sensors Article
An Electrochemical, Low-Frequency Seismic Micro-Sensor Based on MEMS with a Force-Balanced Feedback System Guanglei Li, Junbo Wang *, Deyong Chen, Jian Chen
ID
, Lianhong Chen and Chao Xu
Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China; [email protected] (G.L.); [email protected] (D.C.); [email protected] (J.C.); [email protected] (L.C.); [email protected] (C.X.) * Correspondence: [email protected]; Tel.: +86-135-2008-9501 Received: 14 August 2017; Accepted: 12 September 2017; Published: 13 September 2017
Abstract: Electrochemical seismic sensors are key components in monitoring ground vibration, which are featured with high performances in the low-frequency domain. However, conventional electrochemical seismic sensors suffer from low repeatability due to limitations in fabrication and limited bandwidth. This paper presents a micro-fabricated electrochemical seismic sensor with a force-balanced negative feedback system, mainly composed of a sensing unit including porous sensing micro electrodes immersed in an electrolyte solution and a feedback unit including a feedback circuit and a feedback magnet. In this study, devices were designed, fabricated, and characterized, producing comparable performances among individual devices. In addition, bandwidths and total harmonic distortions of the proposed devices with and without a negative feedback system were quantified and compared as 0.005–20 (feedback) Hz vs. 0.3–7 Hz (without feedback), 4.34 ± 0.38% (without feedback) vs. 1.81 ± 0.31% (feedback)@1 Hz@1 mm/s and 3.21 ± 0.25% (without feedback) vs. 1.13 ± 0.19% (feedback)@5 Hz@1 mm/s (ndevice = 6, n represents the number of the tested devices), respectively. In addition, the performances of the proposed MEMS electrochemical seismometers with feedback were compared to a commercial electrochemical seismic sensor (CME 6011), producing higher bandwidth (0.005–20 Hz vs. 0.016–30 Hz) and lower self-noise levels (−165.1 ± 6.1 dB vs. −137.7 dB at 0.1 Hz, −151.9 ± 7.5 dB vs. −117.8 dB at 0.02 Hz (ndevice = 6)) in the low-frequency domain. Thus, the proposed device may function as an enabling electrochemical seismometer in the fields requesting seismic monitoring at the ultra-low frequency domain. Keywords: electrochemical seismic sensor; microfabrication; negative feedback; repeatability
1. Introduction A seismic sensor functions as a velocity sensor or an accelerometer that senses the ground vibration of the earth, which is widely used in the field of earthquake monitoring, resource exploration, and ocean bottom observation [1–3]. The studies of long-period surface waves and slow seismic processes require better seismic data in the low-frequency domain [4,5]. Seismic sensors include moving-coil seismic sensors [6–8], fiber-optic seismic sensors [9,10], pendulum seismometers [11,12], and micro-electromechanical system (MEMS) accelerometers [13–15]. Conventionally, seismologists have detected low-frequency seismic vibrations by increasing the quality of the inertial mass or decreasing the equivalent spring coefficient, leading to the decrease of the natural frequency, which, however, renders the conventional seismic sensors bulky and vulnerable. Compared to other seismometers, the electrochemical seismic sensors, which is only used for the detection of low seismic signals, are featured with high performance in the low-frequency domain, enabling the measurements of long-period surface waves and slow seismic processes [16–18]. Sensors 2017, 17, 2103; doi:10.3390/s17092103
www.mdpi.com/journal/sensors
Sensors 2017, 17, 2103
2 of 8
The first prototype of the electrochemical seismic sensor was based on “solion,” which was initially developed in the 1950s by the US Navy, targeting the detection of low-frequency acoustic waves [19–22]. Significant improvements of “solion” were realized in Russia where the electrochemical seismic sensors, which were called molecular electronic transducers in the 1990s, were proposed based on four Pt meshes as electrodes insulated by three porous dielectric ceramic spacers [23,24]. However, the conventional mesh weaving approach and ceramic sintering technologies cannot fabricate electrodes with high repeatability leading to issues of low yield and high cost. To resolve these issues, the electrochemical seismic sensors based on MEMS technologies were introduced by He et al. [25–27], which improved the consistency and decreased the cost significantly [27]. However, the proposed electrochemical seismic micro-sensors still suffered from limited working bandwidth; thus, their applications in ocean bottom seismometers were compromised. In this paper, a force-balanced negative feedback system was proposed to extend the working bandwidth of the MEMS-based electrochemical seismometers, which, at the same time, retains the characteristics of high repeatability. In addition, the microfabrication of the sensing electrodes was improved to enhance production yield. The structure of this paper is organized as follows. In Section 2, the structure and the working principle of the MEMS electrochemical seismic sensor with a negative feedback system are described. Section 3 introduces the fabrication and assembling of the MEMS electrochemical seismic sensor. Section 4 provides the characterization of the developed devices. Section 5 concludes the paper. 2. Structure and Working Principle Figure 1 illustrates the schematic of the MEMS electrochemical seismic sensor with force-balanced negative feedback. The electrochemical seismic sensor consists of a sensing unit and a feedback unit. The sensing unit includes porous sensing electrodes immersed in an electrolyte solution as the liquid mass, which is then sealed by two elastic membranes within a plexiglass house. The porous sensing electrodes fabricated by MEMS technologies are arranged in the anode-cathode-cathode-anode setup to improve the linearity of the output signals. The electrolyte, which is a mixture of iodine (I2 ) and potassium iodide (KI), flows through the via-holes in the sensing electrodes. The following reversible electrochemical reactions occur on the surfaces of the anodes and the cathodes, respectively, when a DC bias (0.1–0.3 V) is applied on the electrodes, generating ion concentration gradients between each pair of the anode and the cathode. anode : 3I − − 2e− → I3−
(1)
cathode : I3− + 2e− → 3I −
(2)
where I3− in (1) and (2) is a clathrate combining I2 with I − . In response to external seismic signals, the electrolyte solution moves opposite to the direction of the external vibration due to the inertial force, which causes the variation of the ion concentration gradients between the anode–cathode pairs. These concentration gradients lead to current outputs, which are proportional to the external seismic signals [25]. The feedback unit composes of a feedback circuit (e.g., a pre-amplification, a pre-filter, and a PID adjuster), a feedback magnet and a feedback coil (see Figure 1). The raw output signals generated from the sensing unit are processed by the feedback circuit and then applied to the feedback coil, positioned within the magnetic field of the feedback magnet. The alternating currents in the feedback coil interact with the magnet field to generate feedback forces, which provides an opposite force to counterbalance the movement of the liquid mass. The transfer function of the closed system is presented in the following form: W1 W= × WBPF (3) 1 + W1 × WPID
Sensors Sensors 2017, 2017, 17, 17, 2103 2103
33 of of 88
where = × × (WS, WAC, and WPF represent the transfer functions of the sensor, the where W1 = Ws × WAC × WPF (WS , WAC , and WPF represent the transfer functions of the sensor, amplifying circuit, and the pre-filter, respectively), and and are the transfer functions of the amplifying circuit, and the pre-filter, respectively), and WPID and WBPF are the transfer functions the PID adjuster and the band-pass filter, respectively. of the PID adjuster and the band-pass filter, respectively.
Figure Figure 1. 1. The schematic of the MEMS electrochemical electrochemical seismic seismic sensor sensor which which is is composed composed of of aa sensing sensing unit including porous sensing electrodes immersed in an electrolyte solution sealed in a unit including porous sensing electrodes immersed in an electrolyte solution sealed in a plexiglass plexiglass house unit including a feedback circuit, a feedback magnet, and house by byelastic elasticmembranes membranesand anda feedback a feedback unit including a feedback circuit, a feedback magnet, aand feedback coil. In response to external seismicseismic signals,signals, the electrolyte moves opposite to the direction a feedback coil. In response to external the electrolyte moves opposite to the of the external to ion concentration gradients and raw current between the direction of thevibration, external leading vibration, leading to ion concentration gradients and outputs raw current outputs anode–cathode pairs. The raw output signals generated from the sensing unit were processed by the between the anode–cathode pairs. The raw output signals generated from the sensing unit were feedback and then applied the feedback coil, to generating feedback due tofeedback its interactions processedcircuit by the feedback circuittoand then applied the feedback coil, forces generating forces with theitsfeedback magnet, the movement of the liquid mass. due to interactions withcounterbalancing the feedback magnet, counterbalancing the movement of the liquid mass.
deep negative negativefeedback feedback(W ( 1× × WPID ≫ In aa deep 1) 1) system, system, Equation Equation (3) (3) can be transformed into the following form: form: following W W = BPF . (4) =WPID . (4) As shown in Equation (4), the output signals of the seismic sensor are shaped by WPID . As shown in Equation (4), the output signals of the seismic sensor are shaped by . The The proportional coefficient and the compensation capacitor, two major parameters in the feedback proportional coefficient and the compensation capacitor, two major parameters in the feedback system, can affect the output characterization. system, can affect the output characterization. The pre-filter and the PID adjuster (proportion integration differentiation) in the feedback circuit The pre-filter and the PID adjuster (proportion integration differentiation) in the feedback provides compensations to enhance both the amplitude and the phase margins by adjusting the values circuit provides compensations to enhance both the amplitude and the phase margins by adjusting of the compensation capacitors. The increases of both amplitude and phase margins enable the system the values of the compensation capacitors. The increases of both amplitude and phase margins to function in a manner of deep feedback. Additionally, since the feedback force counterbalances the enable the system to function in a manner of deep feedback. Additionally, since the feedback force movement of the inertial mass in response to environmental vibrations, the dynamic range of the counterbalances the movement of the inertial mass in response to environmental vibrations, the developed electrochemical seismometer is effectively enlarged. dynamic range of the developed electrochemical seismometer is effectively enlarged. 3. Fabrication and Assembling 3. Fabrication and Assembling The sensing electrodes were fabricated by conventional MEMS technologies including lithography, sensing electrodes were fabricated by conventional MEMS where technologies including deep The reactive ion etching, thermal oxidation, sputtering, and wire bonding, detailed processes lithography, deep reactive ion etching, thermal oxidation, sputtering, and wire bonding, can be found in previous publications [26]. Then, the integrated electrodes (see Figure 2A)where were detailed processes can be found in previous publications [26]. Then, the integrated electrodes (see positioned in a plexiglass tube, which was filled by an electrolyte solution and sealed by two elastic Figure 2A) were positioned in a plexiglass tube, which was filled by an electrolyte solution and membranes. The molar ratio of I2 to KI was between 1:50 and 1:100, which ensures that I2 molecules −I2 to KI was between 1:50 and 1:100, which sealed by two elastic membranes. The molar ratio of were fully dissolved in the KI solution in the forms of I3 . ensures that2B I2 introduces molecules the were fully dissolved the KI solution in the sensor forms of Figure assembled MEMSin electrochemical seismic with a. feedback system. Figure 2B introduces the assembled MEMS electrochemical seismic sensor withthea feedback feedback The feedback coil fabricated with 3D printing was immobilized on the frame, while system. The feedback coil fabricated with 3D printing was immobilized on the frame, while the magnet was immobilized on the plexiglass housing. Thus, the feedback magnet moves alongside with feedback the plexiglass Thus, circuit the feedback magnet moves the seismicmagnet sensor, was whileimmobilized the feedbackon coil remains fixed.housing. The feedback including a differential alongside with the seismic sensor, while the feedback coil remains fixed. The feedback circuit amplification unit, a pre-filter unit, and a PID adjuster was immobilized above the seismic sensors. including a differential amplification unit, a pre-filter unit, and a PID adjuster was immobilized above the seismic sensors.
Sensors 2017, 17, 2103 Sensors 2017, 17, 2103 Sensors 2017, 17, 2103
4 of 8 4 of 8 4 of 8
Figure 2. The sensing unit (A) and and the prototype prototype (B) of of the MEMS MEMS electrochemical seismic seismic sensors Figure Figure 2. 2. The The sensing sensing unit unit (A) (A) and the the prototype (B) (B) of the the MEMS electrochemical electrochemical seismic sensors sensors with force-balanced negative feedback. The inertial mass composed of the frame, the electrolyte, the with force-balanced negative feedback. The inertial mass composed of the frame, the electrolyte, with force-balanced negative feedback. The inertial mass composed of the frame, the electrolyte, the membranes, andand the feedback coil. The feedback coil and the membranes, feedback coil. The feedback coil andthe thefeedback feedbackmagnet magnetwere were not not connected connected membranes, and thethe feedback coil. The feedback coil and the feedback magnet were not connected together, which may may have have relative relative motion motion between between the the two two components. components. together, which together, which may have relative motion between the two components.
4. 4. Experimental Experimental Characterization Characterization The characterizations MEMS electrochemical seismic sensors were on The characterizations thethe MEMS electrochemical seismic sensors were conducted on a modified characterizationsofof of the MEMS electrochemical seismic sensors were conducted conducted on aa modified home-developed platform (see Figure 3), which consists of a function generator, a power home-developed platform (see Figure 3), Figure which 3), consists a function a power amplifier, modified home-developed platform (see whichofconsists of agenerator, function generator, a power amplifier, a laser rangefinder, a vibration exciter, and a data acquisition system. The aamplifier, laser rangefinder, a vibration exciter, and a data acquisition system. The excited alternative waves a laser rangefinder, a vibration exciter, and a data acquisition system. The excited excited alternative waves were the generator amplified by power amplifier, were generated from thegenerated function from generator and amplified by and the power amplifier, then alternative waves were generated from the function function generator and amplified by the the which power were amplifier, which were then applied to the vibration exciter to generate vibration signals at a certain frequency. applied to thethen vibration exciter tovibration generate vibration at avibration certain frequency. laser rangefinder which were applied to the exciter tosignals generate signals atThe a certain frequency. The laser rangefinder used the of signals, while the was measure thewas velocities ofmeasure the vibration signals, while thevibration data acquisition NI, The used laserto rangefinder was used to to measure the velocities velocities of the the vibration signals,system while(4472, the data data acquisition system (4472, NI, Austin, TX, USA) was used to measure the output signals. Austin, TX, system USA) was used toAustin, measure theUSA) output signals. acquisition (4472, NI, TX, was used to measure the output signals.
Figure 3. The schematic of the experimental platform where the developed seismic sensor was Figure 3. 3. The experimental platform the developed developed seismic seismic sensor sensor was was Figure The schematic schematic of of the the experimental platform where where the positioned on a vibration exciter, which is excited electrically (function generation and power positioned on a vibration exciter, which is excited electrically (function generation and power positioned on a vibration exciter, which is excited electrically (function generation and power amplifier) amplifier) and detected optically (laser rangefinder). amplifier) andoptically detected(laser optically (laser rangefinder). and detected rangefinder).
Figure Figure 4A 4A shows shows the the frequency frequency characterization characterization results results of of the the MEMS MEMS electrochemical electrochemical seismic seismic Figure 4A shows the frequency characterization results of the MEMS electrochemical seismic sensor sensor without without feedback feedback and and with with feedback feedback plus plus variations variations in in aa proportional proportional coefficient. coefficient. It It was was sensor without feedback and with feedback plus variations in lead a proportional coefficient. Itthe was shown shown that the increase of the proportional coefficient can to (1) an increase in working shown that the increase of the proportional coefficient can lead to (1) an increase in the working that the increase of the proportional coefficient can lead to 4B (1) an increase in the working bandwidth bandwidth bandwidth and and (2) (2) the the possibilities possibilities of of resonance. resonance. Figure Figure 4B shows shows the the frequency frequency characterization characterization and (2) of the possibilities of resonance. Figure 4B shows the frequency characterization results of MEMS results results of MEMS MEMS electrochemical electrochemical seismic seismic sensor sensor with with different different values values of of compensation compensation capacitors. capacitors. It It electrochemical seismic sensor with different values of compensation capacitors. It wasonobserved that was observed that the value of the compensation capacitor had no influence was observed that the value of the compensation capacitor had no influence on the the output output the value ofat the compensation capacitor hadwhich no influence on the output amplitude at the low-frequency amplitude amplitude at the the low-frequency low-frequency domain, domain, which can can decrease decrease the the possibilities possibilities of of resonance resonance due due to to domain, which can decrease the possibilities of resonance due toworking the phasebandwidth compensation effects of the the phase compensation effects of the capacitor. The required can be obtained the phase compensation effects of the capacitor. The required working bandwidth can be obtained capacitor. The required working bandwidth can be obtained by adjusting the appropriate values of the by by adjusting adjusting the the appropriate appropriate values values of of the the proportional proportional coefficient coefficient and and the the compensation compensation capacitor. capacitor. proportional coefficient and the compensation capacitor.
Sensors 2017, 17, 2103 Sensors Sensors 2017, 2017, 17, 17, 2103 2103
5 of 8 55 of of 88
Frequency characterization results of Figure Figure 4. 4. Frequency Frequency characterization characterization results of the the MEMS MEMS electrochemical electrochemical seismic seismic sensors sensors without without feedback, with with feedback feedback plus plus variations variations in in proportional proportional coefficients coefficients (A) (A) and and feedback feedback, and with with feedback feedback plus plus (B). variations in compensation capacitors variations in compensation capacitors (B).
Figure Figure 555 shows shows input input voltages voltages measured measured by by the the laser laser rangefinder rangefinder and and the the output output voltages voltages of of Figure shows input voltages measured by the laser rangefinder and the output voltages of CME 6011 and the proposed device @20 Hz (A)@0.18 mm/s, @1 [email protected] mm/s (B), and @0.016 CME 6011 and the proposed device @20 Hz (A)@0.18 mm/s, @1 [email protected] mm/s (B), and @0.016 CME 6011 and the proposed device @20 Hz (A)@0.18 mm/s, @1 [email protected] mm/s (B), and @0.016 [email protected] [email protected] mm/s mm/s (C). (C). ItItwas was observed that (1) the output voltages of the proposed device device were were smaller smaller [email protected] mm/s (C).It wasobserved observedthat that(1) (1)the theoutput outputvoltages voltages of of the the proposed proposed device were smaller than that of CME 6011 at 20 Hz due to the working bandwidth limitation of the proposed device, (2) than that of CME 6011 at 20 Hz due to the working bandwidth limitation of the proposed device, (2) than that of CME 6011 at 20 Hz due to the working bandwidth limitation of the proposed device, the output voltage of the proposed device was almost the same as that of CME 6011 at 1 Hz, and (3) the output voltage of the proposed device was almost the same as that of CME 6011 at 1 Hz, and (3) (2) the output voltage of the proposed device was almost the same as that of CME 6011 at 1 Hz, the voltage of device was than CME 6011 at Hz to the output output voltagevoltage of the the proposed proposed devicedevice was larger larger than that that of CME 60116011 at 0.016 0.016 Hz due due to the the and (3) the output of the proposed was larger thanof that of CME at 0.016 Hz due to working bandwidth limitation of CME 6011. working bandwidth limitation of CME 6011. the working bandwidth limitation of CME 6011.
Figure 5. Input voltages measured by the laser rangefinder Figure Figure 5. 5. Input Input voltages voltages measured measured by by the the laser laser rangefinder rangefinder and and the the output output voltages voltages of of CME CME 6011 6011 and and the proposed device @20 Hz (A) @0.18 mm/s, @1 [email protected] mm/s (B), and @0.016 [email protected] mm/s (C). the the proposed proposed device device @20 @20 Hz Hz (A) (A) @0.18 @0.18 mm/s, mm/s, @1 @1 [email protected] [email protected] mm/s mm/s (B), (B), and and @0.016 @0.016 [email protected] [email protected] mm/s mm/s (C). (C).
Figure shows the the frequency frequency characteristics characteristics of the developed developed micro micro seismic seismic sensors sensors with with and and Figure 66 shows shows the frequency characteristics of the the developed micro seismic sensors with and without the feedback system and CME 6011. More specifically, the electrochemical seismic micro without the feedback system and CME 6011. More specifically, the electrochemical seismic micro without the feedback system and CME 6011. the electrochemical seismic micro sensors sensors without without feedback feedback produced produced aaa limited limited bandwidth bandwidth at at 0.3–7 0.3–7 Hz. Hz. By sensors without feedback produced limited bandwidth at 0.3–7 Hz. By introducing introducing and and optimizing optimizing the feedback unit, the working bandwidth was extended to 0.005–20 Hz, which was broader the feedback unit, the working bandwidth was extended to 0.005–20 Hz, which was broader than the feedback unit, the working bandwidth was extended to 0.005–20 Hz, which was broader thanthan that that of CME 6011 with a bandwidth of 0.016–30 Hz. that of CME a bandwidth of 0.016–30 of CME 6011 6011 with with a bandwidth of 0.016–30 Hz. Hz. Figure developed micro seismic sensors with and without the shows the sensitivities ofofthe the developed micro seismic sensors with andand without the Figure 77 shows showsthe thesensitivities sensitivitiesof the developed micro seismic sensors with without feedbacks, which were quantified as 2026 ± 12 V/m/s vs. 5856 ± 73 V/m/s at 1 Hz and 1994 ± 8 V/m/s feedbacks, which were quantified as 2026 ± 12 V/m/s vs. 5856 ± 73 V/m/s at 1 Hz and 1994 ± 8 V/m/s the feedbacks, which were quantified as 2026 ± 12 V/m/s vs. 5856 ± 73 V/m/s at 1 Hz and vs. 104 at Hz (see Figure In to micro seismic sensors vs. 5603 5603 104 V/m/s V/m/s at 55 ± Hz (see Figure 7A,B). In comparison comparison toIn micro seismic to sensors without 1994 ± 8 ±±V/m/s vs. 5603 104 V/m/s at7A,B). 5 Hz (see Figure 7A,B). comparison microwithout seismic feedback, the seismic sensors feedback lower sensitivities due to feedback, the micro micro seismic sensors with feedback produced lowerproduced sensitivities duesensitivities to the the feedback feedback sensors without feedback, the micro with seismic sensorsproduced with feedback lower due forces. However, micro seismic sensors with feedback produced much lower total harmonic forces. However, micro seismic sensors with feedback produced much lower total harmonic to the feedback forces. However, micro seismic sensors with feedback produced much lower total distortions compared to without feedbacks (see 7C,D). For the distortionsdistortions comparedcompared to the the micro micro sensors withoutwithout feedbacks (see Figure Figure 7C,D). For instance, instance, the harmonic to thesensors micro sensors feedbacks (see Figure 7C,D). For instance, total harmonic distortions (which was calculated by the ratio of the total mean square root of the total harmonic distortions (which was calculated by the ratio of the total mean square root of the the total harmonic distortions (which was calculated by the ratio of the total mean square root harmonic harmonic wave wave and and the the fundamental fundamental frequency frequency amplitude.) amplitude.) decreased decreased from from 4.34 4.34 ±± 0.38% 0.38% (without (without feedback) feedback) to to 1.81 1.81 ±± 0.31% 0.31% (feedback)@1 (feedback)@1 Hz@1 Hz@1 mm/s mm/s and and 3.21 3.21 ±± 0.25% 0.25% (without (without feedback) feedback) to to
Sensors 2017, 17, 2103
6 of 8
Sensors 2017, 17, 2103 Sensors 2017, 17, 2103
6 of 8 6 of 8
of the harmonic wave and the fundamental frequency amplitude.) decreased from 4.34 ± 0.38% 1.13 ± 0.19% (feedback)@5 Hz@1 mm/s. An increase in the amplitude of 0.25% the output signal lead to to an (without feedback) to 1.81 ± 0.31% (feedback)@1 Hz@1 and 3.21 of ± (without feedback) 1.13 ± 0.19% (feedback)@5 Hz@1 mm/s. An increase in mm/s the amplitude the output signal lead to an increase in feedback force, whichmm/s. confined the movement of the liquid mass within the linear range, 1.13 ± 0.19% (feedback)@5 Anthe increase in theof amplitude the output leadrange, to an increase in feedback force, Hz@1 which confined movement the liquidofmass within signal the linear and decreased total harmonic distortions. increase in feedback force, which confined the movement of the liquid mass within the linear range, and decreased total harmonic distortions. and decreased total harmonic distortions.
Figure 6.Frequency Frequency characterization results results ofthe the MEMSelectrochemical electrochemical seismicsensor sensor withand and Figure Figure 6. 6. Frequency characterization characterization results of of the MEMS MEMS electrochemical seismic seismic sensor with with and device = 6) in comparison to CME 6011. The amplitude is the output voltage of the without feedback (n without = 6) in comparison to CME 6011. The amplitude is the output voltage of the device = 6) in comparison to CME 6011. The amplitude is the output voltage of the without feedback feedback (n (ndevice electrochemicalseismic seismicsensors sensorsatatthe the sameinput input velocityand and differentfrequencies. frequencies. electrochemical electrochemical seismic sensors at the same same input velocity velocity and different different frequencies.
Figure 7. (A,B) The sensitivity of the MEMS electrochemical seismic sensors with and without feedbacks Figure 7. (A,B) The sensitivity of the MEMS electrochemical seismic sensors with and without Figure 7. (A,B) The velocities sensitivityat of theand MEMS seismic and without as a function of input 1 Hz 5 Hz,electrochemical respectively. (C,D) Total sensors harmonicwith distortions of the feedbacks as a function of input velocities at 1 Hz and 5 Hz, respectively. (C,D) Total harmonic feedbacks as a function of input velocities at 1without Hz andfeedbacks 5 Hz, respectively. (C,D) Total harmonic MEMS electrochemical seismic sensors with and as a function of input velocities at distortions of the MEMS electrochemical seismic sensors with and without feedbacks as a function of 1distortions Hz and 5 Hz, respectively. of the MEMS electrochemical seismic sensors with and without feedbacks as a function of input velocities at 1 Hz and 5 Hz, respectively. input velocities at 1 Hz and 5 Hz, respectively.
The thethe self-noise levelslevels were were also conducted in this platform Thetests testsof of self-noise also conducted in this without platforminducing withoutvibrations inducing TheAs tests of in theFigure self-noise levels were alsospectrums conducted in developed this platform without inducing actively. shown 8, the self-noise power of the MEMS electrochemical vibrations actively. As shown in Figure 8, the self-noise power spectrums of the developed MEMS vibrations actively. Aslower shownthan in Figure 8, CME the self-noise power spectrums of the of developed seismic sensors were thatlower of domain less thanMEMS 1 Hz. electrochemical seismic sensors were than6011 that in of the CMEfrequency 6011 in the frequency domain of less electrochemical seismic sensors were lower than that of CME 6011 in the frequency domain of less More the self-noisethe levels were characterized as −165.1 ± 6.1 dB (micro sensors) vs. than 1specifically, Hz. More specifically, self-noise levels were characterized as −165.1 ± 6.1seismic dB (micro seismic than 1 Hz. More specifically, the − self-noise levels were characterized as −165.1 ± 6.1 dB (micro seismic − 137.7 dB (CME 6011) at 0.1 Hz, 151.9 ± 7.5 dB (micro seismic sensors) vs. − 117.8 dB (CME 6011) at sensors) vs. −137.7 dB (CME 6011) at 0.1 Hz, −151.9 ± 7.5 dB (micro seismic sensors) vs. −117.8 dB sensors) vs. −137.7 dB (CME 6011) at 0.1 Hz, −151.9 ± 7.5 dB (micro seismic sensors) vs. −117.8 dB (CME 6011) at 0.02 Hz (ndevice = 6). It was speculated that at a frequency higher than 1 Hz, the noise of (CME 6011) at 0.02 Hz (ndevice = 6). It was speculated that at a frequency higher than 1 Hz, the noise of the seismic sensor mainly comes from mechanical and environmental noises due to packaging. Thus, the seismic sensor mainly comes from mechanical and environmental noises due to packaging. Thus,
Sensors 2017, 17, 2103
7 of 8
0.02 Hz (ndevice = 6). It was speculated that at a frequency higher than 1 Hz, the noise of the seismic Sensors 2017, 17, 2103 7 of 8 sensor mainly comes from mechanical and environmental noises due to packaging. Thus, the increases in self-noise of the proposed may result frommay imperfect packaging technologies in thethe increases in levels the self-noise levels ofdevices the proposed devices result from imperfect packaging comparison sensors.seismic sensors. technologiesto incommercial comparisonseismic to commercial
Figure 8. The self-noise power spectrums of the developed MEMS electrochemical seismic sensors Figure 8. The self-noise power spectrums of the developed MEMS electrochemical seismic sensors with feedback feedback (n (ndevice ==6)6)and with andthe theconventional conventionalCME CME6011 6011where wherethe thedeveloped developed seismic seismic micro micro sensors sensors device device produced lower self-noise levels than CME 6011 in the frequency domain of 0.01–1 Hz. device produced lower self-noise levels than CME 6011 in the frequency domain of 0.01–1 Hz.
5. Conclusions 5. Conclusions A MEMS-based electrochemical seismic sensor with ultra-broad frequency working bandwidth A MEMS-based electrochemical seismic sensor with ultra-broad frequency working bandwidth was demonstrated in this paper. The working bandwidth and the self-noise level of the developed was demonstrated in this paper. The working bandwidth and the self-noise level of the developed device were characterized with feedback parameters optimized. Experimental results showed that device were characterized with feedback parameters optimized. Experimental results showed that the working bandwidth of the developed seismic sensor was extended to 0.005–20 Hz, validating the the working bandwidth of the developed seismic sensor was extended to 0.005–20 Hz, validating the positive effects of the negative feedback. In addition, the developed device produced a lower positive effects of the negative feedback. In addition, the developed device produced a lower self-noise self-noise level at the low frequency domain in comparison to the commercial seismic sensor of level at the low frequency domain in comparison to the commercial seismic sensor of CME6011, CME6011, validating its large dynamic ranges. Thus, this MEMS electrochemical seismic sensor can validating its large dynamic ranges. Thus, this MEMS electrochemical seismic sensor can be potentially be potentially used in the field requesting seismic monitoring at the ultra-low frequency domain used in the field requesting seismic monitoring at the ultra-low frequency domain with large dynamic with large dynamic ranges and low noise levels. ranges and low noise levels. Acknowledgments: This Science Foundation of China (No. 61431910), the Acknowledgments: This work workwas wassupported supportedby bythe theNational National Science Foundation of China (No. 61431910), National High Technology Research and Development Program ofofChina project the National High Technology Research and Development Program China(the (the863 863 Program) Program) with with project No. 2014AA09158, the National Key Scientific Instrument and Equipment Development Project of China with project 2013YQ12035703, and and the the MEMS MEMS Electrochemical Electrochemical Seismic Seismic Sensors Sensors Research Research in in the project No. No. 2013YQ12035703, the Ocean Ocean Bottom Bottom Observation with the project No. 2016YFC0301801. Observation with the project No. 2016YFC0301801. Author Contributions: Guanglei Li performed the experiments and developed the theoretical description of the Author Contributions: Li performed the experiments andanalyzed developed theoreticaldata. description of the method proposed in theGuanglei publication. Lianhong Chen and Chao Xu thethe experimental Junbo Wang, method proposed in the publication. Lianhong Chen and Chao Xu analyzed the experimental data. Junbo Deyong Chen and Jian Chen contributed in writing the paper. Wang, Deyong Chen and Jian Chen contributed in writing the paper. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest.
References References 1. 1. 2. 2. 3. 3. 4. 4.
5.
Havskov, J.; Alguacil, G. Instrumentation in Earthquake Seismology; Springer: Dordrecht, The Netherlands, Havskov, J.; Alguacil, G. Instrumentation in Earthquake Seismology; Springer: Dordrecht, The Netherlands, 2010; pp. 1–5. 2010; pp. 1–5. Niu, F.; Silver, P.G.; Daley, T.M.; Cheng, X.; Majer, E.L. Preseismic velocity changes observed from active Niu, F.;monitoring Silver, P.G.;atDaley, T.M.; Cheng, X.;drill Majer, Preseismic velocity changes observed from active source the Parkfield SAFOD site.E.L. Nature 2008, 454, 204–208. [CrossRef] [PubMed] source monitoring at the Parkfield SAFODavalanche drill site. Nature 2008, 454, 204–208. Herwijnen, A.V.; Schweizer, J. Monitoring activity using a seismic sensor. Cold Reg. Sci. Technol. Herwijnen, A.V.; Schweizer, 2011, 69, 165–176. [CrossRef] J. Monitoring avalanche activity using a seismic sensor. Cold Reg. Sci. Technol. 2011, 69, 165–176. Zeng, R.; Lin, J.; Zhao, Y.J. Development situation of geophones and its application in seismic array Zeng, R.; Lin, J.; Zhao, Y.J.2014, Development situation of geophones and its application in seismic array observation. Prog. Geophys. 29, 2106–2112. observation. Prog. Geophys. 2014, 29, 2106–2112. Liu, Y.J. The Experiments and Studies on Broadband Electric Feedback Seismometer Based on KS-1 Pendulum (In Chinese); Institute of Earthquake Science, China Earthquake Administration: Beijing, China, 2009; pp. 1–2.
Sensors 2017, 17, 2103
5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.
19. 20. 21. 22. 23. 24.
25.
26. 27.
8 of 8
Liu, Y.J. The Experiments and Studies on Broadband Electric Feedback Seismometer Based on KS-1 Pendulum (In Chinese); Institute of Earthquake Science, China Earthquake Administration: Beijing, China, 2009; pp. 1–2. Zhang, G.; Hu, S. Dynamic characteristics of moving-coil geophone with large damping. Int. J. Appl. Electromagn. 2010, 33, 565–571. Bakhoum, E.G.; Cheng, M.H.M. Frequency-selective seismic sensor. IEEE Trans. Instrum. Meas. 2012, 61, 823–829. [CrossRef] Kamata, M. Front end fidelity for seismic acquisition. In Proceedings of the 10th SEGJ International Symposium, Kyoto, Japan, 20–23 November 2011; pp. 1–4. Liang, T.C.; Lin, Y.L. Fiber optic sensor for detection of ground vibrations. Proc. SPIE 2012, 8351, 1–7. Jaroszewicz, L.R.; Krajewski, Z.; Teisseyre, K.P. Usefulness of AFORS—Autonomous fibre-optic rotational seismograph for investigation of rotational phenomena. J. Seismol. 2011, 16, 573–586. [CrossRef] Chistyakov, V.A. Portable seismic sensor. Seism. Instrum. 2011, 47, 8–14. [CrossRef] Bertolini, A.; DeSalvo, R.; Fidecaro, F.; Takamori, A. Monolithic folded pendulum accelerometers for seismic monitoring and active isolation systems. IEEE Geosci. 2006, 44, 273–276. [CrossRef] Milligan, D.J.; Homeijer, B.; Walmsley, R. An ultra-low noise MEMS accelerometer for seismic imaging. In Proceedings of the IEEE Sensors Conference 2011, Limerick, Ireland, 28–31 October 2011; pp. 1281–1284. Hons, M.; Stewart, R.; Lawton, D.; Bertram, M.; Hauer, G. Field data comparisons of MEMS accelerometers and analog geophones. Lead. Edge 2008, 27, 896–903. [CrossRef] Mougenot, D.; Thorburn, N. MEMS-based 3C accelerometers for land seismic acquisition: Is it time? Lead. Edge 2004, 23, 246–250. [CrossRef] R-Sensors LLC (Russia). Molecular-Electronic Broadband Seismometers: CME-60XX. Available online: http://r-sensors.ru/1_products/Descriptions/CME-6011.pdf (accessed on 26 November 2014). Deng, T.; Chen, D.; Wang, J.; Chen, J.; He, W. A MEMS based electrochemical vibration sensor for seismic motion monitoring. J. Microelectromech. Syst. 2014, 23, 92–99. [CrossRef] Levchenko, D.G.; Kuzin, I.P.; Safonov, M.V.; Sychikov, V.N.; Ulomov, I.V.; Kholopov, B.V. Experience in seismic signal recording using broadband electrochemical seismic sensors. Seism. Instrum. 2010, 46, 250–264. [CrossRef] Hurd, R.M.; Lane, R.N. Principlis of very low power electrochemical control devices. J. Electrochem. Soc. 1957, 104, 727–730. [CrossRef] Wittenborn, A.F. Analysis of a logarithmic solion pressure detector. J. Acoust. Soc. Am. 1958, 30, 683. [CrossRef] Collins, J.L.; Richie, W.C.; English, G.E. Solion infrasonic microphone. J. Acoust. Soc. Am. 1964, 36, 1283–1287. [CrossRef] Larcam, C.W. Theoretical analysis of the solion polarized cathode acoustic linear transducer. J. Acoust. Soc. Am. 1965, 37, 664–678. [CrossRef] Abramovich, I.A.; Kharlamov, A.V. Electrochemical Transducers and a Method for Fabricating the Same. U.S. Patent 6576103B2, 10 June 2003. Huang, H.; Carande, B.; Tang, R.; Oiler, J.; Zaitsev, D.; Agafonov, V.; Yu, H. A micro seismometer based on molecular electronic transducer technology for planetary exploration. Appl. Phys. Lett. 2013, 102, 629–632. [CrossRef] He, W.T.; Chen, D.Y.; Wang, J.B.; Chen, J.; Deng, T. Extending upper cutoff frequency of electrochemical seismometer by using extremely thin Su8 insulating spacers. In Proceedings of the IEEE Sensors 2013, Baltimore, MD, USA, 3–6 November 2013; pp. 1–4. He, W.T.; Chen, D.Y.; Li, G.B.; Wang, J.B. Low frequency electrochemical accelerometer with low noise based on MEMS Key. Eng. Mater. 2012, 503, 75–80. Sun, Z.Y.; Chen, D.Y.; Wang, J.B.; Chen, J. A MEMS Based Electrochemical Seismometer with a Novel Integrated Sensing Unit. In Proceedings of the MEMS 2016 Conference, Shanghai, China, 24–28 January 2016; pp. 247–250. © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
An Electrochemical, Low-Frequency Seismic Micro-Sensor Based on MEMS with a Force-Balanced Feedback System Guanglei Li, Junbo Wang *, Deyong Chen, Jian Chen
ID
, Lianhong Chen and Chao Xu
Institute of Electronics, Chinese Academy of Sciences, Beijing 100190, China; [email protected] (G.L.); [email protected] (D.C.); [email protected] (J.C.); [email protected] (L.C.); [email protected] (C.X.) * Correspondence: [email protected]; Tel.: +86-135-2008-9501 Received: 14 August 2017; Accepted: 12 September 2017; Published: 13 September 2017
Abstract: Electrochemical seismic sensors are key components in monitoring ground vibration, which are featured with high performances in the low-frequency domain. However, conventional electrochemical seismic sensors suffer from low repeatability due to limitations in fabrication and limited bandwidth. This paper presents a micro-fabricated electrochemical seismic sensor with a force-balanced negative feedback system, mainly composed of a sensing unit including porous sensing micro electrodes immersed in an electrolyte solution and a feedback unit including a feedback circuit and a feedback magnet. In this study, devices were designed, fabricated, and characterized, producing comparable performances among individual devices. In addition, bandwidths and total harmonic distortions of the proposed devices with and without a negative feedback system were quantified and compared as 0.005–20 (feedback) Hz vs. 0.3–7 Hz (without feedback), 4.34 ± 0.38% (without feedback) vs. 1.81 ± 0.31% (feedback)@1 Hz@1 mm/s and 3.21 ± 0.25% (without feedback) vs. 1.13 ± 0.19% (feedback)@5 Hz@1 mm/s (ndevice = 6, n represents the number of the tested devices), respectively. In addition, the performances of the proposed MEMS electrochemical seismometers with feedback were compared to a commercial electrochemical seismic sensor (CME 6011), producing higher bandwidth (0.005–20 Hz vs. 0.016–30 Hz) and lower self-noise levels (−165.1 ± 6.1 dB vs. −137.7 dB at 0.1 Hz, −151.9 ± 7.5 dB vs. −117.8 dB at 0.02 Hz (ndevice = 6)) in the low-frequency domain. Thus, the proposed device may function as an enabling electrochemical seismometer in the fields requesting seismic monitoring at the ultra-low frequency domain. Keywords: electrochemical seismic sensor; microfabrication; negative feedback; repeatability
1. Introduction A seismic sensor functions as a velocity sensor or an accelerometer that senses the ground vibration of the earth, which is widely used in the field of earthquake monitoring, resource exploration, and ocean bottom observation [1–3]. The studies of long-period surface waves and slow seismic processes require better seismic data in the low-frequency domain [4,5]. Seismic sensors include moving-coil seismic sensors [6–8], fiber-optic seismic sensors [9,10], pendulum seismometers [11,12], and micro-electromechanical system (MEMS) accelerometers [13–15]. Conventionally, seismologists have detected low-frequency seismic vibrations by increasing the quality of the inertial mass or decreasing the equivalent spring coefficient, leading to the decrease of the natural frequency, which, however, renders the conventional seismic sensors bulky and vulnerable. Compared to other seismometers, the electrochemical seismic sensors, which is only used for the detection of low seismic signals, are featured with high performance in the low-frequency domain, enabling the measurements of long-period surface waves and slow seismic processes [16–18]. Sensors 2017, 17, 2103; doi:10.3390/s17092103
www.mdpi.com/journal/sensors
Sensors 2017, 17, 2103
2 of 8
The first prototype of the electrochemical seismic sensor was based on “solion,” which was initially developed in the 1950s by the US Navy, targeting the detection of low-frequency acoustic waves [19–22]. Significant improvements of “solion” were realized in Russia where the electrochemical seismic sensors, which were called molecular electronic transducers in the 1990s, were proposed based on four Pt meshes as electrodes insulated by three porous dielectric ceramic spacers [23,24]. However, the conventional mesh weaving approach and ceramic sintering technologies cannot fabricate electrodes with high repeatability leading to issues of low yield and high cost. To resolve these issues, the electrochemical seismic sensors based on MEMS technologies were introduced by He et al. [25–27], which improved the consistency and decreased the cost significantly [27]. However, the proposed electrochemical seismic micro-sensors still suffered from limited working bandwidth; thus, their applications in ocean bottom seismometers were compromised. In this paper, a force-balanced negative feedback system was proposed to extend the working bandwidth of the MEMS-based electrochemical seismometers, which, at the same time, retains the characteristics of high repeatability. In addition, the microfabrication of the sensing electrodes was improved to enhance production yield. The structure of this paper is organized as follows. In Section 2, the structure and the working principle of the MEMS electrochemical seismic sensor with a negative feedback system are described. Section 3 introduces the fabrication and assembling of the MEMS electrochemical seismic sensor. Section 4 provides the characterization of the developed devices. Section 5 concludes the paper. 2. Structure and Working Principle Figure 1 illustrates the schematic of the MEMS electrochemical seismic sensor with force-balanced negative feedback. The electrochemical seismic sensor consists of a sensing unit and a feedback unit. The sensing unit includes porous sensing electrodes immersed in an electrolyte solution as the liquid mass, which is then sealed by two elastic membranes within a plexiglass house. The porous sensing electrodes fabricated by MEMS technologies are arranged in the anode-cathode-cathode-anode setup to improve the linearity of the output signals. The electrolyte, which is a mixture of iodine (I2 ) and potassium iodide (KI), flows through the via-holes in the sensing electrodes. The following reversible electrochemical reactions occur on the surfaces of the anodes and the cathodes, respectively, when a DC bias (0.1–0.3 V) is applied on the electrodes, generating ion concentration gradients between each pair of the anode and the cathode. anode : 3I − − 2e− → I3−
(1)
cathode : I3− + 2e− → 3I −
(2)
where I3− in (1) and (2) is a clathrate combining I2 with I − . In response to external seismic signals, the electrolyte solution moves opposite to the direction of the external vibration due to the inertial force, which causes the variation of the ion concentration gradients between the anode–cathode pairs. These concentration gradients lead to current outputs, which are proportional to the external seismic signals [25]. The feedback unit composes of a feedback circuit (e.g., a pre-amplification, a pre-filter, and a PID adjuster), a feedback magnet and a feedback coil (see Figure 1). The raw output signals generated from the sensing unit are processed by the feedback circuit and then applied to the feedback coil, positioned within the magnetic field of the feedback magnet. The alternating currents in the feedback coil interact with the magnet field to generate feedback forces, which provides an opposite force to counterbalance the movement of the liquid mass. The transfer function of the closed system is presented in the following form: W1 W= × WBPF (3) 1 + W1 × WPID
Sensors Sensors 2017, 2017, 17, 17, 2103 2103
33 of of 88
where = × × (WS, WAC, and WPF represent the transfer functions of the sensor, the where W1 = Ws × WAC × WPF (WS , WAC , and WPF represent the transfer functions of the sensor, amplifying circuit, and the pre-filter, respectively), and and are the transfer functions of the amplifying circuit, and the pre-filter, respectively), and WPID and WBPF are the transfer functions the PID adjuster and the band-pass filter, respectively. of the PID adjuster and the band-pass filter, respectively.
Figure Figure 1. 1. The schematic of the MEMS electrochemical electrochemical seismic seismic sensor sensor which which is is composed composed of of aa sensing sensing unit including porous sensing electrodes immersed in an electrolyte solution sealed in a unit including porous sensing electrodes immersed in an electrolyte solution sealed in a plexiglass plexiglass house unit including a feedback circuit, a feedback magnet, and house by byelastic elasticmembranes membranesand anda feedback a feedback unit including a feedback circuit, a feedback magnet, aand feedback coil. In response to external seismicseismic signals,signals, the electrolyte moves opposite to the direction a feedback coil. In response to external the electrolyte moves opposite to the of the external to ion concentration gradients and raw current between the direction of thevibration, external leading vibration, leading to ion concentration gradients and outputs raw current outputs anode–cathode pairs. The raw output signals generated from the sensing unit were processed by the between the anode–cathode pairs. The raw output signals generated from the sensing unit were feedback and then applied the feedback coil, to generating feedback due tofeedback its interactions processedcircuit by the feedback circuittoand then applied the feedback coil, forces generating forces with theitsfeedback magnet, the movement of the liquid mass. due to interactions withcounterbalancing the feedback magnet, counterbalancing the movement of the liquid mass.
deep negative negativefeedback feedback(W ( 1× × WPID ≫ In aa deep 1) 1) system, system, Equation Equation (3) (3) can be transformed into the following form: form: following W W = BPF . (4) =WPID . (4) As shown in Equation (4), the output signals of the seismic sensor are shaped by WPID . As shown in Equation (4), the output signals of the seismic sensor are shaped by . The The proportional coefficient and the compensation capacitor, two major parameters in the feedback proportional coefficient and the compensation capacitor, two major parameters in the feedback system, can affect the output characterization. system, can affect the output characterization. The pre-filter and the PID adjuster (proportion integration differentiation) in the feedback circuit The pre-filter and the PID adjuster (proportion integration differentiation) in the feedback provides compensations to enhance both the amplitude and the phase margins by adjusting the values circuit provides compensations to enhance both the amplitude and the phase margins by adjusting of the compensation capacitors. The increases of both amplitude and phase margins enable the system the values of the compensation capacitors. The increases of both amplitude and phase margins to function in a manner of deep feedback. Additionally, since the feedback force counterbalances the enable the system to function in a manner of deep feedback. Additionally, since the feedback force movement of the inertial mass in response to environmental vibrations, the dynamic range of the counterbalances the movement of the inertial mass in response to environmental vibrations, the developed electrochemical seismometer is effectively enlarged. dynamic range of the developed electrochemical seismometer is effectively enlarged. 3. Fabrication and Assembling 3. Fabrication and Assembling The sensing electrodes were fabricated by conventional MEMS technologies including lithography, sensing electrodes were fabricated by conventional MEMS where technologies including deep The reactive ion etching, thermal oxidation, sputtering, and wire bonding, detailed processes lithography, deep reactive ion etching, thermal oxidation, sputtering, and wire bonding, can be found in previous publications [26]. Then, the integrated electrodes (see Figure 2A)where were detailed processes can be found in previous publications [26]. Then, the integrated electrodes (see positioned in a plexiglass tube, which was filled by an electrolyte solution and sealed by two elastic Figure 2A) were positioned in a plexiglass tube, which was filled by an electrolyte solution and membranes. The molar ratio of I2 to KI was between 1:50 and 1:100, which ensures that I2 molecules −I2 to KI was between 1:50 and 1:100, which sealed by two elastic membranes. The molar ratio of were fully dissolved in the KI solution in the forms of I3 . ensures that2B I2 introduces molecules the were fully dissolved the KI solution in the sensor forms of Figure assembled MEMSin electrochemical seismic with a. feedback system. Figure 2B introduces the assembled MEMS electrochemical seismic sensor withthea feedback feedback The feedback coil fabricated with 3D printing was immobilized on the frame, while system. The feedback coil fabricated with 3D printing was immobilized on the frame, while the magnet was immobilized on the plexiglass housing. Thus, the feedback magnet moves alongside with feedback the plexiglass Thus, circuit the feedback magnet moves the seismicmagnet sensor, was whileimmobilized the feedbackon coil remains fixed.housing. The feedback including a differential alongside with the seismic sensor, while the feedback coil remains fixed. The feedback circuit amplification unit, a pre-filter unit, and a PID adjuster was immobilized above the seismic sensors. including a differential amplification unit, a pre-filter unit, and a PID adjuster was immobilized above the seismic sensors.
Sensors 2017, 17, 2103 Sensors 2017, 17, 2103 Sensors 2017, 17, 2103
4 of 8 4 of 8 4 of 8
Figure 2. The sensing unit (A) and and the prototype prototype (B) of of the MEMS MEMS electrochemical seismic seismic sensors Figure Figure 2. 2. The The sensing sensing unit unit (A) (A) and the the prototype (B) (B) of the the MEMS electrochemical electrochemical seismic sensors sensors with force-balanced negative feedback. The inertial mass composed of the frame, the electrolyte, the with force-balanced negative feedback. The inertial mass composed of the frame, the electrolyte, with force-balanced negative feedback. The inertial mass composed of the frame, the electrolyte, the membranes, andand the feedback coil. The feedback coil and the membranes, feedback coil. The feedback coil andthe thefeedback feedbackmagnet magnetwere were not not connected connected membranes, and thethe feedback coil. The feedback coil and the feedback magnet were not connected together, which may may have have relative relative motion motion between between the the two two components. components. together, which together, which may have relative motion between the two components.
4. 4. Experimental Experimental Characterization Characterization The characterizations MEMS electrochemical seismic sensors were on The characterizations thethe MEMS electrochemical seismic sensors were conducted on a modified characterizationsofof of the MEMS electrochemical seismic sensors were conducted conducted on aa modified home-developed platform (see Figure 3), which consists of a function generator, a power home-developed platform (see Figure 3), Figure which 3), consists a function a power amplifier, modified home-developed platform (see whichofconsists of agenerator, function generator, a power amplifier, a laser rangefinder, a vibration exciter, and a data acquisition system. The aamplifier, laser rangefinder, a vibration exciter, and a data acquisition system. The excited alternative waves a laser rangefinder, a vibration exciter, and a data acquisition system. The excited excited alternative waves were the generator amplified by power amplifier, were generated from thegenerated function from generator and amplified by and the power amplifier, then alternative waves were generated from the function function generator and amplified by the the which power were amplifier, which were then applied to the vibration exciter to generate vibration signals at a certain frequency. applied to thethen vibration exciter tovibration generate vibration at avibration certain frequency. laser rangefinder which were applied to the exciter tosignals generate signals atThe a certain frequency. The laser rangefinder used the of signals, while the was measure thewas velocities ofmeasure the vibration signals, while thevibration data acquisition NI, The used laserto rangefinder was used to to measure the velocities velocities of the the vibration signals,system while(4472, the data data acquisition system (4472, NI, Austin, TX, USA) was used to measure the output signals. Austin, TX, system USA) was used toAustin, measure theUSA) output signals. acquisition (4472, NI, TX, was used to measure the output signals.
Figure 3. The schematic of the experimental platform where the developed seismic sensor was Figure 3. 3. The experimental platform the developed developed seismic seismic sensor sensor was was Figure The schematic schematic of of the the experimental platform where where the positioned on a vibration exciter, which is excited electrically (function generation and power positioned on a vibration exciter, which is excited electrically (function generation and power positioned on a vibration exciter, which is excited electrically (function generation and power amplifier) amplifier) and detected optically (laser rangefinder). amplifier) andoptically detected(laser optically (laser rangefinder). and detected rangefinder).
Figure Figure 4A 4A shows shows the the frequency frequency characterization characterization results results of of the the MEMS MEMS electrochemical electrochemical seismic seismic Figure 4A shows the frequency characterization results of the MEMS electrochemical seismic sensor sensor without without feedback feedback and and with with feedback feedback plus plus variations variations in in aa proportional proportional coefficient. coefficient. It It was was sensor without feedback and with feedback plus variations in lead a proportional coefficient. Itthe was shown shown that the increase of the proportional coefficient can to (1) an increase in working shown that the increase of the proportional coefficient can lead to (1) an increase in the working that the increase of the proportional coefficient can lead to 4B (1) an increase in the working bandwidth bandwidth bandwidth and and (2) (2) the the possibilities possibilities of of resonance. resonance. Figure Figure 4B shows shows the the frequency frequency characterization characterization and (2) of the possibilities of resonance. Figure 4B shows the frequency characterization results of MEMS results results of MEMS MEMS electrochemical electrochemical seismic seismic sensor sensor with with different different values values of of compensation compensation capacitors. capacitors. It It electrochemical seismic sensor with different values of compensation capacitors. It wasonobserved that was observed that the value of the compensation capacitor had no influence was observed that the value of the compensation capacitor had no influence on the the output output the value ofat the compensation capacitor hadwhich no influence on the output amplitude at the low-frequency amplitude amplitude at the the low-frequency low-frequency domain, domain, which can can decrease decrease the the possibilities possibilities of of resonance resonance due due to to domain, which can decrease the possibilities of resonance due toworking the phasebandwidth compensation effects of the the phase compensation effects of the capacitor. The required can be obtained the phase compensation effects of the capacitor. The required working bandwidth can be obtained capacitor. The required working bandwidth can be obtained by adjusting the appropriate values of the by by adjusting adjusting the the appropriate appropriate values values of of the the proportional proportional coefficient coefficient and and the the compensation compensation capacitor. capacitor. proportional coefficient and the compensation capacitor.
Sensors 2017, 17, 2103 Sensors Sensors 2017, 2017, 17, 17, 2103 2103
5 of 8 55 of of 88
Frequency characterization results of Figure Figure 4. 4. Frequency Frequency characterization characterization results of the the MEMS MEMS electrochemical electrochemical seismic seismic sensors sensors without without feedback, with with feedback feedback plus plus variations variations in in proportional proportional coefficients coefficients (A) (A) and and feedback feedback, and with with feedback feedback plus plus (B). variations in compensation capacitors variations in compensation capacitors (B).
Figure Figure 555 shows shows input input voltages voltages measured measured by by the the laser laser rangefinder rangefinder and and the the output output voltages voltages of of Figure shows input voltages measured by the laser rangefinder and the output voltages of CME 6011 and the proposed device @20 Hz (A)@0.18 mm/s, @1 [email protected] mm/s (B), and @0.016 CME 6011 and the proposed device @20 Hz (A)@0.18 mm/s, @1 [email protected] mm/s (B), and @0.016 CME 6011 and the proposed device @20 Hz (A)@0.18 mm/s, @1 [email protected] mm/s (B), and @0.016 [email protected] [email protected] mm/s mm/s (C). (C). ItItwas was observed that (1) the output voltages of the proposed device device were were smaller smaller [email protected] mm/s (C).It wasobserved observedthat that(1) (1)the theoutput outputvoltages voltages of of the the proposed proposed device were smaller than that of CME 6011 at 20 Hz due to the working bandwidth limitation of the proposed device, (2) than that of CME 6011 at 20 Hz due to the working bandwidth limitation of the proposed device, (2) than that of CME 6011 at 20 Hz due to the working bandwidth limitation of the proposed device, the output voltage of the proposed device was almost the same as that of CME 6011 at 1 Hz, and (3) the output voltage of the proposed device was almost the same as that of CME 6011 at 1 Hz, and (3) (2) the output voltage of the proposed device was almost the same as that of CME 6011 at 1 Hz, the voltage of device was than CME 6011 at Hz to the output output voltagevoltage of the the proposed proposed devicedevice was larger larger than that that of CME 60116011 at 0.016 0.016 Hz due due to the the and (3) the output of the proposed was larger thanof that of CME at 0.016 Hz due to working bandwidth limitation of CME 6011. working bandwidth limitation of CME 6011. the working bandwidth limitation of CME 6011.
Figure 5. Input voltages measured by the laser rangefinder Figure Figure 5. 5. Input Input voltages voltages measured measured by by the the laser laser rangefinder rangefinder and and the the output output voltages voltages of of CME CME 6011 6011 and and the proposed device @20 Hz (A) @0.18 mm/s, @1 [email protected] mm/s (B), and @0.016 [email protected] mm/s (C). the the proposed proposed device device @20 @20 Hz Hz (A) (A) @0.18 @0.18 mm/s, mm/s, @1 @1 [email protected] [email protected] mm/s mm/s (B), (B), and and @0.016 @0.016 [email protected] [email protected] mm/s mm/s (C). (C).
Figure shows the the frequency frequency characteristics characteristics of the developed developed micro micro seismic seismic sensors sensors with with and and Figure 66 shows shows the frequency characteristics of the the developed micro seismic sensors with and without the feedback system and CME 6011. More specifically, the electrochemical seismic micro without the feedback system and CME 6011. More specifically, the electrochemical seismic micro without the feedback system and CME 6011. the electrochemical seismic micro sensors sensors without without feedback feedback produced produced aaa limited limited bandwidth bandwidth at at 0.3–7 0.3–7 Hz. Hz. By sensors without feedback produced limited bandwidth at 0.3–7 Hz. By introducing introducing and and optimizing optimizing the feedback unit, the working bandwidth was extended to 0.005–20 Hz, which was broader the feedback unit, the working bandwidth was extended to 0.005–20 Hz, which was broader than the feedback unit, the working bandwidth was extended to 0.005–20 Hz, which was broader thanthan that that of CME 6011 with a bandwidth of 0.016–30 Hz. that of CME a bandwidth of 0.016–30 of CME 6011 6011 with with a bandwidth of 0.016–30 Hz. Hz. Figure developed micro seismic sensors with and without the shows the sensitivities ofofthe the developed micro seismic sensors with andand without the Figure 77 shows showsthe thesensitivities sensitivitiesof the developed micro seismic sensors with without feedbacks, which were quantified as 2026 ± 12 V/m/s vs. 5856 ± 73 V/m/s at 1 Hz and 1994 ± 8 V/m/s feedbacks, which were quantified as 2026 ± 12 V/m/s vs. 5856 ± 73 V/m/s at 1 Hz and 1994 ± 8 V/m/s the feedbacks, which were quantified as 2026 ± 12 V/m/s vs. 5856 ± 73 V/m/s at 1 Hz and vs. 104 at Hz (see Figure In to micro seismic sensors vs. 5603 5603 104 V/m/s V/m/s at 55 ± Hz (see Figure 7A,B). In comparison comparison toIn micro seismic to sensors without 1994 ± 8 ±±V/m/s vs. 5603 104 V/m/s at7A,B). 5 Hz (see Figure 7A,B). comparison microwithout seismic feedback, the seismic sensors feedback lower sensitivities due to feedback, the micro micro seismic sensors with feedback produced lowerproduced sensitivities duesensitivities to the the feedback feedback sensors without feedback, the micro with seismic sensorsproduced with feedback lower due forces. However, micro seismic sensors with feedback produced much lower total harmonic forces. However, micro seismic sensors with feedback produced much lower total harmonic to the feedback forces. However, micro seismic sensors with feedback produced much lower total distortions compared to without feedbacks (see 7C,D). For the distortionsdistortions comparedcompared to the the micro micro sensors withoutwithout feedbacks (see Figure Figure 7C,D). For instance, instance, the harmonic to thesensors micro sensors feedbacks (see Figure 7C,D). For instance, total harmonic distortions (which was calculated by the ratio of the total mean square root of the total harmonic distortions (which was calculated by the ratio of the total mean square root of the the total harmonic distortions (which was calculated by the ratio of the total mean square root harmonic harmonic wave wave and and the the fundamental fundamental frequency frequency amplitude.) amplitude.) decreased decreased from from 4.34 4.34 ±± 0.38% 0.38% (without (without feedback) feedback) to to 1.81 1.81 ±± 0.31% 0.31% (feedback)@1 (feedback)@1 Hz@1 Hz@1 mm/s mm/s and and 3.21 3.21 ±± 0.25% 0.25% (without (without feedback) feedback) to to
Sensors 2017, 17, 2103
6 of 8
Sensors 2017, 17, 2103 Sensors 2017, 17, 2103
6 of 8 6 of 8
of the harmonic wave and the fundamental frequency amplitude.) decreased from 4.34 ± 0.38% 1.13 ± 0.19% (feedback)@5 Hz@1 mm/s. An increase in the amplitude of 0.25% the output signal lead to to an (without feedback) to 1.81 ± 0.31% (feedback)@1 Hz@1 and 3.21 of ± (without feedback) 1.13 ± 0.19% (feedback)@5 Hz@1 mm/s. An increase in mm/s the amplitude the output signal lead to an increase in feedback force, whichmm/s. confined the movement of the liquid mass within the linear range, 1.13 ± 0.19% (feedback)@5 Anthe increase in theof amplitude the output leadrange, to an increase in feedback force, Hz@1 which confined movement the liquidofmass within signal the linear and decreased total harmonic distortions. increase in feedback force, which confined the movement of the liquid mass within the linear range, and decreased total harmonic distortions. and decreased total harmonic distortions.
Figure 6.Frequency Frequency characterization results results ofthe the MEMSelectrochemical electrochemical seismicsensor sensor withand and Figure Figure 6. 6. Frequency characterization characterization results of of the MEMS MEMS electrochemical seismic seismic sensor with with and device = 6) in comparison to CME 6011. The amplitude is the output voltage of the without feedback (n without = 6) in comparison to CME 6011. The amplitude is the output voltage of the device = 6) in comparison to CME 6011. The amplitude is the output voltage of the without feedback feedback (n (ndevice electrochemicalseismic seismicsensors sensorsatatthe the sameinput input velocityand and differentfrequencies. frequencies. electrochemical electrochemical seismic sensors at the same same input velocity velocity and different different frequencies.
Figure 7. (A,B) The sensitivity of the MEMS electrochemical seismic sensors with and without feedbacks Figure 7. (A,B) The sensitivity of the MEMS electrochemical seismic sensors with and without Figure 7. (A,B) The velocities sensitivityat of theand MEMS seismic and without as a function of input 1 Hz 5 Hz,electrochemical respectively. (C,D) Total sensors harmonicwith distortions of the feedbacks as a function of input velocities at 1 Hz and 5 Hz, respectively. (C,D) Total harmonic feedbacks as a function of input velocities at 1without Hz andfeedbacks 5 Hz, respectively. (C,D) Total harmonic MEMS electrochemical seismic sensors with and as a function of input velocities at distortions of the MEMS electrochemical seismic sensors with and without feedbacks as a function of 1distortions Hz and 5 Hz, respectively. of the MEMS electrochemical seismic sensors with and without feedbacks as a function of input velocities at 1 Hz and 5 Hz, respectively. input velocities at 1 Hz and 5 Hz, respectively.
The thethe self-noise levelslevels were were also conducted in this platform Thetests testsof of self-noise also conducted in this without platforminducing withoutvibrations inducing TheAs tests of in theFigure self-noise levels were alsospectrums conducted in developed this platform without inducing actively. shown 8, the self-noise power of the MEMS electrochemical vibrations actively. As shown in Figure 8, the self-noise power spectrums of the developed MEMS vibrations actively. Aslower shownthan in Figure 8, CME the self-noise power spectrums of the of developed seismic sensors were thatlower of domain less thanMEMS 1 Hz. electrochemical seismic sensors were than6011 that in of the CMEfrequency 6011 in the frequency domain of less electrochemical seismic sensors were lower than that of CME 6011 in the frequency domain of less More the self-noisethe levels were characterized as −165.1 ± 6.1 dB (micro sensors) vs. than 1specifically, Hz. More specifically, self-noise levels were characterized as −165.1 ± 6.1seismic dB (micro seismic than 1 Hz. More specifically, the − self-noise levels were characterized as −165.1 ± 6.1 dB (micro seismic − 137.7 dB (CME 6011) at 0.1 Hz, 151.9 ± 7.5 dB (micro seismic sensors) vs. − 117.8 dB (CME 6011) at sensors) vs. −137.7 dB (CME 6011) at 0.1 Hz, −151.9 ± 7.5 dB (micro seismic sensors) vs. −117.8 dB sensors) vs. −137.7 dB (CME 6011) at 0.1 Hz, −151.9 ± 7.5 dB (micro seismic sensors) vs. −117.8 dB (CME 6011) at 0.02 Hz (ndevice = 6). It was speculated that at a frequency higher than 1 Hz, the noise of (CME 6011) at 0.02 Hz (ndevice = 6). It was speculated that at a frequency higher than 1 Hz, the noise of the seismic sensor mainly comes from mechanical and environmental noises due to packaging. Thus, the seismic sensor mainly comes from mechanical and environmental noises due to packaging. Thus,
Sensors 2017, 17, 2103
7 of 8
0.02 Hz (ndevice = 6). It was speculated that at a frequency higher than 1 Hz, the noise of the seismic Sensors 2017, 17, 2103 7 of 8 sensor mainly comes from mechanical and environmental noises due to packaging. Thus, the increases in self-noise of the proposed may result frommay imperfect packaging technologies in thethe increases in levels the self-noise levels ofdevices the proposed devices result from imperfect packaging comparison sensors.seismic sensors. technologiesto incommercial comparisonseismic to commercial
Figure 8. The self-noise power spectrums of the developed MEMS electrochemical seismic sensors Figure 8. The self-noise power spectrums of the developed MEMS electrochemical seismic sensors with feedback feedback (n (ndevice ==6)6)and with andthe theconventional conventionalCME CME6011 6011where wherethe thedeveloped developed seismic seismic micro micro sensors sensors device device produced lower self-noise levels than CME 6011 in the frequency domain of 0.01–1 Hz. device produced lower self-noise levels than CME 6011 in the frequency domain of 0.01–1 Hz.
5. Conclusions 5. Conclusions A MEMS-based electrochemical seismic sensor with ultra-broad frequency working bandwidth A MEMS-based electrochemical seismic sensor with ultra-broad frequency working bandwidth was demonstrated in this paper. The working bandwidth and the self-noise level of the developed was demonstrated in this paper. The working bandwidth and the self-noise level of the developed device were characterized with feedback parameters optimized. Experimental results showed that device were characterized with feedback parameters optimized. Experimental results showed that the working bandwidth of the developed seismic sensor was extended to 0.005–20 Hz, validating the the working bandwidth of the developed seismic sensor was extended to 0.005–20 Hz, validating the positive effects of the negative feedback. In addition, the developed device produced a lower positive effects of the negative feedback. In addition, the developed device produced a lower self-noise self-noise level at the low frequency domain in comparison to the commercial seismic sensor of level at the low frequency domain in comparison to the commercial seismic sensor of CME6011, CME6011, validating its large dynamic ranges. Thus, this MEMS electrochemical seismic sensor can validating its large dynamic ranges. Thus, this MEMS electrochemical seismic sensor can be potentially be potentially used in the field requesting seismic monitoring at the ultra-low frequency domain used in the field requesting seismic monitoring at the ultra-low frequency domain with large dynamic with large dynamic ranges and low noise levels. ranges and low noise levels. Acknowledgments: This Science Foundation of China (No. 61431910), the Acknowledgments: This work workwas wassupported supportedby bythe theNational National Science Foundation of China (No. 61431910), National High Technology Research and Development Program ofofChina project the National High Technology Research and Development Program China(the (the863 863 Program) Program) with with project No. 2014AA09158, the National Key Scientific Instrument and Equipment Development Project of China with project 2013YQ12035703, and and the the MEMS MEMS Electrochemical Electrochemical Seismic Seismic Sensors Sensors Research Research in in the project No. No. 2013YQ12035703, the Ocean Ocean Bottom Bottom Observation with the project No. 2016YFC0301801. Observation with the project No. 2016YFC0301801. Author Contributions: Guanglei Li performed the experiments and developed the theoretical description of the Author Contributions: Li performed the experiments andanalyzed developed theoreticaldata. description of the method proposed in theGuanglei publication. Lianhong Chen and Chao Xu thethe experimental Junbo Wang, method proposed in the publication. Lianhong Chen and Chao Xu analyzed the experimental data. Junbo Deyong Chen and Jian Chen contributed in writing the paper. Wang, Deyong Chen and Jian Chen contributed in writing the paper. Conflicts of Interest: The authors declare no conflict of interest. Conflicts of Interest: The authors declare no conflict of interest.
References References 1. 1. 2. 2. 3. 3. 4. 4.
5.
Havskov, J.; Alguacil, G. Instrumentation in Earthquake Seismology; Springer: Dordrecht, The Netherlands, Havskov, J.; Alguacil, G. Instrumentation in Earthquake Seismology; Springer: Dordrecht, The Netherlands, 2010; pp. 1–5. 2010; pp. 1–5. Niu, F.; Silver, P.G.; Daley, T.M.; Cheng, X.; Majer, E.L. Preseismic velocity changes observed from active Niu, F.;monitoring Silver, P.G.;atDaley, T.M.; Cheng, X.;drill Majer, Preseismic velocity changes observed from active source the Parkfield SAFOD site.E.L. Nature 2008, 454, 204–208. [CrossRef] [PubMed] source monitoring at the Parkfield SAFODavalanche drill site. Nature 2008, 454, 204–208. Herwijnen, A.V.; Schweizer, J. Monitoring activity using a seismic sensor. Cold Reg. Sci. Technol. Herwijnen, A.V.; Schweizer, 2011, 69, 165–176. [CrossRef] J. Monitoring avalanche activity using a seismic sensor. Cold Reg. Sci. Technol. 2011, 69, 165–176. Zeng, R.; Lin, J.; Zhao, Y.J. Development situation of geophones and its application in seismic array Zeng, R.; Lin, J.; Zhao, Y.J.2014, Development situation of geophones and its application in seismic array observation. Prog. Geophys. 29, 2106–2112. observation. Prog. Geophys. 2014, 29, 2106–2112. Liu, Y.J. The Experiments and Studies on Broadband Electric Feedback Seismometer Based on KS-1 Pendulum (In Chinese); Institute of Earthquake Science, China Earthquake Administration: Beijing, China, 2009; pp. 1–2.
Sensors 2017, 17, 2103
5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.
19. 20. 21. 22. 23. 24.
25.
26. 27.
8 of 8
Liu, Y.J. The Experiments and Studies on Broadband Electric Feedback Seismometer Based on KS-1 Pendulum (In Chinese); Institute of Earthquake Science, China Earthquake Administration: Beijing, China, 2009; pp. 1–2. Zhang, G.; Hu, S. Dynamic characteristics of moving-coil geophone with large damping. Int. J. Appl. Electromagn. 2010, 33, 565–571. Bakhoum, E.G.; Cheng, M.H.M. Frequency-selective seismic sensor. IEEE Trans. Instrum. Meas. 2012, 61, 823–829. [CrossRef] Kamata, M. Front end fidelity for seismic acquisition. In Proceedings of the 10th SEGJ International Symposium, Kyoto, Japan, 20–23 November 2011; pp. 1–4. Liang, T.C.; Lin, Y.L. Fiber optic sensor for detection of ground vibrations. Proc. SPIE 2012, 8351, 1–7. Jaroszewicz, L.R.; Krajewski, Z.; Teisseyre, K.P. Usefulness of AFORS—Autonomous fibre-optic rotational seismograph for investigation of rotational phenomena. J. Seismol. 2011, 16, 573–586. [CrossRef] Chistyakov, V.A. Portable seismic sensor. Seism. Instrum. 2011, 47, 8–14. [CrossRef] Bertolini, A.; DeSalvo, R.; Fidecaro, F.; Takamori, A. Monolithic folded pendulum accelerometers for seismic monitoring and active isolation systems. IEEE Geosci. 2006, 44, 273–276. [CrossRef] Milligan, D.J.; Homeijer, B.; Walmsley, R. An ultra-low noise MEMS accelerometer for seismic imaging. In Proceedings of the IEEE Sensors Conference 2011, Limerick, Ireland, 28–31 October 2011; pp. 1281–1284. Hons, M.; Stewart, R.; Lawton, D.; Bertram, M.; Hauer, G. Field data comparisons of MEMS accelerometers and analog geophones. Lead. Edge 2008, 27, 896–903. [CrossRef] Mougenot, D.; Thorburn, N. MEMS-based 3C accelerometers for land seismic acquisition: Is it time? Lead. Edge 2004, 23, 246–250. [CrossRef] R-Sensors LLC (Russia). Molecular-Electronic Broadband Seismometers: CME-60XX. Available online: http://r-sensors.ru/1_products/Descriptions/CME-6011.pdf (accessed on 26 November 2014). Deng, T.; Chen, D.; Wang, J.; Chen, J.; He, W. A MEMS based electrochemical vibration sensor for seismic motion monitoring. J. Microelectromech. Syst. 2014, 23, 92–99. [CrossRef] Levchenko, D.G.; Kuzin, I.P.; Safonov, M.V.; Sychikov, V.N.; Ulomov, I.V.; Kholopov, B.V. Experience in seismic signal recording using broadband electrochemical seismic sensors. Seism. Instrum. 2010, 46, 250–264. [CrossRef] Hurd, R.M.; Lane, R.N. Principlis of very low power electrochemical control devices. J. Electrochem. Soc. 1957, 104, 727–730. [CrossRef] Wittenborn, A.F. Analysis of a logarithmic solion pressure detector. J. Acoust. Soc. Am. 1958, 30, 683. [CrossRef] Collins, J.L.; Richie, W.C.; English, G.E. Solion infrasonic microphone. J. Acoust. Soc. Am. 1964, 36, 1283–1287. [CrossRef] Larcam, C.W. Theoretical analysis of the solion polarized cathode acoustic linear transducer. J. Acoust. Soc. Am. 1965, 37, 664–678. [CrossRef] Abramovich, I.A.; Kharlamov, A.V. Electrochemical Transducers and a Method for Fabricating the Same. U.S. Patent 6576103B2, 10 June 2003. Huang, H.; Carande, B.; Tang, R.; Oiler, J.; Zaitsev, D.; Agafonov, V.; Yu, H. A micro seismometer based on molecular electronic transducer technology for planetary exploration. Appl. Phys. Lett. 2013, 102, 629–632. [CrossRef] He, W.T.; Chen, D.Y.; Wang, J.B.; Chen, J.; Deng, T. Extending upper cutoff frequency of electrochemical seismometer by using extremely thin Su8 insulating spacers. In Proceedings of the IEEE Sensors 2013, Baltimore, MD, USA, 3–6 November 2013; pp. 1–4. He, W.T.; Chen, D.Y.; Li, G.B.; Wang, J.B. Low frequency electrochemical accelerometer with low noise based on MEMS Key. Eng. Mater. 2012, 503, 75–80. Sun, Z.Y.; Chen, D.Y.; Wang, J.B.; Chen, J. A MEMS Based Electrochemical Seismometer with a Novel Integrated Sensing Unit. In Proceedings of the MEMS 2016 Conference, Shanghai, China, 24–28 January 2016; pp. 247–250. © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
