Now University of California, Berkeley, scientists have found another unusual effect that could have both good and bad implications for semiconductor devices once they've been shrunk to the nanometer scale.
The discovery also could provide a way to tell whether pieces of rock from outer space came from planets with water.
In a paper appearing in the Aug. 28 issue of Nature, a UC Berkeley team comprised of physicists, chemists and mineralogists reports on the unusual behavior of a semiconducting material, zinc sulphide (ZnS), when reduced to pieces only 3 nanometers across - clumps containing only 700 or so atoms.
They found that when the surface of a ZnS nanoparticle gets wet, its entire crystal structure rearranges to become more ordered, closer to the structure of a bulk piece of solid ZnS.
"People had noticed that nanoparticles often had unexpected crystal structures and guessed it might be due to surface effects," said post-doctoral physicist Benjamin Gilbert of UC Berkeley's Department of Earth & Planetary Science. "This is a clear-cut demonstration that surface effects are important in nanoparticles."
Gilbert and co-author Hengzhong Zhang, a research scientist and physical chemist, suggest that many types of nanoparticles may be as sensitive to water as ZnS.
"We think that, for some systems of small nanoparticles maybe 2 to 3 nanometers across, this kind of structural transition may be common," Zhang said.
"There's a good and bad side to this," Gilbert added. "If we can control the structure of a nanoparticle through its surface, we can expect to produce a range of structures depending on what molecule is bound to the surface. But this also produces unexpected effects researchers may not want."
In addition, Zhang said these effects could have implications for our understanding of extraterrestrial materials and identification of extraterrestrial rocks, especially when the interpretation is being done by a robotic probe. A nanoparticle that formed in a place with water, such as Earth, would have a more ordered surface than a nanoparticle formed in space, where water is not present.
"Nanoparticles are probably widespread in the cosmos, and their surface environments may vary significantly, such as water versus no water, or the presence of organic molecules," said Jillian Banfield, professor of earth and planetary science at UC Berkeley. Banfield has been looking at microscopic and nanoscale particles in rocks, minerals and the environment in general to determine what information they can provide about their origin.
"As essentially all properties of nanoparticles - including spectroscopic ones often measured in the identification of materials - are structure dependent, and we now know that nanoparticle structure depends on the surface environment, it may be important to know how phase, size, structure, and properties relate so that spectra can be correctly interpreted."
Understanding how the characteristics of specific nanomaterials vary with environment could also lead to their use as sensors, for example, for water.
Some nanoparticles, including ZnS, are produced by microbes as a byproduct of metabolism. Banfield found ZnS in the form of a mineral called sphalerite in an abandoned zinc mine in Wisconsin four years ago, a product of sulphate-reducing bacteria. Numerous bacteria produce magnetite particles, or iron oxide, while Banfield has found others that produce uranium oxide. All these particles are small, ranging from nanometers to microns - millionths of a meter - across.
The trick is to distinguish these biogenic nanoparticles from similar nanoparticles formed by geologic processes. The importance of this issue arose several years ago when small inclusions in a meteorite from Mars were interpreted as being of biological origin by some and of geologic origin by others. Similar ambiguity has arisen over how to interpret small inclusions in rocks dating from the early years of the Earth.
Banfield studies naturally occurring nanoparticles of biologic and geologic origin, as well as synthetic ones, in order to understand how structure, properties, and reactivity vary with particle size. The geochemical consequences of size-dependent phenomena may be far reaching, she said.
Zhang developed molecular dynamics models to study the reaction of ZnS nanoparticles to surface binding, and predicted that nanoparticles grown to around 3 nanometers across would be most sensitive to surface water. Feng "Forrest" Huang, a post-doctoral fellow in the Banfield group, developed a method to make ZnS nanoparticles of that size in a methanol solvent. Zhang and Huang worked with Gilbert to observe the structural transition using synchrotron x-ray techniques.
Within the methanol and after it had evaporated, these three-nanometer particles were found through X-ray diffraction to have a somewhat disordered structure at the surface, with only the core of the nanoparticle exhibiting the regular order of bulk ZnS, which is sometimes used as a semiconductor in specialized photoelectronics applications.
When the methanol was spiked with water, however, the ZnS particles developed a much more ordered structure throughout. Only the immediate surface retained a disordered crystal structure.
The UC Berkeley researchers suggest that nano-ZnS has two or more stable structures, depending on what molecules are stuck to the surface. This is not surprising, Gilbert said, because the surface area of nanoparticles is so large compared to the volume that reactions at the surface are likely to affect the entire nanoparticle. In larger materials, the surface/volume ratio is much less, which means the surface has less effect on the interior of the solid.
The team also was able to demonstrate that the nanoparticles undergo reversible structure transformations at room temperature when removed from the methanol solvent and allowed to dry out. That is, when the dry nanoparticles are again immersed in methanol, they revert to their original structure.
"This result demonstrates that these nanoparticles are not trapped in a metastable state, but can respond to changes in their surface environments," Banfield said.
"To our knowledge, these are the first surface-driven room temperature transitions observed in nanoparticles," she added. "In methanol, the nanoparticles are highly distorted, but water addition removes this distortion. If alternative ligands or solvents can be found that stabilize alternative variants, there may be ways to generate uncommon structures through surface binding after the nanoparticle is synthesized."
The research team plans to continue to investigate how and why the crystal structure of ZnS nanoparticles changes with the surface environment. In particular, they hope to find out how the crystal rearranges itself so easily at room temperature, and how long it takes.
The work was sponsored by the National Science Foundation and the U.S. Department of Energy.
Journal
Nature