Optimising the design of building blocks for self-assembly of discrete clusters

Abstract

Self-assembly is the spontaneous organisation of matter into an ordered state.
Significant progress has been made in the fabrication of synthetic components
for self-assembly, opening up routes to building blocks for the production of
functional materials and nanomachines. The information required to assemble a
target structure can be encoded into the building blocks. For assembly of an
equilibrium state, the target must be thermodynamically stable and the pathway
must avoid kinetic traps. The design of building blocks must address both these
requirements.

In this work a generic model is introduced which, through an explicit
representation of interactions, is able to express many approaches to
self-assembly. The model consists of hard cubic particles, whose faces are
patterned with attractive patches. A hybrid, dynamical Monte Carlo protocol is
developed to simulate self-assembly of such inhomogeneous systems efficiently,
accounting for both internal rearrangements and relative diffusion rates of
aggregates.

Using this single model, different self-assembly strategies are assessed,
ranging from simple approaches with only one type of building block, to more
complex strategies using multiple components and hierarchical paths. The
important case of fully addressable targets, where all components of the
structure are unique and have a specific location, is then examined in more
detail. Firstly, a new metric is introduced to quantify the problem of
competition between partly assembled fragments, which is a prominent source of
kinetic traps in addressable clusters. Principles are established for
minimising this problem. Secondly, a scheme for globally optimising the
interactions amongst a set of particles is developed to maximise the performance
of building blocks of a given complexity. This also makes it possible to
determine the level of complexity required for a given target to assemble
reliably.

The computational tools and general principles established in this work should
be applicable in a wide range of self-assembly problems.