Anisotropic Elastic Modulus of Oriented Regioregular Poly(3

Dec 23, 2015 - Kiarash Vakhshouri , Brandon H. Smith , Edwin P. Chan , Chenchen ... Joshua I. Scott , Xiao Xue , Ming Wang , R. Joseph Kline , Benjami...
1 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Macromolecules

Anisotropic Elastic Modulus of Oriented Regioregular Poly(3hexylthiophene) Films Omar M. Awartani,† Bingxiao Zhao,† Tyler Curie,† R. Joseph Kline,‡ Mohammed A. Zikry,† and Brendan T. O’Connor*,† †

Department of Mechanical and Aerospace Engineering, North Carolina State University, Raleigh, North Carolina 27695, United States ‡ Material Measurement Laboratory, National Institute of Standards and Technology, Gaithersburg, Maryland 20899, United States S Supporting Information *

ABSTRACT: Specific morphological features of polymer semiconductors are often promoted in devices to optimize optoelectronic behavior. Less studied is the role of morphology on the mechanical properties of the film, such as elastic modulus, which is an important property for the development of flexible and stretchable devices. To gain insight into the morphological origin of elasticity in polymer semiconductors and its relationship to charge transport, we study the anisotropic in-plane elastic modulus of strain-aligned regioregular poly(3-hexylthiophene) (P3HT) films and compare the results to previously measured field effect charge mobility. The film morphology is varied through the amount of applied strain and post strain thermal annealing. Morphological characterization includes UV−vis optical spectroscopy and X-ray diffraction. The elastic modulus is measured using a buckling-based measurement technique. The elastic modulus of the film is found to decrease as the film is plastically strained. Thermally annealing the strained films results in a large in-plane elastic modulus anisotropy, where the modulus increases in the direction of backbone alignment and decreases in the transverse direction. The measured elastic modulus is compared to the film morphology, showing a dependence on both in-plane polymer chain alignment and local aggregate order. Comparing the elastic modulus to field effect mobility shows that they are not necessarily correlated, which has important implication for flexible organic electronic device design.

■

INTRODUCTION One of the advantages of polymer semiconductors is their mechanical flexibility allowing for novel device applications where conventional inorganic semiconductors are difficult to implement.1−4 Critical to the success of flexible (and stretchable) polymer semiconductor devices is a detailed understanding of the mechanical response of the constituent materials. Similar to optoelectronic performance in polymer semiconductors, mechanical behavior has been shown to depend significantly on molecular structure5−7 and film morphology.8−11 Determining the morphological features that dictate the mechanical properties of polymer semiconductors and relating them to optoelectronic behavior will aid the development of flexible and physically robust organic electronics. A method to gain insight into the mechanical properties of polymers is through processing oriented films with long-range in-plane chain alignment.12−16 The anisotropic structural features can then be related to the anisotropic mechanical behavior highlighting morphological features that impact the mechanical response of the film. Oriented polymers have traditionally been exploited for the development of mechanically stronger materials in structural applications.12 Studies © XXXX American Chemical Society

have shown that the elastic modulus can increase significantly in the direction of chain alignment,13,14,17,18 resulting in polymers with very high stiffness to weight ratios under certain loading conditions.19 Similarly, oriented polymer semiconductors have been used to correlate microstructure with electronic properties20−23 and have been exploited to enhance device performance.24 For example, aligning polymer semiconductors in the plane of the film is a common approach to maximize charge carrier mobility in thin film transistors, where charge transport is typically favored in the direction of backbone alignment.22,24 Oriented polymer semiconductors also have anisotropic optoelectronic characteristics and have been used for polarized electroluminescence in organic light-emitting devices25 and polarization-sensitive photovoltaic devices.26,27 Correlating the anisotropic mechanical and optoelectronic properties in oriented polymer semiconductor films will provide a deeper understanding of their relationships and provide guidelines for improved flexible device design.28−30 In this study, we investigate the elastic modulus of oriented regioregular poly(3-hexylthiophene) (P3HT) films. P3HT is Received: December 10, 2015

A

DOI: 10.1021/acs.macromol.5b02680 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules chosen as it is one of the most widely studied polymer semiconductor materials.31−33 The in-plane orientation of P3HT films can also be well controlled through the application of large physical strains.22,34 Previously, mechanical properties of regiorandom poly(3-alkylthiophenes) were investigated showing increased stiffness and strength in oriented bulk samples (∼40 μm thick).29 Here, we focus on regioregular P3HT at device relevant film thicknesses along with detailed morphological characterization. The in-plane orientation of the P3HT films is controlled by the applied strain, and the local order is varied through thermal annealing at 180 °C for 10 min.22,34 In previously demonstrated strain-aligned P3HT films, charge mobility was shown to increase significantly in the direction of backbone alignment and decrease in the transverse direction, resulting in a large mobility anisotropy. The increase in mobility was attributed to an increase in charge transport along the polymer backbone. However, when the strain-aligned films were thermally annealed, the mobility in the direction of strain dropped considerably along with the mobility anisotropy. This drop in mobility was primarily attributed to a reduction of interaggregate polymer tie chains that provided efficient charge carrier coupling between aggregates. In these films, the morphology and charge mobility both vary considerably, providing an opportunity to gain insight into potential relationships between these features and the film’s compliance. Below, the morphology of strain-aligned films is reported followed by characterization of the film elasticity. The film morphology is characterized with UV−vis optical spectroscopy and grazing incidence X-ray diffraction (GIXD). The elastic moduli of the films are measured using a buckling-based metrology method. Relationships between film morphology, elastic modulus, and charge mobility in the oriented P3HT films are then discussed. We show that as the polymer chains are strain aligned and the charge mobility along the strain direction increases, the elastic modulus decreases. Once the strained films are thermally annealed, where charge mobility is found to drop, a large elastic modulus anisotropy develops with the stiffness increasing considerably in the direction of chain alignment. By comparing the film morphology to the elasticity and charge mobility, we find that there are microstructural features that correlate with both properties as well as features that primarily impact only one of the properties.

Figure 1. UV−vis absorbance under polarized light parallel (para) and perpendicular (perp) to the applied strain direction for films strained by 0%, 50%, and 100% for (a) as-cast films and (b) thermally annealed films.

from 0 for films with no overall in-plane preferential alignment of the polymer backbone to 1 for complete uniaxial alignment of the polymer backbone in the plane of the film. For the films under study, S2 varies from 0 for the unstrained films to 0.58 for a film strained by 100% and thermally annealed at 180 °C, as shown in Figure 2d. In addition to the average in-plane orientation of the polymer chains in the film, the absorbance can provide additional details of the local morphology including aggregate width, aggregate packing order, and percent aggregated polymer.37,38 In order to

■

RESULTS AND DISCUSSION Morphology Characterization. The morphology of the strained films is first characterized by UV−vis optical spectroscopy under linear polarized incident light. As shown in Figure 1, as the film is strained, the absorbance of polarized light parallel to the strain direction is found to increase relative to the absorbance of polarized light perpendicular to the strain direction. Given that the primary optical transition dipole of P3HT is along the backbone,35 this indicates that the polymer backbone aligns in the direction of applied strain, with an illustration of the alignment shown in Figure 4a.22 Thermally annealing the strained films further increases the in-plane alignment of the backbone, as shown in Figure 1b.23 To quantify the level of in-plane alignment, we use a 2-dimensional order parameter defined as S2 = (R − 1)/(R + 1), where R is the dichroic ratio of the film.23 Here, the dichroic ratio is defined as R = Apara/Aperp, where Apara is the maximum absorbance of polarized light parallel to the strain direction, and A perp is the maximum absorbance of polarized light perpendicular to the strain direction. Generally, S2 varies

Figure 2. (a, b) Weakly coupled H-aggregate model fitting parameters of the strain-aligned P3HT films including (a) Gaussian line width (σ) and (b) free-exciton bandwidth (W) for as-cast (AC) and thermally annealed (AN) films. (c) Estimated percent P3HT aggregate for absorbed light polarized light parallel (para) and perpendicular (perp) to the strain direction and the estimated percent aggregate for the entire film. (d) The two-dimensional order parameters for the strained films as measured by UV−vis spectroscopy. B

DOI: 10.1021/acs.macromol.5b02680 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules quantify the differences in aggregate order for the various films, we fit the absorbance to a weakly coupled H-aggregate model,37,39,40 with fits given in Figures S1 and S2. This model has been previously described in detail,38−40 and here we focus on the model fits to the data. The fitting parameters of the model include the exciton bandwidth (W) and the Gaussian line width (σ). W can be related to the average conjugation length of the P3HT aggregate chains, where a lower W indicates an increase in aggregate width (illustrated in Figure 4b).41 σ is related to the energetic disorder, where a lower σ indicates greater aggregate order both intra- and intermolecularly.37 The percent aggregate P3HT in the film is also estimated using the weakly coupled H-aggregate model, following a previously described method.40 W, σ, and percent aggregate for the strained as-cast and thermally annealed films are given in Figure 2a−c. Generally, we find that as the film is strained the P3HT aggregates have reduced local order (inferred from σ) and smaller aggregate width (inferred from W). The aggregates with chains that align in the direction of strain have larger width but slightly lower local order compared to aggregates that remain more misaligned. Once the strained films are thermally annealed, the aggregate order improves and is similar to the unstrained films. The width of the highly oriented aggregates increases significantly while the more misaligned polymer aggregates remain similar to the corresponding as-cast film. Finally, we find that there is a slight increase in total percent aggregate as the film is strained for both as-cast and annealed films, as shown in Figure 2c. Additional details on how σ and W relate to film morphology and how the percent aggregation in the oriented films was determined are provided in the Supporting Information. The P3HT crystal stacking characteristics are measured with grazing incidence X-ray diffraction (GIXD). Two-dimensional (2D) image plate data for the 0% and 100% strained films for both as-cast and annealed conditions are given in Figure S3. From the diffraction images, a clear in-plane diffraction anisotropy develops consistent with alignment of the polymer chains in the direction of strain. While the in-plane orientation distribution cannot be quantitatively determined with the image plate data, the results are consistent with previous measurements of similar films where both 2D images and highresolution in-plane φ-scans of the (100) scattering peak were reported.22 The previous φ-scans show that in films strained by 50% or more the vast majority of P3HT crystallites are oriented such that the polymer backbone is within 30° of the strain alignment direction. To compare the out-of-plane orientation distribution, pole figures that consider the (100) scattering peak with out-of-plane angle (ω) are constructed from the GIXD image plate data for the unstrained and 100% strained films, shown in Figure 3. Given that there is significant crystalline alignment in the strained films, we focus on the out-of-plane orientation of the polymer chains aligned in the direction of strain (scattering vector nominally perpendicular to the strain direction). It should be noted that Figure 3 is not corrected for the variation in measured solid-angle with ω or the variation in the in-plane orientation distribution of polymer chains between films.42 Thus, the measurement does not provide the fraction of crystals at a given tilt angle but can be used to compare the relative distribution of crystal orientations in the films. The pole figures show that the unstrained films have a preferential edgeon stacking behavior with a similar orientation distribution for as-cast and thermally annealed processing conditions. An illustration of edge-on packing is shown in Figure 4b. In the

Figure 3. (a) Pole figure of the (100) scattering peak for the unstrained films (0%) and 100% strained films with incident X-ray parallel to the strain direction that are as-cast (AC) and thermally annealed (AN).

Figure 4. (a) Top-view schematic of strain-aligned films showing the semicrystalline nature of the films and the alignment of the aggregate P3HT in the direction of strain. (b) Example illustration of an edge-on stacking P3HT aggregate showing the aggregate width and relative orientation of the conjugated ring plane with respect to the substrate plane. (c) Schematic of the buckling method employed in this study to measure the in-plane anisotropy of the elastic modulus. (d) Microscope images of the buckled films for the 100% strained and thermally annealed film, loaded parallel (θ = 0°, E0) and perpendicular (θ = 90°, E90) to the strain direction.

strained films, a broad out-of-plane orientation distribution is found for the strain-aligned P3HT crystals, and there is not a significant out-of-plane stacking preference for either the as-cast or thermally annealed films.42 It is important to note that the C

DOI: 10.1021/acs.macromol.5b02680 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules P3HT aggregates are not the same as crystals.43 Nevertheless, we assume a similar out-of-plane orientation distribution exists for the P3HT aggregates, allowing us to combine aspects of the morphological information provided by UV−vis spectroscopy and GIXD. Elastic Modulus Characterization. The elastic modulus of the films is measured using a buckling based metrology method.44 Briefly, this consists of applying a compressive strain to the film while on a host elastomer substrate, which in this study is polydimethylsiloxane (PDMS). The modulus mismatch between the film and substrate results in a buckling instability with a characteristic buckling wavelength. The elastic modulus of the film is then given by44 ⎛ 1 − ν 2 ⎞⎛ λ ⎞ 3 f ⎟ ⎟⎜ Ef = 3Es⎜ 2 ⎝ 1 − νs ⎠⎝ 2πh ⎠

minimum occurs at draw ratios of approximately 1.5.16 In the thermally annealed P3HT films, E0 is found to increase with strain as compared to the unstrained films, while E90 behaves similar to the strained as-cast counterpart. This results in an increasing in-plane elastic modulus anisotropy with greater inplane polymer alignment. The angular dependence of the elastic modulus for the 100% strained, thermally annealed film is given in Figure 6. In

(1)

where Ef is the elastic modulus of the thin film, Es is the elastic modulus of the substrate, νf and νs are the Poisson’s ratios for the thin film and substrate, respectively, λ is the average buckling wavelength, and h is the thickness of the film. This approach has been previously shown to compare well to flexural modulus measured in bulk plastic samples.44 The elastic modulus anisotropy in the oriented films is measured by compressing the film in different directions relative to the strain axis. A schematic of the buckling method and characteristic buckling patterns from a strain-aligned P3HT film on PDMS is given in Figure 4. The measured elastic modulus along the strain direction (E0) and in the transverse direction (E90) is given in Figure 5 with

Figure 6. Elastic modulus of 100% strained and thermally annealed P3HT film as a function of compression angle (θ) with respect to the strain direction. The model fit is made using eq 2. Inset: the finite element model (FEM) predictions of the buckling wavelength for the film for loading parallel (θ = 0°) and perpendicular (θ = 90°) to the strain direction.

oriented polymer films, the elastic modulus at angles away from the primary chain alignment axis may be found anywhere from greater than to less than the primary loading directions (i.e., E0 and E90)16,45 and thus is an important feature to characterize. Here we find the angular dependence of the modulus varies continuously between the high modulus of E0 and low modulus at E90. For an oriented film that is transversely isotropic, the elastic modulus (Eθ) as a function of loading angle (θ) is given by46,47 2ν ⎞ cos 4(θ) ⎛ 1 sin 4(θ) 1 = +⎜ − E ⎟ sin 2(θ ) cos2(θ ) + Eθ E0 E0 ⎠ E90 ⎝ G12 (2)

where G12 is the shear modulus and νE is the Poisson ratio of the polymer for a tensile load applied to the strain direction (i.e., extension direction). The angular dependence of the elastic modulus given by eq 2 is fit to the experimental data in Figure 6, assuming νE = 0.33 and using G12 as a fitting parameter, which is found to be 0.33 GPa. While the oriented P3HT film may not be strictly transversely isotropic due to potential stiffness differences normal to the plane of the film, the large out-of-plane orientation distribution of the polymer’s conjugation ring plane suggests that this is a reasonable approximation. The applicability of eq 1 to quantify the in-plane elastic modulus for a transversely isotropic film is also verified by comparing the results with finite element analysis (FEA) modeling. Details of the FEA model are given in the Supporting Information. For the unstrained films, using the experimentally determined elastic modulus of P3HT and PDMS, the predicted buckling wavelength was 4.0 μm, the same as the average experimentally obtained value. For the 50% and the 100% strained film, the experimentally determined transversely

Figure 5. Elastic modulus of as-cast (AC) and thermally annealed (AN) P3HT films, parallel (para) and perpendicular (perp) to orientation direction, as a function of order parameter (S2) obtained from UV−vis absorbance measurements.

various applied strains. For the 0% strained P3HT films (S2 = 0), the in-plane elastic modulus does not show a measurable anisotropy, as expected. Annealing the unstrained film does not significantly change the measured modulus, which is consistent with the similar absorbance and diffraction behavior. As the film is strained, both E0 and E90 are found to decrease with increasing applied strain. For the 100% strained films (S2 = 0.58), E0 and E90 begin to diverge with the modulus along the direction of strain beginning to increase. This behavior is observed in some tensile drawn (i.e., large physically strained) polymers, including polyethylene, where the modulus along the alignment direction has been shown to decrease with strain until reaching a minimum and subsequently increasing with further chain alignment.15,16 For low-density polyethylene this D

DOI: 10.1021/acs.macromol.5b02680 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules isotropic moduli, E0 and E90, were used in the FEA model. The FEA model accurately predicts the measured wavelength anisotropy, with the predicted buckling pattern of the 100% strained film shown in Figure 6 (inset). The differences between the predictions and the experimentally determined wavelength were within 3% for both the 50% and 100% stained films in both primary loading directions. These results support that eq 1 accurately determines the elastic moduli for the different loading directions in transversely isotropic films. Correlating Structure with Mechanical and Electrical Properties. In the strain-aligned films, a clear change in film stiffness is observed. This change in stiffness is accompanied by a number of morphological changes. Perhaps the most drastic change in the morphology of the films upon strain is the alignment of the polymer backbone in the plane of the film. However, polymer alignment, over the range considered here, is not a major driving force for increasing film stiffness if the oriented P3HT aggregates are not well ordered. In fact, while the polymer chains are aligning in the strained as-cast films, we see a decrease in the elastic modulus due to the morphological disorder introduced by the applied strain. To gain insight into the morphological features driving the elastic modulus, we compare to the elastic modulus with the calculated σ, W, and percent aggregate along the appropriate in-plane measurement direction. A significant correlation is only found between elastic modulus and W, as shown in Figure 7, with a correlation

illuminate a possible correlation. In addition, the influence of the out-of-plane stacking orientation of the P3HT crystals on the in-plane elastic modulus cannot be resolved in the measured films. It is expected that the π-stacking and alkylstacking directions will have different moduli. However, the out-of-plane stacking orientation distributions for the oriented as-cast and annealed films are very similar, limiting the ability to resolve potential differences. Finally, it is important that these results are put into context of the P3HT charge transport characteristics. In previous reports on strain-aligned regioregular P3HT films,22,23 the saturated field effect mobility measured in a transistor configuration was found to increase in the direction of applied strain for as-cast films, increasing from approximately 0.015 cm2 V−1 s−1 for an unstrained film to 0.046 cm2 V−1 s−1 for a 100% strained film.23 However, when thermally annealing the film, a pronounced decrease in field effect mobility was found, dropping to approximately 0.012 cm2 V−1 s−1.23 Comparing the mobility to the measured elastic modulus, the more compliant material has greater charge mobility in the direction of backbone alignment, suggesting that film stiffness and charge mobility are not necessarily correlated. In a perfect crystal, under deformation potential scattering theory, there is a direct relationship between charge mobility and bulk modulus.50,51 However, typical film disorder results in a charge transport bottleneck that is not captured by this model.51 This is evidenced by the drop in mobility for the thermally annealed oriented films, which are found to have higher local order and higher elastic modulus. The drop in mobility in the thermally annealed films was previously attributed to the formation of well-defined P3HT grains reducing the density of molecular tie chains that provided an efficient charge transport link between aggregates. This unique feature appears to have a large impact on charge mobility but not on film stiffness. In the strainaligned as-cast films, the mobility improves despite increased aggregate disorder that lowers the elastic modulus. Since increasing the local order through thermal annealing also results in tie-chain removal, it is difficult to isolate the impact of local aggregate order on charge mobility in this study. If the tiechain density was maintained in the thermally annealed films, the improved aggregate order may improve charge mobility through improved electronic coupling, maintaining a correlation between charge mobility and elasticity. Nevertheless, these results suggest that reducing aggregate order while maintaining efficient intermolecular coupling is a potential route to achieve high charge mobility and compliant films. Further research is required to determine the extent to which this is broadly applicable to recently developed donor−acceptor polymers that are characterized by highly planar and rigid backbones and films that are not necessarily highly crystalline.50 Finally, these results also highlight an opportunity to exploit the anisotropic modulus of oriented films for the development of flexible electronic devices that require in-plane charge transport, such as field effect transistors. In such applications, devices may be designed such that the stiffer and higher charge mobility axis is transverse to the primary flexing direction. Thus, the more compliant direction in the film will experience larger strain under flexure reducing local stresses in the device.

Figure 7. Elastic modulus of aligned as-cast (AC) and thermally annealed (AN) P3HT films as a function of exciton bandwidth (W) measured with polarized light parallel (||) and perpendicular (⊥) to the strain direction.

coefficient to a linear fit of −0.87. No clear correlation is observed with σ or percent aggregate, given in Figure S4. These results suggest that the aggregate width plays a significant role in the elastic modulus of the film, perhaps through an increase in backbone rigidity due to improved electronic coupling along the backbone. The data points that deviate low in this correlation are the strained as-cast films with the elastic modulus measured parallel to the strain direction, which corresponds to the most disordered aggregates (i.e., highest σ). In addition to W, the alignment of the polymer backbone is a contributing factor in the measured elastic modulus for the thermally annealed films. In these films, σ and W do not change substantially in the strained films, and the increase in measured elastic modulus is driven by the increase in chain alignment. It is important to note that while a correlation was not found between modulus and percent aggregate, the range of percent aggregate was not large, and expanding this range may

■

CONCLUSION In this study we considered the elastic modulus of strainaligned P3HT films. As the films are strained, we found that the elastic moduli of the films decrease, and little elastic modulus E

DOI: 10.1021/acs.macromol.5b02680 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

measured to be 125 nm for a 0% strained P3HT film, 102 nm for a 50% strained films, and 88 nm for a 100% strained film. The thickness of the 0% strained film is based on thickness measurements of 10 films. The thicknesses of the strained films were obtained using the equation t2 = t1/(1 + ϵ)1−ν, where t2 is the final thickness, t1 is the initial thickness, ϵ is the strain percentage, and ν is Poisson’s ratio taken as 0.5. This estimate was verified for accuracy with controlled strained films measured with VASE. Measuring the elastic modulus of aligned P3HT with varying compression offset angle (θ) was performed by buckling the film on a circular PDMS slab, illustrated in Figure 4. The elastic modulus of the PDMS was measured using an Instron 5943 tensile tester and found to be 0.68 MPa. The uncertainty in the reported elastic modulus is based on the standard deviation of the mean buckling wavelength measured over three to five films.

anisotropy was observed until high levels of strain are applied (100% strain). Once the strained films were annealed, a large modulus anisotropy was found. This anisotropy is primarily due to the increase in elastic modulus in the direction of backbone alignment. Considering the morphological changes in the film, not only was the alignment of the backbone in the direction of strain a contributing factor but also the P3HT aggregate character, and more specifically the width and local order of the aggregates. These morphological features are all coupled, which limits a quantitative predictive relationship. However, these results qualitatively highlight the relative role of local aggregate order and chain alignment on the elastic modulus of the film. These results have important implications for transistor design, where it has been shown that the field effect mobility increases in the direction of alignment in as-cast films (low in-plane modulus) and decreases with thermal annealing (high in-plane modulus).22,23 This suggests that aligned films may have opportunities to improve charge mobility while not necessarily increasing film stiffness. The anisotropic mechanical and electrical properties also provide significant opportunities to design flexible devices with reduced internal stress concentrations.

■

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.macromol.5b02680. Weakly coupled H-aggregate model fits to experimental absorbance for each of the films under different polarizations, interpretation of the model fits, plots of elastic modulus vs Gaussian line width (σ) and percent aggregate, 2-D GIXD image plate data, and more details on the finite-element analysis modeling (PDF)

EXPERIMENTAL METHODS

Film Preparation. The P3HT was obtained from Sigma-Aldrich51 and had a molecular weight of 53 kg/mol and a polydispersity of 1.2. P3HT films were cast from a 20 mg/mL solution in 1,2dichlorobenzene prepared in a nitrogen-filled glovebox. The solution was spun-cast at room temperature on octyltrichlorosilane (OTS)treated silicon substrates at 1000 rpm for 60 s. All substrates used in this study were cleaned by sonication in deionized water, acetone, and isopropanol for 10 min each followed by UV-ozone treatment for 10 min. The substrates were then thoroughly rinsed with water and dried with nitrogen. OTS surface modification was performed using a previously described process.34 In order to align the P3HT film in a controlled fashion, the films were transferred from the OTS-Si substrate to a slab of PDMS (20:1 elastomer to cross-linker ratio) attached to a custom strain stage. This was done by laminating the PDMS onto the P3HT film followed by fast removal of the substrate. The low surface energy of the OTS-Si substrate as well as the fast substrate removal promoted adhesion of the P3HT to the PDMS slab.52 The PDMS was then strained and held at the desired strain. The aligned P3HT film was then laminated back to an OTS-treated glass or silicon substrate for further characterization. The films is transferred off the PDMS onto the second substrate by applying a small shear load and slowly pulling away the PDMS stamp.52 The final films were measured either without further treatment (as-cast) or thermally annealed at 180 °C for 10 min (annealed films). Morphology Characterization. UV−vis absorbance measurements were performed using a Jazz spectrometer from Ocean Optics. Absorbance measurements of a glass substrate were used as reference. The absorbance measurements were made with linear polarized light, either parallel or perpendicular to the strain direction. GIXD measurements were performed at the Stanford Synchrotron Radiation Lightsource (SSRL) on beamline 11-3. The 2D GIXD images were collected on an area detector (MAR345 image plate), with 12.735 keV X-ray beam and an incidence angle of ≈0.12°. The samples were aligned with their in-plane orientation either perpendicular or parallel to the incident beam. The sample chamber was purged with helium during the scattering experiments to reduce beam damage and background scattering. Elastic Modulus. To measure the elastic modulus, the films now on secondary OTS-treated Si substrates are transferred back to a PDMS substrate. The film composite is placed under small compressive strain until buckling instability is observed under an optical microscope. The thickness of the P3HT films was measured using a variable angle spectroscopic ellipsometer (VASE) and was

■

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected] (B.T.O.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This research work was supported by the National Science Foundation Award CMMI-1200340. Portions of this research were carried out at the Stanford Synchrotron Radiation Lightsource, a Directorate of SLAC National Accelerator Laboratory, and an Office of Science User Facility operated for the U.S. Department of Energy Office of Science by Stanford University. We thank Dr. Michael F. Toney for assistance with the X-ray diffractions measurements. We also thank Dr. Michael Dickey for assistance with the tensile test measurements.

■

REFERENCES

(1) Li, K.; Zhen, H.; Niu, L.; Fang, X.; Zhang, Y.; Guo, R.; Yu, Y.; Yan, F.; Li, H.; Zheng, Z. Adv. Mater. 2014, 26, 7271−7278. (2) Kaltenbrunner, M.; White, M. S.; Głowacki, E. D.; Sekitani, T.; Someya, T.; Sariciftci, N. S.; Bauer, S. Nat. Commun. 2012, 3, 770. (3) Leo, K. Nat. Photonics 2011, 5 (12), 716−718. (4) Sekitani, T.; Zschieschang, U.; Klauk, H.; Someya, T. Nat. Mater. 2010, 9 (12), 1015−1022. (5) O’Connor, B.; Chan, E. P.; Chan, C.; Conrad, B. R.; Richter, L. J.; Kline, R. J.; Heeney, M.; McCulloch, I.; Soles, C. L.; DeLongchamp, D. M. ACS Nano 2010, 4 (12), 7538−7544. (6) Savagatrup, S.; Printz, A. D.; O’Connor, T. F.; Zaretski, A. V.; Rodriquez, D.; Sawyer, E. J.; Rajan, K. M.; Acosta, R. I.; Root, S. E.; Lipomi, D. J. Energy Environ. Sci. 2015, 8 (1), 55−80. (7) Bruner, C.; Dauskardt, R. Macromolecules 2014, 47, 1117−1121. (8) Awartani, O.; Lemanski, B. I.; Ro, H. W.; Richter, L. J.; DeLongchamp, D. M.; O’Connor, B. T. Adv. Energy Mater. 2013, 3 (3), 399−406. (9) Zhao, B.; Zikry, M. A. Polymer 2015, 57, 1−11. F

DOI: 10.1021/acs.macromol.5b02680 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules (10) Savagatrup, S.; Printz, A. D.; Rodriquez, D.; Lipomi, D. J. Macromolecules 2014, 47 (6), 1981−1992. (11) Tummala, N. R.; Bruner, C.; Risko, C.; Bredas, J.-L.; Dauskardt, R. H. ACS Appl. Mater. Interfaces 2015, 7, 9957−9964. (12) Blumentritt, B. F. J. Appl. Polym. Sci. 1979, 23, 3205−3217. (13) Biswas, P. K.; Sengupta, S.; Basu, A. N. Colloid Polym. Sci. 1986, 264 (2), 128−130. (14) Ye, C.; Singh, G.; Wadley, M.; Karim, A.; Cavicchi, K.; Vogt, B. Macromolecules 2013, 46, 8608−8615. (15) Ward, I. M. Proc. Phys. Soc., London 1962, 80, 1176−1188. (16) Raumann, G.; Saunders, W. Proc. Phys. Soc., London 1961, 77, 1028−1037. (17) Rueda, D. R.; Balta Calleja, F. J.; Garcia Pena, J.; Ward, I. M.; Richardson, A. J. Mater. Sci. 1984, 19 (8), 2615−2621. (18) Coran, A. Y.; Boustany, K.; Hamed, P. J. Appl. Polym. Sci. 1971, 15 (10), 2471−2485. (19) Capaccio, G.; Ward, I. M. Polym. Eng. Sci. 1975, 15 (3), 219− 224. (20) Jimison, L. H.; Toney, M. F.; McCulloch, I.; Heeney, M.; Salleo, A. Adv. Mater. 2009, 21 (16), 1568−1572. (21) Lee, M. J.; Gupta, D.; Zhao, N.; Heeney, M.; McCulloch, I.; Sirringhaus, H. Adv. Funct. Mater. 2011, 21 (5), 932−940. (22) O’Connor, B.; Kline, R. J.; Conrad, B. R.; Richter, L. J.; Gundlach, D.; Toney, M. F.; DeLongchamp, D. M. Adv. Funct. Mater. 2011, 21 (19), 3697−3705. (23) O’Connor, B. T.; Reid, O. G.; Zhang, X.; Kline, R. J.; Richter, L. J.; Gundlach, D. J.; Delongchamp, D. M.; Toney, M. F.; Kopidakis, N.; Rumbles, G. Adv. Funct. Mater. 2014, 24, 3422−3431. (24) Tseng, H.-R.; Phan, H.; Luo, C.; Wang, M.; Perez, L. A.; Patel, S. N.; Ying, L.; Kramer, E. J.; Nguyen, T.-Q.; Bazan, G. C.; Heeger, A. J. Adv. Mater. 2014, 26 (19), 2993−2998. (25) Whitehead, K. S.; Grell, M.; Bradley, D. D. C.; Jandke, M.; Strohriegl, P. Appl. Phys. Lett. 2000, 76 (20), 2946. (26) Zhu, R.; Kumar, A.; Yang, Y. Adv. Mater. 2011, 23 (36), 4193− 4198. (27) Awartani, O.; Kudenov, M. W.; O’Connor, B. T. Appl. Phys. Lett. 2014, 104 (9), 093306. (28) Cao, Y.; Smith, P.; Heeger, A. J. Polymer 1991, 32 (7), 1210− 1218. (29) Moulton, J.; Smith, P. Polymer 1992, 33 (11), 2340−2347. (30) Reyes-Martinez, M. A.; Ramasubramaniam, A.; Briseno, A. L.; Crosby, A. J. Adv. Mater. 2012, 24 (41), 5548−5552. (31) Sirringhaus, H.; Tessler, N.; Friend, R. Science 1998, 280 (5370), 1741−1744. (32) Sirringhaus, H.; Brown, P. J.; Friend, R. H.; Nielsen, M. M.; Bechgaard, K.; Spiering, A. J. H. Nature 1999, 401, 685−688. (33) Kim, Y.; Cook, S.; Tuladhar, S.; Choulis, S. A.; Nelson, J.; Durrant, J. R.; Bradley, D. D. C.; Giles, M.; McCulloch, I.; Ha, C.-S.; Ree, M. Nat. Mater. 2006, 5, 197−203. (34) Gargi, D.; Kline, R. J.; DeLongchamp, D. M.; Fischer, D. A.; Toney, M. F.; O’Connor, B. T. J. Phys. Chem. C 2013, 117 (34), 17421−17428. (35) Gurau, M. C.; Delongchamp, D. M.; Vogel, B. M.; Lin, E. K.; Fischer, D. A.; Sambasivan, S.; Richter, L. J. Langmuir 2007, 23 (14), 834−842. (36) Biniek, L.; Leclerc, N.; Heiser, T.; Bechara, R.; Brinkmann, M. Macromolecules 2013, 46 (10), 4014−4023. (37) Spano, F. C. J. Chem. Phys. 2005, 122 (23), 234701. (38) Spano, F. C. Chem. Phys. 2006, 325 (1), 22−35. (39) Clark, J.; Silva, C.; Friend, R.; Spano, F. Phys. Rev. Lett. 2007, 98 (20), 206406. (40) Clark, J.; Chang, J. F.; Spano, F. C.; Friend, R. H.; Silva, C. Appl. Phys. Lett. 2009, 94 (16), 163306. (41) Gierschner, J.; Huang, Y. S.; Van Averbeke, B.; Cornil, J.; Friend, R. H.; Beljonne, D. J. Chem. Phys. 2009, 130 (4), 044105. (42) DeLongchamp, D. M.; Kline, R. J.; Herzing, A. Energy Environ. Sci. 2012, 5 (3), 5980. (43) Duong, D. T.; Toney, M. F.; Salleo, A. Phys. Rev. B: Condens. Matter Mater. Phys. 2012, 86 (20), 1−5.

(44) Stafford, C. M.; Harrison, C.; Beers, K. L.; Karim, A.; Amis, E. J.; VanLandingham, M. R.; Kim, H.-C.; Volksen, W.; Miller, R. D.; Simonyi, E. E. Nat. Mater. 2004, 3 (8), 545−550. (45) Ward, I. M. Structure and Properties of Oriented Polymers, 2nd ed.; Springer: Netherlands, 1997. (46) Hadley, D. W.; Pinnock, P. R.; Ward, I. M. J. Mater. Sci. 1969, 4 (2), 152−165. (47) Landel, R. F.; Nielsen, L. E. Mechanical Properties of Polymer and Composites, 2nd ed.; Marcel Dekker Inc.: New York, 1993. (48) Street, R. A.; Northrup, J. E.; Salleo, A. Phys. Rev. B: Condens. Matter Mater. Phys. 2005, 71, 165202. (49) Wang, C. C.; Jimison, L. H.; Goris, L.; McCulloch, I.; Heeney, M.; Ziegler, A.; Salleo, A. Adv. Mater. 2010, 22, 697. (50) Sirringhaus, S. Adv. Mater. 2014, 26, 1319. (51) Certain commercial equipment, or materials are identified in this paper in order to specify the experimental procedure adequately. Such identification is not intended to imply recommendation or endorsement by the National Institute of Standards and Technology, nor is it intended to imply that the materials or equipment identified are necessarily the best available for the purpose. (52) Carlson, A.; Kim-Lee, H.-J.; Wu, J.; Elvikis, P.; Cheng, H.; Kovalsky, A.; Elgan, S.; Yu, Q.; Ferreira, P. M.; Huang, Y.; Turner, K. T.; Rogers, J. A. Appl. Phys. Lett. 2011, 98, 264104.

G

DOI: 10.1021/acs.macromol.5b02680 Macromolecules XXXX, XXX, XXX−XXX