Dynamic-template-directed multiscale assembly for large-area coating of highly-aligned conjugated polymer thin films

Solution processable semiconducting polymers have been under intense investigations due to their diverse applications from printed electronics to biomedical devices. However, controlling the macromolecular assembly across length scales during solution coating remains a key challenge, largely due to the disparity in timescales of polymer assembly and high-throughput printing/coating. Herein we propose the concept of dynamic templating to expedite polymer nucleation and the ensuing assembly process, inspired by biomineralization templates capable of surface reconfiguration. Molecular dynamic simulations reveal that surface reconfigurability is key to promoting template–polymer interactions, thereby lowering polymer nucleation barrier. Employing ionic-liquid-based dynamic template during meniscus-guided coating results in highly aligned, highly crystalline donor–acceptor polymer thin films over large area (>1 cm2) and promoted charge transport along both the polymer backbone and the π–π stacking direction in field-effect transistors. We further demonstrate that the charge transport anisotropy can be reversed by tuning the degree of polymer backbone alignment.

Wherein AS and AP denote the total membrane surface area and the projected area respectively.
Consequently, the energy difference between these two configurations (ΔE) can be calculated, AS for AAO structure can be estimated as, Where r is the pore radius (0.1 μm), l is the membrane thickness (60 μm) and dpore is the pore density (10 13 m -2 ). Now we can calculate R, = = 2 (1 + ( − )) = 2 (1 + 10 13 2 × ×0.1×10 −6 ×(60 − 0.1)×10 −6 ) = 378 (11)  Simulation on a DPP2T-TT 18-unit polymer chain (average length of the actual polymer used) was also performed to investigate the conformation of the actual polymer chains in chloroform solvent. This is to confirm that the polymer backbone is fully exposed to IL, in order to justify the removal of alkyl chains in the oligomer simulation. Lastly, simulation on pure chloroform molecules and ILs was performed to calculate the buffer distributions of IL molecules, in order to compare with the IL distribution around the polymers. Simulation details are summarized in the All the simulations were set up using the AMBERTools14 and performed with AMBER14 software 5 using the general AMBER force field (GAFF). All the partial charges were derived using the AM1/BCC method. The same partial charges from the monomer DPP2T-TT were used for each unit of the DPP2T-TT polymer chain. Previous studies have shown that the GAFF can accurately predict transport and thermodynamic properties of ILs 6 . The system energy was first minimized with 2000 steps of steepest-descent method and then with 8000 steps of conjugatedgradient method. Subsequently, the systems were slowly heated from 0 to 10 K in NVT and NPT ensemble for 1 ns, respectively. Then the systems were heated from 10 K to 298 K in NPT ensemble for 2 ns. Cartesian coordinate restraints were applied on all heavy atoms during the minimization and heating stages with a force constant of 20 kcal/mol. Before the production runs, the systems were equilibrated without any restraints in NPT ensemble (1 atm, 298 K) for 1 ns.
All the simulations were performed in NPT ensemble (1 atm, 298 K) with periodic boundary conditions. Particle-mesh Ewald 7 method was used to treat the electrostatic interactions with a 10 Å cutoff distance. The SHAKE algorithm 8  Simulations were performed on the Blue Waters petascale computing facility.
Distribution Probability Analysis. For the monomer/dimer simulations, all the production trajectories were processed with CPPTRAJ 10 so that the ILs layers were positioned at the center of simulation boxes. The distance r was defined as the center of mass distances between the polymer backbones and ILs in the direction orthogonal to the initial CHCl3/ILs interface. The normalized distribution probability of polymers p(r) were calculated with MDTraj 1.7.2 package 11 . Excess probability was defined to directly compare the probability of the polymer at the ILs interface and the bulk. The bulk excess probability is 1.
Radius distribution function analysis. The radius distribution functions ( ) of cation and anion around the function groups of DPP2T-TT dimeric backbone were calculated from the dynamic ILs simulations. The radius distribution function is defined as where is the distance to the backbone atom, n is the number of cations or anions within dr region from r to r+dr, is the bulk ion concentrations, 2 2 is the area of hemisphere with a radius of r TEM electron diffraction-based two-dimensional orientation mapping procedure. We recorded electron diffraction patterns to cover 100 meshes (10 by 10) on a 3mm diameter TEM grid with 100 μm×100 μm mesh size using a selected beam depicted in Figure S7-a. We translated the sample and acquired the diffraction patterns (DPs) in a step-by-step manner to scan over the entire designated area on the thin film. Within each mesh, the step size was 10 μm to yield 100 scans per mesh. A total of 100 meshes were scanned to cover an area of 1 mm 2 (Fig. S7). Within each mesh of 0.01 mm 2 , we plotted the 2D orientation distribution as a histogram of 100 scans, scattered in terms of the rotation angle of the π-π stacking peak relative to an arbitrarily defined fixed axis. The histograms from five representative meshes are shown in Figure S7-b. For plotting the 2D orientation distribution over 100 meshes of 1 mm 2 , we used the average rotation angle of the π-π stacking peak from each mesh, and constructed the histogram shown in Figure 3e. The average rotation angle was normalized as zero in this plot.
Constructing the 2D orientation color map shown in Figure 3e requires grouping similar diffraction patterns. We applied image cross-correlation analysis to cluster the DPs from each mesh 12 . Since the brightness of the DPs can vary from mesh to mesh, the similarity of images is compared using normalized cross-correlation factor making the analysis independent of the image intensity: where IA(x, y) and IB(x, y) are the intensities of a pixel (x, y) in images A and B, respectively, and I A ̅ and I B ̅ are the mean intensities of images A and B, respectively 13 . The construction of the 2D orientation color map requires applying a correlation factor (γ) threshold for grouping similar images (similar orientation judged from the π-π stacking rotational angle) into distinct clusters.
The cross-correlation threshold was chosen by trial and error and it was set at >95.5% for binning γ of similar values. We estimate that this threshold value can distinguish DPs with angular spread within 4 degrees of the bin average. Our orientation mapping method should be broadly applicable for quantifying local orientation in-plane in a wide range of crystalline and semi-crystalline thin film systems.
Interestingly, we observed single-crystal-like hexagonal diffraction patterns with very sharp peaks occasionally during mapping (Fig. S-7c). Similar patterns were reported for TA-PPE conjugated polymer nanowire crystals (a derivative of poly(para-phenylene ethynylene) with thioacetate end groups) 14 . This may be due to the broad distribution in molecular weight of the DPP2T-TT polymer synthesized.
Polymer synthesis. The conjugated polymer DPP2T-TT was synthesized following a previously published procedure 15  where h, C, ρ, L, v and Qevap are thickness, concentration, density, meniscus length, coating speed and evaporation rate respectively 20 . Calculations (provided below) illustrates that evaporation rate on IL is approximately 50% faster than on OTS. This is consistent with experimental static advancing contact angle values of chloroform on OTS (θOTS) and on IL (θIL), where θOTS is larger than 30° but chloroform completely wets the AAO/IL hybrid (θIL~0°).