$12\text{\ensuremath{-}}\ensuremath{\mu}\mathrm{m}$-Pitch Electromechanical Resonator for Thermal Sensing

$12 − μ m$ -Pitch Electromechanical Resonator for Thermal Sensing

Ludovic Laurent, Jean-Jacques Yon, Jean-Sébastien Moulet, Michael Roukes, and Laurent Duraffourg

Phys. Rev. Applied 9, 024016 – Published 15 February 2018

Abstract

We provide here a demonstration of $12 − μ m$ -pitch nanoelectromechanical resonant infrared sensors with fully integrated capacitive transduction. A low-temperature fabrication process is used to manufacture torsional resonator arrays. An $H$ -shaped pixel with $9 − μ m$ -long nanorods and $(250 \times 30) − {nm}^{2}$ cross section is designed to provide high thermal response whose experimental measurements reach up to $1024 Hz / nW$ . A mechanical dynamic range of over 113 dB is obtained, which leads to an Allan deviation of $σ_{A} = 3 \times 10^{- 7}$ at room temperature and 50-Hz noise bandwidth ( $σ_{A} = 1.5 \times 10^{- 7}$ over 10 Hz). These features allow us to reach a sensitivity of $(8 - 12) − μ m$ radiation of $27 pW / \sqrt{Hz}$ leading to a noise-equivalent temperature difference (NETD) of 2 K for a 50-Hz noise bandwidth ( $NETD = 1.5 K$ at 10 Hz). We demonstrate that the resolution is no more set by the phonon noise but by the anomalous phase noise already encountered in flexural nanoresonators. By both improving the temperature coefficient of frequency of a factor 10 and using a readout electronics at the pixel level, these resonators will lead to a breakthrough for uncooled infrared detectors. We expect that the NETD will rapidly drop to 180 mK with electronics close to the pixel. As a result of the features of our torsional resonators, an alternative readout scheme of pixels is suggested.

Received 24 May 2017
Revised 28 August 2017

DOI:https://doi.org/10.1103/PhysRevApplied.9.024016

Physics Subject Headings (PhySH)

Metrology Optoelectronics Optomechanics Organic sensors

Nanomechanical devices Solid-state detectors

Infrared techniques

Condensed Matter, Materials & Applied Physics

Authors & Affiliations

Ludovic Laurent^1,2, Jean-Jacques Yon^1,2, Jean-Sébastien Moulet^1,2, Michael Roukes³, and Laurent Duraffourg^1,2,*

¹Université Grenoble Alpes, F-38000 Grenoble, France
²CEA, LETI, Minatec Campus, F-38054 Grenoble, France
³Departments of Physics, Applied Physics, and Bioengineering, Kavli Nanoscience Institute, California Institute of Technology, MC 149-33, Pasadena, California 91125, USA

^*Corresponding author. laurent.duraffourg@cea.fr

Article Text (Subscription Required)

Click to Expand

Supplemental Material (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 9, Iss. 2 — February 2018

Subject Areas

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
SEM images of the bolometer. (a) SEM image of the typical matrix of microelectromechanical pixels. (b) SEM picture of a typical $H$ -shaped pixel; nanorod $length = 1.5 μ m$ , $width = 250 nm$ , and $thickness = 180 nm$ (insulation arm $length = 8.6 μ m$ ). (c) SEM picture of another $H$ -shaped pixel with a thinner nanorod to enhance the thermal insulation; rod $thickness = 30 nm$ . (d) SEM image of a butterfly-shaped pixel with longer rods.
Reuse & Permissions
Figure 2
Schematic of the MEMS and NEMS resonator and readout scheme. Capacitive MEMS and NEMS resonator embedded in a downmixed phase-locked loop. The electrostatic actuation is done at $2 ω_{r} + Δ ω$ , and the plate is polarized at $ω_{r} + Δ ω$ to get a downmixed signal at $Δ ω$ that feeds the lock-in amplifier input. A transimpedance circuit is specifically designed for measuring tiny capacitance variations due to the torsional vibration of the plate.
Reuse & Permissions
Figure 3
Typical electromechanical responses. (a) Amplitude of a typical frequency response for a device with a mechanical stiffness close to $κ \sim 2.0 \times 10^{- 11} N m$ (dots). (b) Phase of a typical frequency response of the same device. We extract a quality factor of 1800 from a Lorentzian fit (light blue line), considering first and last quarters of the spectrum as background (the slight discrepancy observed at resonance is due to the presence of nonlinear terms, which are not included in the fit). (c) Electromechanical response according to frequency and actuation voltage $V_{ac}$ ( $V_{pol} = 10 V$ ). As expected, the voltage at resonance is proportional to $V_{ac}^{2}$ . (d) The amplitude according to the frequency and polarization of the plate $V_{pol}$ ( $V_{ac} = 6.5 V$ ). As expected, the voltage at resonance is proportional to $V_{pol}$ .
Reuse & Permissions
Figure 4
Frequency response. IR illumination corresponds to 26-nW incident power. This power provokes 21-Hz frequency shift (i.e., $R_{f} = 720 W^{- 1}$ ). The acquisition is performed in the closed-loop scheme for an integration time of 1 ms. Inset: Similar measurement done with 17-nW incident power on one of the best devices (20-Hz frequency shift is observed in this case).
Reuse & Permissions
Figure 5
Allan deviation measurement ( $V_{pol} = 10 V$ and $V_{ac} = 1.75 V$ ). The red star is the frequency deviation at 7-kHz integration bandwidth ( $σ_{A} = 3.5 \times 10^{- 6}$ ), and the blue star is the frequency deviation at 50-Hz integration bandwidth $σ_{A} = 3 \times 10^{- 7}$ ; 50 Hz and 7 kHz correspond to 20 ms and $70 μ s$ , respectively [35]. The minimum frequency noise ( $σ_{A} = 1.5 \times 10^{- 7}$ ) is reached for longer integration times between 50 and 200 ms. Inset: Similar measurement done with the best device [ $σ_{A} (70 μ s) = 2.5 \times 10^{- 6}$ and $σ_{A} (20 ms) = 2 \times 10^{- 7}$ ].
Reuse & Permissions
Figure 6
The FOM versus pitch. Comparison with work recently reported in the literature. Symbols: spheres correspond to devices based on frequency detection; pentagons are resistive detection. (a) The FOM tends to drop, demonstrating a continuous improvement of resonant MEMS bolometers. The FOM of our device is 1 order of magnitude higher than the values exhibited by the resistive sensor [1]. (b) Extrapolated FOM of our thermal sensor according to cumulative assumptions: (i) integration of our sensor on a CMOS circuit, (ii) pixelwise readout rate, and (iii) improvement of the thermal response through TCF enhancement ( $\times 10$ ).
Reuse & Permissions

Physical Review Applied

12−μm-Pitch Electromechanical Resonator for Thermal Sensing

Ludovic Laurent, Jean-Jacques Yon, Jean-Sébastien Moulet, Michael Roukes, and Laurent Duraffourg

Phys. Rev. Applied 9, 024016 – Published 15 February 2018

Abstract

Physics Subject Headings (PhySH)

Authors & Affiliations

Article Text (Subscription Required)

Supplemental Material (Subscription Required)

References (Subscription Required)

Issue

Subject Areas

Access Options

Authorization Required

Other Options

Download & Share

Images

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Log In

Search

Article Lookup

Paste a citation or DOI

Enter a citation

$12 − μ m$ -Pitch Electromechanical Resonator for Thermal Sensing