sensors Article

Contact Pressure Level Indication Using Stepped Output Tactile Sensors Eunsuk Choi 1 , Onejae Sul 2 , Juyoung Kim 1 , Kyumin Kim 1 , Jong-Seok Kim 1 , Dae-Yong Kwon 1 , Byong-Deok Choi 1 and Seung-Beck Lee 1,2, * 1

2

*

Department of Electronic Engineering, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 133-791, Korea; [email protected] (E.C.); [email protected] (J.K.); [email protected] (K.K.); [email protected] (J.-S.K.); [email protected] (D.-Y.K.); [email protected] (B.-D.C.) Institute of Nano Science and Technology, Hanyang University, 222 Wangsimni-ro, Seongdong-gu, Seoul 133-791, Korea; [email protected] Correspondence: [email protected]; Tel.: +82-2-2220-1676; Fax: +82-2-2294-1676

Academic Editor: Vittorio M. N. Passaro Received: 29 February 2016; Accepted: 4 April 2016; Published: 9 April 2016

Abstract: In this article, we report on a novel diaphragm-type tactile pressure sensor that produces stepwise output currents depending on varying low contact pressures. When contact pressures are applied to the stepped output tactile sensor (SOTS), the sensor’s suspended diaphragm makes contact with the substrate, which completes a circuit by connecting resistive current paths. Then the contact area, and therefore the number of current paths, would determine the stepped output current produced. This mechanism allows SOTS to have high signal-to-noise ratio (>20 dB) in the 3–500 Hz frequency range at contact pressures below 15 kPa. Moreover, since the sensor’s operation does not depend on a material’s pressure-dependent electrical properties, the SOTS is able to demonstrate high reproducibility and reliability. By forming a 4 ˆ 4 array of SOTS with a surface bump structure, we demonstrated shear sensing as well as surface (1 ˆ 1 cm2 ) pressure mapping capabilities. Keywords: tactile sensor; output characteristics

tactile sensor array;

spatially digitized electrode;

stepped

1. Introduction Recently, tactile pressure sensors fabricated over an arbitrary substrate, or artificial skins, are being developed for possible applications ranging from touch sensitive robotic interfaces to biomedical instrumentation [1–4]. As pressure sensors, they are required to have high pressure sensitivity, low pressure threshold [5,6] and wide sensing range [7,8]. At the same time, the sensors are required to have uniformity and reliability with high spatial resolution [9], and they consist a system with multiple arrays of these sensors. In addition to these functional requirements, for biomimetic operation, they aim to mimic the function of the human skin, emulating the human mechanoreceptors having higher sensitivity for change in relative magnitude of pressure than absolute values for higher differential pressure sensitivity at very low pressure ranges [10]. In recent years, many pressure sensors have been reported that utilize various aspects of nanomaterials and microstructures to produce high pressure sensitivity showing output signals with amplitude proportional to the strength of the applied stimulus [11–22]. Although the reported sensors demonstrate high sensitivity, most lack actual applicability since sensors based on nanomaterials, such as carbon nanotubes or graphitic oxides, have issues with uniformity and reproducibility [23,24]. Especially, capacitance type sensors have issues with cross-talk and noise when measuring low pressure levels at high spatial resolution [25,26]. In this article, we demonstrate the operation of a stepped output tactile sensor (SOTS) which produces stepwise changes in current outputs depending on analog tactile pressure inputs. Sensors 2016, 16, 511; doi:10.3390/s16040511

www.mdpi.com/journal/sensors

Sensors 2016, 16, 511

2 of 14

In this article, we demonstrate the operation of a stepped output tactile sensor (SOTS) which 14 produces stepwise changes in current outputs depending on analog tactile pressure inputs.2 of The sensor’s operation, in principle, does not depend on the pressure dependent resistivity of an active sensing material but mostly on the geometry layouton ofthe the pressure contact electrodes. basic concept The sensor’s operation, in principle, does notand depend dependentThe resistivity of an followssensing from amaterial previous where spatially separated contact electrodes were usedThe to detect active butreport mostly on the geometry and layout of the contact electrodes. basic consecutive deflection of a diaphragm depending on the applied pressure. The output from concept follows from a previous report where spatially separated contact electrodes were used toa particular electrodedeflection corresponded to a specificdepending applied pressure, producing a digitized detect consecutive of a diaphragm on the applied pressure. Theoutput outputsignal from One drawback in this sensor wasto that the number of output electrodes had toabe increased if the a[27]. particular electrode corresponded a specific applied pressure, producing digitized output pressure detection range and resolution were to be increased, making the sensor impractical for array signal [27]. One drawback in this sensor was that the number of output electrodes had to be increased type application require too many electrodes formaking integration. Here, we reduced if the sensor pressure detectionwhich rangewould and resolution were to be increased, the sensor impractical thearray output electrode to two perwould sensorrequire while still maintaining the basic operationalHere, concept for type sensor numbers application which too many electrodes for integration. we by utilizing a parallel distribution of resistive current paths. When a suspended diaphragm, with a reduced the output electrode numbers to two per sensor while still maintaining the basic operational conducting layer, deflects withdistribution tactile pressure, it makes contact withWhen the substrate, and completes concept by utilizing a parallel of resistive current paths. a suspended diaphragm,a circuit by connecting a number of resistive current paths. Therefore the resulting output current with a conducting layer, deflects with tactile pressure, it makes contact with the substrate, and reflects the level of applied pressure. Although the sensing mechanism is simple, SOTS produces completes a circuit by connecting a number of resistive current paths. Therefore the resulting outputa highly reliable current with the number and the spacing of the current pathsisdetermining the current reflectsoutput the level of applied pressure. Although the sensing mechanism simple, SOTS detection resolution and pressure range. The sensor demonstrates high reproducibility with high produces a highly reliable output current with the number and the spacing of the current paths signal-to-noise (SNR) of more than 20 dB inrange. 3~500 The Hz frequency range. Also by forming a4×4 determining theratio detection resolution and pressure sensor demonstrates high reproducibility SOTShigh array, it was possible to(SNR) demonstrate ofin spatially distributed over a with signal-to-noise ratio of moredetection than 20 dB 3~500 Hz frequencycontact range. pressures Also by forming 2 surface area. With further developments it will be possible to utilize the SOTS system for robotic 1 cm a 4 ˆ 4 SOTS array, it was possible to demonstrate detection of spatially distributed contact pressures tactile andarea. electric skin prosthesis. over a 1interfaces cm2 surface With further developments it will be possible to utilize the SOTS system for Sensors 2016, 16, 511

robotic tactile interfaces and electric skin prosthesis. 2. Stepped Output Tactile Sensor 2. Stepped Output Tactile Sensor 2.1. Concepts and Operating Mechanism 2.1. Concepts and Operating Mechanism Figure 1a shows a schematic diagram of the device. It consists mainly of two parts, the upper Figure 1a shows a schematic diagram of the device. It consists mainly of two parts, the upper diaphragm layer and the bottom substrate. The upper layer consists of a polyethylene terephthalate diaphragm layer and the bottom substrate. The upper layer consists of a polyethylene terephthalate (PET) film with a polydimethylsiloxane (PDMS) ridge structure on top [27–29]. On the bottom of the (PET) film with a polydimethylsiloxane (PDMS) ridge structure on top [27–29]. On the bottom of the PET film, Pt contact electrodes were fabricated to act as the two contact electrodes. On the substrate, PET film, Pt contact electrodes were fabricated to act as the two contact electrodes. On the substrate, parallel resistors were placed from the center of the spacer defined pit outwards, which will act as parallel resistors were placed from the center of the spacer defined pit outwards, which will act as digitized current paths. digitized current paths.

Figure 1. The schematic illustration showing the operating mechanism of the stepped output tactile Figure 1. The schematic illustration showing the operating mechanism of the stepped output tactile sensor (SOTS): (a) The schematic diagram of SOTS; (b) the 3D cross-sectional image under applied sensor (SOTS): (a) The schematic diagram of SOTS, (b) the 3D cross-sectional image under applied pressure, showing short-circuited multiple resistors. The contact areas and equivalent circuits of SOTS pressure, showing short-circuited multiple resistors. The contact areas and equivalent circuits of SOTS when small (c) or large (d) pressure was applied. when small (c) or large (d) pressure was applied.

When tactile pressure is applied, the upper PET diaphragm deflects and makes contact with the central resistor at a threshold pressure (Figure 1b), completing a circuit with a corresponding

Sensors 2016, 16, 511

3 of 14

Sensors 2016, 16, 511

3 of 14

When tactile pressure is applied, the upper PET diaphragm deflects and makes contact with the central resistor at a threshold pressure (Figure 1b), completing a circuit with a corresponding current output (Figure 1c). Initial below the threshold level will notwill be detected by the sensor, current output (Figure 1c).tactile Initialpressures tactile pressures below the threshold level not be detected by the since the threshold pressure will be pressure level forforthe sensor, since the threshold pressure willthe be minimum the minimum pressure level theintended intendedtactile tactile sensing application. This will contribute to reducing sensor power consumption because no power will be consumed during duringtactile tactile pressures below threshold. If required, the threshold can be consumed pressures below threshold. If required, the threshold pressurepressure can be lowered lowered by reducing spacer thickness or by increasing the diaphragm the by reducing the spacerthe thickness or by increasing the diaphragm diameter.diameter. IncreasingIncreasing the pressure pressureresults further in morecurrent resistor current paths being contacted in sequence, which to a further in results more resistor paths being contacted in sequence, which leads to a leads stepwise stepwisein increase current1d). (Figure 1d). Since each of current corresponds to a designated increase currentin(Figure Since each value ofvalue current plateauplateau corresponds to a designated range range of contact the between distance the between thedetermines resistors determines pressure detection of contact pressure,pressure, the distance resistors the pressurethe detection resolution of resolution thecloser sensor. Theare, closer are, the higher the resolution. the sensor. of The they the they higher the resolution. 2.2. Fabrication Process

Figure 2 shows illustrations illustrations of of the the device devicefabrication fabricationprocess. process.For Forthe theupper upperlayer layerfabrication, fabrication, a amold-transfer mold-transfertechnique techniquewas wasused. used.AA75 75μm-thick µm-thickSU-8 SU-8master masterlayer layer was was patterned patterned to form a negative of the film, that will form thethe diaphragm, waswas placed on the top topridge ridgestructure structure(Figure (Figure2a). 2a).Then Thenthe thePET PET film, that will form diaphragm, placed top of the master and was in laid a Petri after which PDMS wasPDMS poured on poured top (Figure on top of SU-8 the SU-8 master andlaid was in dish, a Petri dish, after which was on 2b). top The SU-8 facing surface of the PET was treated with reactive ion etching (performed at 5 mTorr (Figure 2b). The SU-8 facing surface of the PET was treated with reactive ion etching (performed of at pressure, 30 pressure, sccm of CF 100 W of RF power, 10 min) to form structures. The 5 mTorr of 304 gas sccmflow, of CF 4 gas flow, 100 W for of RF power, for 10nanobrush min) to form nanobrush PET nanobrushes, in the SEM image of SEM Figure 2b, promoted adhesion of PDMS to structures. The PETshown nanobrushes, shown in the image of Figure the 2b, promoted thethe adhesion of PET by increasing the contact surface area between PET film and PDMS layer. The PDMS solution the PDMS to PET by increasing the contact surface area between PET film and PDMS layer. percolated the 30percolated µm gap between filmbetween and the master form thethe topmaster skin oftothe sensor The PDMSinto solution into thethe 30 PET μm gap the PETtofilm and form the (Figure curing(Figure the PDMS, 100 nm thick the Pt electrodes, gap of µm, were with formed using top skin2c). of After the sensor 2c). After curing PDMS, 100with nmathick Pt50 electrodes, a gap of optical lithography and sputter deposition (Figure 2d). Finally, the top layer was peeled off (Figure 2e). 50 μm, were formed using optical lithography and sputter deposition (Figure 2d). Finally, the top The substrate fabrication process begins with the defining process of WOx begins spatially digitized parallel layerbottom was peeled off (Figure 2e). The bottom substrate fabrication with the defining of resistors on the oxidized Si substrate (Figure 2g). The resistors were formed by defining the pattern WOx spatially digitized parallel resistors on the oxidized Si substrate (Figure 2g). The resistors were using electron beam the lithography in PMMA andbeam sputtering W (50 nm thickness) oxygen atmosphere. formed by defining pattern using electron lithography in PMMA andinsputtering W (50 nm The deflection pit (2, atmosphere. 4, 8 mm diameters) was defined optical lithography in 2-µm-thick SU-8 thickness) in oxygen The deflection pit (2,using 4, 8 mm diameters) was defined using optical negative resist which was used as the spacer layer (Figure 2h). lithography in 2-μm-thick SU-8 negative resist which was used as the spacer layer (Figure 2h).

Figure 2. 2. The SU-8 master master for for the the top top ridge ridge structure; structure; (b) (b) PDMS PDMS Figure The fabrication fabrication process process of of SOTS: SOTS: (a) (a) SU-8 molding using usingPET PET nanobrushs inset showing SEM image; (c) complete curing; molding nanobrushs with with inset showing the SEM the image; (c) complete curing; (d) deposition (d) deposition of the top electrode; (e) peel-off; (f) fabricated top layer; (g) deposition of spatially of the top electrode; (e) peel-off; (f) fabricated top layer; (g) deposition of spatially digitized resistor; digitized resistor; (h) SU-8 (i) spacer formation; (i) deposition of contact and bonding the (h) SU-8 spacer formation; deposition of contact electrodes; and (j)electrodes; bonding the top(j)layer and the top layer and the bottom substrate. bottom substrate.

Sensors 2016, 16, 511

4 of 14

Sensors 2016, 16, 511

4 of 14

The electrodes (20 nm Cr/100 nm Au) used to make contact with the top Pt layer were patterned The electrodes (20 nm Cr/100 nm Au) used to make contact with the top Pt layer were patterned on theSensors SU-82016, surface 16, 511 by optical lithography and thermal evaporation (Figure 2i). Finally the fabricated 4 of 14 on the SU-8 surface by optical lithography and thermal evaporation (Figure 2i). Finally the fabricated top layer and bottom substrate were combined using conducting paste (Figure 2j). top layer and bottom substrate were combined using conducting paste (Figure 2j). The3aelectrodes nmimage Cr/100 of nm Au)spacer used to make contact with the top Pt layer were patternedwere Figure shows a(20 SEM WOx xparallel parallel resistors. resistors Figure 3a shows a SEM image ofthe the spacer and and the the WO resistors. TheThe resistors were on the SU-8 surface by optical lithography and thermal evaporation (Figure 2i). Finally the fabricated 150 µm length, withwith 5 µm width. Multiple resistors µm gaps gapsstarting startingfrom from pit 150inμm in length, 5 μm width. Multiple resistorswere werearranged arranged with with 33 μm top layer and bottom substrate were combined using conducting paste (Figure 2j). centerpittocenter edge.toFigure 3b shows a SEMa image of theoftop with a 75a µm high and 5050µm edge. Figure 3b shows SEM image the layer top layer with 75 μm high and μmwide wideridge Figure 3a shows a SEM image of the spacer and the WOx parallel resistors. The resistors were ´4 ×Ωm, −4 Ωm, which formed 90 kΩ resistors. ridge The structure. The resistivity of WO x resistors were 4.5 10 structure. resistivity of WO resistors were 4.5 ˆ 10 which formed 90 kΩ resistors. The full 150 μm in length, with 5 μmxwidth. Multiple resistors were arranged with 3 μm gaps starting from 2 in area. 2 The full size of the devices was 2 × 2 cm size of devices wasFigure 2 ˆ 23b cmshows in area. pitthe center to edge. a SEM image of the top layer with a 75 μm high and 50 μm wide ridge structure. The resistivity of WOx resistors were 4.5 × 10−4 Ωm, which formed 90 kΩ resistors. The full size of the devices was 2 × 2 cm2 in area.

Figure 3. The images SOTS:(a) (a)The TheSEM SEM image image of WO x resistors and SU-8 Figure 3. The SEMSEM images of of SOTS: ofspatially spatiallydigitized digitized WO x resistors and SU-8 spacer with 2 μm height and 2 mm diameter; (b) the SEM image of top structure composed of the spacer with 2 µm height and 2 mm diameter; (b) the SEM image of top structure composed of the PDMS ridge and the PET substrate, with false coloring. PDMS ridge3.and PET substrate, with falseSEM coloring. Figure Thethe SEM images of SOTS: (a) The image of spatially digitized WOx resistors and SU-8

2.3.

spacer with 2 μm height and 2 mm diameter; (b) the SEM image of top structure composed of the PDMS ridge and the PET substrate, with false coloring. Pressure Detection Characteristics

2.3. Pressure Detection Characteristics

In general, the sensitivity of a pressure sensor is defined as d(AOutput/A0)/dP, where AOutput is the

In the sensitivity ofsensor a pressure sensor is defined AOutput is 2.3.general, Pressure Detection Characteristics output quantity of a pressure such as current, resistanceasord(A capacitance, and A0where is its initial Output /A0 )/dP, value. However case, since the output were inresistance steps, the definition was inadequate. the output quantity in ofour a pressure sensor suchlevels as current, or capacitance, and A0 isThus, its initial In general, the sensitivity of a pressure sensor is defined as d(AOutput/A0)/dP, where AOutput is the have modified the definition to fit our sensor operation as follows: value.we However in our case, since the output levels were in steps, the definition was inadequate. output quantity of a pressure sensor such as current, resistance or capacitance, and A0 is its initialThus, we have modified definition tofit as follows: value. Howeverthe in our case, since the output levels were R0 operation 1 sensor 1 in steps, 1the definition was inadequate. Thus, I our S      R0 (1) we have modified the definition to fit our sensor operation as follows: I 0 P Rresistor P Rresistor  P ∆I 1 R0 1 1 S“ ˆ I “1 ˆ R0 (1) R0 ˆ ∆P 1 “resistance, 1 ˆ Rresistor Rresistor ∆P where, I0 and R0 refer to theI0sensor’s and refers to the resistance S  ∆Poff-state  current  and  RRresistor (1) 0 I 0 4).ΔI P is the Rresistor Pcurrent Rresistor  P due to individual resistors, of the parallel resistors (see Figure step of variation where, I and R refer to the sensor’s off-state current and resistance, andpressure Rresistorincrement refers to the 0 0 which were 11 mA (=1 V/90 kΩ) when the 1 V bias voltage was applied. ΔP is the where, I0 and R0 refer to the sensor’s off-state current and resistance, and Rresistor refers to the resistance resistance of the parallel resistors (see Figure 4). ∆I is the step of current variation due to required to activate the next resistor, which depends on the spacing between the resistors.individual From of the parallel resistors (see Figure 4). ΔI is the step of current variation due to individual resistors, resistors, which 11 mA V/90sensitivity kΩ) when V bias voltage was either applied. ∆P is the pressure Equation (1),were it is clear that(=1 higher canthe be 1achieved by reducing the resistance of a which were 11 mA (=1 V/90 kΩ) when the 1 V bias voltage was applied. ΔP is the pressure increment resistor or their spacing. If thethe resistance due to the resistors were minimized, then the total resistance increment required to activate next resistor, which depends on the spacing between the resistors. required to activate the next resistor, which depends on the spacing between the resistors. From the sensor be dominated bysensitivity the resistance of and Pt by electrodes, which would not be of Fromof Equation (1),itwould itis isclear clear that higher beAu achieved reducing the resistance Equation (1), that higher sensitivity can can be achieved by reducing eithereither the resistance of a desirable. Thus reduction of the spacing, and therefore, the pressure variation was more desirable. a resistor or their spacing. tothe theresistors resistorswere were minimized, then the total resistance resistor or their spacing.IfIfthe theresistance resistance due due to minimized, then the total resistance of theofsensor would be be dominated of Au Auand andPtPtelectrodes, electrodes, which would not be the sensor would dominatedby bythe theresistance resistance of which would not be desirable. Thus reduction of the spacing, and therefore, the pressure variation was more desirable. desirable. Thus reduction of the spacing, and therefore, the pressure variation was more desirable.

Figure 4. The graphical definition of sensitivity of the SOTS.

Figure 4. The graphicaldefinition definition of thethe SOTS. Figure 4. The graphical ofsensitivity sensitivityofof SOTS.

Sensors 2016, 16, 511

5 of 14

∆P can be defined analytically starting from the diaphragm model [30]. We use the diaphragm model as a reference for the device concept and so the model does not correspond faithfully to the actual behavior of the diaphragm. But we may analyze the behavior of our device, such as sensitivity and its dependence on geometrical parameters, based on this model. The deflection of the diaphragm w at the distance r from the center of a diaphragm is given as: Pp1 ´ νq2 w“k Et3

˜ˆ ˙ ¸2 d 2 2 ´r 2

(2)

where E is the Young’s modulus, ν is the Poisson’s ration, t is the thickness, and d is the diameter of the diaphragm, with a proportionality constant k. If the diaphragm touches the bottom of the pit, then the radius of contact rcontact is determined by the pressure P: ˜ˆ ˙ ¸´2 d 2 P“ ˆ ´ rcontact 2 2 kp1 ´ νq2 Et3 h

(3)

here h is the height of the pit. Assume that Pn is the pressure when the diaphragm makes contact with the nth resistor at rn and that Pn+1 for the next resistor at rn+1 . Then the distance between the nth and the next resistor is ∆r = rn+1 – rn . Then the pressure variation is ∆P = Pn+1 – Pn = Pn+1 (1 – Pn /Pn+1 ) determined by the pressure ratio, Pn /Pn+1 : ˛2 ¨ ˛2 2 2 ´ r 2 n `1 ‹ Pn prn ` ∆rq ´ rn ‹ ˚ ˚ “ ˝ ´ ¯2 ‚ “ ˝1 ´ ´ ¯2 ‚ Pn`1 d d 2 2 ´ r ´ r n n 2 2 ¨ ´ ¯2 d 2

(4)

From Equation (4), minimization of ∆r leads to reduction of ∆P and maximization of sensitivity S. Therefore, the pressure sensitivity of SOTS can be controlled by adjusting the spacing between neighboring resistors. This feature was well demonstrated by the following experimental results. Figure 5a shows the contact pressure dependent output of SOTSs with different resistor spacing, 43 µm (SOTS1) and 3 µm (SOTS2), with identical pit diameter of 8 mm. Having the same diameter means that the diaphragm deflection in both sensors would be the same for equally applied contact pressures. It can be seen that the change in output current, with applied pressure, of SOTS2 was higher than SOTS1. The change in output current for the same change in applied pressure was ~6 times higher for SOTS2. This was due to the higher density of resistors (6 times more) making contact in SOTS2 than in SOTS1 for the same applied pressures. At the moment of contact, the output current from SOTS2 shows a jump from 0 to 1.35 mA at 1 V supply voltage, which means that the initial contact signal was made with 39 to 41 resistors, making the diameter of the contact area 312~332 µm. The initial rise in the output current of SOTS1, up to 1.6 kPa of applied pressure, was attributed to about 6 to 8 resistors being contacted at the same time with the contact area gradually increasing. Beyond this pressure, steps in output currents were observed with increase in pressure, as highlighted in the inset of Figure 5a. Clear steps of ~15 µA were observed at 3 kPa and 5 kPa indicating that the sensors may be designed to produce differential current steps at predesignated applied pressures. This stepped output was not observed for SOTS2 due to the higher number of resistors making contact which resulted in a continuous rise in the output with applied pressures, similar to conventional pressure sensors relying on piezoresistive outputs of their active sensing materials. Therefore, by varying the spacing between the resistors, SOTS was able to change its output mode from stepped to continuous.

Sensors 2016, 16, 511 Sensors 2016, 16, 511

6 of 14 6 of 14

Figure The operating basic operating characteristics of (a) SOTS. (a) the dependent pressure dependent output Figure 5. The5.basic characteristics of SOTS. the pressure output characteristics characteristics of SOTS1 (resistor spacing = 43 μm) and SOTS2 (resistor spacing = 3 μm) with identical of SOTS1 (resistor spacing = 43 µm) and SOTS2 (resistor spacing = 3 µm) with identical pit diameter pit diameter of 8 mm with the inset showing the optical image of SOTS1 and expanded output of 8 mm with the inset showing the optical image of SOTS1 and expanded output characteristics of characteristics of SOTS1, and (b) the pressure dependent current measurements of SOTS2 depending SOTS1; and (b) the pressure dependent current measurements of SOTS2 depending on the 2, 4, 8 mm on the 2, 4 ,8 mm of pit diameter. of pit diameter.

The threshold pressures and sensitivities were 0.2, 1.0, 1.4 kPa and 17.8, 5.7, 1.8 kPa−1 for SOTS2 device with 8, sensitivity 4, 2 mm pit may diameters around 5 kPa, by respectively. The result implies the sensor’s The pressure also be controlled the pit diameter. Figure 5bthat shows the pressure pressure sensing range can be tailored by changing the pit diameter. The sensitivity of SOTS2 device of dependent output current of SOTS2 type devices with different pit diameters. Since the diameter around 5 kPa compares with the high sensitivity values of previously reported sensors: 0.15 kPa−1 the pit determines the pressure dependent deflection depth of the PET diaphragm, it will determine [12]; 0.2 kPa−1 [14]; 0.14 kPa−1 [17], 5.54 kPa−1 [22].

the pressure sensitivity. The threshold pressures and sensitivities were 0.2, 1.0, 1.4 kPa and 17.8, 5.7, 1.8 kPa´1 for SOTS2 2.4. Repeatability Test & Detection of Vibration and Grating device with 8, 4, 2 mm pit diameters around 5 kPa, respectively. The result implies that the sensor’s Figure 6a shows the output current of SOTS2 during 30 cycles of 10 kPa pressure application. pressure sensing range can be tailored by changing the pit diameter. The sensitivity of SOTS2 device The magnitude of the output had only 3% deviation. Such high stability of output magnitude comes around 5 kPa with the high sensitivity values of previously reported sensors: 0.15 kPa´1 [12]; from the compares reliable contact switching between the top electrode and the spatially digitized resistors ´1 [14]; 0.14 kPa´1 [17], 5.54 kPa´1 [22]. 0.2 kPa [10,29]. The output current levels may be reduced by reducing the device supply voltage for lower power operation. The frequency response was investigated by applying vibrational stimulus using a

2.4. Repeatability Test & aDetection of Vibration and Grating piezoactuator, with 3 μm vertical displacement, at the top ridge structure of SOTS, as shown in the

inset of 6a Figure 6b. The in pressure would result in 30 thecycles time dependent in application. contact Figure shows the variation output current of SOTS2 during of 10 kPavariation pressure area, and therefore the number of resistors making contact. Figure 6b shows the fast Fourier The magnitude of the output had only 3% deviation. Such high stability of output magnitude transform data of the SOTS2 output depending on vibrational pressure application at the frequency comes from the reliable contact switching between the top electrode and the spatially digitized range of 3~500 Hz. The results show that it was possible to determine the frequency of the applied resistors [10,29]. The output currentmagnitude levels may be20 reduced bythan reducing thefor device supply voltage for pressures, and that the response was dB larger the noise the entire frequency lowerrange. power operation. frequency was investigated applying stimulus The fluctuationThe observed underresponse 5 dB at all frequency regionsby was assumedvibrational to be caused by usingwhite a piezoactuator, with a 3 µmpeak vertical displacement, the top ridge SOTS, as shown noise, the low frequency under 10 Hz is fromatbackground basestructure line. Thisof demonstrated in thethat inset of Figure 6b.frequency, The variation pressure would result the time dependent variation at the measured whichin was similar to the range of in human mechanoreceptor sensing in range [31], SOTS showedthe high sensitivity and the possibility to detect Figure low level contact area, and therefore number of resistors making contact. 6bsurface showsvibrations. the fast Fourier alsoofattempted to detect contact sheer signals using SOTS. The PDMS ridge structures the transformWe data the SOTS2 output depending on vibrational pressure application at the on frequency SOTS surface functions to convert lateral shear into vertical vibrations, similar to human fingerprint range of 3~500 Hz. The results show that it was possible to determine the frequency of the applied structures, The shear force was applied on top of the SOTS2 surface PDMS ridge structures at pressures, and that the response magnitude was 20 dB larger than the noise for the entire frequency 1.6 mm/s scanning speed. Figure 6c shows the output signals in the frequency domain and the inset range. The fluctuation observed under 5 dB at all frequency regions was assumed to be caused by shows the time domain. From the frequency domain, we could identify 2 Hz periodicity, which was whitea noise, the low frequency peak underby10the Hz is from background basespeed. line. Considering This demonstrated close match to the periodicity predicted grating period and scanning the that at the measured frequency, which was similar to the range of human mechanoreceptor sensing output current level, only one or two resistors were in contact indicating that the applied pressures rangewere [31],inSOTS showed high sensitivity and the possibility to detect low level surface vibrations. the range of, or slightly above ~0.2 kPa, which demonstrates the high sensitivity of the SOTS at low amplitude vibrational We also attempted to detectpressures. contact sheer signals using SOTS. The PDMS ridge structures on the SOTS surface functions to convert lateral shear into vertical vibrations, similar to human fingerprint structures, The shear force was applied on top of the SOTS2 surface PDMS ridge structures at 1.6 mm/s scanning speed. Figure 6c shows the output signals in the frequency domain and the inset shows the time domain. From the frequency domain, we could identify 2 Hz periodicity, which was a close match to the periodicity predicted by the grating period and scanning speed. Considering the output current level, only one or two resistors were in contact indicating that the applied pressures were in the range of, or slightly above ~0.2 kPa, which demonstrates the high sensitivity of the SOTS at low amplitude vibrational pressures.

Sensors 2016, 16, 511 Sensors 2016, 16, 511

7 of 14 7 of 14

Figure 6. The The repeatable repeatablepressure pressuresensing sensingoperation operationSOTS2. SOTS2.(a)(a) The measured output depending The measured output depending on on repeated application 10 kPa; (b)frequency the frequency response the frequency of Hz, 3~500 Hz, repeated application of 10ofkPa, (b) the response in theinfrequency range range of 3~500 where where theshows inset the shows the measurement setup; (c) the fast transform Fourier transform of the sensorobtained output the inset measurement setup, (c) the fast Fourier of the sensor output obtained duringscan surface scan ofgrating a silicon grating with structure with 800 µm period, theshows left inset during surface of a silicon structure 800 μm period, where the where left inset the shows the SEM of structure grating structure (scale is 1 mm) right insetshows shows the the real-time SEM image of image grating (scale bar is bar 1 mm) andand thethe right inset real-time measurement measurement data. data.

3. SOTS SOTSArray Array 3.1. 3.1. Concept Concept and and Fabrication Fabrication To pressure distribution, an array of sensors are required. Thus, weThus, have improved To detect detectsurface surface pressure distribution, an array of sensors are required. we have the basic sensor design to facilitate device integration reducing device area and removing improved the basic sensor design multiple to facilitate multiple devicebyintegration by reducing device area contacts to the top electrodes. Figure 7a shows the optical images and the operating mechanism of the and removing contacts to the top electrodes. Figure 7a shows the optical images and the operating revised SOTS. the Pt SOTS. top electrodes not separatedwere as shown in Figureas1shown and were used to1 mechanism of First, the revised First, thewere Pt top electrodes not separated in Figure complete the circuit connection between the ground and spatially separated contact electrodes [27]. and were used to complete the circuit connection between the ground and spatially separated contact This removed the additional step to fabricate contact electrodes on the SU-8 spacer surface (as in electrodes [27]. This removed the additional step to fabricate contact electrodes on the SU-8 spacer Figure device assembly process. Second, the contact were arranged in surface2i), (as simplifying in Figure 2i),the simplifying the device assembly process. Second,electrodes the contact electrodes were

arranged in a spiral to reduce the overall size of the individual sensor while maintaining high

Sensors 2016, 16, 511

8 of 14

Sensors 2016, 16, 511

8 of 14

a spiral to reduce the overall size of the individual sensor while maintaining high resistance of the resistance of the resistors to within 4 mm2. And finally, the WOx resistors with 7.4 kΩ were placed resistors to within 4 mm2 . And finally, the WOx resistors with 7.4 kΩ were placed under the spacer under the spacer enhancing mechanical and chemical stability. The diameter of the spacer pit was enhancing mechanical and chemical stability. The diameter of the spacer pit was 1 mm, and the gap 1 mm, and the gap between the spatially digitized contact electrodes (CEs) were 70 μm, preventing between the spatially digitized contact electrodes (CEs) were 70 µm, preventing simultaneous contacts simultaneous contacts at low pressure levels. The diaphragm layer was replaced with a 100 μm thick at low pressure levels. The diaphragm layer was replaced with a 100 µm thick PDMS instead of PET, to PDMS instead of PET, to lower the Young’s modulus which gives enhanced sensitivity to low level lower the Young’s modulus which gives enhanced sensitivity to low level pressures. Figure 7b shows pressures. Figure 7b shows the SEM image of the bump structure on top of the PDMS layer. The bump the SEM image of the bump structure on top of the PDMS layer. The bump improves the pressure improves the pressure sensitivity by localizing the contact stimulus to the pit area [32,33]. Figure 7c sensitivity by localizing the contact stimulus to the pit area [32,33]. Figure 7c shows an optical image shows an optical image of the integrated 4 × 4 SOTS array on a single chip. The overall size of the of the integrated 4 ˆ 4 SOTS array on a single chip. The overall size of the sensor array was 2 ˆ 2 cm2 , sensor array was 2 × 2 cm2, and 16 unit sensors were located within 1 × 1 cm2. 2 and 16 unit sensors were located within 1 ˆ 1 cm .

Figure ofof thethe 4 ˆ4 4× SOTS array. (a) (a) TheThe contact areaarea when low Figure 7. 7. The The images imagesand andoperating operatingmechanism mechanism 4 SOTS array. contact when and high pressure are applied, and equivalent circuits diagrams; (b) the SEM image of PDMS bump low and high pressure are applied, and equivalent circuits diagrams; (b) the SEM image of PDMS structure with 375 µm375 diameter and 40 and µm height; and (c) and the optical the 4 of ˆ the 4 SOTS bump structure with μm diameter 40 μm height, (c) the image opticalofimage 4 × 4array. SOTS

array.

Figure 8 shows illustrations of the fabrication process for the 4 ˆ 4 SOTS array. For the top layer Figure 8 shows illustrations of the fabrication process the 4master × 4 SOTS array. For100 the µm top layer we used a molding technique starting with a 40 µm thickfor SU-8 (Figure 8a). thick we used a molding technique starting with a 40 μm thick SU-8 master (Figure 8a). 100 μm thick PDMS PDMS was coated and cured (Figure 8b). After sputtering 100 nm Pt electrode, the top layer with a was coated and was cured (Figure 8b). After sputtering 100 nm Pt electrode, the top with a bump bump structure peeled off (Figure 8c,d). To fabricate the bottom substrate, the layer pattern of resistors structure was (Figure 8c,d). To fabricate the bottom substrate, the atmosphere pattern of resistors was formed bypeeled using off optical lithography, and tungsten sputtered in oxygen (Figurewas 8e). formed by using opticalelectrode lithography, tungsten sputtered oxygen atmosphere (Figure The spatially digitized and and contact electrode were in patterned, and 20 nm Cr/ 1008e). nm The Au spatially digitized electrode and contact electrode were patterned, and 20using nm Cr/ 100 nm Au was was evaporated (Figure 8f). Then the 2 µm thick SU-8 spacer was formed optical lithography evaporated (Figurethe 8f).top Then the 2 bottom μm thick SU-8 spacer was formed using(Figure optical8h). lithography (Figure 8g). Finally layer and substrate was aligned and bonded (Figure 8g). Finally the top layer and bottom substrate was aligned and bonded (Figure 8h).

Sensors 2016, 16, 511

9 of 14

Sensors 511 Sensors 2016, 2016, 16, 16, 511

99 of of 14 14

Figure 8. The fabrication process of the 4 × 4 SOTS array. (a) SU-8 master for bump structure molding; (b) PDMS coating and curing; sputter Pt master electrode; (d) peel-off; (e)molding; sputter Figure 8. The fabrication process (c) of the 4 ˆ 4deposition SOTS array.of(a)top SU-8 for bump structure Figure 8. The fabrication process of the 4 × 4 SOTS array. (a) SU-8 master for bump structure molding; resistors; (f) (c) deposition of spatiallyofdigitized electrodes (d) andpeel-off; contact (e) electrodes; deposition of WOx and (b) PDMS coating curing; sputter deposition top Pt electrode; sputter (b) PDMS coating and curing; (c) sputter deposition of top Pt electrode; (d) peel-off; (e) sputter deposition of WO deposition spatially digitized electrodes and contact electrodes; (g) Su-8 spacer formation; (h) (f) bonding of theof top layer and the bottom substrate. x resistors; deposition of WOx resistors; (f) deposition of spatially digitized electrodes and contact electrodes; (g) Su-8 spacer formation; (h) bonding of the top layer and the bottom substrate. (g) Su-8 spacer formation; (h) bonding of the top layer and the bottom substrate.

3.2. Operating Characteristics 3.2. Operating Figure 9aCharacteristics shows the response of four sensors in the array with 0.1 V supply voltage. As the 3.2. Operating Characteristics applied pressure gradually increased, wesensors were able observe of about μA at Figure 9a shows the response of four in thetoarray withfour 0.1 Vcurrent supplysteps voltage. As the13 applied Figure 9a shows the response of four sensors in the array with 0.1 V supply voltage. As the designated pressures betweenwe 6~14 kPa. The sensor response results the pressure gradually increased, were able to nonlinear observe four current steps of aboutfrom 13 µA atnonlinearly designated applied pressure gradually increased, we were able to observe four current steps of about 13 μA at deflectingbetween diaphragm with uniformly spaced The sensorsdeflecting showed pressures 6~14making kPa. Thecontact nonlinear sensor response resultsresistors. from the nonlinearly designated pressures between 6~14 kPa. The nonlinear sensor response results from the nonlinearly reproducible resultscontact underwith repeated pressure cycles (Figure with showed about 0.5% distribution in diaphragm making uniformly spaced resistors. The9b) sensors reproducible results deflecting diaphragm making contact with uniformly spaced resistors. The sensors showed current. It can pressure be observed results fromabout 32 sensors from two in array devices under repeated cyclesthat (Figure 9b) with 0.5% distribution current. It candemonstrates be observed reproducible results under repeated pressure cycles (Figure 9b) with about 0.5% distribution in relatively high uniformity of threshold and current magnitude at eachhigh pressure level step that results from 32 sensors from two pressure array devices demonstrates relatively uniformity of current. It can be observed that results from 32 sensors from two array devices demonstrates (Figure 9c). threshold pressure and current magnitude at each pressure level step (Figure 9c). relatively high uniformity of threshold pressure and current magnitude at each pressure level step (Figure 9c).

Figure 9. Cont.

Figure 9. Cont.

Sensors 511 Sensors 2016, 2016, 16, 16, 511 511 Sensors 2016, 16,

10 of 14 10 of of 14 14 10

Figure 9. characteristics of of 444 ˆ SOTS arrays. (a) The pressure dependent measurement Figure 9. The The operating operating characteristics characteristics SOTSarrays. arrays.(a) (a)The Thepressure pressuredependent dependent measurement measurement Figure of ×× 444SOTS results of four unit sensors sensors showing showing the the fully results of of four four unit fully stepped stepped output output characteristics characteristics and and operating operating uniformity; uniformity; results (b) the output depending on repeated application of 20 kPa of contact pressure; (c) the distribution of (b) the the output output depending depending on on repeated repeated application application of of 20 20 kPa kPa of of contact contact pressure; pressure; (c) (c) the the distribution distribution of of (b) pressure threshold and current magnitude at each current step of 32 sensors in two array devices. pressure threshold and current magnitude at each current step of 32 sensors in two array devices. pressure threshold and current magnitude at each current step of 32 sensors in two array devices.

3.3. Read-Out Read-Out Circuit Circuit 3.3. A readout readoutcircuit circuitwas wasprepared preparedtoto to read all of the 16 sensors sensors without any cross-talk, in sequence sequence read allall of of thethe 16 16 sensors without anyany cross-talk, in sequence (see A readout circuit was prepared read without cross-talk, in (see Figure Figure 10). A single-pole, single-pole, single–throw (SPDT) switch (ADG733, Analog Devices Inc., Norwood, Norwood, Figure 10). A single-pole, single–throw (SPDT) switch (ADG733, Analog Devices Inc., Inc., Norwood, MA, (see 10). A single–throw (SPDT) switch (ADG733, Analog Devices MA, USA) USA) wasasused used asselection row selection selection switch, and high highamplifier precision(LMP7701, amplifierTexas (LMP7701, Texas USA) was used a row switch, and high precision Instruments, MA, was as aa row switch, and precision amplifier (LMP7701, Texas Instruments, Dallas, TX, USA) was used as an amplifier. When the driving voltage (V DRV the Dallas, TX, USA) wasTX, usedUSA) as anwas amplifier. When the drivingWhen voltage reference voltage Instruments, Dallas, used as an amplifier. the(Vdriving voltage (VDRV ),), the DRV ), the reference voltage (VREF REF)) and and the feedback feedback resistor (R0.5 FB)) V were at7.4 V, 0.5respectively, V and and 7.4 7.4 kΩ, kΩ, respectively, the (V the feedback resistor (RFB ) were at 0 V,(R andat kΩ, therespectively, voltage output reference voltage (V the resistor FB were 00 V, 0.5 V the REF ) and voltage output (V OUT ) was given as: (V ) was given as: voltage OUT output (VOUT) was given as:

RFB 7.4kk 7.4 RR FB FB V pV 0.5`0.5 0.5 7.4k VVOUT ((VVDRV V VVREF  0.5 0.5 OUT REF´ DRV´ REFq)) VOUT “VV “ 0.5 REF REF DRV REF Rsensor R RR RR sensor sensor sensor sensor sensor

(5) (5)

where, R Rsensor sensor is isisthe the measured resistance of unit sensor. The five five levels levels of of detectable were where, themeasured measuredresistance resistanceof ofaaaunit unitsensor. sensor. The detectable pressures pressures were were where, R sensor converted to the output voltage of 0.5, 1, 1.5, 2, and 2.5 V. converted to the output voltage of 0.5, 1, 1.5, 2, and 2.5 V. converted to the output voltage of 0.5, 1, 1.5, 2, and 2.5 V.

Figure 10. 10. The The read-out read-out circuit circuit for for the the44ˆ SOTS array. array. Figure ×× 44 SOTS

3.4. Pressure Pressure Mapping Mapping Capabilities Capabilities 3.4.

Sensors 2016, 16, 511

11 of 14

3.4. Pressure Mapping Capabilities Sensors 2016, 16, 511 11 of 14 The surface pressure distribution sensing characteristics of the SOTS array is shown in Figure 9. The optical images were taken before the sensor was wired for measurement. The test objects placed The surface pressure distribution sensing characteristics of the SOTS array is shown in Figure 9. on the sensor array surface were before spheres radius of 8 and 20 mm, and The a rectangular The optical images were taken thewith sensor was wired for measurement. test objects eraser placed with 2 contact area which weighed 40, 5 and 2 g, respectively. Once the objects were placed on 4 ˆ 10 onmm the sensor array surface were spheres with radius of 8 and 20 mm, and a rectangular eraser with the sensor surface, the five levels of 40, detectable were converted to were the output 4 × 10 array mm2 contact area which weighed 5 and 2 g,pressures respectively. Once the objects placed voltage on thereadout sensor array surface, the five levels of detectable pressuresthe were converted to the output voltage by the circuit and the intermediate levels between sensor positions were interpolated circuit and at theeach intermediate levelsThe between the sensor positions were area interpolated basedbyonthe thereadout measured levels sensor point. detected pressurized contact for the hard based on the measured levelswider at eachdue sensor point. The detected pressurized resulting contact area hard spheres in Figure 11a,b became to the PDMS layer deformation in for thethe distribution spheres in Figure 11a,b became wider due to the PDMS layer deformation resulting in the distribution 2 of vertical pressure laterally. Considering that 2 g placed over 1 cm area equals 200 Pa, the measured of vertical pressure laterally. Considering that 2 g placed over 1 cm2 area equals 200 Pa, the measured distribution of 8~14 kPa for the assumed contact points shows that the bump structure acted to localize distribution of 8~14 kPa for the assumed contact points shows that the bump structure acted to the contact pressure enabling detection of such low contact pressures. This demonstrated that the localize the contact pressure enabling detection of such low contact pressures. This demonstrated that SOTSthe array was able to able measure the distribution of contact pressures between kPa, within within the SOTS array was to measure the distribution of contact pressures between0~14 0~14 kPa, 2 1 ˆ 1the cm12 surface area. Figure 11d shows detection of dynamic pressure distribution using the readout × 1 cm surface area. Figure 11d shows detection of dynamic pressure distribution using the circuit (1 V supply voltage), which demonstrates that the SOTS also capable of detecting readout circuit (1 V supply voltage), which demonstrates that thearray SOTSwas array was also capable of detecting dynamic changes in contact pressures. dynamic changes in contact pressures.

Figure 11. The pressure (voltage) distribution convertedfrom from SOTS array’s read-out voltage, Figure 11. The pressure (voltage) distribution map map converted thethe SOTS array’s read-out voltage, spheres (a) mm 20 mm and mmradius; radius, and and (c) with 4 × 410ˆmm contact whenwhen spheres withwith (a) 20 and (b)(b)8 8mm (c) rectangular rectangulareraser eraser with 10 mm contact area were pressed on the 4 SOTSarray; array. (d) (d) The The voltage of of SOTS array when the sphere area were pressed on the 4 ˆ4 4× SOTS voltagecolor colormap map SOTS array when the sphere was in circular motion. optical imagesshow show the the SOTS SOTS array wiring. was in circular motion. AllAll optical images arraybefore before wiring.

Sensors 2016, 16, 511

12 of 14

The power consumption levels of SOTS may be tailored by controlling the supply voltage and the value of resistors. If we compare the power consumption levels between SOTS2 and the integrated SOTS, by reducing the supply voltage and increasing resistor value the output power during operation was reduced by 1/1000 to ~5 µW at saturation pressures. It may be possible to further reduce the output power by utilizing the stepped nature of the output current. Since a designated current level determines the measured pressure, the output current may be reduced to levels where the measurement noise does not exceed the step current distinguishing each detected pressure levels. The sensor is in open circuit mode when no pressure was detected and the current fluctuation was at 10´12 A levels meaning that the power consumption may be lowered to 10´12 W levels while still maintaining high SNR levels. The ability to distinguish small differences in pressures at low contact pressure ranges may allow SOTS to function as a minimum pressure level indicator for systems that utilizes predesignated contact pressures as triggers for a desired action command. Also, high sensitivity to low frequency contact vibrations may also allow SOTS to determine the differences in contact shear characteristics of materials with varying surface roughness enabling robotic tactile interfaces and electronic skin prosthesis. 4. Conclusions We have developed a tactile pressure sensor that produces stepwise output currents when contact pressures are applied. The sensor’s operation relies on the pressure dependent deflection of the polymer diaphragm making contact with resistive current paths on the substrate surface. The stepped output current depended on the number of short-circuited resistors, which reflected the tactile pressure level. The SOTS showed that they can have variable sensitivity and sensing range just by adjusting the number and spacing between the resistive current paths. They also showed SNR above 20 dB at contact pressures below 15 kPa. A 4 ˆ 4 SOTS array with a surface bump structure was fabricated to demonstrate 1 ˆ 1 cm2 areal pressure mapping capabilities. Acknowledgments: This research was supported by the Basic Science Research Program (2012R1A6A1029029 & 2014R1A1A2053599) through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT & Future Planning and partly by the Nano Material Technology Development Program (2012M3A7B4035198 & 2012M3A7B4035201) through NRF funded by the Ministry of Education. Author Contributions: Eunsuk Choi designed the device, and measured the output characteristics of device, and wrote the paper, Onejae Sul designed the measurement system, and analyzed the experimental result. Juyoung Kim and Kyumin Kim fabricated the device. Jong-Seok Kim, Dae-Young Kwon and Byong-Deok Choi designed and fabricated the read-out circuit. Seung-Beck Lee conceived and led the research, analyzed the experimental results, and wrote the paper. Conflicts of Interest: The authors declare no conflict of interest.

References 1.

2.

3.

4. 5.

Raspopovic, S.; Capogrosso, M.; Petrini, F.M.; Bonizzato, M.; Rigosa, J.; Di Pino, G.; Carpaneto, J.; Controzzi, M.; Boretius, T.; Fernandez, E.; et al. Restoring natural sensory feedback in real-time bidirectional hand prostheses. Sci. Transl. Med. 2014, 6, 222ra19. [CrossRef] [PubMed] Kim, J.; Lee, M.; Shim, H.J.; Ghaffari, R.; Cho, H.R.; Son, D.; Jung, Y.H.; Soh, M.; Choi, C.; Jung, S.; et al. Stretchable silicon nanoribbon electronics for skin prosthesis. Nat. Commun. 2014, 5, 1–11. [CrossRef] [PubMed] Odhner, L.U.; Jentoft, L.P.; Claffee, M.R.; Corson, N.; Tenzer, Y.; Ma, R.R.; Buehler, M.; Kohout, R.; Howe, R.D.; Dollar, A.M. A compliant, underactuated hand for robust manipulation. Int. J. Robot. Res. 2014, 33, 736–752. [CrossRef] Westebring-van der Putten, E.P.; Goossens, R.H.M.; Jakimowicz, J.J.; Dankelman, J. Haptics in minimally invasive surgery-A review. Minim. Invasive Ther. Allied Technol. 2008, 17, 3–16. [CrossRef] [PubMed] Myles, K.; Binseel, M.S. The Tactile Modality: A Review of Tactile Sensitivity and Human Tactile Interfaces; U.S. Army Research Laboratory: Aberdeen, MA, USA, 2007.

Sensors 2016, 16, 511

6. 7. 8. 9. 10. 11. 12.

13. 14. 15. 16.

17.

18.

19.

20.

21.

22.

23. 24. 25. 26. 27.

13 of 14

Bruce, M.F.; Sinclair, D.C. The relationship between tactile thresholds and histology in the human finger. J. Neurol. Neurosurg. Psychiatry 1980, 43, 235–242. [CrossRef] [PubMed] Johansson, L.; Kjellberg, A.; Kilbom, A.; Hagg, G.M. Perception of surface pressure applied to the hand. Ergonomics 2010, 42, 1274–1282. [CrossRef] [PubMed] Brennum, J.; Kjeldsen, M.; Kai Jensen, K.; Jensen, T.S. Measurements of human pressure-pain thresholds on fingers and toes. Pain 1989, 38, 211–217. [CrossRef] Vallbo, A.B.; Johansson, R.S. Properties of cutaneous mechanoreceptors in the human hand related to touch sensation. Hum. Neurobiol. 1984, 3, 3–14. [PubMed] Weber, E.H. E.H.Weber on the Tactile Senses, 2nd ed.; Ross, H.E., Murray, D.J., Eds.; Erlbaum (UK) Taylor & Francis: London, UK, 1996. Wang, X.; Gu, Y.; Xiong, Z.; Cui, Z.; Zhang, T. Silk-molded flexible, ultrasensitive, and highly stable electronic skin for monitoring human physiological signals. Adv. Mater. 2014, 26, 1336–1342. [CrossRef] [PubMed] Pang, C.; Lee, G.-Y.; Kim, T.; Kim, S.M.; Kim, H.N.; Ahn, S.-H.; Suh, K.Y. A flexible and highly sensitive strain-gauge sensor using reversible interlocking of nanofibers. Nat. Mater. 2012, 11, 795–801. [CrossRef] [PubMed] Zhu, B.; Niu, Z.; Wang, H.; Leow, W.R.; Wang, H.; Li, Y.; Zheng, L.; Wei, J.; Huo, F.; Chen, X. Microstructured graphene arrays for highly sensitive flexible tactile sensors. Small 2014, 10, 3625–3631. [CrossRef] [PubMed] Su, B.; Gong, S.; Ma, Z.; Yap, L.W.; Cheng, W. Mimosa-inspired design of a flexible pressure sensor with touch sensitivity. Small 2015, 11, 1886–1891. [CrossRef] [PubMed] Viry, L.; Levi, A.; Totaro, M.; Mondini, A.; Mattoli, V.; Mazzolai, B.; Beccai, L. Flexible three-axial force sensor for soft and highly sensitive artificial touch. Adv. Mater. 2014, 26, 2659–2664. [CrossRef] [PubMed] Mannsfeld, S.C.B.; Tee, B.C.-K.; Stoltenberg, R.M.; Chen, C.V.H-H.; Barman, S.; Muir, B.V.O.; Sokolov, A.N.; Reese, C.; Bao, Z. Highly sensitive flexible pressure sensors with microstructured rubber dielectric layers. Nat. Mater. 2010, 9, 859–864. [CrossRef] [PubMed] Park, S.; Kim, H.; Vosgueritchian, M.; Cheon, S.; Kim, H.; Koo, J.H.; Kim, T.R.; Lee, S.; Schwartz, G.; Chang, H.; et al. Stretchable energy-harvesting tactile electronic skin capable of differentiating multiple mechanical stimuli modes. Adv. Mater. 2014, 26, 7324–7332. [CrossRef] [PubMed] Kim, S.Y.; Park, S.; Park, H.W.; Park, D.H.; Jeong, Y.; Kim, D.H. Highly sensitive and multimodal all-carbon skin sensors capable of simultaneously detecting tactile and biological stimuli. Adv. Mater. 2015, 27, 4178–4185. [CrossRef] [PubMed] Lee, J.S.; Shin, K.-Y.; Cheong, O.J.; Kim, J.H.; Jang, J. Highly sensitive and multifunctional tactile sensor using free-standing ZnO/PVDF thin film with graphene electrodes for pressure and temperature monitoring. Sci. Rep. 2015, 5, 7887. [CrossRef] [PubMed] Kang, D.; Pikhitsa1, P.V.; Choi, Y.W.; Lee, C.; Shin, S.S.; Piao, L.; Park, B.; Suh, K.-Y.; Kim, T.; Choi, M. Ultrasensitive mechanical crack-based sensor inspired by the spider sensory system. Nature 2014, 516, 222–226. [CrossRef] [PubMed] Wang, Y.; Wang, L.; Yang, T.; Li., X.; Zang, X.; Zhu, M.; Wang, K.; Wu, D.; Zhu, H. Wearable and highly sensitive graphene strain sensors for human motion monitoring. Adv. Funct. Mater. 2014, 24, 4666–4670. [CrossRef] Wang, J.; Jiu, J.; Nogi, M.; Sugahara, T.; Nagao, S.; Koga, H.; He, P.; Suganuma, K. A highly sensitive and flexible pressure sensor with electrodes and elastomeric interlayer containing silver nanowires. Nanoscale 2015, 7, 2926–2932. [CrossRef] [PubMed] Jariwala, D.; Sangwan, V.K.; Lauhon, L.J.; Marks, T.J.; Hersma, M.C. Carbon nanomaterials for electronics, optoelectronics, photovoltaics, and sensing. Chem. Soc. Rev. 2013, 42, 2824–2860. [CrossRef] [PubMed] Lee, S.-M.; Kim, J.-H.; Ahn, J.-H. Graphene as a flexible electronic material: Mechanical limitations by defect formation and efforts to overcome. Mater. Today 2015, 18, 336–344. [CrossRef] Terzic, E.; Terzic, J.; Nagarajah, R.; Alamgir, M. A Neural Network Approach to Fluid Quantity Measurement in Dynamic Environments; Springer-Verlag: London, UK, 2012; pp. 11–37. Yousefa, H.; Boukallel, M.; Althoefer, K. Tactile sensing for dexterous in-hand manipulation in robotics–A review. Sens. Actuators A Phys. 2011, 167, 171–187. [CrossRef] Choi, E.; Sul, O.; Hwang, S.; Cho, J.; Chun, H.; Kim, H.; Lee, S.-B. Spatially digitized tactile pressure sensors with tunable sensitivity and sensing range. Nanotechnology 2014, 25, 425504. [CrossRef] [PubMed]

Sensors 2016, 16, 511

28. 29.

30. 31. 32. 33.

14 of 14

Scheibert, J.; Leurent, S.; Prevost, A.; Debrégeas, G. The role of fingerprints in the coding of tactile information probed with a biomimetic sensor. Science 2009, 323, 1503–1506. [CrossRef] [PubMed] Oddo, C.M.; Beccai, L.; Wessberg, J.; Wasling, H.B.; Mattioli, F.; Carrozza, M.C. Roughness encoding in human and biomimetic artificial touch: Spatiotemporal frequency modulation and structural. Sensors 2011, 11, 5596–5615. [CrossRef] [PubMed] Timoshenko, S.; Woinowsky-Krieger, S. Theory of Plates and Shells, 2nd ed.; McGraw-Hill: New York, NY, USA, 1959; pp. 51–78. Johnson, K.O. The roles and functions of cutaneous mechanoreceptors. Curr. Opin. Neurobiol. 2001, 11, 455–461. [CrossRef] Lee, H.-K.; Chang, S.-I.; Yoon, E. A flexible polymer tactile sensor: Fabrication and modular expandability for large area deployment. IEEE J. Microelectromech. Syst. 2006, 15, 1681–1686. [CrossRef] Peng, P.; Rajamani, R.; Erdman, A.G. Flexible tactile sensor for tissue elasticity measurements. IEEE J. Microelectromech. Syst. 2009, 18, 1226–1233. [CrossRef] © 2016 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/).