Solvent Polarity Effect on Chain Conformation, Film Morphology, and

Aug 20, 2010 - ... and electronic properties which are critical to device performance. ...... Joke Vandenbergh , Jeroen Dergent , Bert Conings , T.V.V...
0 downloads 0 Views 2MB Size
Download PDF
11746

J. Phys. Chem. B 2010, 114, 11746–11752

Solvent Polarity Effect on Chain Conformation, Film Morphology, and Optical Properties of a Water-Soluble Conjugated Polymer Zhihua Xu,† Hsinhan Tsai,‡ Hsing-Lin Wang,*,‡ and Mircea Cotlet*,† Center for Functional Nanomaterials, BrookhaVen National Laboratory, Upton, New York 11973, and Chemistry DiVision, Los Alamos National Laboratory, New Mexico, 87545 ReceiVed: June 1, 2010; ReVised Manuscript ReceiVed: July 23, 2010

The solvent polarity effect on chain conformation, film morphology, and photophysical properties of a nonionic water-soluble conjugated polymer (WSCP), poly[2,5-bis(diethylaminetetraethylene glycol)phenylene vinylene] (DEATG-PPV) is investigated in detail. The combination of stationary absorption and photoluminescence (PL) spectroscopy, time-resolved PL spectroscopy, and fluorescence correlation spectroscopy methods enables us to probe the chain conformation of DEATG-PPV, down to the level of a single chain when working with extremely diluted solutions. The use of correlated atomic force microscopy and confocal fluorescence lifetime imaging microscopy measurements of drop-casted DEATG-PPV films reveals the intrinsic relationship between chain conformation, film morphology, and optical properties. Depending on solvent polarity, DEATG-PPV presents extended, coiled, and collapsed chain conformations in solutions, which lead to distinct morphology and optical properties in solid films. Our work presents a pathway to control and characterize the film morphologies of WSCPs toward the optimal performance of various optoelectronic devices. Introduction In recent years, water-soluble conjugated polymers (WSCPs) have drawn increasing interest due to their great potential to fabricate highly sensitive biological sensors1-3 and optoelectronic devices4 using environmentally friendly procedures. For example, WSCPs have been successfully used as active layers of organic light-emitting diodes (OLEDs),5,6 electron injection materials of multilayer OLEDs,7-11 or as light-emitting and electrolyte materials in polymer light-emitting electrochemical cells12-14 and electron donor materials in organic photovoltaics (OPVs).15,16 Unlike conventional conjugated polymers (CPs) which are only soluble in organic solvents, WSCPs render their water solubility by attaching hydrophilic side chains to highly hydrophobic π-conjugated backbones, for example, poly(phenylene vinylene), polythiophenes, and polyfluorenes. Their hydrophilic side chains may contain charged groups such as sulfonates, carboxylates, phosphonates, and ammonium salts or neutral groups such as hydroxyl and ethylene glycol (EG).17,18 It is well-known that the film morphology, i.e., the way that conjugated polymer chains aggregate or pack into solid film, has a large impact on their optical and electronic properties which are critical to device performance.19 For example, Nguyen and Schwartz et al. have reported that increasing chain aggregation in conjugated polymer poly[2-methoxy-5-(20-ethylhexyloxy)-1,4-phenylene vinylene] (MEH-PPV) films significantly reduces both the photoluminescence (PL) quantum yield and electroluminescence quantum efficiency (EQE) of the OLEDs based on MEH-PPV films.20 It has also been reported that the power conversion efficiency of OPVs heavily relies on the degree of polymer chain aggregation and the way in which chains pack in films.21 Like the other CPs, the optoelectronic properties of WSCPs are also dependent on film morphology. * To whom correspondence should be addressed. Tel.: 631-344-7778, Fax: 631-344-7765, E-mail: [email protected] (M.C.); Tel.: 505-677-9944, Fax: 505-667-0440, E-mail: [email protected] (H.-L.W.). † Brookhaven National Laboratory. ‡ Los Alamos National Laboratory.

Jen and co-workers have reported that by optimizing the film morphology of a nonionic WSCP utilized as electron injection material in a white OLED, poly[9,9-bis(6′-(diethanolamino)hexyl)fluorene], the power efficiency of the device can be improved by a factor of 4.22 Compared to the widely explored topic of controlling film morphology and optoelectronic properties in the case of conventional CPs, for example, by tuning processing parameters such as polymer concentration,23 solvent polarity,24 solvent evaporation time,25 and annealing temperature,26 there are only a few scattered reports on this important topic for WSCPs.27,28 It has been generally accepted that the conformation of polymer chains in their precursor solution can be retained to some extent in the solution-casted thin films and that film morphology and polymer chain conformation are intrinsically correlated.29,30 Although different scattering methods such as dynamic light scattering and small angle neutron scattering have been successfully used to study polymer chain conformation and aggregation in solution,31,32 it remains a challenge to fully characterize the single-chain conformation in extremely diluted solutions. With the advent of laser-induced single molecule methods capable of detecting and characterizing the fluorescence of single chromophores, it has become possible to interrogate conjugated polymers at the level of a single chain.33-35 Here we report control of chain conformation, film morphology, and optical properties for a nonionic WSCP, poly[2,5bis(diethylaminetetraethylene glycol)phenylene vinylene] (named herein DEATG-PPV), by tuning the polarity of the solvent. We show that polymer-solvent interactions have a predominant impact on the chain conformation of this particular nonionic WSCP and that the chain conformation in turn determines the morphology and the optical properties of drop-casted thin films. By combining stationary absorption and PL spectroscopy, timeresolved PL spectroscopy, and fluorescence correlation spectroscopy (FCS), we were able to investigate the chain conformation of DEATG-PPV at various concentrations, from single chain in diluted solutions to aggregates in concentrated (semi-

10.1021/jp105032y  2010 American Chemical Society Published on Web 08/20/2010

Water-Soluble Conjugated Polymer DEATG-PPV diluted) solutions and thin films, thus providing a detailed picture of the evolution of chain conformation. Correlated atomic force microscopy (AFM) and confocal fluorescence lifetime imaging microscopy (FLIM) were used to investigate the relationship between film morphology and optical properties of DEATGPPV thin films. By combining the two methods into a single experiment, we could access topographical information such as film roughness and aggregate size, concomitantly with undisturbed spatially resolved PL information (intensity, lifetime), thus providing a detailed picture of the intrinsic relationship among chain conformation, film morphology, and optical properties of the nonionic WSCP. Our work presents a pathway to control and characterize the film morphologies of WSCPs toward the optimal performance of various optoelectronic devices. Materials and Methods Polymer Synthesis. Synthesis of conjugated polymer DEATGPPV is described in detail elsewhere36 and starts off with dihydroquinone and 2-[2-(2-chloroethoxy)ethoxy]ethanol. The product was then treated with 2-chlorotriethyleamine and followed by choloromethylation using formaldehyde and HCl. Finally, we used a Gilch reaction to polymerize the monomer to form polymer DEATG-PPV. Solvents used in this work were of spectroscopic grade (Sigma Aldrich), and they were used without further purification. For confocal FLIM and AFM measurements, thin film samples were prepared by drop-casting 10 µg/mL DEATG-PPV solutions on freshly cleaned cover glasses. The DEATG-PPV solutions with chloroform and methanol were covered to slow down solvent evaporation to keep similar processing time with the water solution. Ensemble Spectroscopy. UV-vis absorption and photoluminescence spectra of DEATG-PPV solutions were measured by a Lambda 35 UV/vis spectrometer (PerkinElmer) and a Cary Eclipse Fluorescence spectrophotometer (Varian, Inc.), respectively. PL quantum yields of DEATG-PPV in various solvents were calculated against Rodamine 110 in water. Time-resolved PL spectroscopy in solution was measured by the time-correlated single-photon-counting (TCSPC) method using a FluoTime 200 spectrometer (Picoquant) in combination with the 460 nm output of a frequency-doubled Ti:Sapphire laser system (Newport Spectra Physics, Tsunami, 85 fs pulse width, 80 MHz repetition rate). Decays were measured by a detection system consisting of a monochromator, a microchannel-plate photomultiplier (Hamamatsu), and a PicoHarp 300 TSCPC analyzer, delivering on average an instrument response of 45 ps. Lifetimes were estimated by iterative reconvolution of the instrumental response function with an exponential model function, with the fit judged according to literature procedures.37,38 Lifetime contributions within a given PL decay were calculated as weighted amplitudes. Fluorescence Microscopy. Fluorescence correlation spectroscopy (FCS) measurements of DEATG-PPV in various solvents were performed on a home-built scanning-stage confocal fluorescence microscope based on an inverted Olympus IX81 coupled with the same laser system described above. Diluted DEATG-PPV was contained in a home-built sealed chamber to prevent solvent evaporation. The laser beam entering the microscope was reflected by a dichroic mirror (Di01-R442, Semrock) and focused at the sample by a 100×, 1.4 NA, oilimmersion objective lens (Olympus Japan). Fluorescence from the solution samples was collected by the same lens, filtered from laser excitation by the dichroic and by a band-pass filter (FF01-583/120, Semrock), spatially filtered by a 75 µm pinhole, split by a 50/50 beam splitter, and finally refocused onto two

J. Phys. Chem. B, Vol. 114, No. 36, 2010 11747 single-photon-counting avalanche photodiodes (APD, MPD Picoquant). The signals from the APDs were cross-correlated by a PicoHarp 300 TCSPC analyzer (PicoQuant) and analyzed with the Symphotime software (Picoquant). Data were analyzed against a three-dimensional diffusion model as described in the article below. Calibration of the instrument for FCS measurements was performed by diluted rhodamine 110 in water (D ) 430 µm2 s-1).39 Confocal fluorescence lifetime imaging (FLIM) microscopy of DEATG-PPV films was performed using the same microscope, this time by using a piezoscanning stage (Physics Instrumente) interfaced with the PicoHarp300 TCSPC analyzer. For FLIM experiments, the fluorescence was detected by a single APD. FLIM images were acquired with the Symphotime software (Picoquant). Fluorescence spectra were acquired by a Spectra Pro 2300i monochromator coupled to a back-illuminated CCD camera (Roper Scientific) by directing the collected fluorescence via a side port of the microscope and through a 100 µm pinhole. After each FLIM measurement, topographical images of DEATG-PPV films were recorded from the same scanned area by using an AFM head (NanoSurf EasyScan 2) integrated on the top of the scanning stage of the confocal microscope. AFM images were recorded in the tapping mode with a high-frequency silicon tip (Vista T300R, 300 kHz). Results Ensemble Spectroscopy of DEATG-PPV in Various Solvents. DEATG-PPV comprises a PPV backbone and side chains with three ethylene glycol (EG) repeating units and a tertiary amine end group (see chemical structure in Figure 1a). Incorporation of EG units renders solubility (more than 5 mg/ mL) in a variety of common solvents such as tetrahydrofuran (THF), dichloromethane (DCM), chloroform, methanol, and water.36 Here we focus on the interaction between DEATGPPV and three solvents with distinct polarity: chloroform, methanol, and water. The UV-vis absorption and PL spectra of DEATG-PPV in these three solvents (10 µg/mL concentration) are shown in Figures 1b and 1c, respectively. They feature broad absorption and PL bands centered at around 440 and 540 nm, respectively, associated with π-π* transitions originating from the PPV conjugated backbone. The main effect of the solvent on the spectroscopy of DEATG-PPV is spectral broadening, with a full width at the half-maximum (FWHM) increasing from chloroform to methanol and to water (see Figures 1b and 1c). For example, for PL spectra, the FWHM increases from chloroform, ∼80 nm, to methanol, ∼87 nm, and is observed as a broadening only at the blue spectral side, and to water, ∼100 nm. This evolution of the PL spectra of DEATGPPV in the three solvents is accompanied by a decrease in the PL quantum yield as follows: chlorophorm 34%, methanol 24%, water 12%. PL spectra of DEATG-PPV measured at various concentrations, from 0.4 to 10 µg/mL, showed no significant changes for chloroform and methanol, neither in shape nor in peak position. However, for water we observed a slight blue spectral shift with the increase of concentration (see Figure S1). Similarly, the time-resolved PL decays measured at various concentrations, from 0.4 to 10 µg/mL, in chloroform and in methanol showed no concentration dependency (see Figures 2a and 2b and Table 1). For these solvents, the multiexponential PL decay profile could be modeled by lifetimes (contribution) with values of 1.2 ns (14%), 0.5 ns (73%), and 0.12 ns (13%) in chloroform and 1.3 ns (12%), 0.5 ns (63%), and 0.1 ns (25%) in methanol. By contrary, the time-resolved PL profile of DEATG-PPV in water changed with the polymer concentration. Table 1 describes the polymer concentration dependency for

11748

J. Phys. Chem. B, Vol. 114, No. 36, 2010

Xu et al.

Figure 2. Concentration-dependent time-resolved photoluminenscence (PL) decays of DEATG-PPV in (a) chloroform, (b) methanol, and (c) water. (d) Time-resolved PL decays of 10 µg/mL DEATG-PPV in different solvents. Black, chloroform; red, methanol; blue, water. Panels a-d also include the response function of the instrument depicted in pink.

TABLE 1: Fitting Parameters of Time-Resolved PL Decay Curves of DEATG-PPV Solutionsa solution

R1

R2

R3

τ1 (ns)

τ2 (ns)

τ3 (ns)

chloroform (10 µg/mL) methanol (10 µg/mL) water (10 µg/mL) water (2 µg/mL) water (0.4 µg/mL)

0.14 0.12 0.21 0.18 0.13

0.73 0.63 0.43 0.43 0.36

0.13 0.25 0.36 0.39 0.51

1.21 1.26 1.44 1.41 1.29

0.54 0.46 0.57 0.53 0.43

0.12 0.10 0.074 0.068 0.042

a

Figure 1. (a) Chemical structure of poly[2,5-bis(diethylaminetetraethylene glycol)phenylene vinylene] (DEATG-PPV). (b) and (c) Normalized absorption and PL spectra, respectively, of 10 µg/mL DEATG-PPV in various solvents. Chloroform, black line; methanol, red dashed line; water, blue dashed line.

all three PL lifetimes modeling the PL decay in water. When comparing lifetimes measured in various solvents at the same DEATG-PPV concentration (10 µg/mL), from chloroform to methanol and to water, the shortest lifetime decreased in value (0.12 to 0.07 ns) and increased in contribution (see Table 1). Fluorescence Correlation Spectroscopy (FCS). FCS measurements of diluted solutions of DEATG-PPV, from 0.04 µg/ mL above, were used to characterize the size, brightness, and polymer particle number based on the following threedimensional diffusion model:39,40

G(t) )

(

1 t 1+ N τD

)

-1

(

1+

ω2xy t ωz2 τD

)

-1/2

(1)

Here G(t) is the autocorrelation of the PL intensity, N is the average number of fluorescent particles, in this case isolated

Amplitudes reported here are calculated as weighted amplitudes.

Figure 3. Fluorescence correlation spectroscopy (FCS) of DEATGPPV in various solvents. (a) Concentration dependence of particle brightness, B, for DEATG-PPV in chloroform (black), methanol (red), and water (blue). (b) FCS curves and fits according to a 3D-diffusion model (eq 1) for DEATG-PPV in very diluted solutions of chloroform (red), methanol (black), and water (red).

DEATG-PPV chains or aggregates, which are present, on average, in the focal volume, ωxy, ωz are the radial and axial focal radii, and τD is the diffusion time. The instrumental paramenters ωxy, ωz were estimated based on the diffusion coefficient measured for rhodamine 110 in water.39 Particles sizes, RH, were estimated by the Stokes-Einstein equation, D ) KB T/6 π η RH, with KB Boltzmann’s constant, T temperature, η viscosity, and D diffusion coefficient with D ) ω2xy/4τD. The particle brightness was estimated as B ) 〈I〉/N, with 〈I〉 the average PL intensity signal detected in the FCS experiment. The particle concentration was estimated as [C] ) N/Veff with Veff focal volume. Figure 3a shows the dependency of the particle brightness, B, as a function of particle concentration, [C], for all three solvents. For chlorophorm and water, FCS

Water-Soluble Conjugated Polymer DEATG-PPV

J. Phys. Chem. B, Vol. 114, No. 36, 2010 11749

Figure 4. Registered atomic force microscopy (AFM) topographical images (33 µm × 33 µm area), confocal fluorescence intensity images, and confocal fluorescence lifetime (FLIM) images from 33 µm × 33 µm areas of DEATG-PPV thin films processed with chloroform (a, d, g, respectively), methanol (b, e, h, respectively), and water (c, f, i, respectively). Note the presence of polymer aggregates as intense small islands for all three films and especially in the AFM images (panels a-c).

data from Figure 3a suggest plateau-like dependency in the region of low concentration, 6 nM and below for chloroform and 10 nM and below for water. For these concentrations, assuming that particles are composed of single chains, the calculated hydrodynamic radius of the single polymer chain is RH ∼ 9.9 nm for chloroform and RH ∼ 5.4 nm for water. For methanol, FCS data from Figure 3a indicates a monotonic increase of B with particle concentration, from 1.5 nM and above, suggesting polymer particles in methanol are composed of several polymer chains. For the lowest concentration in methanol at which a measurable FCS signal can still be attained, 1. 5nM, the estimated size for the polymer particle is RH ∼ 17 nm. Correlated AFM and Confocal FLIM of Drop-Casted DEATG-PPV Films. Drop-casted DEATG-PPV films processed from chloroform have the highest surface roughness, a few hundred nanometers presenting large aggregates and lamellar structures (see AFM image from Figure 4a). The correlated confocal fluorescence intensity and FLIM images from Figures 4d and 4g identify these micrometer-sized aggregates as bright islands with high PL signal and somewhat quenched lifetimes (see green colored regions in the FLIM image from Figure 4g.) PL spectra registered from such

Figure 5. PL spectra corresponding to polymer aggregates observed in films processed from chloroform (black, peak at 550 nm), methanol (red, peak at 550 nm), and water (blue, peak at 580 nm). Integrated PL lifetime distributions calculated from the FLIM images from Figures 4g, 4h, and 4i and corresponding to thin films processed from chloroform (black, mean value 0.89 ns, width 0.26 ns), methanol (red, mean value 0.89 ns, width 0.21 ns), and water (red, mean values 1.05 ns, width 0.38 ns).

aggregates feature a main peak at 550 nm (see Figure 5a) similar to the PL spectrum in solution. Interestingly, the integrated PL lifetime distribution shown in Figure 5b and corresponding to the entire area of the chloroform-processed film subjected to AFM and FLIM (33 × 33 µm) is narrow and symmetric, with a mean lifetime value close to the average lifetime obtained for the chloroform solution data, suggesting the presence of weakly interacting aggregates in the chloroform-processed film.38,41

11750

J. Phys. Chem. B, Vol. 114, No. 36, 2010

SCHEME 1: Proposed Chain Conformations Adopted by DEATG-PPV in Diluted Solutions of Chloroform (a, Left Side), Methanol (b, Left Side), and Water (c, Left Side) and Their Evolution into Semidiluted Solutions and Thin Films (Right Side of a, b, and c)a

a

PPV backbone is drawn in red, EG side chains in blue.

The methanol-processed film is the smoothest, with a surface roughness less than 100 nm (see AFM in Figure 4b), displaying few aggregates that have photophysical behavior similar to those observed for the chlorophorm-processed film: PL spectra peaking at 550 nm (see Figure 5a), and quenched PL lifetimes as observed by the FLIM image in Figure 4h. The integrated PL distribution of the area of the methanol-processed film subjected to AFM and FLIM (33 × 33 µm) is similar to that of the chloroform-processed film, both in shape and mean value (see Figure 5b). The most heterogeneous is the water-processed film. It has a rough surface morphology and contains many aggregates of various sizes (see AFM image, Figure 4c). This water-processed film has an integrated PL intensity by far lower that that of the chloroform and methanol-processed films (see Figure S2, Supporting Information), the most heterogeneous FLIM image (see Figure 4i), and the broadest and also bimodal integrated PL lifetime distribution (see Figure 5b). PL spectra recorded from aggregates featured by the water-processed film are spectrally red-shifted, with a peak at 580 nm (see Figure 5a). This indicates that both weak and strong interacting aggregates are present in the water-precessed film. Discussion Chain Conformation in Solution. The spectroscopic data of DEATG-PPV in diluted solutions can be explained by assuming that this particular nonionic WSCP adopts different chain conformations in different solvents: extendend polymer chain conformation in chloroform, coiled chain conformation in methanol, and collapsed chain conformation in water (see Scheme 1a-c, left side). Ultimately, these conformations are being transferred into thin (drop-casted) films (see below). We explain the spectroscopic and morphological data of DEATGPPV in diluted and semidiluted solutions and in thin film by invoking these chain conformations introduced above together with a model in which the conjugated polymer is viewed as a composition of many conjugated segments (several chro-

Xu et al. mophores) of various lengths linked by chemical and/or structural defects in the polymer chain.19,42 In this model, longer segments have lower energy than shorter segments. The overall contribution from all segments will result in inhomogeneous broadening of the polymer absorption spectrum. For PL spectrum, when optical excitation with sufficient energy is employed so that it can excite all polymer segments, shorter segments will transfer their excitation energy to longer segments of lower energy from which PL will be emitted. In diluted chlorophorm, DEATG-PPV adopts an extended conformation as proposed in Scheme 1a, left side. With the increase of polymer concentration, e.g., toward semidiluted solutions, π-π interaction between PPV backbones can drive isolated and extended polymer chains to form planar aggregates with the chain conformation almost unchanged (see Scheme 1a, right side). However, the relatively long EG side chains prevent π-conjugated backbones from stacking closely with each other to form ordered, dense aggregates, but more likely associations in the form of loose aggregates composed of few weakly interacting polymer chains causing minimal perturbation to the spectroscopy of the single polymer chain.38 In this assumption, stationary and time-resolved spectroscopic data for DEATGPPV in chlorophorm are concentration-independent, as reported herein. In the more hydrophilic solvent methanol, the hydrophobic PPV backbone adopts a coiled conformation to reduce polymer-solvent interaction (see Scheme 1b, left side). This hydrophobic interaction facilitates the polymer chains to aggregate at higher concentration, but their irregular coiled chain conformation will prevent formation of ordered and dense aggregates; hence, we propose that a similar loose aggregate composed of weakly interacting chains and with spectroscopy similar to that of a single polymer chain exists in semidiluted solutions in methanol. The similarity between the absorption spectra of DEATG-PPV in chloroform and methanol diluted solutions indicates that the π-conjugation length of chromophores (segments) is not significantly affected by the chain coiling in methanol. However, a broadening of the PL spectrum of DEATG-PPV in methanol at the blue spectral side when compared to that in chloroform indicates that some of the highenergy conjugation segments do not couple as well with lowenergy segments as they do in chlorophorm. Such a hypothesis has been proposed previously for MEH-PPV dissolved in chloroform where energy transfer along the conjugated polymer backbone was found to be weakened by the collapse of the chain and by tetrahedral defects.43 The extremely unfavorable interaction between DEATG-PPV backbone and water forces the polymer to adopt a collapsed chain conformation, which can be regarded as a single-chain aggregate (see Scheme 1c, left side). With the increase of polymer concentration, π-π interactions and strong hydrophobic interactions drive the collapsed chains to form ordered and dense aggregates. Both blue and red broadening of absorption and PL spectra were observed for DEATG-PPV in water, and they cannot be simply explained by a suppressed energy transfer. Blue broadening in both absorption and PL spectra indicates a shortening of conjugation length for some segments resulting from more collapsed chain conformation in water. The red broadening suggests the introduction of low-energy interchain excitons which result from strong π-π interaction between different polymer chains or intersegment excited states which originate from the π-stacking of conjugated segments in a single collapsed chain. PL spectra of DEATG-PPV in water are redshifted when the polymer concentration is reduced from 10 to

Water-Soluble Conjugated Polymer DEATG-PPV 0.4 µg/mL to diminish interchain aggregation, indicating that in fact intersegment excited states are more responsible for the red broadening of the PL spectra in diluted water solutions. In short, the characteristics of the absorption and PL spectra of DEATG-PPV in water support the collapsed chain conformation depicted in Figure 5c, left. The chain conformations proposed by us are also consistent with the values of the PL quantum yield of DEATG-PPV observed for the three solvents (34% chloroform, 24% methanol, 12% water) because increased structural defects (quenching sites) resulting from chain coiling and collapsing leads to a lower PL quantum efficiency in methanol and water compared to chloroform. Time-resolved PL spectroscopic studies further support the existence of the proposed chain conformations. The shortest PL lifetime component (0.12-0.07 ns) observed in all three solvents can be assigned to quenching of segments/chromophores by chemical/structural defects present in the polymer chain or to interchain/intersegment interactions. The long lifetime components (1.2-1.4 ns) reflect PL emitted by unquenched segments. As shown in Table 1, with the increasing of solvent polarity from chloroform to methanol and to water, the shortest lifetime component increases in contribution while decreasing in value, suggesting an increased number of structural defects or increased interchain/intersegment interactions. This is consistent with the extended, coiled, and collapsed conformation proposed for chloroform, methanol, and water, respectively. Direct evidence for the proposed chain conformations adopted by DEATG-PPV is given by the FCS measurements at very diluted concentrations. The difference in hydrodynamic radius of single chains of DEATG-PPV in chloform (9 nm) and water (5.4 nm) measured by FCS provides direct evidence of the extended and collapsed conformations. However, for methanol, a hydrodynamic radius for the single chain of DEATG-PPV cannot be estimated by FCS, most probably because even at very diluted concentrations, for which a sizeable FCS signal can still be obtained, this polymer still forms multiple-chain aggregates (see Figure 3a). Nevertheless FCS data in methanol support the assumption of a coiled conformation in this particular solvent. This is because FCS data in methanol indicate that multiple chains have a larger hydrodynamic radius value compared to the single chains in chloroform (extended conformation) and water (collapsed conformation). This suggests that in extremely diluted methanol solution, coiled chains can bind together through diffusion-controlled collision to form aggregates, mostly loose aggregates comprised of noninteracting chains/segments. As mentioned before, an increase in DEATG-PPV concentration for chloroform and methanol from diluted (0.4 µg/mL) to semidiluted (10 µg/mL) concentrations does not change the PL spectroscopy (shape, peak wavelength) and time-resolved PL profile, suggesting that the chain conformation is concentration independent for both solvents and that at high concentrations we deal with loose aggregates of weakly interacting polymer chains preserving the spectroscopy of the single chain (see Figures 5a and b). The evolution of chain conformation/optical properties with increasing concentration is different in water, clearly manifested by the blue shift of the PL spectrum (see Figure S1) and by the increase of the overall time-resolved PL profile (see Figure 2d). One possible explanation is that an increase in polymer concentration leads to the collapse of the PPV chains to form a cylindrical micelle structure (see Figure 5c, right), similar to that of some other WSCPs with an enhanced π-π interchain interaction resulting from backbone-solvent

J. Phys. Chem. B, Vol. 114, No. 36, 2010 11751 interaction.44 The chain conformation in these aggregates remains extended and the relatively weak π-π interchain interaction to some extent replaces the strong intersegment interaction in individual collapsed chains, resulting in an overall increase of the PL lifetimes and a blue shift of the PL spectra with increasing concentration. Nevertheless, the red-shifted and broad PL spectrum together with the strongly contributing short PL lifetime detected at semidiluted concetrations (10 µg/mL) of DEATG-PPV in water do suggest that interchain interaction in the cylindrical aggregates formed in water are stronger than in both planar aggregates in chloroform and disordered aggregates in methanol. Chain Conformation in Thin Films. With further increase of polymer concentration during the formation of solid thin films, the polymer chains aggregate and form closely packed structures. However, the chain conformations remain unchanged based on the results of film morphology and optical microscopy measurements. The rough surface topography resolved by AFM along with the large aggregates and lamellar structures of the chloroform-processed thin film is consistent with the extended chain conformation and planar aggregates formed in solution (see Figure 4a). Regular chain packing and moderate-to-weak interchain interaction in the lamellae can be visualized by the locally ordered patterns in the confocal FLIM image from Figure 4c. Chain interactions within aggregates remain weak, as demonstrated by the PL spectra in film (see Figure 5a) which are similar to those in solution (see Figure 1) and by the narrow and symmetric integrated PL lifetime distribution from film (see Figure 5c, black colored line) exhibiting a mean PL lifetime consistent with that of chloroform solution data. The relative smooth film morphology of the methanolprocessed film (see Figure 4b) is consistent with the coiled chain conformation in solution, which prevents the formation of ordered aggregates. This smooth film morphology results in uniform PL intensity (see Figure 4e), uniformly distributed PL lifetimes as observed by the FLIM image (see Figure 4h), a narrow and symmetric integrated PL lifetime distribution (see Figure 5c, red-colored line) with a mean PL lifetime consistent with the average lifetime of the methanol solution data, and a PL spectrum similar to the solution case (see Figure 5a). The closely packed (dense) aggregates formed in water solution correlate well with the relatively small-sized but bright aggregates in the corresponding water-processed thin film (see Figures 4c, 4f, and 4d). These aggregates are featured with redshifted PL spectra (see Figure 5a), as opposed to the aggregates observed in chloroform and methanol, high PL intensity, and short lifetimes, resulting from strong π-π interchain interaction and dense chain packing. The bimodal broad integrated PL lifetime distribution from Figure 5b further confirms the heterogeneous character of the water-processed film given by these strongly interacting aggregates. Conclusions The film morphology and optical properties of the nonionic water-soluble conjugated polymer DEATG-PPV can be manipulated by tuning the solvent polarity which dominates the polymer chain conformation. DEATG-PPV has an extended chain conformation in chloroform and forms loosely packed planar aggregates with increasing polymer concentration, leading to a rough morphology and lamellar structures in the drop-casted film, with heterogeneous PL intensity and lifetime distribution. The coiled chain conformation of DEATG-PPV in methanol results in smooth film morphology, homogeneous PL lifetime distribution, and high overall PL intensity. Due to the unfavor-

11752

J. Phys. Chem. B, Vol. 114, No. 36, 2010

able interaction between water and DEATG-PPV main chains, polymer chains collapse in water and form cylindrical micellar aggregates. With increasing polymer concentration, DEATGPPV forms densely packed aggregates in the case of films processed from water. These films feature red-shifted PL spectra, short PL lifetimes and low PL intensity. In summary, dissolving WSCP in solvents with distinct polarity leads to a wide range of chain conformations, as manifested by the resulting film morphologies and optical properties. The uniform morphology and weak interchain interaction of methanol-processed thin films have major implications for optoelectronic devices. Moreover, a certain degree of chain aggregation may facilitate charge injection and charge transport which are critical for the performance of both OLEDs and OPVs. By fine-tuning of solvent polarity such as solventmixing, highly optimized film morphologies of WSCPs suited for various thin film devices may be realized. Acknowledgment. Research was carried out at the Center for Functional Nanomaterials, Brookhaven National Laboratory (BNL), which is supported by the U.S. Department of Energy, Office of Basic Energy Sciences, under Contract No. DE-AC0298CH10886 (M.C. with user H.-L.W.). We thank Dr. D. Nkypanchuk and Mr. S. Baker from BNL for help with the integration of the AFM and FLIM experiments at BNL. Supporting Information Available: Concentration-dependent PL spectra of DEATG-PPV water solutions(Figure S1) and PL intensity of DEATG-PPV films processed from different solutions (Figure S2). This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) Gaylord, B. S.; Heeger, A. J.; Bazan, G. C. J. Am. Chem. Soc. 2003, 125, 896. (2) Liu, B.; Bazan, G. C. J. Am. Chem. Soc. 2004, 126, 1942. (3) Nilsson, K. P. R.; Herland, A.; Hammarstrom, P.; Inganas, O. Biochemistry 2005, 44, 3718. (4) Hoven, C. V.; Garcia, A.; Bazan, G. C.; Nguyen, T. Q. AdV. Mater. 2008, 20, 3793. (5) Huang, F.; Wu, H. B.; Wang, D.; Yang, W.; Cao, Y. Chem. Mater. 2004, 16, 708. (6) Mikroyannidis, J. A.; Barberis, V. P.; Cimrova, V. J. Polym. Sci., Part A: Polym. Chem. 2007, 45, 1016. (7) Wu, H. B.; Huang, F.; Mo, Y. Q.; Yang, W.; Wang, D. L.; Peng, J. B.; Cao, Y. AdV. Mater. 2004, 16, 1826. (8) Ma, W. L.; Iyer, P. K.; Gong, X.; Liu, B.; Moses, D.; Bazan, G. C.; Heeger, A. J. AdV. Mater. 2005, 17, 274. (9) Yang, R. Q.; Wu, H. B.; Cao, Y.; Bazan, G. C. J. Am. Chem. Soc. 2006, 128, 14422. (10) Garcia, A.; Yang, R.; Jin, Y.; Walker, B.; Nguyen, T. Q. Appl. Phys. Lett. 2007, 91, 153502. (11) Hoven, C.; Yang, R.; Garcia, A.; Heeger, A. J.; Nguyen, T. Q.; Bazan, G. C. J. Am. Chem. Soc. 2007, 129, 10976. (12) Cimrova, V.; Schmidt, W.; Rulkens, R.; Schulze, M.; Meyer, W.; Neher, D. AdV. Mater. 1996, 8, 585. (13) Edman, L.; Pauchard, M.; Liu, B.; Bazan, G.; Moses, D.; Heeger, A. J. Appl. Phys. Lett. 2003, 82, 3961.

Xu et al. (14) Cheng, C. H. W.; Boettcher, S. W.; Johnston, D. H.; Lonergan, M. C. J. Am. Chem. Soc. 2004, 126, 8666. (15) Kawai, T.; Yamaue, T.; Onoda, M.; Yoshino, K. Synth. Met. 1999, 102, 971. (16) Yang, J. H.; Garcia, A.; Nguyen, T. Q. Appl. Phys. Lett. 2007, 90, 103514. (17) Khan, A.; Muller, S.; Hecht, S. Chem. Commun. 2005, 584. (18) Tan, C. Y.; Pinto, M. R.; Kose, M. E.; Ghiviriga, I.; Schanze, K. S. AdV. Mater. 2004, 16, 1208. (19) Schwartz, B. J. Annu. ReV. Phys. Chem. 2003, 54, 141. (20) Nguyen, T. Q.; Martini, I. B.; Liu, J.; Schwartz, B. J. J. Phys. Chem. B 2000, 104, 237. (21) Blom, P. W. M.; Mihailetchi, V. D.; Koster, L. J. A.; Markov, D. E. AdV. Mater. 2007, 19, 1551. (22) Zhang, Y.; Huang, F.; Chi, Y.; Jen, A. K. Y. AdV. Mater. 2008, 20, 1565. (23) Quan, S. Y.; Teng, F.; Xu, Z.; Qian, L.; Hou, Y. B.; Wang, Y. S.; Xu, X. R. Eur. Polym. J. 2006, 42, 228. (24) Traiphol, R.; Sanguansat, P.; Srikhirin, T.; Kerdcharoen, T.; Osotchan, T. Macromolecules 2006, 39, 1165. (25) Li, G.; Yao, Y.; Yang, H.; Shrotriya, V.; Yang, G.; Yang, Y. AdV. Funct. Mater. 2007, 17, 1636. (26) Kim, Y.; Choulis, S. A.; Nelson, J.; Bradley, D. D. C.; Cook, S.; Durrant, J. R. Appl. Phys. Lett. 2005, 86, 063502. (27) Mwaura, J. K.; Pinto, M. R.; Witker, D.; Ananthakrishnan, N.; Schanze, K. S.; Reynolds, J. R. Langmuir 2005, 21, 10119. (28) Lukkari, J.; Salomaki, M.; Aaritalo, T.; Loikas, K.; Loiko, T.; Laiho, T.; Kankare, J. Langmuir 2002, 18, 8496. (29) Nguyen, T. Q.; Doan, V.; Schwartz, B. J. J. Chem. Phys. 1999, 110, 4068. (30) Huser, T.; Yan, M.; Rothberg, L. J. Proc. Natl. Acad. Sci. U.S.A. 2000, 97, 11187. (31) Wang, D. L.; Lal, J.; Moses, D.; Bazan, G. C.; Heeger, A. J. Chem. Phys. Lett. 2001, 348, 411. (32) Li, Y. C.; Chen, C. Y.; Chang, Y. X.; Chuang, P. Y.; Chen, J. H.; Chen, H. L.; Hsu, C. S.; Ivanov, V. A.; Khalatur, P. G.; Chen, S. A. Langmuir 2009, 25, 4668. (33) Barbara, P. F.; Gesquiere, A. J.; Park, S. J.; Lee, Y. J. Acc. Chem. Res. 2005, 38, 602. (34) Sartori, S. S.; De Feyter, S.; Hofkens, J.; Van der Auweraer, M.; De Schryver, F.; Brunner, K.; Hofstraat, J. W. Macromolecules 2003, 36, 500. (35) Lin, H. Z.; Tabaei, S. R.; Thomsson, D.; Mirzov, O.; Larsson, P. O.; Scheblykin, I. G. J. Am. Chem. Soc. 2008, 130, 7042. (36) Wang, C. C.; Tsai, H. H.; Shih, H. H.; Jeon, S. H.; Xu, Z. H.; Williams, D.; Iyer, S.; Sanchez, T. C.; Wang, L.; Cotlet, M.; Wang, H. L. ACS Appl. Mater. Interfaces 2010, 2, 738. (37) Maus, M.; Rousseau, E.; Cotlet, M.; Schweitzer, G.; Hofkens, J.; Van der Auweraer, M.; De Schryver, F. C.; Krueger, A. ReV. Sci. Instrum. 2001, 72, 36. (38) Wang, C. C.; Gao, Y.; Shreve, A. P.; Zhong, C.; Wang, L.; Mudalige, K.; Wang, H. L.; Cotlet, M. J. Phys. Chem. B 2009, 113, 16110. (39) Gendron, P. O.; Avaltroni, F.; Wilkinson, K. J. J. Fluor. 2008, 18, 1093. (40) Haustein, E.; Schwille, P. Annu. ReV. Biophys. Biomol. Struct. 2007, 36, 151. (41) Yu, J.; Hu, D. H.; Barbara, P. F. Science 2000, 289, 1327. (42) Samuel, I. D. W.; Crystall, B.; Rumbles, G.; Burn, P. L.; Holmes, A. B.; Friend, R. H. Chem. Phys. Lett. 1993, 213, 472. (43) Traiphol, R.; Charoenthai, N.; Srikhirin, T.; Kerdeharoen, T.; Osotchan, T.; Maturos, T. Polymer 2007, 48, 813. (44) Burrows, H. D.; Tapia, M. J.; Fonseca, S. M.; Pradhan, S.; Scherf, U.; Silva, C. L.; Pais, A. A. C. C.; Valente, A. J. M.; Schillen, K.; Alfredsson, V.; Carnerup, A. M.; Tomsic, M.; Jamnik, A. Langmuir 2009, 25, 5545.

JP105032Y