Abstract

A pplications of M agnetoelectric M aterials for S olid -S tate D evices B S J Karthik Gururangan In 1947, three men, William Shockley, Walter Brattain, and John Barden, revolutionized the world of electronics by developing the first operational transistor. The transistor made the creation of the computer and other digital electronics possible, ushering in a new era of technology and earning the three men the 1956 Nobel Prize in Physics. Armed with the transistor, scientists and engineers (notably, those from Intel Corporation) developed faster, smaller, and more efficient microprocessors and RAM storage devices throughout the 1980s and 1990s. The development of solid-state storage devices and hard drives is of utmost importance for our increasingly electronic world. Within the last decade, computers have reached unprecedented operation speeds, but we are approaching a plateau. Recently, scientists have found that we cannot overcome this hurdle simply by fitting more and more transistors onto a chip; we must actually change the fabric of the device by creating new materials in a fashion that takes advantage of their inherent quantum electronic properties. Currently, hard drives operate by using the principle of magnetoresistance. Magnetoresistance describes the phenomenon of a magnetic polarization affecting the resistance of a material. Figure 1. Diagram of parallel and anti-parallel magnetic tunnel junction (MTJ) configurations (Tsymbal, 1999) In Figure 1, we can see that the magnetic tunnel junction (MTJ) consists of a thin insulator sandwiched by two magnets. The insulator has a thickness on the order of a few nanometers; therefore, electrons from one magnet can tunnel across the dielectric to the other magnet. Tunneling essentially means that every particle has a non-zero probability of traversing a solid boundary. While the mechanics of tunneling are not well understood, we do know that it is due to the wave-particle duality of matter; thus, tunneling is a purely quantum phenomenon and can only happen to very small particles such as electrons or quarks. When electrons traverse the boundary, they create a current and we can measure a resistance across the MTJ device. The number of electrons that will tunnel across the insulator is dictated by the relative orientations of the two magnets. Suppose that there are only two magnetic orientations: “left” and “right.” If both magnets have the same orientation, we call that the parallel state. If they point in opposite directions, they are in the anti-parallel state. Tunneling is highly encouraged in the anti-parallel state and suppressed in the parallel configuration (Nishimura, 2002). Thus, we can conclude that for a given potential, the anti- parallel state has a very low resistance while the parallel state has a high resistance. In hard disk drives, a read head applies a set voltage over many of these MTJ bits, measuring the resulting resistance of each one. It interprets a high resistance as a “1” state and a low resistance as a “0” state; thus, magnetoresistance directly leads to a digital interpretation. In the case of RAM (random-access memory), these bits are constantly being changed and re- interpreted over and over. The problem is that imparting these magnetic polarizations (a process called writing) utilizes an external magnetic field and is energetically inefficient given how many times the polarization changes. Recently, there has been a huge interest in a new class of materials called multiferroics. These materials exhibit two or more so-called “ferroic parameters”: ferroelectricity, ferromagnetism, and ferroelasticity. Current research suggests that using specific multiferroic materials in place of the magnets in Figure 1 can revolutionize computer memory systems and usher in a new era of quantum electronic technology. Before we can understand what the different ferroic parameters are and how they occur, we need to understand crystalline materials and crystal structure. When one thinks of crystals, one often pictures 1 • B erkeley S cientific J ournal • S ynthetics • S pring 2014 • V olume 18 • I ssue 2

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call