See also: self-assembly
Self-assembly is a bottom-up process during which particles of any kind come together to form a specific structure. Common examples of such a process include the three-dimensional formation of proteins from amino acids in a living organism and the often six-fold symmetric formation of crystals in snowflakes.
Directed self-assembly refers to a self-assembly process where the initial particles and their environment have been engineered to promote the formation of a specific structure.
Soft matter physics relates particularly to this subject since it provides a good model system for the study of self-assembly: colloidal suspensions of microspheres which can be used as the building blocks of more complex assemblies. The size range of microspheres can be as large as a few microns, which makes them fairly easy to image with conventional techniques such as optical microscopy and confocal microscopy; a recently introduced technique, digital holographic microscopy, is also being used and under development for the study of dynamic processes.
Thermodynamics predicts that a system in equilibrium will relax in its lowest energy state. With that in mind, the game of directed self-assembly translates into engineering a system such that its ground state coincides with the desirable structure. It is possible for a system to be in a metastable state with a local energy minimum; if the lifetime of this state is long enough compared to the timescales relevant for the use of the self-assembled structure, this approach can also be used.
Thermodynamics can be very helpful in designing systems for directed self-assembly, since it provides a theoretical toolcase for the prediction of the probability of occurrence of a certain configuration based on properties as general as the number of different types of interactions between particles and the range of the associated interparticle energies (see Design principles for self assembly with short ranged interactions).
Thermodynamics underlies all physical processes, so in a sense all the methods described below can be eventually explained in terms of thermodynamical principles.
Geometry and Surface Forces
Geometry can set constraints on the motion of particles, making the desired structure more likely to self-assemble. Geometrical constraints are sometimes combined with surface forces which can be very strong at interfaces; in this case, the geometry of the space available to the particles is engineered so that the magnitude and direction of the associated surface forces will guide them in a desirable way.
An example of how geometry can affect the yield of a self-assembly process is currently being explored for the creation of tetramers from a suspension of colloidal particles of two different sizes. The underlying idea is quite simple: a tetramer of particles can be formed by one small sphere in the center, on the surface of which are three larger spheres. If the larger spheres are too large compared to the small one, then it is impossible for three of them to fit, whereas if the larger spheres are too small compared to the small one, then it is highly likely that more than three of them will fit. It has been proven mathematically and shown experimentally that there is an optimal size ratio between the diameters of the two sphere types which results in the assembly of exactly three large spheres on the surface of one small sphere 3.
The combination of geometry with capillary forces is commonly used for the directed self-assembly of thin films of crystals of particles on a flat surface. In such a process, a clean glass slide is immersed in a dense colloidal suspension and slowly pulled out. As the glass surface moves, a thin layer of the suspension full of particles protrudes from the liquid surface and carries the particles to the glass surface. If the surface moves slowly enough and the suspension is dense enough, this can result in the self-deposition of the spheres on a hexagonal lattice with the (1,1,1) plane parallel to the glass surface .
The interaction between two particles depends strongly on the properties of their surfaces. This can be used to engineer attractive or repulsive interactions. For example, in a colloidal system particles can be coated with surface charges. One of the simplest examples is the creation of colloidal crystals in a suspension where all particles have the same charge (see Photonic Properties of Strongly Correlated Colloidal Liquids). In this case the particles try to maintain the furthest distance from each other which is allowed by the volume of the surrounding medium and so they hover in the medium at periodically spaced locations. Even though electrostatics has only provided us with two types of electric charges, this technique can be surprisingly versatile with the addition of salt in the system, which allows control of the intensity of the electrostatic interactions since the concentration of free charges affects the Debye length.
Such a system is currently being explored for the self-assembly of particle clusters with a specific number of constituents. Small negatively charged spheres are mixed with larger positively charged spheres; the large spheres are attracted to the small spheres and park on their surface, and due to the existence of free ions in the suspension large spheres attracted to the same small sphere are not repelled from each other, allowing for the attachment of more than one positively charged sphere on the surface of a negatively charged sphere .
Another way to taylor interparticle interactions which offers greater variety relies on the use of DNA strands. By coating some spheres with half of a DNA strand and some with the complementary half, and assuming that the temperature of the system is higher than the binding energy between the DNA strands so that spheres can move Brownianly and find each other, it is possible to create a suspension where different pairs of particles are attracted to each other but indifferent to other particles which are coated with the half of another, non-complimentary DNA strand. This method is also being actively studied for the self-assembly of complex particle clusters .
Keyword in references:
 Jiang, Bertone, Hwang, and Colvin, Chem. Mater. 1999, 11, 2132-2140
 Leunissen et al, Soft Matter, 2009, 5, 2422-2430
 Schade et al, 2011 (in preparation)