A simple chemical reduction process has been used to replicate intricate natural networks of silica at a relatively low temperature. The equally elaborate product is made of silicon — electronics' golden boy.
Living organisms can produce amazingly complex materials — from bones to butterfly wings — that are not only functional but often beautiful1,2. A dream of materials scientists is to understand and mimic nature, and thereby to obtain better man-made materials3,4. This remains a thorny problem, and few technologically relevant materials have so far been obtained solely by mimicking biological processes; conventional fabrication steps are typically needed too.
On page 172 of this issue, Bao et al.5 provide a powerful new tool for modifying biologically derived or inspired materials. They show how intricate glass skeletons, obtained from common algae, can be converted into silicon while maintaining their complex structure. Silicon is arguably the 'gold standard' among electronic materials, and this approach is akin to the magic touch of a modern Midas. It should allow a variety of intricate glass structures, both natural and artificial, to be transformed into silicon, and could have applications in sensing, electronics and optoelectronics.
Biological structures such as bones, shells and spines contain both inorganic and organic components. Nature uses these composites to achieve desirable mechanical properties in a material that can be formed under ambient conditions. The inorganic components tend to be limited to a few minerals such as phosphates, carbonates and silicates. The presence of silicates is perhaps not surprising: these materials combine oxygen and silicon, the two most abundant elements in Earth's crust. A common silicate is silica, SiO2, which is found both in biological settings and in man-made materials such as cement and everyday glass. Silica has a rich chemistry that can be adapted to obtain a variety of structures6.
With the notable exception of the optical fibres that are ubiquitous in telecommunications, silica oddly does not have a starring role in modern technology. In electronic circuits, for instance, the semiconductor silicon takes the lead, and silica is relegated to a supporting role of protecting, insulating and coating. But silica's abundance as a natural material is crucial in this context for another, often forgotten reason: the central player, silicon, is extracted from it. Heating silica at about 2,000 °C in the presence of carbon produces silicon. This can be processed further to obtain silicon wafers, which are among the purest available materials, and are the foundation for most of electronics.
Can silica be converted to silicon under milder conditions? Moreover, can this be achieved while preserving a complex shape? Bao et al.5 provide a simple answer to these questions. They react silica with magnesium gas at 650 °C to obtain a solid that contains both silicon and magnesia (magnesium oxide) phases. Silicon is not very volatile at these temperatures and is partially trapped by the magnesia, with which it is intertwined, and so it does not move significantly during this reaction. Consequently, the original shape of the silica is largely preserved. The magnesia can be selectively removed by bathing the solid in hydrochloric acid, leaving behind a silicon replica of the original silica structure.
Bao and colleagues demonstrate their approach with several types of diatom. Because of their abundance, these unicellular, photosynthesizing algae are important in soil, freshwater and seawater ecosystems. Diatoms fortify their cell walls with silica, and so have also been used to study biomineralization7. Their exoskeletons exhibit a variety of beautifully intricate, species-specific shapes and patterns (Fig. 1).
The authors use their technique to convert these silica skeletons into silicon. Their results show that the overall shapes, with their intricate pores and channels, are conserved. Also, as oxygen is removed from the structure, fewer atoms remain. This introduces new, nanometre-scale pores into the replica. For some applications — those that require solid silicon — this would be a disadvantage. For others it can be desirable for two reasons. First, the presence of the pores significantly increases the surface area of the final material, which can be important for applications such as sensing. Second, the final structure is composed of nanoscale crystals of silicon, which can have useful optical properties. In particular, unlike bulk silicon, nanocrystalline silicon can fluoresce efficiently8,9.
Bao et al.5 test both of these effects. First, they attach wires to a single diatom replica and use it as a microsensor for nitric oxide. Their results suggest that the diatom-derived silicon structure can provide a much more efficient sensor element than other, more conventional approaches. Second, they find that their silicon replicas can fluoresce after being partially oxidized in water.
This work indicates that many different silicon materials can be derived from diatoms, as well as other silica structures harvested from nature. More generally, it complements other recently developed 'templating' techniques. Solids structured on nanometre or micrometre length scales can now be obtained through a variety of simple self-assembly routes. These templates are then filled with a different material of interest before the template is removed. Silica templates are common because they are so easy to prepare, and silicon has been used as an infill material10 because of its technological importance and its high refractive index, which is useful in some applications in photonics.
A templating approach obviously produces the 'negative' of the original template. To obtain a duplicate of the template, the procedure must be repeated11: the negative structure is used as a mould, infiltrated a second time and removed. The mould material must be chosen carefully for all of these steps to be completed successfully. In contrast, Bao et al.5 have provided a method for copying a structure into silicon directly, albeit with the introduction of nanometre-scale pores. In combination with other self-assembly tricks, this opens the way to a variety of exciting new fabrication strategies. The implications might encompass not just biologically derived materials, but also many man-made creations.
Vukusic, P. & Sambles, J. R. Nature 424, 852–855 (2003).
Aizenberg, J. et al. Science 309, 275–278 (2005).
Heuer, A. H. et al. Science 255, 1098–1105 (1992).
Mann, S. & Ozin, G. A. Nature 382, 313–318 (1996).
Bao, Z. et al. Nature 446, 172–175 (2007).
Iler, R. K. The Chemistry of Silica (Wiley, New York, 1979).
Sumper, M. Science 295, 2430–2433 (2002).
Canham, L. T. Appl. Phys. Lett. 57, 1046–1048 (1990).
Littau, K. A., Szajowski, P. J., Muller, A. J., Kortan, A. R. & Brus, L. E. J. Phys. Chem. 97, 1224–1230 (1993).
Chomski, E. & Ozin, G. A. Adv. Mater. 12, 1071–1078 (2000).
Jiang, P., Bertone, J. F. & Colvin, V. L. Science 291, 453–457 (2001).
About this article
Bioinspired Nanoparticle-Assembly Route to a Hybrid Scaffold: Designing a Robust Heterogeneous Catalyst for Asymmetric Dihydroxylation of Olefins
European Journal of Inorganic Chemistry (2015)