Nanowires as Sensitive Mechanical Sensors

The world’s most sensitive mechanical transducers and some of its most common including cantilevers used in atomic force microscopy (AFM) – are all fabricated from Si using top-down methods. The best of these cantilevers have demonstrated a thermally-limited force sensitivity of 10 attonewton in a 1 Hz bandwidth [1]. Improvement of this sensitivity would benefit a slew of applications ranging from mass detection, to cantilever magnetometry, to ultra-sensitive forms of scanning force microscopy and force-detected magnetic resonance. Ultimately the ability to make more sensitive measurements could pave the way for scientific breakthrough in a variety of fields. Currently, typical transducer fabrication processes involve optical or electron-beam lithography, chemical or plasma etching, and a release step. Even smaller structures can be milled out using focused ion beam techniques. The fluctuation-dissipation theorem and the Euler-Bernoulli beam equations imply that for a given material and temperature, long, narrow, and thin cantilevers are the most sensitive transducers. In fact, a review of real transducers confirms this trend. The ultimate force resolution of such devices, which inevitably have large surface-to-volume ratios, is limited by surface imperfections.

New developments in the growth of inorganic nanowires (NWs), however, are starting to change the status quo. Researchers can now grow nanometer-scale structures from the bottom-up with unprecedented mechanical properties. Unlike traditional cantilevers and other top-down structures, which are etched or milled out of a larger block of material, bottom-up structures are assembled unit-by-unit to be almost defect-free on the atomic-scale with perfectly terminated surfaces. This near perfection gives NWs a much smaller mechanical dissipation than their top-down counterparts while their high resonance frequencies allow them to couple less strongly to common sources of noise.

NW oscillators can be grown from a wide range of materials and diameters range from tens to hundreds of nanometers and lengths reach up to tens of microns. The ability to grow NW heterostructures presents additional flexibility. Until now, the mechanical properties of very few types of NWs have been measured and thermal motion has only been reported for Si NWs. Preliminary results show that dissipation in Si and GaAs NWs is extremely low [2, 3], making this system an excellent candidate for the next generation in ultra-sensitive force transduction. Detailed characterization of NW oscillators and their application as ultra-sensitive force transducers remain open lines of inquiry. The possibility of finding NW tranducers with force sensitivities several orders of magnitude better than the state-of-the-art is a distinct possibility.

Displacement detection is a crucial component of any mechanical sensor. A mechanical oscillator such as a cantilever or membrane is merely a transducer, i.e. an element which transforms a force into a displacement. For a force to be measured, the resulting displacement must be detected. Various techniques are used for the displacement detection of traditional micromechanical oscillators including optical, microwave, capacitive, magnetic, or piezoelectric schemes.

Displacement detection of bottom-up resonators has proven more complicated due to their small size. The sensitivity of convenient optical techniques such as beam deflection or interferometry suffers as the dimensions of the mechanical resonator become smaller than the wavelength of light. Recently, however, Nichol et al. demonstrated a polarization-enhanced interferometry technique capable of detecting the thermal motion of a Si NW with a diameter less than 100 nm [2].  This method succeeds because NWs, which are typically many microns long, exhibit enhanced reflectivity when the incident light is polarized along their long axis. Once the thermal motion of a mechanical transducer can be measured, the combined system can be used as a thermally limited force sensor  ̶  a system whose minimum detectable force is solely determined by the spectral density of its thermal fluctuations.

Applications of SPM requiring extremely fine force sensitivity have benefited from the development of ultra-sensitive cantilevers. In particular, non-contact AFM in the “pendulum geometry”, where the cantilever hangs perpendicular to the scanned surface, allows for the use of extremely soft – and therefore sensitive – cantilevers. Measurements of this type include high-sensitivity AFM in ultra-high vacuum (UHV), spectroscopy measurements of small friction forces, and Kelvin probe force microscopy. Magnetic resonance force microscopy (MRFM) is perhaps the most prominent example of a technique that relies on exquisite force sensitivity. Over the last 20 years, researchers have been making steady progress improving the sensitivity of force-detected magnetic resonance: it presently surpasses the sensitivity of conventional, inductive nuclear magnetic resonance detectors by 8 orders of magnitude [4].

In 2009, a group led by Dr. Dan Rugar (IBM Almaden), demonstrated the promise of these developments by using MRFM to capture 3D images of individual virus particles [5]. Later, related work showed that MRFM could provide further information on the nanoscale by applying techniques such as nuclear double resonance [6], by imaging organic nanolayers [7], and by measuring nuclei inside a nanoscale semiconductor sample [8,9]. This new MRFM technique – dubbed nanoscale magnetic resonance imaging (nano-MRI) – has the unique capability to image the interior of nanoscale objects non-invasively and with intrinsic chemical selectivity. Despite the tremendous improvements made in the last few years, important obstacles must still be overcome in order for the technique to become a useful tool for biologists and materials scientists. The possibility of extending MRFM to atomic resolution – whereby molecules could be imaged atom-by-atom in 3D – appears within reach, though it remains a technically challenging prospect.

2012 saw an intriguing development with the first report of force-detected nuclear magnetic resonance using a Si NW cantilever as the force sensor [10,11]. This result represents one of the first applications of bottom-up resonators as ultra-sensitive sensors. The authors demonstrate the low-temperature detection of a statistical polarization of 1H in nanoscale sample of polystyrene with a nearly thermally limited force sensitivity of 1.9 aN2/Hz. Force sensitivity is a major limitation on the resolution of nano-MRI and improving it will almost certainly involve using such bottom-up cantilevers. Therefore, this achievement could be the first in a series of developments pushing MRFM closer to attaining its goal of 3D molecular imaging with atomic resolution.

We have realized such NW force sensors and have integrated them into a low-temperature SPM apparatus. This task required the construction of scanning probe microscope customized for NW cantilevers. Our ability to scan such force sensors is now opening are variety of scanning force measurements including measurements of weak lateral forces, atomic-scale friction, and vectorial force sensing [12,13].

[1] D. Rugar et al., Nature 430, 329 (2004); J. D. Teufel et al., Nature Nanotech. 4, 820 (2009).
[2] J. M. Nichol et al., Appl. Phys. Lett. 93, 193110 (2008).
[3] R. Budakian, 4th NanoMRI Conference, Monte Verità, Switzerland, 25 July, 2012.
[4] M. Poggio and C. L. Degen, Nanotechnology 21, 342001 (2010).
[5] C. L. Degen et al., Proc. Nat. Acad. Sci. U.S.A. 106, 1313 (2009).
[6] M. Poggio et al., Phys. Rev. Lett. 102, 087604 (2009).
[7] H. J. Mamin et al., Nano Letters 9, 3020 (2009).
[8] Fei Xue et al., Phys. Rev. B 84, 205328 (2011).
[9] P. Peddibhotla et al., Nature Phys. 9, 631 (2013).
[10] J. M. Nichol et al., Phys. Rev. B 85, 054414 (2012).
[11] J. M. Nichol et al., Phys. Rev. X 3, 031016 (2013).
[12] A. Gloppe et al., Nature Nanotech. 9, 920 (2014).
[13] N. Rossi et al., Nat. Nanotechnol. AOP, 17 October 2016 (DOI 10.1038/nnano.2016.189).