Lab on a Chip View Article Online

Published on 31 January 2014. Downloaded by University of Illinois at Chicago on 04/02/2014 22:51:02.

PAPER

View Journal

Iontronic microdroplet array for flexible ultrasensitive tactile sensing† Cite this: DOI: 10.1039/c3lc50994j

Baoqing Nie,a Ruya Li,a James D. Brandtb and Tingrui Pan*a An iontronic microdroplet array (IMA) device, using an ultra-large interfacial capacitance at the highly elastic droplet–electrode contact, has been proposed for flexible tactile sensing applications. The transparent IMA sensors consist of an array of nanoliter droplets sandwiched between two polymeric membranes with patterned transparent electrodes, forming the electrical double layers with remarkable unit-area capacitance. Under external loading, the membrane deformation results in the circumferential expansion at the highly elastic droplet–electrode contact, which offers a completely new capacitive sensing scheme with a dramatic increase in sensitivity. Under the simple device architecture, the IMA has achieved device sensitivity of 0.43 nF kPa−1 and a minimal detectable pressure of 33 Pa, the highest Received 29th August 2013, Accepted 1st December 2013 DOI: 10.1039/c3lc50994j

reported values for its dimension. In addition, the hysteresis of the droplet deformation has been reduced by introducing a layer of hydrophobic coating to the conductive electrode surface, ensuring a fast mechanical response (on the order of several milliseconds). To demonstrate the utility of the transparent flexible IMA sensor, it has been successfully mounted onto a fingertip setting to map different surface topologies

www.rsc.org/loc

and embedded into a wristband to resolve dynamic pressure waves throughout cardiovascular cycles.

Introduction Droplet-based microfluidic devices have received tremendous attention in a variety of areas, such as biochemical analysis and material synthesis and signal sensing, for their high sensitivity, mechanical flexibility and simple fabrication.1–6 In one such implementation, a mercury droplet is sandwiched between two planar electrodes covered with insulating material with high dielectric permittivity.2 As a result, the mercurybased sensor offers high electrical-to-mechanical sensitivity (of 2.24 μF kPa−1), and moreover, it can be microfabricated through standard CMOS processing, enabling system integration and scalable packaging. In addition, a droplet-enabled surface acoustic wave (SAW) device has been reported as a urea-concentration detector.3 The concentration of a urea droplet is determined by measuring the impedance changes of the droplet under the urease catalytic reactions. Furthermore, an inertial sensor is introduced by utilizing the interfacial instability of aqueous droplets on superhydrophobic surfaces. The external accelerations are determined from the positioning of the droplets which can be detected by an array of parallel microelectrodes.4 a

Micro-Nano Innovations (MiNI) Laboratory, Department of Biomedical Engineering, University of California, Davis, USA. E-mail: [email protected] b Department of Ophthalmology, Eye Centre, Health System, University of California, Davis, USA † Electronic supplementary information (ESI) available: Experimental setup. Mathematical derivation for eqn 1. Circuitry for IMA electrical signal measurement. Video of real-time pulse signal recording. See DOI: 10.1039/c3lc50994j

This journal is © The Royal Society of Chemistry 2014

Flexible tactile sensing devices have been extensively researched for their potential utility in medicine and robotics, in which a wide spectrum of sensing mechanisms have been introduced and investigated, including resistive, piezoelectric and capacitive sensing.7–18 Among those, elastomeric resistive materials have been frequently employed in building such arrays due to their low cost and easy manufacturability.7–10 Alternatively, a piezoelectric nanomaterial has been successfully incorporated into a pressure sensitive matrix with submillimetre resolution for tactile imaging.14 Capacitive sensing leads another trend in the development of artificial skins, given its overwhelming success in consumer electronics.15–22 Conventional capacitive tactile arrays, using a parallel plate configuration, have been devised for both pressure and force mapping.15–18 However, several major technical challenges still remain in current artificial skin research, including low device sensitivity, susceptibility to environmental noise, in addition to complex signal processing. Latest developments in iontronic materials (e.g., ionic liquids) have renewed interest in utilizing ionic solutions to regulate the electronic transport at the electrolyte–electrode interface.23–26 Upon solid–liquid contact, an electric double layer (EDL) is immediately established with an ultrahigh unit-area capacitance of the order of 10 μF cm−2 at the atomic interface, which is more than a thousand times greater than that of the solid-state counterpart (up to 2 nF cm−2).18,27,28 It has been shown that the iontronic device can be utilized in field effect transistor applications, where the EDL exhibits ultrahigh-density charge accumulation while avoiding reactions

Lab Chip

View Article Online

Paper

Published on 31 January 2014. Downloaded by University of Illinois at Chicago on 04/02/2014 22:51:02.

due to its electrochemical inertness within the potential windows.29,30 Latest advances in synthetic chemistry have yielded low-viscosity yet high-conductivity ionic liquids (ILs) in addition to thermal stability under ambient conditions, which can be of broad interest for electronic and sensing applications.31 For example, ILs have been investigated in optoelectronic devices by utilizing the highly conductive EDL at the

Lab on a Chip

IL/charged solid interface, which produces large photocurrent responses under light excitation.32 In another effort, ILs as strain gauges have been utilized in a microfluidic pressure sensing system, in which the device possesses a long-term operating period due to the thermal stability of the ILs.33 In addition, ILs have been explored in gaseous sensing devices by utilizing their high absorbability to a wide variety of gases.34,35 In this paper, we present a novel iontronic microdroplet array (IMA), using the interfacial capacitive sensing principle, for flexible tactile sensing applications. Fig. 1a illustrates the transparent IMA device, of which each sensing element comprises a nanoliter ionic droplet, sandwiched between two flexible substrates with patterned transparent electrodes. Fig. 1b exhibits the interfacial capacitive sensing principle. The direct ionic droplet–electrode contact immediately establishes the EDL, which possesses a remarkable interfacial capacitance (of the order of 10 μF cm−2). Under external loads, the flexible surfaces experience mechanical deformations, resulting in circumferential expansion of the droplet–electrode contact. The corresponding capacitive change over the increased contact area can be detected electronically. This new EDL sensing mechanism is the latest addition to the conventional capacitive sensing principle, mainly configured in parallel plates or interdigitated comb drives.15,17,21,36 Benefiting from the ultrahigh unit-area capacitance of the EDL, we can achieve a device sensitivity of 0.43 nF kPa−1 and a minimal detectable pressure of 33 Pa at a 3 × 3 × 0.2 mm3 packaging, which to the best of our knowledge is the most sensitive capacitive pressure sensor of its dimension, though one report shows a sensitivity of 0.8 nF kPa−1 with a much larger footprint of 6 × 6 mm2.18 Moreover, the low viscosity of the ionic liquid and the high contrast wettability control on the electrode surface enable fast mechanical response (on the order of several milliseconds), in comparison with that of solid-state thin film-based tactile sensing devices.8,10,17,18 In addition, the IMA device exhibits high stability over more than 20 000 cycles of pressure/force loads. Furthermore, the simple device architecture is optically transparent and can be massively produced at a low cost. As a demonstration of the utility of the iontronic devices, the IMA sensors have successfully resolved surface topologies and detected dynamic wrist pulses throughout cardiovascular cycles.

Methods Microfabrication process

Fig. 1 The photograph and sensing principle of the iontronic microdroplet sensing device. a, A photograph of an IMA sensor array of 12 × 12 elements with a spatial resolution of 1 mm with our lab logo as the background, indicating the high level of transparency. b, A cross-sectional sketch of the interfacial capacitive sensing principle. The EDL capacitance is simultaneously established at the ionic droplet–electrode contact (top). The interfacial contact area expands circumferentially, which is caused by the mechanical deformation of the membrane under external loads and results in an increase in the EDL capacitance (bottom). External voltage (V) has been applied to the top and bottom electrodes for the EDL capacitance measurement.

Lab Chip

The IMA device consists of two flexible films with micropatterned transparent electrodes and one spacing layer hosting the ionic droplet sensing units. The fabrication process starts with micropatterning of conductive indium– tin-oxide (ITO, ~100 nm thick) electrodes on flexible polyethylene terephthalate (PET) films (of various thickness from 75 μm to 175 μm) using standard photolithography followed by wet etching.37,38 In the subsequent step, a dry-film photoresist (50 μm, PerMX3050, DuPont) is thermally laminated onto the ITO-patterned PET substrate. Following a soft bake at 115 °C for 5 min, it is then exposed to selective UV light in

This journal is © The Royal Society of Chemistry 2014

View Article Online

Published on 31 January 2014. Downloaded by University of Illinois at Chicago on 04/02/2014 22:51:02.

Lab on a Chip

Paper

a mask aligner (365 nm, 220 mJ cm−2, ABM, Inc.). In the subsequent step, the dry-film is post-baked at 95 °C for 2 min and developed in an ultrasonic bath with propylene glycol monomethyl ether acetate (PGMEA >99.5%, Sigma-Aldrich) for 30 s, leaving the micropillar patterns on the substrate. To accurately position the microdroplets, a surface wettability patterning technique has been utilized. The ITO-patterned substrate is first activated with hydroxyl groups for 30 s in an oxygen plasma at 90 W (FEMTO, Diener). Then, a hydrophobic oligomer layer of polydimethylsiloxane (PDMS) is contact-printed onto both electrode surfaces for 2 hours, using a PDMS stamp made from a mixture of a base and a curing agent in a 15 : 1 weight ratio (Sylgard 184, Dow Corning).39 As a result, a nanometre-thick layer of PDMS oligomers is selectively deposited, forming high-contrast surface-energy patterns on the electrode surfaces, as described in our previous studies.40 Subsequently, using a microfluidic impact printing technique, nanoliter droplets (~3 nL) of the ionic liquid will be sequentially deposited onto an array of hydrophilic microdots formed by the wettability patterning.41 Prior to the final assembly, two electrode films are aligned face-to-face with the conductive patterns positioned orthogonally to each other, forming a grid of capacitance at the crossover points where the ionic droplet array sits in. The top and bottom layers are then bonded together after the oxygen-plasma activation of hydroxyl groups of the PDMS oligomer layers (30 s exposure at 90 W).

Measurement setup Experimental investigations of the device sensitivity have been conducted on individual sensing units of the iontronic microdroplet array. The measurement stage is comprised of a force gauge with 1 mN resolution (DFS, Chatillon) mounted on a computer-controlled step motor (VT-80 PImiCos) with a spatial resolution of 400 nm, from which mechanical loads and displacements can be controlled and monitored simultaneously. The pressure values are calculated based on the ratio of the applied force to the surface area of the membrane in each sensing unit. The corresponding capacitive changes are directly recorded through an LCR meter (4284A, Agilent) (Fig. S1, ESI†). Each sensitivity measurement has been conducted twice on two identical sensing devices. For the characterization of the response time, an electromagnetically driven pin actuator (Panasonic KX-P1150), powered by a pulsed

voltage signal from 1 to 100 Hz, has been used to apply periodic contact pressure to the sensor surface. The output signals of the IMA device are measured with a custom readout circuitry (Fig. S2, ESI†).

Results Iontronic microdroplet fluid Iontronic capacitive sensing is established upon forming an ionic–electronic interface at the droplet–electrode contact. The droplet sensing fluid has to satisfy several design criteria, including high conductivity (providing ultrahigh EDL capacitance and low electrical loss), low viscosity (ensuring short response time), electrochemical stability (no electrochemical reaction under the operating voltage) and environmental stability (maintaining the physical properties over the operating period). Three types of ionic fluids have been considered, such as aqueous electrolytes (e.g., NaCl electrolyte solution), organic solvent solutions (e.g. KClO4–PEO), and ionic liquids, which are commonly investigated in electrochemical processes.42–44 Aqueous and solventbased electrolytes are typically highly evaporative under ambient conditions and it becomes extremely challenging to maintain constant electrical performance, as both the volume and the physical properties change over time.45 As an emerging alternative, ionic liquids, consisting of an organic anion or cation, exhibit high electrical conductivity, low volatility, and tuneable viscosity.46,47 In addition to their wide electrochemical window, ILs are ideal candidates for microdroplet sensors. Table 1 lists the physical properties of four types of imidazolium-based ILs, a group of ILs which show excellent ionic conductivity and good chemical stability.46 As can be seen, these ILs have wide electrochemical windows, ranging from 2.6 V to 5.7 V, and possess negligible vapour pressures (more than a thousand times lower than that of water).48–50 In addition, the low melting points of the ILs (typically less than 0 °C) ensure they remain in a liquid state under room temperature conditions.50–52 Interestingly, the electrical conductivities of the ILs are inversely related to their dynamic viscosities, which allows for their use in ultrasensitive and highly responsive tactile sensors.53 Therefore, the ionic liquid of 1-ethyl-3-methylimidazolium tricyanomethanide, with the highest conductivity (18 mS cm−1) and lowest

Table 1 Physical properties of four different imidazolium-based ILs

ILs Electrochemical window [V] Melting point [°C] Vapor pressure [Pa] Molar weight [g mol−1] Conductivity [mS cm−1] Viscosity [Pa s]

1-Butyl-3-methylimidazolium 1-Butyl-3-methylimidazolium 1-Ethyl-3-methylimidazolium 1-Ethyl-3-methylimidazolium hexafluorophosphate tetrafluoroborate tetrafluoroborate tricyanomethanide 5.7

4.2

2.6

2.9

−61 10−11 284 1.4 312

−82 N/A 225.8 3.5 180

15 N/A 197.8 14 32

−11 N/A 201.23 18 18

This journal is © The Royal Society of Chemistry 2014

Lab Chip

View Article Online

Paper

viscosity (18 Pa s) among the iontronic fluid of the ILs surveyed in Table 1, was selected as the working fluid in the iontronic droplet sensors.

Published on 31 January 2014. Downloaded by University of Illinois at Chicago on 04/02/2014 22:51:02.

Interfacial capacitance As mentioned previously, the EDL presents a remarkable unit-area capacitance at the nanoscopic interface between the electrode and the electrolyte droplet. Unlike solid-state capacitors, it is established by mobile electrons in a conductive surface and counter-ions migrating in the adjacent liquid environment, and the value can be determined by the surface charge density and the Debye length.54 In particular, the EDL capacitance is frequency-dependent with several associated intermolecular interaction mechanisms (e.g., interfacial polarization).55 The frequency dependence of the EDL is characterized by using a LCR meter to determine the unit-area capacitance of a symmetric ITO/IL/ITO structure in the sub-MHz spectrum. Prior to the measurement, an IL (of 0.3 μL) droplet is sandwiched between two ITO-coated PET films, of which the conductive ITO layer is 100 nm in thickness. Under an AC excitation voltage of 0.5 V, the device is connected to the LCR meter in a bipolar configuration. Fig. 2 plots the frequency responses of the EDL for the ionic droplets on both hydrophobic-modified and unmodified electrode surfaces. As can be seen, the EDL exhibits the maximal unit-area capacitance around 10 μF cm−2 at DC, and slowly decreases with the rise in frequency until 20 kHz (6 μF cm−2). It then drastically declines beyond this turning point, which is mainly attributed to the low-frequency dispersion of ionic conductions.26,55 Interestingly, the modified ITO surface expresses a slightly higher unit-area capacitance than that of the unmodified one, possibly due to removal of a native oxide layer by the plasma treatment and improved interactions between ionic and electronic charges.56

Fig. 2 Frequency dependence of the interfacial EDL capacitance at the electronic–ionic interface. Blue triangles indicate the measurement results from IL on the original ITO/PET surface, while the red ones show those measured from IL on the hydrophobic-modified ITO/PET surface.

Lab Chip

Lab on a Chip

Theoretical analysis The device sensitivity of the iontronic microdroplet sensor can be modelled both mechanically and electrically. As mentioned previously, under the external load, the suspended membranes deform elastically over the droplets, and accordingly, the droplet–electrode contact experiences circumferential expansion. The measured EDL capacitance can be directly related to the area of the droplet–electrode contact, as the invariant unit-area capacitance can be experimentally determined. On the other hand, the mechanical deformation of the membrane can be well defined in the classic mechanic theory.57 It is worth noting that the interfacial capacitive sensing principle offers an ultrahigh capacitive sensitivity, which is more than a thousand times greater than that of the solid-state counterpart, arising mainly from the nanoscopic charge separation in EDL, yielding ultrahigh overall device sensitivity.17,18,27,28 The relationship between the measurable capacitive change (ΔC) and the contact pressure applied (ΔP) can be derived from the new interfacial capacitive sensing principle (see eqn S1–S4, ESI† for detailed mathematical derivations):  P  P 2  C  C0       K  K  

(1)

where C0 indicates the initial capacitance, K = 5ET 3h/(1 − υ 2)a 4 is a constant derived from the design parameters, including the width (a) and height (h) of each sensing cell, and the membrane properties, including Young's modulus E, thickness T and Poisson ratio υ. The gravitational effect has been neglected in our consideration, as the microdroplet dimensions are considerably less than the capillary length (of ~1.8 mm).58 Experimental characterization Fig. 3 shows the experimental measurements of the device sensitivity with various geometrical designs (i.e., spatial resolutions and membrane thicknesses), in which the measurement results (dots) with the corresponding fitting curves (dashed lines) are plotted against the theoretical predictions from eqn (1) (solid lines). As a result, the device sensitivity can be calculated from the slope of each ΔP–ΔC curve. As expected, the sensitivity exhibits a strong dependence (minor 4th power) on the spatial resolution. As shown in Fig. 3a, by varying the spatial resolution from 1 mm to 3 mm with a constant membrane thickness of 75 μm, the sensitivity can be improved from 3.9 pF kPa−1 to 433.7 pF kPa−1 (a greater than 100-fold increase), and the device achieves the highest sensitivity (of 77.7 pF kPa−1) with a large initial capacitance (of 2.2 nF), in comparison with the highest reported values in the literature; a sensitivity of 3.2 pF MPa−1 and initial capacitance of 14 pF at a spatial resolution of 2 mm, to the best of our knowledge.17 In addition, the minimal detectable pressure of 33 Pa is characterized for the sensor with the highest sensitivity. Moreover, the membrane thickness plays another notable role in the device performance, as the sensitivity is inversely

This journal is © The Royal Society of Chemistry 2014

View Article Online

Lab on a Chip

Paper

Published on 31 January 2014. Downloaded by University of Illinois at Chicago on 04/02/2014 22:51:02.

Response time Experiments have been conducted to characterize the response time of the IMA devices. A pulsed contact pressure (of ~1.4 kPa) in the frequency range of 1 Hz to 100 Hz has been applied to the device surface through an electromagnetically driven pin actuator.41 Both the driving voltages to the actuator and the capacitive changes are recorded. As shown in Fig. 4a, the capacitive changes of the sensor repeats in the same frequency to the corresponding voltages applied to the pin actuator, suggesting that the sensor can respond to the pressure at frequencies of up to 100 Hz. It is worth noting that the distortion of the recorded pressure signals is likely attributed to the open-loop operation of the load applied by the actuator (i.e., the rapid rise edge and slow recovery phase of the capacitive readings).59 Repeatability tests and influence of temperature

Fig. 3 Experimental investigations of the device sensitivity. a, The spatial resolution (by varying the pixel sizes from 1 to 3 mm with a constant membrane thickness of 75 μm). b, The membrane thickness (by altering the thickness from 75 to 175 μm at a fixed pixel size of 2 mm), where the measurement results (dots) with the corresponding fitting curves (dashed lines) are plotted against the theoretical predictions from eqn (1) (solid lines).

related to the 3rd power of the thickness. As plotted in Fig. 3b, by adjusting the membrane thickness from 75 μm to 175 μm, with a fixed spatial resolution of 2 mm, the thinner membrane (75 μm thick) shows a higher sensitivity of 77.7 pF kPa−1, while the membranes of the thicker devices (175 μm thick) exhibit a lower sensitivity of 7.8 pF kPa−1. Furthermore, the targeted dynamic range can be tuned by the geometrical constraints. For instance, in the most sensitive design (3 mm in resolution and 75 μm in thickness), the maximal pressure is around 7 kPa, while in the design of 1 mm in resolution and 75 μm in thickness, the maximal pressure can be extended up to 200 kPa. When the applied pressure goes beyond the measurement range, the response could become highly non-linear and saturated. Overall, the spatial resolution and the membrane thickness of the IMA could be the determining factors in the sensor performance (i.e., device sensitivity and dynamic range), allowing highly customizable sensors for a wide range of specifications and applications.

This journal is © The Royal Society of Chemistry 2014

To investigate the mechanical reliability and robustness of the IMA devices, repeatability tests were conducted by recording the capacitive changes of a single sensing element as a function of press-and-release cycles. As shown in Fig. 4b, the sensor maintains relatively constant capacitive changes with less than 3% variation even after 20 000 cycles of pressure/force loads, illustrating the mechanical robustness and reliability of the IMA sensing devices. Moreover, the environmental thermal influence on the device electrical performance has been investigated. In principle, both the unit-area EDL capacitance and the volume of the IL droplets can be affected by the temperature, resulting in changes in the overall device capacitance.40 Fig. 4c illustrates the frequency spectra of the initial capacitance with a liquid–solid contact area of 4.2 mm2 over different temperatures from 5 °C to 50 °C. As can be seen, the temperature variation has a minor impact on the interfacial capacitance, i.e., less than 10% increase over the tested temperature range. In addition, the minimal capacitive change over the temperature fluctuation was observed at a frequency of 1 kHz (a 4% change in total), which could be taken into consideration in the IMA operation. In addition, any bending or stretching could likely result in a change in the initial capacitance as the flexible membranes and the droplets underneath are mechanically deformed. In such a case, we would need to adjust the sensing curve with a new initial setting (i.e., initial capacitance) for the subsequent pressure and force measurements. Surface topology mapping To demonstrate the utility of the iontronic devices, we have applied the IMA sensors to resolve surface topology and to record dynamic arterial pulses. The two IMA devices have been devised to map the static surface topology, 6 × 6 and 12 × 12 arrays with pixel resolutions of 1.5 mm and 2.0 mm, respectively. By placing a polymeric stamp and a weight (of 363 g) on the top of the surface, the capacitance value of each sensing unit can be scanned and processed with a readout circuitry (Fig. S2, ESI†). Fig. 5 shows the surface topology

Lab Chip

View Article Online

Published on 31 January 2014. Downloaded by University of Illinois at Chicago on 04/02/2014 22:51:02.

Paper

Lab on a Chip

Fig. 4 Characterization of the time-resolved response, mechanical repeatability and temperature influence of the IMA devices. a, The timeresolved sensor response measurements under repetitive mechanical loads over frequencies from 1 Hz to 100 Hz (red curves indicate the input voltage to drive the electromagnetic pin actuator, and the blue curves are the output capacitance measured from a single sensing unit of the IMA device). b, Capacitive changes as a function of repetitive cycles of external pressure (>20 000), suggesting the mechanical reliability and robustness of the IMA device. c, The influence of temperature variation. The frequency spectra of the initial capacitance of the sensing devices with the solid–liquid contact area of 4.2 mm2 are measured under the temperatures varying from 5 °C to 50 °C, where the corresponding the capacitive changes are found to be less than 10%.

measurement results (Fig. 5a–b) and the corresponding stamps made of PDMS elastomer (Fig. 5c), from which the pressure distribution has been clearly resolved. The accurate mapping of the spatial pressure distribution is highly reliant on the large EDL capacitance. Unlike the capacitance in classic solid-state capacitive sensors, the novel interfacial capacitance in each sensing element is of the order of a few nF, allowing to largely reduce interference from the environment, e.g. stray capacitance and electrical field from the neighbouring sensing units.60 Although the pixel resolution is currently limited by the printed droplet size, further improvement in microdroplet dispensing can obtain a higher sensor resolution. To further extend the flexibility and

Lab Chip

adaptability of the IMA devices to artificial tactile applications, we have configured a sensor array to detect fine surface topology, such as Braille letters. As can be seen in Fig. 5d, the custom IMA consists of 2 × 3 pixels with a spatial resolution of 2.3 mm (to match with the standard Braille letters), and can be worn in a fingertip set up. For fingertip reading of Braille texts, a gentle contact pressure is applied to the text surface by the finger. The raised dotted impressions of each Braille character cause membrane deformation in the corresponding droplet sensing units, which can be subsequently detected by the changes in the interfacial capacitance. As shown in the bottom panel of Fig. 5d, using the fingermounted IMA device, the letter of “BRAILLE” has been

This journal is © The Royal Society of Chemistry 2014

View Article Online

Published on 31 January 2014. Downloaded by University of Illinois at Chicago on 04/02/2014 22:51:02.

Lab on a Chip

Paper

Fig. 5 Application of the IMA devices to surface topology mapping. a, A 6 × 6 pixel array and b, a 12 × 12 pixel array are built to map the surface topology from c, a plastic stamp (scale bar: 1 cm in a–c). d, A 2 × 3 IMA array, mounted onto a fingertip, is configured to resolve raised dots of the “BRAILLE” characters in a Braille textbook. (Note: a red dye is introduced to the transparent ionic droplet for ease of observation).

successfully resolved, in which each pressure reading is converted to a digital colorimetric scale. Digital recording of the tactile sensing can be further processed and transmitted into audible readings, and thus, it can be of potential use for Braille education for visually impaired patients.

Wrist pulse recording Furthermore, we have utilized the ultrahigh device sensitivity and rapid response time of the iontronic droplet sensors for non-invasive cardiovascular pressure recording. An IMA device consisting of a 5 × 5 array with a spatial resolution of 3 mm was positioned in contact with the skin above the radial artery and fixed with a plastic wristband (Fig. 6a). Realtime pulse recording was performed by scanning all the sensing elements covering a skin area of 15 × 15 mm2 at the sampling frequency of 1 kHz in each unit. The IMA device enables two important functions in pulse recording. First of all, the sensor spatially maps the pulse on the skin surface, from which the sites of the maximal pressure variations can be located. Comparing the pressure mapping results (Fig. 6b) with the sensor position (Fig. 6a), the pressure sensing units right above the radial artery provide the highest capacitive recordings (marked as I, II, III) as expected, and thus, closely reflect the cardiovascular pressure readings using the tonometry principle.61 In the following step, the pressure waveforms are continuously tracked from these optimal sensing positions. Fig. 6c shows the continuous pulse recordings from the three pixels. As can be seen, the maximal pulse variation is around 1.2 kPa recorded by Sensor II. Fig. 6d provides a close-up analysis of a single cardiac cycle, which has been characterized into three peaks (P1, P2 and P3). These maxima are caused by traveling waves of the systolic phase and diastolic phase of blood pressure conducted in the

This journal is © The Royal Society of Chemistry 2014

elastic cardiovascular vessels62 (Movie S1, ESI†). Clinically significant parameters, such as the radial augmentation index, AI, (=P2/P1) and the reflection index, RI, (=P3/P1), can be directly extracted and computed from the maximal pulse recordings, which can be potentially used to screen arterial compliance and arteriolar tone.63,64 Moreover, the radial pulse waveforms recorded at the optimal sites can be further processed to estimate the central aortic pressure and cardiac output, which reflect important cardiovascular events and health states.65 Though a similar measurement has been conducted recently through a single-channel capacitive sensor, the IMA offers the combined advantage of simultaneous pressure mapping of the optimal recording area and continuous tracking of the blood pressure waveform, in addition to its flexible transparent packaging.18 In this fashion, the IMA can serve as a flexible sensing device that is highly attractive for the emerging wearable health monitoring applications, in comparison with conventional invasive cardiovascular monitoring.

Conclusions An iontronic tactile sensing array, referred to as an iontronic microdroplet array (IMA), has been reported using the novel droplet-enabled interfacial capacitive sensing principle. As an emerging alternative to the existing solid-state capacitive sensors, the IMA utilizes a highly capacitive EDL interface upon the electrode–electrolyte contact as the sensing element to achieve ultrahigh mechanical-to-electrical sensitivity (of 0.43 nF kPa−1) and fine pressure resolution (of 33 Pa) in a 3 × 3 × 0.2 mm3 package, in comparison with the highest reported sensitivity of 0.8 nF kPa−1 with a much larger footprint (of 6 × 6 mm2). The device sensitivity and dynamic range are highly customizable regarding various design parameters,

Lab Chip

View Article Online

Published on 31 January 2014. Downloaded by University of Illinois at Chicago on 04/02/2014 22:51:02.

Paper

Lab on a Chip

Fig. 6 Real-time wrist pulse measurements. a, Illustration of a transparent IMA device of a 5 × 5 array (inset) with each pixel size 3 × 3 mm2 embedded in a flexible wrist band for non-invasive pressure wave recording. (Note: a red dye is introduced into the transparent ionic droplet for ease of observation). b, The spatial distributions of the pulse intensities mapped by the IMA matrix. The highest pressure variation readings, marked as I, II and III, are located above the radial artery, corresponding to the positions of the three units in a. c, Time-resolved strongest pulses recorded by the marked sensing units in a and b. d, A close-up view of one pulse signal recorded by sensor II in c, in which P1, P2 and P3 represent the three consecutive peaks of the recorded pressure wave in each cardiovascular cycle.

e.g., spatial resolution, membrane thickness and chamber height, and therefore, it can be readily configured for various pressure sensing applications. In addition, the fluidic nature of the sensors enables rapid mechanical responses (of the order of a few milliseconds). Moreover, the IMA sensor exhibits high repeatability (less than 3% variation in capacitive readings) over more than 20 000 external load cycles. In addition, the simple device is optically transparent and can be produced on a massive scale with high reliability yet low cost. To illustrate the utility of the IMA sensor, it has been successfully applied to map surface pressure distributions as well as to scan Braille characters in a fingertip configuration. We further extend the ultrahigh device sensitivity and rapid response time of the iontronic droplet sensor to non-invasive cardiovascular pressure recording, from which the optimal pulse taking position has been determined and several key physiological indicators can be extracted and analysed in real time. Overall, the iontronic microdroplet sensors

Lab Chip

inspired by our sense of touch could potentially enable a highly transformative platform of tactile sensing for a wide range of emerging applications, including robotics, medical prosthetics, surgical instruments, video gaming and wearable computing.

Acknowledgements This work is in part supported by the National Science Foundation (ECCS-0846502 and ECCS-1307831) to TP. BN and RL acknowledge the fellowship support from China Scholarship Council (CSC). Authors would also like to acknowledge the samples of ITO-coated membranes generously provided by Southwall Technology and Mianyang Prochema Commercial Co., Ltd. In addition, we would like to thank Yuzhe Ding for help with the microdroplet printing setup and Xing Dong for assistance in the circuit design.

This journal is © The Royal Society of Chemistry 2014

View Article Online

Lab on a Chip

Published on 31 January 2014. Downloaded by University of Illinois at Chicago on 04/02/2014 22:51:02.

Notes and references 1 T. R. Pan and W. Wang, Ann. Biomed. Eng., 2011, 39, 600–620. 2 E. G. Bakhoum and M. H. M. Cheng, J. Microelectromech. Syst., 2010, 19, 443–450. 3 T. Sugita and J. Kondoh, presented in part at the Sensors 2010 IEEE, 1–4 Nov., 2010. 4 H. S. Zeng and Y. Zhao, Sens. Actuators, B, 2011, 154, 33–40. 5 Z. Barikbin, M. T. Rahman, P. Parthiban, A. S. Rane, V. Jain, S. Duraiswamy, S. H. S. Lee and S. A. Khan, Lab Chip, 2010, 10, 2458–2463. 6 D. Witters, K. Knez, F. Ceyssens, R. Puers and J. Lammertyn, Lab Chip, 2013, 13, 2047–2054. 7 S. J. R. Mohsin, I. Tiwana and N. H. Lovell, Sens. Actuators, A, 2012, 179, 17–31. 8 K. Takei, T. Takahashi, J. C. Ho, H. Ko, A. G. Gillies, P. W. Leu, R. S. Fearing and A. Javey, Nat. Mater., 2010, 9, 821–826. 9 T. Someya, T. Sekitani, S. Iba, Y. Kato, H. Kawaguchi and T. Sakurai, Proc. Natl. Acad. Sci. U. S. A., 2004, 101, 9966–9970. 10 C. Pang, G. Y. Lee, T. I. Kim, S. M. Kim, H. N. Kim, S. H. Ahn and K. Y. Suh, Nat. Mater., 2012, 11, 795–801. 11 B. C. K. Tee, C. Wang, R. Allen and Z. N. Bao, Nat. Nanotechnol., 2012, 7, 825–832. 12 J. Engel, J. Chen, Z. F. Fan and L. Chang, Sens. Actuators, A, 2005, 117, 50–61. 13 C. Y. Li, P. M. Wu, S. Lee, A. Gorton, M. J. Schulz and C. H. Ahn, J. Microelectromech. Syst., 2008, 17, 334–341. 14 W. Wu, X. Wen and Z. L. Wang, Science, 2013, 340, 952–957. 15 H. K. Kim, S. Lee and K. S. Yun, Sens. Actuators, A, 2011, 165, 2–7. 16 H.-K. Lee, S.-I. Chang and E. Yoon, J. Microelectromech. Syst., 2006, 15, 1681–1686. 17 D. J. Lipomi, M. Vosgueritchian, B. C. Tee, S. L. Hellstrom, J. A. Lee, C. H. Fox and Z. Bao, Nat. Nanotechnol., 2011, 6, 788–792. 18 G. Schwartz, B. C.-K. Tee, J. Mei, L. Appleton, D. Hwan Kim, H. Wang and Z. Bao, Nat. Commun., 2013, 4, 1859. 19 M. Y. Cheng, C. L. Lin, Y. T. Lai and Y. J. Yang, Sensors, 2010, 10, 10211–10225. 20 M. Zirkl, A. Sawatdee, U. Helbig, M. Krause, G. Scheipl, E. Kraker, P. A. Ersman, D. Nilsson, D. Platt, P. Bodo, S. Bauer, G. Domann and B. Stadlober, Adv. Mater., 2011, 23, 2069–2074. 21 A. D. Mazzeo, W. B. Kalb, L. Chan, M. G. Killian, J. F. Bloch, B. A. Mazzeo and G. M. Whitesides, Adv. Mater., 2012, 24, 2850–2856. 22 S. C. B. Mannsfeld, B. C. K. Tee, R. M. Stoltenberg, C. V. H. H. Chen, S. Barman, B. V. O. Muir, A. N. Sokolov, C. Reese and Z. N. Bao, Nat. Mater., 2010, 9, 859–864. 23 J. H. Cho, J. Lee, Y. Xia, B. Kim, Y. Y. He, M. J. Renn, T. P. Lodge and C. D. Frisbie, Nat. Mater., 2008, 7, 900–906. 24 K. H. Lee, M. S. Kang, S. P. Zhang, Y. Y. Gu, T. P. Lodge and C. D. Frisbie, Adv. Mater., 2012, 24, 4457–4462.

This journal is © The Royal Society of Chemistry 2014

Paper

25 M. Kawasaki and Y. Iwasa, Nature, 2012, 489, 510–511. 26 H. T. Yuan, H. Shimotani, A. Tsukazaki, A. Ohtomo, M. Kawasaki and Y. Iwasa, Adv. Funct. Mater., 2009, 19, 1046–1053. 27 M. J. Madou, Fundamentals of Microfabrication: The Science of Miniaturization, 2002. 28 H. K. Lee, J. Chung, S. I. Chang and E. Yoon, J. Microelectromech. Syst., 2008, 17, 934–942. 29 J. H. Cho, J. Lee, Y. He, B. Kim, T. P. Lodge and C. D. Frisbie, Adv. Mater., 2008, 20, 686–690. 30 J. T. Ye, S. Inoue, K. Kobayashi, Y. Kasahara, H. T. Yuan, H. Shimotani and Y. Iwasa, Nat. Mater., 2010, 9, 125–128. 31 M. Armand, F. Endres, D. R. MacFarlane, H. Ohno and B. Scrosati, Nat. Mater., 2009, 8, 621–629. 32 B. Li, S. Dalgleish, Y. Miyoshi, H. Yoshikawa, M. M. Matsushita and K. Awaga, Appl. Phys. Lett., 2012, 101, 17. 33 C. Y. Wu, W. H. Liao and Y. C. Tung, Lab Chip, 2011, 11, 1740–1746. 34 N. Dossi, R. Toniolo, A. Pizzariello, E. Carrilho, E. Piccin, S. Battiston and G. Bontempelli, Lab Chip, 2012, 12, 153–158. 35 K. Ohsawa, H. Takahashi, K. Noda, T. Kan, K. Matsumoto and I. Shimoyama, A gas sensor based on viscosity change of ionic liquid, Micro Electro Mechanical Systems, IEEE 24th International Conference, 2011, 525528. 36 K. Oda, H. Takao, K. Terao, T. Suzuki, F. Shimokawa, I. Ishimaru and F. Oohira, Sens. Actuators, A, 2012, 181, 61–69. 37 S. W. Zhao, H. L. Cong and T. R. Pan, Lab Chip, 2009, 9, 1128–1132. 38 S. W. Zhao, A. Chen, A. Revzin and T. R. Pan, Lab Chip, 2011, 11, 224–230. 39 Y. Z. Ding, S. Garland, M. Howland, A. Revzin and T. R. Pan, Adv. Mater., 2011, 23, 5551–5556. 40 B. Q. Nie, S. Y. Xing, J. D. Brandt and T. R. Pan, Lab Chip, 2012, 12, 1110–1118. 41 Y. Z. Ding, E. Huang, K. S. Lam and T. R. Pan, Lab Chip, 2013, 13, 1902–1910. 42 A. Guerfi, M. Dontigny, P. Charest, M. Petitclerc, M. Lagace, A. Vijh and K. Zaghib, J. Power Sources, 2010, 195, 845–852. 43 A. Lewandowski, A. Olejniczak, M. Galinski and I. Stepniak, J. Power Sources, 2010, 195, 5814–5819. 44 A. Lewandowski and A. Swiderska-Mocek, J. Power Sources, 2009, 194, 601–609. 45 L. Scheflan and M. B. Jacobs, The Handbook of Solvents, Robert E. Krieger Publishing Company, Huntington, New York, 1973. 46 A. Fernicola, B. Scrosati and H. Ohno, Ionics, 2006, 12, 95–102. 47 X. L. Suojiang Zhang, Q. Zhou, X. Li, X. Zhang and S. Li, Ionic Liquids: Physicochemical Properties, 2009. 48 J. M. S. S. Esperanca, J. N. C. Lopes, M. Tariq, L. M. N. B. F. Santos, J. W. Magee and L. P. N. Rebelo, J. Chem. Eng. Data, 2010, 55, 3–12. 49 O. Aschenbrenner, S. Supasitmongkol, M. Taylor and P. Styring, Green Chem., 2009, 11, 1217–1221.

Lab Chip

View Article Online

Published on 31 January 2014. Downloaded by University of Illinois at Chicago on 04/02/2014 22:51:02.

Paper

50 Y. Yoshida, K. Muroi, A. Otsuka, G. Saito, M. Takahashi and T. Yoko, Inorg. Chem., 2004, 43, 1458–1462. 51 H. Olivier-Bourbigou and L. Magna, J. Mol. Catal. A: Chem., 2002, 182, 419–437. 52 M. Galinski, A. Lewandowski and I. Stepniak, Electrochim. Acta, 2006, 51, 5567–5580. 53 J. G. Huddleston, A. E. Visser, W. M. Reichert, H. D. Willauer, G. A. Broker and R. D. Rogers, Green Chem., 2001, 3, 156–164. 54 K. B. Oldham, J. Electroanal. Chem., 2008, 613, 131–138. 55 S. Ono, S. Seki, R. Hirahara, Y. Tominari and J. Takeya, Appl. Phys. Lett., 2008, 92, 103313. 56 D. J. Milliron, I. G. Hill, C. Shen, A. Kahn and J. Schwartz, J. Appl. Phys., 2000, 87, 572–576. 57 S. P. Timoshenko and S. Woinowsky-Krieger, Theory of Plates and Shells, 1959.

Lab Chip

Lab on a Chip

58 P.-G. de Gennes, F. Brochard-Wyart and D. Quere, Capillarity and Wetting Phenomena: Drops, Bubbles, Pearls, Waves, Springer, 2003. 59 A. Boglietti, M. Chiampi, M. Tartaglia, S. Cattaneo, M. Contessa and A. Garramone, IEEE Trans. Magn., 1989, 25, 3587–3589. 60 W. H. Ko, Sens. Actuators, 1986, 10, 303–320. 61 C. C. Tyan, S. H. Liu, J. Y. Chen, J. J. Chen and W. M. Liang, IEEE Trans. Biomed. Eng., 2008, 55, 288–297. 62 I. B. Wilkinson, N. H. Mohammad, S. Tyrrell, I. R. Hall, D. J. Webb, V. E. Paul, T. Levy and J. R. Cockcroft, Am. J. Hypertens., 2002, 15, 24–30. 63 S. Motohiro and K. Kazuomi, Ther. Adv. Cardiovasc. Dis., 2008, 2, 25–35. 64 S. Wallberg-Jonsson, K. Caidahl, N. Klintland, G. Nyberg and S. Rantapaa-Dahlqvist, Scand. J. Rheumatol., 2008, 37, 1–5. 65 S. A. Hope, D. B. Tay, I. T. Meredith and J. D. Cameron, Diabetes Care, 2004, 27, 746–751.

This journal is © The Royal Society of Chemistry 2014

Iontronic microdroplet array for flexible ultrasensitive tactile sensing.

An iontronic microdroplet array (IMA) device, using an ultra-large interfacial capacitance at the highly elastic droplet-electrode contact, has been p...
3MB Sizes 2 Downloads 0 Views