57th Annual Report on Research 2012 Under Sponsorship of the ACS Petroleum Research Fund

We are exploring the viability of using self-assembled vertically oriented arrays to create solar cells. We discovered a gravity driven self-assembly method that produces high density, vertically oriented arrays of anisotropic microparticles. Silica nanotubes with partial Au cores were used as a model system. These particles formed structures as large as 10 cm2 in about 15 minutes with over 60% of the particles were vertically orientated. The formed structures depended highly on nanotube center of mass and surface density. An offset center of mass (caused by the partial gold core) oriented the particles as they sedimented, and when on the surface it assisted in rotation out of the surface plane into standing orientations. A minimum surface density of particles was needed to form columnar structures, since neighboring particles stabilized the arrays. In the past year, we explored the direct positioning and shaping of arrays using microwells and strategies to integrate our arrays onto substrates. Microwell patterned surfaces were examined to direct and specifically locate arrays. These wells concentrated the nanotubes, increased the percent standing, and stabilized assemblies. Vertically aligned structures were confined to square wells, 25 to 400 square microns in size with 2 to 10 micron spacings, patterned over a 0.01 cm2 area. The nanotubes diffused across the surface, fell into the wells (thereby gaining locally high concentrations), and oriented perpendicular to the substrate. Most wells showed greater than 90% standing, and we observed numerous 100% wells. In general, arrays formed in microwells displayed standing percentages that were 20 percentage points higher than those formed on an open surface. The walls of the microwells supported standing nanotubes. Solid particles (without an offset center of mass) showed 70 ± 10% standing in wells which were 20% smaller than their length. In a well that was wider than the particle length, however, these wires could easily lie down and only 30 ± 10% standing was observed compared to 5% on an open surface. Microwells also facilitated sample drying. Structures aligned within microwells were shielded from convective flows, which catastrophically disrupted assemblies formed on open surfaces. Dried assemblies are ideal for the further processing steps (e.g., metal evaporation as a counter electrode) needed to produce a solar cell. We have studied several strategies to either attach or transfer dried structures to surfaces. Arrays formed on conductive substrates had metal electro deposited between nanotubes to anchor them, though, this method proved challenging as the height of the metal plated was non-uniform. To achieve smoother plating, this strategy will incorporate elevated temperatures and new plating solutions. Metal films were evaporated over assembled nanotubes, alternatively, followed by conductive adhesives (e.g., silver paint) attaching this film to a new substrate (e.g., glass slide). Subsequent dissolution of the photoresist released the metal film from the assembly substrate transferring an inverted array. Lastly, polydimethylsiloxane (PDMS) was cured over a dried array. This was then peeled from the original substrate. As they were completely embedded within PDMS, we observed full transfer of the assembled structures. Bending produced no observed damage to the aligned tubes. Measured standing percentages were 92 ± 3% with a particle density 60% higher than before drying. This density correlates to a packing density of 45% of the maximum possible for an array of close packed tubes. We believe the increase comes from additional tubes being guided into the wells as the substrate dries. To fabricate a solar cell, the information learned above will be combined with further experiments. Particles produced with semiconductor coatings (e.g., TiO2, SnO2, ZnO2) in conjugation with alternate cores (e.g., Ag, CdSe) should have photoactivity. Finding suitable selective etchants to create these particles with offset centers of mass has been challenging. By using microwells, however, we are now able to assemble unetched nanowires into columnar arrays, so etching is avoided. Nanowires will be aligned into vertically-oriented arrays and contacted to a substrate. These arrays will be optically analyzed to obtain absorptive, reflective, and transmission characteristics. Electrical analysis (conductivity vs. voltage) will be preformed on both arrays and individual nanowires in the presence and absence of light. If successful, then our assembly strategy could provide an inexpensive route to solar cells, specifically positioned with existing lithographic processes, on any desired substrate.