Collaborative Disappearing Act Trumps Levitation As Useful Trick For Future Electronics

Paul Riechers
Writer’s Comment: To strengthen public funding and support, scientists must effectively convey the progress and promise of new scientific ideas to the public. In Dr. Meisenheimer’s UWP 104E Science Writing course, I had the opportunity to refine this skill through the Comparison Paper assignment. This paper aims to compare two inherently technical subjects in a way that is both accessible and interesting to a lay reader. Specifically, I chose to compare the usefulness of single-electron tunneling and Casimir effects, which are both consequences of quantum mechanics, for future electronic devices. One of the greatest difficulties I had in writing this paper was in narrowing the scope and limiting the technicalities. After much research, revision, and discussion, I was able to pick out what really matters—not only to me as a physicist, but also to a larger audience. The next challenge was to develop evolving metaphors that are insightful, but not oversimplified. I hope that the outcome of my struggles provides an example of effective science writing for a lay audience.
—Paul Riechers

Instructor’s Comment: At some point in my ten years of teaching, a student from every science major represented on campus has approached me during my UWP 104E Scientific Writing course to let me know, confidentially, that his or her particular major faces much more difficult challenges in translating scientific concepts for a lay reader than any other major. I am inclined to agree only when it comes to some of the stranger reaches of theoretical physics, and that’s why Paul Riecher’s paper on the Casimir effect and quantum tunneling was such a pleasure to read. For this assignment, I ask students to identify the most challenging concept an everyday reader must understand to make sense of the paper’s overall comparison argument and explain it; in the case of the topic Paul chose, almost all the concepts were challenging, abstract, even absurd. Nevertheless, he developed a variety of visuals and metaphors and kept the essay entertaining with a cheerfulness of phrasing that shows not only his enjoyment of physics, but of sharing what he understands with others. He’s a great popularizer. Just imagine how well he must understand the underlying physics himself!
—Don Meisenheimer, University Writing Program

Competing Magic Tricks
The interatomic interaction that can levitate tiny switches, mirrors [1], and sensors [2] offers some neat tricks for a future generation of computers, but its preference to perform as a solo act limits its usefulness for information processing. However, since physical limits and manufacturing techniques prevent further downsizing of current technology, such magical effects must be utilized to significantly advance electronic components. With the help of these strange interactions, the next generation of basic computational elements will not only be smaller, but will also be smarter than those employed today. The interaction known as the Casimir effect, which mediates levitation, is one of the quirky effects that begin to dominate the dynamics of closely spaced material bodies as the distance between them diminishes to less than one hundredth the width of a human hair. Another such effect that becomes relevant as electronic components shrink below these dimensions to meet society’s demands for smarter and faster computers is quantum tunneling, in which a confined electron performs a Houdini-like maneuver to emerge in a new confinement region across a supposedly impenetrable barrier. While both effects vie for prominence in this realm, tunneling allows for more collaborative array elements, useful for information processing [3]. 

Although they may seem magical, the Casimir effect and quantum tunneling each reflect the fundamental quantum mechanical nature of nature. Quantum mechanics insists that everything behaves like a wave at small length scales. As strange as wavelike particles may seem, they are not nearly as curious as the undulations of energetic fields including the so-called empty space. 

(Not So) Empty Space 

Physicists now understand the vacuum of empty space to be a dynamic and energetic field fully able to interact with its inhabitants. The energy of the field distributes itself into a continuum of wavelengths, where each wavelength represents a unique energy. Like the waves of the sea, the waves of the vacuum constantly fluctuate and interact, but, like the constant sea level, the net movement of the waves yields an essentially constant vacuum energy, referred to as the zero-point energy of the vacuum.

Imagine placing two clusters of atoms together, as shown in Figure 1, to confine a small region of the vacuum. The situation is analogous to scooping up a cup-full of ocean water, thereby constraining the activity of the once boisterous water. In each situation, the large waves and irregularities characteristic of the larger body, along with their corresponding energies, are excluded from the confined region. Panels (a) and (b) in Figure 1 show examples of forbidden wavelengths with dashed lines. Only harmonic waves, like those in panels (c) through (f), which fit neatly in the cavity with an integer number of crests, contribute to the modified zero-point energy in the cavity. The resultant energy differential across the boundaries of the confined region produces the so-called Casimir forces that act on the boundaries themselves as the cavity either deflates or inflates to seek equilibrium. Such effects are always present when extremely small distances separate materials, but serious efforts to exploit them in electronic devices are relatively new [4].

Walking Through Walls

While Casimir effects depend on the wavelike nature of space, quantum tunneling relies on the wavelike nature of particles, electrons in particular. Tunneling also acts across small distances, but, unlike Casimir effects, tunneling does not specifically require empty space to intervene between nodes. For example, the gap between two metals in a single electron tunneling junction (SETJ) is bridged by an insulating organic molecule. Each metal confinement region, the lighter clusters in Figure 2, acts like an open room where an electron is free to move around. However, the organic molecule, the darker clusters in Figure 2, is unwilling to host any electrons, and acts like a brick wall to separate the two regions. With an applied voltage increasing across the junction, the electron is, in effect, guzzling coffee until it is literally bouncing off the walls. As shown in Figure 2, the added energy gives an electron an extreme form of the jitters, causing its wavefunction, which describes its position in space, to spread out as time progresses from panels (a) through (c). Once the energy across the junction has reached a threshold voltage, as in panel (c), the electron’s wavefunction has seeped all the way through the organic molecule to the other confinement region. At this point, if the crazed electron decides to run headlong into the once formidable brick wall, it will completely penetrate the barrier and emerge unscathed in the other room, finally clearing the way for the next tunneling event to occur. In relation to information processing, the flow of coffee into the first region represents a flow of information, and the electron processes and transmits that information through the various stages of the tunneling process just presented. 

Tunneling is a relatively old trick for electronics and might seem passé compared to the new capabilities enabled by Casimir effects, yet tunneling is robust. Although Casimir effects offer great energy efficiency, we can better control tunneling effects, which will soon allow us to array tunneling junctions into powerful processors.

How To Pull Energy Out of a Hat

Since Casimir effects draw from the zero-point energy of the vacuum, they offer great energy efficiency for future electronic devices. This is particularly true in the regime of nanoelectromechanical systems (NEMS), which incorporate mechanical, optical, and electrical functions in a single integrated circuit [4]. For example, repulsive Casimir effects can levitate electromechanical components so that they do not lose any energy to friction or heat. Even more profound than the Casimir effect’s ability to minimize power consumption is its ability to generate power from the seemingly bottomless well of vacuum generosity. Dr. Klimchitskaya and his team at the Center of Theoretical Studies and Institute for Theoretical Physics, in Germany, have found that self-sustaining Casimir oscillators can utilize both the attractive and repulsive Casimir effects to harvest zero-point energy and mechanically power NEMSs [1]. Tunneling, on the other hand, will never be able to generate more power than it consumes. Someone must supply the coffee. However, single electron tunneling is still extremely efficient compared with current technology since information transfer requires only the energy to move a single electron literally just a small fraction of a hair.

Some Sticky Situations 

Although tunneling may cost an additional quantum of energy, we can control its effects better than we can control Casimir effects. Some important control parameters to consider are operating temperature, spacing between neighboring components, and doping concentration, which is a measure of the chemical impurities pumped into a material to change its behavior [5]. 

In SETJs, electrons only follow the prescribed tunneling behavior of Figure 2 below a certain temperature, above which the rambunctious electrons are influenced more by random thermal fluctuations [5] than by the coffee we use to stimulate their desired behavior. While the temperature knob must be turned down to a minimum, distance and doping can be varied across a large range along which electrons respond predictably, allowing scientists to easily control the timing of tunneling events.

Casimir effects are slightly less picky about temperature, but spacing and doping sensitivities make Casimir effects nearly erratic. Most notably, Casimir effects can cause undesirable stiction when spacing is not optimal, wherein nearby parallel surfaces collapse and stick together. According to Aaron Katzenmeyer, from the electrical engineering department of the University of California at Davis, stiction is hard to control and causes irreversible damage to NEMSs [6]. Although scientists continue to advance their control over Casimir effects [1], progress is slow. In fact, repulsive Casimir effects are so hard to obtain that their existence was experimentally proven only last year [7]. In contrast, tunneling has not only been proven, but it has also proven easily controllable at low temperatures.

Hurrays For Arrays
The ability to control tunneling effects allows SETJs to be arrayed into powerful processing units. Although Casimir arrays have been proposed [2], our inability to control Casimir effects severely limits implementation of such theoretical proposals. Therefore, large numbers of SETJs are more realizable. However SETJ arrays offer more than power in number. The nature of tunneling in SETJs allows for collaborative networks, called cellular neural networks (CNN) to process information at an emergent level of sophistication [8]. The electrical characteristics of a SETJ are very similar to those of a brain’s neuron [3] and the emergent information processing abilities of the networks are consequentially more analogous to thoughts than the typical bit-by-bit manipulation of today’s computers. Someday, your computer, based on SETJ CNNs, could really have a mind of its own, and perhaps you can sip coffee together and agree that tunneling really made the magic happen for a new era of electronic devices.


[1]    G. L. Klimchitskaya, U. Mohideen, and V. M. Mostepankenko, “Pulsating Casimir force,” J. Phys. A: Math. Theor., vol. 40, pp.  F841–F847, 2007. doi: 10.1088/1751-8113/40/34/F03
[2]    F. Capasso, J. Munday, D. Iannuzzi, and H. B. Chan, “Casimir forces and quantum electrodynamical torques: physics and nanomechanics,” IEEE J. Selected Topics in Quantum Electronics, vol. 13, no. 2, March/April 2007.
[3]    R. Kiehl, “Information processing in nanoscale arrays: DNA assembly, molecular devices, nano-array architectures,” presented at ICCAD ’06, San Jose, CA, Nov. 5–9, 2006.
[4]    H. J. De Los Santos, “Nanoelectromechanical quantum circuits and systems,” In Proc. IEEE, vol. 91, no. 11, Nov. 2003.
[5]    A. K. Kikombo, T. Hirose, T. Asai, and Y. Amemiya, “Non-linear phenomena in electronic systems consisting of coupled single-electron oscillators,” Chaos, Solitons and Fractals, vol. 37, pp. 100–107, 2008.
[6]    A. Katzenmeyer, V. P. Logeeswaran, B. Tekin, and M. S. Islam, “Impact of Casimir force in molecular electronic switching junctions,” In Proc. 2nd IEEE International Nanoelectronics Conference, vol. 1–3, pp. 166–169, March 24–27, 2008.
[7]    J. N. Munday, F. Capasso, and V. A. Parsegian, “Measured long-range repulsive Casimir–Lifshitz forces,” Nature, vol. 457, pp. 170–173, 2009.
[8]    N. Li, H. Lu, “Single-electron tunneling depressing synapse for cellular neural networks,” Neural Comput. & Applic., vol. 17, pp. 111–118, 2008