingolymer and the AuNPs surfaces may exist in these composites. It isnown that some metals interact strongly with the sulfur atom inolythiophenes, and that the solvent and the preparation methoday affect this interaction [32]. It is possible that the interaction

etween the polymer and the AuNPs in the nanocomposites syn-hesized here is stronger than in films prepared by simple mixturesf the isolated components, resulting in a loss of polymer crys-allinity induced by the presence of the nanoparticles. Also, it is

P3HT-3T-SWCNT films; and (c) P3HT/SWCNT-Au, P3HT-2T-SWCNT-Au and P3HT-3T-SWCNT-Au films.

important to mention that, due to this strong interaction, significantphase segregation is not expected in the nanocomposites.

The presence of SWCNTs in the copolymers and compositeswas not clearly indicated by XRD analysis, but better evidence fortheir presence can be seen in the HRTEM images (Fig. 2). Fig. 2ashows that a simple physical mixture of pristine P3HT and pristineSWCNTs gives samples with considerable amounts of amorphousmaterial and aggregates. In the P3HT/SWCNT composite the tubesand bundles of nanotubes were more easily identified (Fig. 2b). Thisis consistent with the debundling and better dispersion of carbonnanotubes in the polymer medium, due to the stronger interactionbetween the materials in this sample. For the copolymers P3HT-2T-SWCNT and P3HT-3T-SWCNT, the identification of tubes waseven easier, and it is now possible to see the presence of blocks oforganized nanotubes (Fig. 2c,d).

The presence of AuNPs in the nanocomposites was also readilyseen in the HRTEM images (Fig. 3). All the samples contained a largeamount of AuNPs, of varying sizes and with a wide size distribution.

J.N. de Freitas et al. / Synthetic Metals 176 (2013) 55– 64 59

Scheme 2. Synthesis of nanocomposites containg P3HT, SWCNTs and (�) AuNPs.

60 J.N. de Freitas et al. / Synthetic Metals 176 (2013) 55– 64

F s), (b)1

TF

pp(

ig. 2. HRTEM images of (a) P3HT + SWCNT (physical mixture of isolated component00 nm.

he size distribution histograms for these samples can be seen inig. S3.

Most of the nanoparticles synthesized in the presence of theristine P3HT show diameters ranging from 2–6 nm (P3HT-Au sam-le, Fig. 3a), although some regions containing larger nanoparticlessize 8–12 nm) could also be found (Fig. 3b). Interestingly, when

Fig. 3. HRTEM images of (a,b) P3HT-Au, (c) P3HT-Au + SWCNT, (d) P3HT/SWCNT-Au

P3HT/SWCNT, (c) P3HT-2T-SWCNT and (d) P3HT-3T-SWCNT. All scale bars indicate

the nanoparticles were synthesized in the presence of pristineP3HT plus pristine SWCNTs, the formation of larger, non-spherical

structures was observed (size 8–20 nm, Fig. 3c). In both samplesthe polymer was “free” to interact with the growing nanoparti-cles. This result therefore suggests that the carbon nanotubes mayplay a role in the mechanism of nucleation or growth of the gold

, (e) P3HT-2T-SWCNT and (f) P3HT-3T-SWCNT. All scale bars indicate 50 nm.

thetic Metals 176 (2013) 55– 64 61

ncp

csIcwctAs

tuZfdoouban

3

Ftisfdcamdb

btaimppgnlasa

taoacPatd

c

Fig. 4. TGA (a,c) and first derivative curves (b,d) of (----) pristine SWCNTs, (-*-) P3HT,(-�-) P3HT/SWCNT, (-�-) P3HT-2T-SWCNT, (-�-) P3HT-3T-SWCNT, (-♦-) P3HT-Au,

J.N. de Freitas et al. / Syn

anoparticles, or that the presence of “free” SWCNTs drasticallyhanges the conformation, swelling behavior or diffusion of theolymer chains in solution.

The AuNPs synthesized in the presence of the P3HT/SWCNTomposite (Fig. 3d) display a narrower size distribution and themallest nanoparticles (diameters 2–10 nm) amongst all samples.n comparison, the nanoparticles synthesized in the presence ofopolymer P3HT-2T-SWCNT have a much broader size distribution,ith diameters ranging from 2–13 nm, with a few larger parti-

les (17–25 nm) (Fig. 3e). Also, the synthesis of nanoparticles inhe presence of P3HT-3T-SWCNT (Fig. 3f) leads to the formation ofuNPs with diameters of 2–10 nm and fewer particles with a largerize (13–17 nm).

In this present study, the formation of heterogeneous struc-ures with broad size distribution was observed in all samples,nlike the narrow size distribution (diameter ∼10 nm) reported byhai and McCullough [24] for P3HT-capped AuNPs. These authorsound that the concentration of P3HT affected the nanoparticle sizeistribution. It was also reported that, as the concentration of olig-thiophene increased, the nanoparticle size also increased [33]. Inur work, perhaps more important than the amount of polymersed is the presence of SWCNTs in the reaction medium. It is possi-le that either the nanotubes interact with the polymer chains andffect the coordination of AuNPs by the polymer, or that the carbonanotubes play a direct role in the growth of the AuNPs.

.3. Thermogravimetric analysis

TGA and first derivative curves for all the materials are shown inig. 4. Under an inert atmosphere, the pristine SWCNTs degrade inhe temperature range 400–640 ◦C. The temperature of the max-mum weight loss rate (Tmax) of the two main decompositionteps appear at 470 ◦C and 540 ◦C, and are attributed to the loss ofree carbon nanotubes and bundles, respectively [9]. Pristine P3HTegrades in the temperature range 270–615 ◦C, in a two-step pro-ess (Tmax at 380 ◦C and 510 ◦C). The beginning of the mass loss isttributed to the loss of the hexyl side group attached to the aro-atic thiophene backbone and, as the temperature increases, the

egradation is accelerated as the pyrolysis of the aromatic back-one of the polymer chains occurs [34].

P3HT/SWCNT and P3HT-2T-SWCNT show a similar thermalehavior. These materials degrade in the range ∼330–690 ◦C, withwo main processes, with their Tmax at 480 ◦C and 625 ◦C, whichre higher than the Tmax observed for the pristine materials. Thisncrease is not unusual; indeed, an increase in the temperature of

aximum weight loss has been observed by other authors after dis-ersion of 1% of carbon nanotubes within P3HT [34]. This increase isostulated to indicate an enhanced thermo-oxidative stability, trig-ered by interactions between the conducting polymer and carbonanotubes. Saini et al. [26] observed that composites of P3HT with

ow concentrations of MWCNTs have a better thermal stability, as result of the more uniform dispersion of the nanotubes and theubsequent increase in interactions between the polymeric chainsnd the graphitic nanotube walls.

The P3HT-3T-SWCNT material decomposes in the tempera-ure range 270–690 ◦C. The degradation processes occur with Tmax

t 370 ◦C, 470 ◦C and 630 ◦C. The first process is similar to thatf pristine P3HT and may indicate the presence of a significantmount of “free” P3HT segments in this sample. The second pro-ess is coincident with the degradation processes observed for the3HT-2T-SWCNT copolymer and P3HT/SWCNT composite, and isttributed to the decomposition of P3HT chains in intimate con-

act with SWCNTs. Lastly, the third process is attributed to theecomposition of carbon nanotubes.

The amount of SWCNTs incorporated into the composite andopolymers could not be estimated from the TGA curves, since the

(-�-) P3HT/SWCNT-Au (-©-) P3HT-2T-SWCNT-Au, (-�-) P3HT-3T-SWCNT-A and(—) P3HT/Au + SWCNTs, carried out under an inert atmosphere.

polymer and the carbon nanotube samples degrade in similar tem-perature ranges. Since the carbon nanotube component representsonly ∼10% of the sample, they are not easy to detect.

When AuNPs were incorporated into these systems, a decreasein the thermal stability of the polymer and carbon nanotubessegments was observed. All the materials started to decompose

62 J.N. de Freitas et al. / Synthetic Metals 176 (2013) 55– 64

Table 2TGA analysis of the synthesized materials: initial degradation temperature (Tinit),final degradation temperature (Tfinal), maximum weight loss rate temperature (Tmax)and concentration of AuNPs estimated from the thermograms.

Material Tinit/(◦C) Tfinal/(◦C) Tmax/(◦C) Au content

SWCNT 400 600 470, 540 –P3HT 270 615 380, 510 –P3HT/SWCNT 330 690 470, 625 –P3HT-2T-SWCNT 330 690 480, 625 –P3HT-3T-SWCNT 270 690 370, 470, 630 –P3HT-Au 200 560 300, 440, 505 25%P3HT/SWCNT-Au 240 560 300, 450, 500 55%P3HT-2T-SWCNT-Au 150 580 190, 270, 460, 530 33%

addie

ccts

tTnttscottSicoicgsna

3

ttUos

siitopsptc

Fig. 5. Normalized (a,b) absorption and (c,d) photoluminescence spectra of1 mg mL−1 toluene solutions of (-*-) P3HT, (-�-) P3HT/SWCNT, (-�-) P3HT-2T-SWCNT, (-�–) P3HT-3T-SWCNT, (----) P3HT + SWCNT (physical mixture of isolated

P3HT-3T-SWCNT-Au 160 580 270, 450, 530 44%P3HT-Au + SWCNT 150 580 190, 270, 460, 530 32%

t temperatures between 140 and 200 ◦C, and were completelyegraded by 550–600 ◦C. It was also evident that the process ofegradation seemed to occur in three main steps. From the change

n degradation profile it is thus possible to infer that AuNPs areffectively interacting with the P3HT chains and SWCNTs.

At temperatures above 700 ◦C, the residue is expected to beomposed mostly of gold. From the difference between the residueontents before and after incorporation of AuNPs, one can estimatehe amount of nanoparticles loaded into the nanocomposites, ashown in Table 2.

It is possible to infer some preliminary conclusions based onhe amount of AuNPs incorporated. From the results displayed inable 2, it can be seen that the P3HT-Au and P3HT-Au + SWCNTanocomposites were loaded with ∼25% and 32% of AuNPs, respec-ively. This suggests that the presence of “free”, pristine SWCNTs inhe reaction medium affects the amount of AuNPs loaded into theystem. In the materials containing SWCNTs grafted to the polymerhain, or previously incorporated into polymer matrix, the amountsf AuNPs loaded into the nanocomposite are even higher (from 33%o 55%). The different AuNP uptake in these nanocomposites canhus be related to the distinct types of interaction between theWCNTs and polymer segments. The polymer/carbon nanotubesnteraction may lead to different conformations of the polymerhains, interfere with the diffusion rate, or may change the coveringf the carbon nanotubes surfaces by the polymer (i.e., by chang-ng the degree of nanotube surface exposure). All these factorsould change the interaction between the P3HT with the growingold nuclei, thus altering both the amount and size of the synthe-ized nanoparticles. The existence of a physical adsorption of theanoparticles onto the nanotube surface is also possible and wouldlso change their interaction with the P3HT segments.

.4. Optical characteristics

Fig. 5 shows the normalized absorption and emission spectra ofhe synthesized materials dispersed in toluene. The main absorp-ion band, with a maximum wavelength (�max) at ∼430 nm in theV–vis spectra of all the materials, was attributed to the absorptionf a �–�* transition in the conjugated polymer chain of the P3HTegments in solution [35].

In Fig. 5a, it can be seen that the absorption characteristics ofamples containing P3HT and SWCNTs is similar and that it is dom-nated by the polymer characteristics, independent of the type ofnteraction between the polymer and carbon nanotubes. In addi-ion, there is no significant shift in the �max upon incorporationf the carbon nanotubes in the composite or copolymers. This isrobably related to the low concentration of the SWCNTS in these

amples. Other authors have obtained similar data for compositesrepared with small amounts (below 10%) of SWCNTs [8,27] andhis has been attributed to the lack of ground state interactions andharge transfer between the polymer and the nanotubes.

components), (-♦-) P3HT-Au, (-�-) P3HT/SWCNT-Au, (-©-) P3HT-2T-SWCNT-Au,(-�-) P3HT-3T-SWCNT-Au and (—) P3HT/Au + SWCNT.

Fig. 5b shows the absorption characteristics of the nanocom-posites containing AuNPs. In previous reports, strong absorptionbroadening was observed for P3HT-Au [23,24] or P3HT-C60-Au [36]solutions, which was attributed to the plasmon resonance effect. Inthis work, a discrete broadening of the absorption in the region500–700 nm for the ternary composites was observed. This broad-ening is more evident in the P3HT/SWCNT-Au sample due to thehigher concentration of AuNPs in this sample.

As observed in the absorption spectra, the emission charac-teristics are dominated by P3HT characteristics, with a maximumemission at ∼575 nm. There is no significant variation of this valueafter addition of SWCNTs or AuNPs. A discrete shoulder located at∼615 nm can also be seen for the P3HT and P3HT/SWCNT samples,in this case attributed to the emission of aggregates. This shoulderhas a lower intensity for P3HT-2T-SWCNT and P3HT-3T-SWCNTcopolymers, indicating a lower concentration of aggregated poly-mer species in these samples (Fig. 5a). This result is in agreementwith the increase observed in the FWHM in the XRD data. Thestronger interaction between the SWCNTs and P3HT units in thecopolymers resulting from the grafting process appears to increasethe spacing between the polymeric chains in these materials. In theP3HT/SWCNT composite, the weaker interaction obtained throughphysical mixing leads to the formation of regions richer in polymeraggregates, as in the case of the P3HT sample. A similar behaviorwas also observed in the spectra of the ternary nanocomposites,compared to that of the P3HT-Au sample (Fig. 5b).

Fig. 6 displays the non-normalized emission spectra of toluenedispersions of the pristine P3HT sample and ternary nanocom-posites. The emission intensity increases more than one order of

J.N. de Freitas et al. / Synthetic

F((

msbicoIasoSioessihP

dlta

dssidAatsi

4

PrtPtpSai

[

[

[

[

[14] B.K. Kuila, K. Park, L.M. Dai, Macromolecules 43 (2010) 6699.

ig. 6. Photoluminescence spectra of toluene solutions of (-*-) P3HT, (-♦-) P3HT-Au,-�-) P3HT/SWCNT-Au, (-©-) P3HT-2T-SWCNT-Au, (-�-) P3HT-3T-SWCNT-Au and—) P3HT/Au + SWCNT.

agnitude after incorporation of AuNPs. Increased photolumine-cence after the addition of Au nanoclusters to P3HT has previouslyeen reported and was explained by the polymer chain separation

nduced by the presence of the nanoparticles [37]. In our work,hain separation must be an important effect, resulting in the lackf polymer crystalline patterns, as shown by the XRD data (Fig. 1c).t is interesting to note that the emission intensity does not increases the concentration of AuNPs in the sample increases. Instead, iteems to show non-linear behavior. An intermediate concentrationf relatively small and heterogeneous AuNP samples (P3HT-2T-WCNT/Au and P3HT-3T-SWCNT/Au) seems to lead to the highestntensities of the emission signals. This is probably related to anptimum polymer chain separation in these samples. In the pres-nce of a much larger nanoparticle concentration (P3HT/SWCNT/Auample) or much larger nanoparticle structures (P3HT/Au + SWCNTample), the propensity for phase separation may be too great, lead-ng to the formation of regions rich in P3HT agglomerates, whichave a lower luminescence yield (i.e., they behave like a pristine3HT film).

Further investigations into the effects of the incorporation ofifferent concentrations of SWCNTs into the polymer matrix may

ead to a better control of the characteristics of the metal nanopar-icles, and may even allow the synthesis of a great variety of sizesnd shapes of nanoparticles.

Other properties of the nanocomposites, such as electrical con-uctivity, mechanical stability and catalytic activity should also betudied in the future. The application of these nanocomposites inolar cells is under development. The use of these nanocompos-tes as sensitizers in dye-sensitized solar cells did not improve theevice, as was observed for nanocomposites based on P3HT, C60 anduNPs [36]. Perhaps a more interesting approach for these materi-ls will be their use as a “buffer layer”, instead of by application inhe active layer, to take advantage of the high conductivity of theseamples; as previously described, AuNPs can be successfully usedn this way [38,39].

. Conclusion

In this study, new ternary nanocomposites consisting of3HT, SWCNTs and AuNPs were synthesized using a simpleoute. In the first step, SWCNTs were functionalized with 2-hiophenecarboxaldehyde or 3-thiophenecarboxaldehyde via therato reaction. Pristine SWCNTs or functionalized SWCNTs werehen reacted with 3-hexylthiophene in the presence of FeCl3 to

roduce a composite (P3HT/SWCNT) or copolymers (P3HT-2T-WCNT and P3HT-3T-SWCNT) containing 3-hexylthiophene unitss the main chain and carbon nanotubes as side groups. HRTEMmages of the composite and copolymers revealed a debundling and

[

[

Metals 176 (2013) 55– 64 63

better dispersion of carbon nanotubes in the polymer medium, incomparison with a simple physical mixture of pristine P3HT andpristine SWCNTs. The physical wrapping of polymer chains ontothe walls of the nanotubes, expected in the composite and copoly-mers, decreased the formation of aggregated polymer species (asobserved in emission characteristics). This effect was more pro-nounced in the copolymers, since the interaction between P3HTand the SWCNTs by grafting should be stronger than in physi-cal mixing. Nevertheless, the most significant difference betweenthe copolymers containing chemically grafted SWCNTs and thephysically mixed P3HT/SWCNT composite was observed duringthe synthesis of AUNPs in the second step. The AuNPs were syn-thesized using a two-phase, one-pot reaction, in the presence ofP3HT, P3HT/SWCNT, P3HT-2T-SWCNT or P3HT-3T-SWCNT, to giveternary nanocomposites. The combination of TGA analysis andHRTEM images showed that the nancomposites contained differ-ent loadings of AuNPs with different sizes and size distributions.The presence of SWCNTs and the different types of interactionsof this material with P3HT in the nancomposites (physical mixingor grafting) affected the formation (size and shape) and uptake ofthe AuNPs. In addition, the presence of the nanoparticles appearsto disrupt the polymeric crystalline phase, as observed from XRDanalysis. The profile of the absorption and emission spectra of thesenanocomposites are mainly dictated by the properties of the P3HTsegments, although the intensity of emission signals increasedmore than one order of magnitude after incorporation of the AuNPs.

Acknowledgements

The authors thank the IBSA Project(fellowship Pró-África490.490/2007-7), FAPESP (fellowship 2009/15428-0), CNPq, INEOand the National Research Foundation (NRF, South Africa) for finan-cial support and LME/LNNano/CNPEM for the technical supportduring the HRTEM work.

Appendix A. Supplementary data

Supplementary data associated with this arti-cle can be found, in the online version, athttp://dx.doi.org/10.1016/j.synthmet.2013.05.026.

References

[1] Q. Liu, J. Mao, Z. Liu, N. Zhang, Y. Wang, L. Yang, S. Yin, Y. Chen, Nanotechnology19 (2008) 115601.

[2] S. Berson, R. Bettingbies, S. Bailly, S. Guillerez, B. Jousselme, Adv. Funct. Mater.17 (2007) 3363.

[3] S. Chaudhary, H. Lu, A.M. Müller, C.J. Bardeen, M. Ozkan, Nano Lett. 7 (2007)1973.

[4] S. Ren, M. Bernardi, R.R. Lunt, V. Bulovic, J.C. Grossman, S. Gradecak, Nano Lett.11 (2011) 5316.

[5] J.N. de Freitas, G. Conturbia, L.C.P. Almeida, F.S. Freitas, A.D. Gonc alves, A.F.Nogueira, Global J. Inorg. Chem. 2 (2011) 157.

[6] I. Osaka, R.D. McCullough, Acc. Chem. Res. 41 (2008) 1202.[7] R.D. McCullough, Adv. Mater. 10 (1998) 93.[8] A.F. Nogueira, B.S. Lomba, M.A. Soto-Oviedo, C.R.D. Correia, P. Corio, C.A. Fur-

tado, I.A. Hümmelgen, J. Phys. Chem. C 111 (2007) 18431.[9] U. Dettlaff-Weglikowska, J.M. Benoit, P.W. Chiu, R. Graupner, S. Lebedkin, S.

Roth, Curr. Appl. Phys. 2 (2002) 497.10] M.M. Stylianakis, J.A. Mikroyannidis, E. Kymakis, Sol. Energy Mater. Sol. Cells

94 (2010) 267.11] A.A. Stefopoulos, C.L. Chochos, M. Prato, G. Pistolis, K. Papagelis, F. Petraki, S.

Kennou, J.K. Kallitsis, Chem. Eur. J. 14 (2008) 8715.12] A.P. Le, T.M. Huang, P.T. Chen, A.C.M. Yang, J. Polym. Sci. B Pol. Phys. 49 (2011)

581.13] B. Philip, J. Xie, A. Chandrasekhar, J. Abraham, V.K. Varadan, Smart Mater. Struct.

13 (2004) 295.

15] F. Boon, S. Desbief, L. Cutaia, O. Douheret, A. Minoia, B. Ruelle, S. Clement, O.Coulembier, J. Cornil, P. Dubois, R. Lazzaroni, Macromol. Rapid Commun. 31(2010) 1427.

16] T. Nakahodo, H. Fujihara, J. Synth. Org. Chem. Jpn. 71 (2013) 39.

6 thetic

[[

[

[

[[

[[

[

[

[[

[

[

[

[[

[

[

[1052.

4 J.N. de Freitas et al. / Syn

17] K. Kim, D.L. Carroll, Appl. Phys. Lett. 87 (2005) 203113.18] M. Park, B.D. Chin, J.-W. Yu, M.-S. Chun, S.-H. Han, J. Ind. Eng. Chem. 14 (2008)

382.19] K. Topp, H. Borchert, F. Johnen, A.V. Tunc, M. Knipper, E. von Hauff, J. Parisi, K.

Al-Shamery, J. Phys. Chem. A 114 (2010) 3981.20] D.H. Wang, D.Y. Kim, K.W. Choi, J.H. Seo, S.H. Im, J.H. Park, O.O. Park, A.J. Heeger,

Angew. Chem. Int. Ed. 50 (2011) 5519.21] M. Maggini, G. Scorrano, M. Prato, J. Am. Chem. Soc. 115 (1993) 9798.22] M.R. Andersson, D. Selse, M. Berggren, H. Jarvinen, T. Hjertberg, O. Inganas, O.

Ennerstrom, J.E. Oserholm, Macromolecules 27 (1994) 6503.23] L. Zhai, R.D. McCullough, J. Mater. Chem. 14 (2004) 141.24] R.K. Singh, J. Kumar, A. Kumar, V. Kumar, R. Kant, R. Singh, Sol. Energy Mater.

Sol. Cells 94 (2010) 2386.25] V. Saini, Z. Li, S. Bourdo, E. Dervishi, Y. Xu, X. Ma, V.P. Kunets, G.J. Salamo, T.

Viswanathan, A.R. Biris, D. Saini, A.S. Biris, J. Phys. Chem. C 113 (2009) 8023.

26] H.J. Kim, R.B. Koizhaiganova, M.R. Karim, G.H. Lee, T. Vasudevan, M.S. Lee, J.

Appl. Pol. Sci. 118 (2010) 1386.27] S.L. Liu, J. Yue, R.J. Wehmschulte, Nano Lett. 2 (2002) 1439.28] T. Erb, U. Zhokhavets, G. Gobsch, S. Raleva, B. Stühn, P. Schilinsky, C. Waldauf,

C.J. Brabec, Adv. Funct. Mater. 15 (2005) 1193.

[

[

Metals 176 (2013) 55– 64

29] S. Hugger, R. Thomann, T. Heinzel, T. Thurn-Albrecht, Colloid Polym. Sci. 282(2004) 932.

30] A.W. Musumeci, G.G. Silva, J.W. Liu, W.N. Martens, E.R. Waclawik, Polymer 48(2007) 1667.

31] E.O. Sako, H. Kondoh, I. Nakai, A. Nambu, T. Nakamura, T. Ohta, Chem. Phys.Lett. 413 (2005) 267–271.

32] J.H. Youk, J. Locklin, C.J. Xia, M.K. Park, R. Advincula, Langmuir 17 (2001) 4681.33] A.R. Adhikari, M. Huang, H. Bakhru, M. Chipara, C.Y. Ryu, P.M. Ajayan, Nano-

technology 17 (2006) 5947.34] S. Hotta, S.D.D.V. Rughooputh, A.J. Heeger, F. Wudl, Macromolecules 20 (1987)

212.35] J.N. de Freitas, M.A. Mamo, M. Maubane, W.A.L. van Otterlo, N.J. Coville, A.F.

Nogueira, J. Power Sources 215 (2012) 99.36] P.G. Nicholson, V. Ruiz, J.V. Macpherson, P.R. Unwin, Chem. Commun. 12 (2005)

37] F.-C. Chen, J.-L. Wu, C.-L. Lee, Y. Hong, C.-H. Kuo, M.H. Huang, Appl. Phys. Lett.95 (2009) 013305.

38] R. Morioka, K. Yasui, M. Ozawa, K. Odoi, H. Ichikawa, K. Fujita, J. Photopolym.Sci. Technol. 23 (2010) 313.