Nano and biology

Self-assembly: structures that build themselves

There are two ways to make something very small. You can start big and carve away material, which is how chips are patterned. Or you can design small components so that they arrange themselves into the structure you want. The second route, self-assembly, is how living cells build almost everything, and it became one of nanoscience's central ideas.

The principle

The Wikipedia article on self-assembly defines it as a process in which a disordered system of pre-existing components forms an organised structure or pattern as a consequence of specific, local interactions among the components themselves, without external direction. When the components are molecules, the process is called molecular self-assembly. The article distinguishes static self-assembly, in which the ordered state forms as the system approaches equilibrium and lowers its free energy, from dynamic arrangements that need a continuing input of energy and are often better called self-organised.

The driving forces are mostly weak ones: hydrogen bonds, electrostatic attraction between opposite charges, the tendency of oily groups to avoid water, stacking between flat aromatic rings, and van der Waals forces. Individually each is feeble and reversible. Collectively, and arranged by design, they can hold together remarkably regular structures. Reversibility is part of the point: a component that lands in the wrong place can come loose and try again, so the system can find its ordered state rather than freezing into the first arrangement it meets.

Examples across chemistry and materials science

The same article lists important examples in materials science: molecular crystals, colloids, lipid bilayers, phase-separated polymers and self-assembled monolayers. It also names the folding of polypeptide chains into proteins and of nucleic acids into their working forms as self-assembled biological structures. Each example sits in a different corner of the literature. Self-assembled monolayers, single layers of molecules that organise themselves on a metal surface, fill the surface-chemistry journals; block copolymers that separate into regular nanoscale domains fill the polymer journals; colloidal particles that pack into ordered crystals fill the materials titles.

Peptides that assemble

Short peptides turned out to be unusually good building blocks, and three papers mark the progress of that idea.

In 1993, in the Proceedings of the National Academy of Sciences, Zhang and colleagues described a sixteen-residue peptide made of alternating charged and uncharged amino acids. Their abstract reports that on the addition of salt the peptide spontaneously assembled into a macroscopic membrane that did not dissolve on heating, in acidic or alkaline solutions, or when exposed to several denaturing agents and protein-digesting enzymes. Scanning electron microscopy showed a network of interwoven filaments, which the authors put at approximately 10 to 20 nm in diameter, and they suggested that complementary ionic bonds between the side chains were probably an important part of its stability.

Ten years later, in Science, Reches and Gazit reported that a very short peptide, a two-residue motif built from the amino acid phenylalanine, assembled into discrete and stiff nanotubes. Their abstract describes using the tubes as moulds: silver ions were reduced inside them, and the peptide was then removed by enzymatic degradation, leaving discrete metal nanowires. The same building block made from the mirror-image form of phenylalanine gave nanotubes that resisted enzymatic breakdown.

By 2017 the field had moved from structure to function. An article in Science by Tao and colleagues argued that self-assembled nanostructures of short peptides are promising candidates for bioinspired nanoscale semiconductors. The abstract explains that ordered stacking of aromatic rings and networks of hydrogen bonds can create quantum-confined structures within the assemblies, lowering their band gaps into the semiconductor range, and that the semiconducting behaviour can be tuned, doped and functionalised. The authors present these materials as a possible bridge between inorganic semiconductors and biological systems.

Why self-assembly matters for nanoscience

Top-down methods such as lithography are precise but expensive, and they struggle below a few tens of nanometres. Self-assembly works naturally at the scale of molecules and can build enormous numbers of identical structures in parallel, in a beaker, at room temperature. Its weakness is control: designing components that assemble into exactly one intended structure, without defects, is hard. Much of the literature therefore combines the two approaches, using lithography to define a template and self-assembly to fill in the fine detail.

Two of the most successful self-assembling systems use biological molecules. DNA, whose base-pairing rules make it almost programmable, is the subject of DNA as a building material. Peptides, which fold and stack into fibres, tubes and gels, are the subject of peptides at the nanoscale. For the landmark 2001 example of a self-assembled peptide scaffold directing mineral growth, see the page on 2001–2002.