Guihua Xiaoa,
Yingjuan Suna,
Wentao Xua,
Yuan Lin*a,
Zhaohui Sua and
Qian Wangab
aState Key Laboratory of Polymer Physics and Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, 5625 Renmin Street, Changchun, Jilin 130022, P. R. China. E-mail: linyuan@ciac.ac.cn
bDepartment of Chemistry and Biochemistry, University of South Carolina, Columbia, South Carolina 29208, USA
First published on 3rd September 2015
We report a facile and robust route to fabrication of large area patterns of poly(3-hexylthiophene) (P3HT) by a controlled evaporative self-assembly (CESA) technique using mixed solvents. The alignment, self-assembly and patterning were achieved simultaneously in one step. Chain orientation via this method compares favorably with that attained by mechanical forces.
Although mixed solvents have been applied to improve the homogeneity of ink-jet-printed deposits,14–18 and tune the self-assembly morphology of conjugated co-polymers in the solution state which lies in the solubility and selectivity of polymer chains in different ratio of the poor/good solvents,19,20 few studies have employed mixed solvents in CESA, an open system accompanied by drying process, due to the great complexity. Lin et al. used the CESA method to obtain the hierarchically structures of conjugated polymers in a single solvent at first, then used another solvent vapour annealing process to rearrange of nanodomains.21 Herein, we report an extremely facile route to obtain unprecedently ordered conjugated polymer assemblies via CESA from mixed solvents. The pattern formation is governed by the motion of the CL, which is determined by the competition between the pinning and depinning forces. As shown in Scheme 1c, the pinning forces are consisted of the surface tension of the film (γf) and the friction force (f), preventing shrinkage of the droplet. The depinning force is the capillary force caused by surface tension of the droplet (γL).22 Apparently, only the surface tension of the solvents directly contributes to the depinning force and may significantly influence assembled patterns. Meanwhile, evaporating rate determines solvent loss and thus the amounts of deposits at the CL, which will indirectly change the friction force. Considering mixed solvents composed of two solvents with different surface tensions may cause surface tension gradient due to heterogeneous evaporation, which may result in patterns with instabilities.23 Therefore, solvents with similar surface tensions were employed to suppress Marangoni flow and obtain regular structures.24–26 Besides the surface tension of solvents, the evaporating rate and solubility also have significant influence on CESA. Evaporating rate determines the solvent loss and thus the amounts of solute deposit at the CL, which will directly affect the pattern formation.27 As for conjugated polymers, the evaporating rate also decides whether they have enough time to take preferred orientation and self-organize (i.e., crystallize) into nanowires. Meanwhile the solubility of conjugated polymers in solvents influences the aggregation state of polymer chains, and thus would govern the mobility of the assembling objects at the CL. Solvents with better solubility can enhance the mobility of polymer chains, which is favorable to the orientation and assembly/crystallization process. Therefore, evaporating rate and the solubility of solvents have decisive influence on the orientation and crystallization process of conjugated polymer chains. By using a mixed solvent system, the CESA process can be optimized: polymer chains can get highly aligned and self-assemble into nanowires in the ordered stripes.
P3HT was selected as a nonvolatile solute in our study. It represents as promising materials for the elaboration of OFETs and OPVs due to the combination of good solubility, facile processability, high charge carrier motilities and good environmental stability.28 The cylinder-on-flat confined geometry was constructed in our previous report (Scheme 1a).13 P3HT solution (30 μL) with a concentration of 0.1 mg mL−1 was injected, which upon evaporation, afforded a number of ordered stripes on the substrate, as shown schematically in Scheme 1b. In our previous report, regular stripe patterns with the low molecular orientation can be obtained by using chloroform (CF) pre-aged solution, and the long time required. For the mixed solvents, fresh prepared CF & methylbenzene (MB) mixture was utilized. They have similar surface tensions but different boiling points and solubilities (Table S1†). Considering the fast evaporation rate of CF, the ratio changing of CF/MB as a function of time during the drying process was monitored through a Gas Chromatography (Fig. S1†). After drying 12 min, the ratio of MB/CF increases to 39.5. The remaining content of CF was extremely few in the late drying process, so the outermost region could best reflect the nature of mixed solvents ratio, then the outermost region X1 was chosen to study in the following work. Unlike irregular stripes with fingering instabilities formed using fresh CF (Fig. 1a, d and g), ordered stripes can be obtained for MB (Fig. 1c, f and i). In these stripes, fingering instabilities have been greatly suppressed, which matches the expectation made by Lin group.9 As for the CF/MB mixture, the regularity of the stripes is dramatically enhanced (Fig. 1b, e and h), even over a scale of several hundred micrometers (Fig. S2†).
Quantitative information concerning chain orientation within the stripe can be obtained from polarized Raman spectroscopy (Fig. 2a–c). There are some difference in Raman spectra at 1440–1450 cm−1 peak before and after assembly, which corresponding to the CC symmetry stretch vibration of thiophene ring (Fig. S3†). This feature was related to the ordered structures and further confirmed the assembly. P3HT chains align parallel to the CL in order to maximize the interfacial coverage, as indicated in previous report. Raman anisotropy (R) is defined as the ratio of IZZ to IXX, R = IZZ/IXX. The perfectly oriented fraction f of Fraser's model can be obtained from the value of R.29,30 In Fig. 2a and b, stripes from mixed solvents show larger R and f values than these from pure solvents. With the increase of the volume fraction of MB, the values of R and f increase initially and then decrease, which arrive at their peak values when the volume fraction of MB is ca. 50%. Within the stripes of the outermost region, the Raman intensity in the ZZ configuration is much larger than that in other configurations (Fig. 2c). This indicates an uniaxial alignment with P3HT chains mainly aligning parallel to the CL. A perfectly oriented fraction f of 72% can be obtained, which is significantly higher than reported results.6,31 During the evaporation process, an outward flow diffuses from the bulk to replenish the evaporation losses at the perimeter, which will take P3HT molecules from the bulk solution (Scheme 1e at t0). More P3HT chains are driven to the edges and align parallel to the CL (Scheme 1e at t1). When it reaches a certain threshold value of concentration, P3HT molecules begin to form nanocrystals and deposit on the substrate (Scheme 1e at t2). As a result, more and more P3HT nanocrystals deposit at the contact line and self-assemble into 1D nanowires induced by the strong intermolecular π–π stacking interaction (Scheme 1e at t3). Furthermore, the TEM image could demonstrate that P3HT molecules do not form nanowires in the fresh prepared mixed solvents (Fig. S4†). Therefore, deposition not only provides a pinning force, but also serves as a nucleation point for P3HT nanowires. Within these stripes, P3HT nanowires take an orientation perpendicular to the CL. The total drying time and each movement of different ratios of CF/MB were monitored (Fig. S5†). We found that the total drying time of CF/MB 1/1 fabrication on the 1.5 × 1.5 cm2 substrate was about 16 min and the time of each stripe formation was about 1.5 seconds in the outermost region.
Here the process differs fundamentally from previous reports that pre-aged solutions were applied in the CESA operation. Instead, the interplay of two solvents governs both the nucleation and the assembly processes. CF shows better solubility for P3HT than MB and it has a rather high evaporation rate. During the CESA process, P3HT chains do not have enough time to take preferred orientation and crystallize into nanowires. As solvent evaporates, solutes would get concentrated at the edges by the outward capillary flow.9,11 P3HT chains would aggregate or even sediment from the MB due to a poor solubility. A balanced solvent system would ensure the mobility of P3HT molecule within the bulk solution, yet be able to promote the alignment and crystallization at the CL. Therefore, highly-ordered hierarchical structures can be obtained at an optimized volume fraction of two solvents (ca. 1:1 in our case). Current measurements were performed at a constant applied voltage on the stripes formed on the SiO2 substrates from different solvents. Two Au electrodes with the gap of 80 μm were deposited onto the stripes in the outer region. Conductance (S) of stripes from different solvents follows the order: CF/MB (1/1) mixture (1.61 × 10−7) > MB (3.80 × 10−8) > CF (2.02 × 10−9) (Fig. 2d). Electrical conductivity from CF/MB (1/1) mixture is two orders higher in magnitude than that of pure CF. It is well known that the strong anisotropy in charge transport of P3HT, where the charge carrier along the conjugated backbone. As expected that the higher conductivity was obtained by the CF/MB (1/1) mixed solvents having the higher oriented fraction, this trend is consistent with the extent of chain orientation and nanowires' formation, both of which are favorable to charge transportation.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: 10.1039/c5ra10426b |
This journal is © The Royal Society of Chemistry 2015 |