Single crystals make high-tech possible.
For example, silicon—the (current) king of the semiconductor world—is grown as single crystals for its myriad uses in high-tech applications.
That’s because single crystals have an atomic order that gives materials the superior properties that high-tech applications demand. In comparison, polycrystalline materials are comprised of individual crystals, or grains, that have spaces in between them—spaces that act as defects.
These grain boundaries alter the way materials behave. Though that’s not necessarily a bad thing, those changes often give the material properties that are incompatible with high-tech applications.
So in the high-tech world, the score is clear:
Single crystals: 1.
Polycrystalline materials: 0.
Single crystals of silicon are grown by melting the material and then seeding it with a single crystal, which convinces other atoms in the melted silicon to join the crystal’s neat and orderly structure.
However, although this process works well, heating some materials just isn’t an option—especially for compositions that easily decompose or transform or don’t melt uniformly.
Now more materials may be heading into that high tech world, however, thanks to new research from Lehigh University researchers.
The Lehigh University team—led by ACerS member Himanshu Jain and including ACerS member Volkmar Dierolf, along with authors Dmytro Savytskii and Brian Knorr—devised a new fabrication method that could extend the reach of single crystals by ditching the need for melting.
“Many promising materials could unleash their full potential if they were available in a single crystal form,” the authors write in their new Scientific Reports paper. And they want to unleash the full potential of an interesting class of promising materials—chalcogenide glasses.
In particular, the team grew antimony trisulfide (Sb2S3) single crystals in Sb-S-I glasses. Antimony trisulfide has ferroelectric properties and a full range of potential applications in solar cells, microwave devices, switching sensors, and thermoelectric and optoelectronic devices, the authors write in the paper.
So how did they do it? Instead of melting the entire material, the scientists instead used a laser to heat up a small, controlled area of the chalcogenide glass to its crystallization temperature.
The method keeps the material below its melting temperature, causing it to transform directly from the solid state to a crystal structure, with no intermediate gas or liquid stage.
By heating instead of cooling towards crystal formation, the team is able to more closely control crystallization—applying just enough heat organizes a single crystal instead of multiple.
“Once we make the single-crystal line, we backtrack to get additional parallel single-crystal lines and eventually a single-crystal-layer surface on top of the glass,” Jain says in a Lehigh press release. “We can stitch these lines to convert the entire glass surface into a single crystal.”
The trick is to grow the crystals small and fast—using the laser to heat only a small portion of the material, and quickly grow the crystallization nucleus into a single crystal by moving the laser just fast enough to stay ahead of the crystal-forming front. That rate doesn’t allow formation of additional crystals, resulting in growth of a single crystal within the glass.
But getting the process rate just right wasn’t as simple as it sounds—Jain says the scientists found that there was a sweet spot to how quickly the laser could move to optimally form a single crystal.
While normally going slower affords more control, the Lehigh scientists found that going too slow had a surprising result. “Below a critical scanning rate, multiple grains grew simultaneously,” Jain says. “Speeding up the scanning rate prevented this unwanted situation, but at very high scanning rate, of course, no crystals formed.”
Jain says the team used a CCD camera to monitor the glass to crystal transformation in real-time. “From optical contrast, we know the part of glass that has crystallized. However, this observation cannot unequivocally establish that it is a single crystal.”
So the team also analyzed their results post-crystallization using electron diffraction and microscopy color mapping to detect the orientation of atomic configurations within the material.
In addition to optimizing the process rate, the authors also found that adding SbI3 as a stabilizer made single crystal formation even more successful because it eased glass formation and inhibited additional crystal nucleation, they report.
Jain says that when Sb2S3 glass is heated with SbI3 stabilizer, “the composition of glass in front of the crystal becomes richer in SbI3 and the becomes even more stable, thereby preventing unwanted additional nucleation.”
In addition to the large potential this new single crystal fabrication method affords to a variety of materials, it’s scalable. “There is really no significant barrier for large-scale fabrication,” Jain says. “The important consideration is to maintain good control of the laser system and glass translation stage.”
The open-access paper, published in Scientific Reports, is “Demonstration of single crystal growth via solid-solid transformation of a glass” (DOI: 10.1038/srep23324).