Particle integration across scales using self-assembly and transfer

Abstract

The integration of functional particles into planar devices facilitates a more efficient use of material, a simple definition of very small features, the use of otherwise incompatible materials and possibly the fabrication of novel device types that take advantage of the unusual properties of nanoparticles. Bulk synthesis is the most efficient way to produce such particles. However, in general the products are disordered, whereas devices usually require ordered arrangements. Self-assembly processes can order large numbers of particles in parallel on surfaces, but require specific chemistry and patterned substrates, both of which often conflict with other fabrication requirements. In addition to particle order, also the particle-substrate junction is often critical for device performance, and it remains challenging to create well-defined particle-surface interfaces reliably. In this contribution we show that through careful control of particle transport, self-assembly techniques can be made efficient and compatible with standard fabrication processes. Particles having diameters of 100 �m, 500 nm and 60 nm were assembled in templates using different methods: Large particles were fluidized and trapped in holes by gravitational assembly, smaller latex particles were forced into position by capillary forces. To increase yield and precision, particle transport was controlled intricately by means of the template geometry and the process parameters, such as temperature, velocity and suspension properties. To separate the assembly process and the actual integration, we rely on adhesion forces that commonly govern particle behavior in the dry state. Adhesion holds assembled particles in position at the end of any self-assembly process. If the adhesive forces are properly tailored, they can be strong enough to hold the ordered particles after assembly and allow their inspection, cleaning or modification, but at the same time weak enough to release the particles onto a target substrate when brought into intimate contact (see Figure 1). Thus the assembly process does not need to be carried out on the actual target substrate, but can be done on a specialized template. This separation gives us access to the surfaces that will later form the particle-substrate junction so that we can tailor the junction properties for each application envisaged. Examples shown in our contribution include glass particles that were connected to the substrate through a thin glass layer, tin beads with a conductive tin bridge, and resist beads that provide a protective, well-defined circular contact area.