The National Institute for Computational Sciences

A Humble Substance with Big Possibilities for Electronics

Researchers Use Kraken to Understand the Promising Electronic Properties of a Common Lubricant and Catalyst

By Hanneke Weitering

The hunt for new and improved materials is at the forefront of today’s physics research and is key to the advancement of countless fields such as medicine, energy systems, electronics, aeronautics and much more. Novel materials research involves manipulating the structure of various materials that we already use, such as plastics and metals, to improve their performance. From more durable medical implants to see-through smart phones, the possibilities for novel materials are endless.

A Rival to Graphene in Electronics

Recently a great deal of attention in novel materials research has been focused on graphene — a one-atom thick, two-dimensional sheet of carbon with a unique honeycomb structure that allows this ultra-thin material to also be incredibly strong. This lightweight and transparent material is up to 200 times stronger than steel and has potential applications in many fields such as bioengineering, solar power and optical electronics (transparent touch-screen devices). However, other novel materials— specifically, semiconductors — may prove to be even better candidates for these electronic applications. Graphene may be ultra-strong and lightweight, but because it is a conductor and not a semiconductor, its applications for electronics are limited.

Chris Van de Walle and Hartwin Peelaers, a member of Van de Walle's Computational Materials research team at the University of California, Santa Barbara, have been investigating the potential of a different material for use in electronics — molybdenum disulfide (MoS2) .

MoS2 is a unique solid lubricant that is already widely used for mechanical purposes, like preventing wear and tear due to friction in machinery and thereby reducing energy consumption. It also serves as a catalyst in petroleum refining. The material consists of plate-like layers of molecules just like graphene. However, unlike graphene, MoS2 is actually a semiconductor, giving it a major advantage for electronics. Semiconductors are useful because their electrical conductivity can be easily controlled and manipulated by a method called “doping,” in which scientists add impurities to the semiconductor to manipulate its electrical properties. Scientists have yet to exploit the semiconducting properties of MoS2, but Van de Walle hopes to find some new and groundbreaking uses for the material that will give graphene a run for its money.

Thin layers of MoS2 can be used to construct a variety of devices that have already been conceived using graphene. However, unlike graphene, MoS2 exhibits a band gap, an attribute that is advantageous for electronics. A band gap in a solid is the difference in electron energies between those electrons that are stationary (in the valence band) and those that are free to venture and conduct electricity (in the conduction band). The presence and size of a band gap will determine how well a material conducts electric current. Materials with no band gap are good conductors of electricity, whereas materials with a large band gap— insulators — do not conduct electricity well. Semiconductors like MoS2 have a very small band gap. Although conductors are better at carrying current than semiconductors, semiconductors are advantageous because their conductivity can be easily controlled and manipulated with doping.

Field-effect Transistor (FET)

The FET is one example of a common electronic device that could be manufactured using MoS2. Transistors are tiny devices that are used to amplify and switch electric current in circuits. These little contraptions are a fundamental building block of nearly all modern electronic devices — computers, cell phones, radios, televisions, calculators and much more. They work somewhat like a water faucet, able to not only start and stop the flow of current but also to control the amount of current flowing out. Transistors come in many forms, but all types serve the same function. FETs use the electric field created by an input, or “gate,” voltage to modulate the electrical current coming out of the device.

“Graphene has been proposed as an excellent material for devices such as field-effect transistors because of its two-dimensional nature and high carrier mobility,” Van de Walle says. “However, such devices require the material to be semiconducting [that is, it should have a band gap]. Pure graphene does not have a band gap, and therefore additional modifications are required — for instance, the formation of nanoribbons, strips of graphene with ultra-thin widths —to open up a band gap.”

Van de Walle explains that “because of these complications, it is attractive to investigate other materials that have properties similar to graphene but that do have a band gap. MoS2 is one such material. Field-effect transistors based on MoS2 have already been demonstrated.”

Because MoS2 is intrinsically a semiconductor, no modifications need to be made to create a band gap; it naturally has one already. By using MoS2 rather than graphene in the fabrication of electronics, scientists can save time and also lower the cost of manufacturing.

While the possibilities for molybdenum disulfide seem promising, more knowledge about its electronic properties must be generated to effectively exploit the potential of this material. Van de Walle and his team of researchers seek to understand the relationships between composition, structure, and electronic and optical properties of MoS2 in hopes of finding new and useful implementations for the material.

The Essentialness of Supercomputing for Calculations

“Our work is based on first-principles calculations; that is to say, it is based entirely on a quantum mechanical description of the material, without any input from experiment,” Van de Walle says. Rather than drawing conclusions from experimental results, Van de Walle’s models are based on calculations of the interactions among all the microscopic particles that make up his theoretical system. The interactions between the atoms that make up MoS2 are governed by the interactions of their constituent parts — the atomic nuclei and the electrons. Any other physical phenomena that occur in the system can be traced back to these basic interactions. This approach of calculating properties of a system from first principles calculations is known as density functional theory, or DFT.

The physics involved in DFT first-principles calculations is actually relatively simple and straightforward. Interactions between the atomic nuclei and the electrons are well understood and can be calculated from basic physics principles. What makes this approach so difficult is not the complexity of the problem but rather the immensity of the problem. Real-life systems contain billions upon trillions of subatomic particles (if not more), and calculating their interactions could take a lifetime to do by hand. With help from the NICS Kraken supercomputer, Van de Walle and his team were able to complete their calculations in just a matter of months.

“To get accurate results for a material such as MoS2, we need to use an advanced functional [a so-called "hybrid functional"]. This makes the calculations very computationally demanding, and the high-performance capabilities of Kraken were crucial,” said Van de Walle.

To get an idea of the required computational time, Van de Walle's team used approximately 5,300,000 SUs (service units, or compute hours) on Kraken since January of 2013, to tackle a range of problems using similar first-principles approaches.

Investigating Band Structure

Van de Walle explains that the band structure diagram provides necessary information about how electrons behave in the material as well as how light interacts with it. To utilize MoS2 for electronic devices, scientists must first understand how electrons behave in the material — that is, how they move around and conduct electricity. Before MoS2 can be used for solar panels and smart phones, scientists must first comprehend exactly how light interacts with the material. This information about electron and light interactions with MoS2 can be derived from Van de Walle’s band structure diagram.

“The lower bands in the picture [below 0 eV] are the valence bands, and these are completely filled with electrons. The higher bands are conduction bands, which are unoccupied. If additional electrons are present in the material [for instance, because of doping, or because of the application of a gate voltage in a transistor], the conduction bands will become filled with electrons,” Van de Walle says. The electrons in the conduction bands are responsible for conducting current.

The aforementioned band gap is the energy difference between the top of the valence band and the bottom of the conduction band of the material. This energy difference is what makes MoS2 a semiconductor.

“The band structure also provides information about the optical properties,” Van de Walle explains. “In order for light to be absorbed, the photon energy [determined by the wavelength of the light] should be at least equal to the band gap. The energy of the photon can then excite an electron from the valence band into a higher-energy state in the conduction band.” This means that the material will not absorb all incoming photons (“particles” of light), but it will absorb only photons with energies greater than or equal to the band gap (the band gap is an energy range and not a physical distance). When an electron in the valence band absorbs incoming light, it becomes excited and pops up into the conduction band, where it can then help conduct electricity.

With the band structure diagram complete, more investigations concerning electron behavior and mobility are to follow. Electron mobility is determined by the effective mass (a quality that can also be determined from the band structure diagram) and by the scattering behavior of electrons navigating through a material.

“Electrons can be scattered by defects or impurities in the material, but even in perfectly pure material, scattering of electrons will occur due to interactions with lattice vibrations. We are therefore currently investigating these interactions,” Van de Walle says.

More Research Needed

Although much remains to be discovered about the properties of MoS2, the material clearly has countless potential applications for modern electronics. Improvements to transistors and solar panels are only the beginning — from new and improved touch-screen devices and ultra-thin, transparent optical displays to lithium-ion batteries capable of recharging in minutes rather than hours, the possibilities for MoS2 are vast. With more research, soon scientists will be able to fully exploit the potential of this novel material.

Article posting date: 14 July 2013

Related Links

About NICS: The National Institute for Computational Sciences (NICS) operates the University of Tennessee supercomputing center, funded in part by the National Science Foundation. NICS is a major partner in NSF’s Extreme Science and Engineering Discovery Environment, known as XSEDE. The Remote Data Analysis and Visualization Center (RDAV) is a part of NICS.