The Casimir effect and NEMS sensors

Citation

(2011), "The Casimir effect and NEMS sensors", Sensor Review, Vol. 31 No. 1. https://doi.org/10.1108/sr.2011.08731aab.004

The Casimir effect and NEMS sensors

Article Type: Nanosensor update From: Sensor Review, Volume 31, Issue 1

Predicted in 1948 by Dutch Theoretical Physicist Hendrik Casimir, the Casimir effect is a small, attractive force that acts between two close, parallel, uncharged conducting plates (Figure 7). It is due to quantum vacuum fluctuations in the ground state of the electromagnetic field, whereby virtual particle-antiparticle pairs continually form “out of nothing” and then quickly vanish. The magnitude of this force (F) decreases rapidly with distance and varies with the separation of the plates (L) according to the inverse of the fourth power of L, i.e. F∝L⁻⁴. At the nanoscale, this becomes the dominant force between uncharged conductors and, depending on surface geometry and other factors, a separation of 10 nm can produce the equivalent of one atmosphere pressure (101.3 kPa).

A decade or so ago, this effect was little more than a scientific curiosity but today it is a potential technological problem. If effective NEMS devices, including certain types of nanosensors, are to be developed, it becomes necessary to control the quantum forces that come into play at these scales. Indeed, this was recognised back in 1995 when a group from the University of Illinois noted that the ever-falling dimensions of micro-electromechanical system (MEMS) was such that Casimir forces needed to be taken into account and argued that they are in part responsible for the “stiction” effects observed in some MEMSs. While some workers have subsequently suggested that the elimination of Casimir effects is necessary for the fabrication of NEMS, others see it as potentially useful and propose exploiting it in families of nanosensors. In any event, it needs to be fully characterised and although physicists developed a mathematical formula in the 1960s that, in principle, describes the effects of Casimir forces on any number of objects with any shape, in the vast majority of cases the formula remained impossibly hard to solve. A breakthrough came in 2006 when MIT’s Mehran Kardar demonstrated a way to solve the formula for a plate and a cylinder. Then, in 2010, another MIT group described a means of solving Casimir force equations for any number of objects with any conceivable shape. In parallel with this, recent DARPA-funded work by the ANL has conducted a detailed characterisation of the Casimir effect. Working closely with Indiana and Purdue Universities, the National Institute of Standards & Technology and the Los Alamos National Laboratory, the ANL group now has an understanding of how component and material properties can influence the Casimir effect. Interestingly, a MEMS device (Figure 8) was used during this work and measured the Casimir force through changes in capacitance. While this work is important in trying to eliminate the force’s attractive properties, the ANL team hopes also to make it repulsive. A repulsive force acting at the nanoscale would allow the design of novel NEMS devices capable of frictionless motion, perhaps through “nanolevitation” and in March 2010, Michael Levin of Harvard University’s Society of Fellows, together with the MIT researchers, described the first arrangement of materials that enable Casimir forces to cause repulsion in a vacuum.

Given this understanding and the ability to control and perhaps exploit Casimir effects, what are the prospects for sensors? This ability will be central to fabricating the next generations of MEMS sensors, such as the accelerometers and gyroscopes used in mobile phones, where the displacements of moving elements will shrink to sub-micron levels as manufacturers seek to reduce further the product footprints and material usage. Enhanced yield is expected to arise from the elimination of stiction. At the nanoscale, a class of NEMS structures that appears to offer strong sensing prospects is the nanoresonator. Effectively, the nanoscale version of silicon and quartz microresonators, these have been shown to exhibit unprecedented levels of sensitivity to mass loading. A device fabricated from silicon carbide by Michael Roukes at CalTech in 2005 had the ability to detect mass changes as low as 7 zg (7×10⁻²¹ g), which is roughly the mass of a single protein molecule, and in 2008, workers from the Spanish Research Centre of Nanoscience and Nanotechnology reported a SWCNT-based resonator with a mass resolution of 1.4 zg. It is anticipated that these devices could form the heart of new generations of mass spectrometers and resonant chemical sensors but to make this possible, some means of zeroing the devices following a mass measurement is required. The Spanish team has accomplished this by applying a current of few nA through the nanotube for several minutes. Adsorbed atoms are removed via heating and/or electromigration and as a result, the resonant frequency returns to its initial value. Although NEMS resonator technology has the potential to detect individual protein molecules, in order to distinguish between different proteins with similar masses, yoctogram (10⁻²⁴ g) levels of sensitivity would be necessary. Theoretical studies at Cornell University, published in 2010, suggest that this may well be realised in the near future.

Figure 7 Schematic representation of the Casimir effect

Figure 8 The MEMS device used during the characterisation of the Casimir effect at ANL

While these devices will benefit from an understanding of the Casimir effect, little work has so far aimed to exploit it, despite being widely proposed in the recent literature. One example, however, is work by a group from the University of Groningen in The Netherlands who propose using it in a nanoswitch which might ultimately lead to a novel family of nanosensors. This is based on AIST, an alloy of antimony, indium, silver and tellurium, which reversibly switches from a crystalline to an amorphous state when heated by a laser. The team deposited AIST on an aluminium-coated silicon wafer and held it between 40 and 120 nm from a gold sphere in an ultra-high vacuum. When AIST was in an amorphous form, the measured Casimir force was ∼100 pN but increased by 20-25 per cent when the AIST was in its crystalline form. This is because the crystalline phase is more reflective, so it confines the electromagnetic fluctuations more effectively and so increases the Casimir force. The switch would be physically moved by altering the state of the AIST and so changing the strength of the Casimir force.

Reflecting the growing interest in and potential applications of the Casimir effect, the European Science Foundation set up a research network “New trends and applications of the Casimir effect” in 2008. This will run for five years and aims to bring together European workers in this field. While much of the research is of a theoretical nature, elucidating some of the subtleties of Casimir phenomena will certainly contribute to harnessing it in future generations of nanosensing devices.