An Ionic Liquid-Based Stretchable Sensor for Measuring Normal and Shear Force

Abstract

Soft and stretchable force sensors are widely used for health monitoring, robotics, prosthetics, and other applications. Soft force sensors with the capability of measuring both normal and shear force could offer even greater functionality and provide more information, particularly in the field of biomechanics. In this work, a new solid-state force sensor is proposed that can measure both normal and shear forces at the same time. The soft and stretchable sensor was fabricated using an ionic liquid (IL)/polymer network. Two separate IL-based polymer membranes were used to detect normal and shear forces. Sensor architecture and electrical wiring for normal, shear, and combined sensing were developed, and various material compositions for different sensor layers were investigated to find the combination that could achieve the optimum sensor performance. A basic material formulation for carbon nanotube-based electrodes, the IL/polymer network, and polymeric insulation layers was proposed. To configure a combined (normal and shear) sensor, separate sensors for normal and shear deformations were first designed and investigated. Later, a combined sensor was fabricated using a mold via screen printing, photocuring, and thermal curing. The combined sensor was evaluated under different force conditions. The results show that the sensor can reliably measure normal and shear forces. Moreover, the findings demonstrate a way to successfully modulate the sensitivity for normal and shear sensing by varying the material composition or geometric configuration, which provides flexibility for application-specific designs.

Introduction

There has been a growing interest in soft and stretchable force sensors over the last few decades. Their applications in soft robotics,¹ wearable devices,² and health monitoring³ are spurring new developments in sensor research. Soft sensors that can be integrated into automobiles,⁴ packaging,⁵ medical devices,⁶ and protective kits⁷ are also being studied. Rigid sensors are not able to achieve the necessary mechanical pliability needed for many applications where bending and flexing are required; when used in wearable devices, these sensors can also cause discomfort. Stretchable and flexible sensors overcome many limitations of rigid sensors as they can conform and stretch without obstructing the primary movement of a device.⁸ In particular, soft sensors are reported in numerous biomedical applications, including simulated electronic skin,^9,10 biological assessment devices,^11,12 and drug delivery platforms.^13,14

Flexible and stretchable sensors are also crucial for various robot applications that require small-scale devices,¹⁵ the ability to grip objects,¹⁶ and the ability to operate in underwater¹⁷ and harsh environments.¹⁸ The primary sensing mechanism of the soft sensor is typically based on the mechanical strains observed by the sensing elements.^19,20 Various elastomers were used in the fabrication of soft sensors due to their viscoelastic properties.²¹

Ionic liquids (ILs), which are largely made of organic ions and are liquid at room temperature, have recently been used to fabricate solid-state soft sensors.^22,23 Incorporating ILs into a polymer network induces pressure sensitivity in the polymer, and the electrical resistivity of an IL/polymer network changes with applied deformation^24,25; several pressure sensors have been proposed by applying this principle.^26,27 ILs are suitable for pressure sensors because of their high ionic conductivity and electrochemical stability,²⁸ and they can be easily blended with different polymers to obtain various mechanical properties. Since a small amount of IL can generate sensitivity in the IL/polymer network, the network is predominantly able to retain the original mechanical properties of the polymer.

Therefore, it is possible to fabricate sensors with different hardness, stretchability, and strength while utilizing a low concentration of the IL. ILs have also been mixed with monomers to develop a sensitive thermoset polymer.²⁹ Moreover, the IL content in the sensors is a controllable parameter, since the sensitivity depends on the IL ratio in the IL/polymer network. This can play a very important role in sophisticated applications in biotechnology and biomechanics.

Numerous studies have been conducted on soft pressure sensors.³⁰ Carbon nanotube (CNT)-based polymer composites have been widely proposed for fabricating piezoresistive stretchable sensors.³¹ Recently, with the incorporation of ILs, elastomeric soft sensors were developed; these sensors provide another avenue of research. Although the proposed IL-based polymeric sensors showed reliable performance,^32,33 they were mostly characterized in tensile, compressive, or bending actions. Moreover, while several soft shear sensors have been reported, their sensitivities and other properties are not easily modifiable.^34,35

There are opportunities for advancement on soft polymeric sensors for simultaneous normal and shear sensing. Sensors with the capability of measuring normal and shear force could be vital for human–machine interfacing, and soft shear sensors could be crucial for robotics, wearables, and prosthetics. Biomedical applications can be a major area to implement these sensors. For example, clinical gait analysis to study walking patterns could benefit from the integration of a shear sensor, as shear force data can offer more information about the biomechanical process, and it can be useful in the assessment of podiatric conditions. It can also help in gaining a better understanding of tire mechanics and can aid in the design of sporting equipment and consumer health monitoring devices.

Quite a few studies have been conducted on normal and shear force sensors. However, many of the suggested sensors do not have adequate flexibility and stretchability for wearable applications.^34,36,37 In addition, a significant number of studies have been published on soft and flexible sensors that are capable of measuring both shear and normal forces.^38–40 These studies mostly proposed multilayered sensor architectures, and the sensors performed reliably. One aspect that many of the proposed studies lack is the development of controllable parameters to achieve various sensor specifications (different sensitivities, mechanical properties, and chemical properties) depending on the application.

In the present study, an IL-based soft sensor is proposed that measures both the normal and shear forces. In addition to providing electrochemical stability, introducing IL opens up a controllable parameter in the sensor. ILs could be incorporated at different concentrations with different base polymers to achieve various electrical and mechanical properties to allow the dynamic range of the IL-based sensor to be modulated.³² To develop the proposed sensor, various acrylate monomer compositions were investigated, and additives to the monomer (such as a photoinitiator and crosslinkers of various weight ratios) were also examined for reliable viscoelastic properties. The sensor was fabricated with a shear force-sensitive and a normal force-sensitive layer by incorporating ILs into the base monomer formulation and was evaluated for different sensor configurations and load conditions. The fabricated sensor was found to perform reliably under normal and shear forces.

The sensors in the existing studies in the literature provide very limited manufacturing flexibility. Generally, they are developed in simple two-dimensional geometries.^38,39 For many of the existing sensors, any geometric complexity (especially in three dimensions) would be challenging to handle. While advanced manufacturing methods, such as additive manufacturing (AM), can be used to fabricate structures with complex geometries and can provide customizability, many of the proposed sensors would face challenges in AM due to material incompatibility. The ability to implement versatile manufacturing techniques for sensor fabrication provides design and manufacturing flexibility. Although the proposed sensor in this report was fabricated using a mold and screen-printing process, it can also be 3D printed. The 3D printing of similar IL-based sensors for normal force measurement was previously reported by the authors.^41,42

In the future, the authors plan to work on 3D printing of the proposed sensor. While 3D printing would provide customizability and manufacturability for complex geometries, normal and shear data from the soft sensor could provide valuable insights into the areas of human–machine interfaces and biomechanics. This study is believed to open many doors for future research and applications. The unique contributions of this work compared with others in the literature include the following:

Mechanical stretchability and flexibility of the sensor to make it compatible with wearable devices.

Electrochemical stability provided by IL that results in reliable sensor performance.

Controllable parameters to adjust the electrical, mechanical, and chemical properties by varying the IL concentration and base polymer.

Manufacturing versatility, as AM provides design flexibility and could be used for the fabrication of sensors with complex geometries.

Materials and Methods

Sensor design

The fabricated soft and stretchable sensor, which is shown in Figure 1b, is capable of detecting normal and shear forces. Figure 1a and c shows a 3D model and an exploded view, respectively, of the multilayer multimaterial sensor. The sensor has insulation layers at the top and bottom, two intermediate pressure-sensitive layers made from a blend of an IL and polymer, and electrodes made from a CNT-based polymer composite.

FIG. 1.

Sensing technology: (a) 3D model of a multilayer sensor; (b) fabricated sensor; (c) exploded view of the 3D model (IL-based layers are used for detecting deformations in the Z and X axes); (d) sectional view (A–A in the 3D model); (e) X and Z taxels used to detect shear and normal deformations, respectively, and simplified equivalent electrical circuits for both taxels. IL, ionic liquid.

After polymerization, all of the layers become flexible and stretchable. In each sensor are three electrodes. The first two are placed in the same horizontal layer with an IL/polymer network between them to create a horizontal tactile pixel (taxel), as indicated by the purple dashed lines in Figure 1d and e; this taxel is referred to as an X-taxel. The third electrode is placed below one of the electrodes in the X-taxel with an IL/polymer network between them, as shown by the red dashed lines in Figure 1d and e; this vertical taxel is referred to as a Z-taxel.

The electrical resistance of the IL/polymer network varies with the applied deformation or force, and the sensitivity of the IL/polymer network depends on the IL concentration in the network. As a more sensitive IL/polymer network is needed to detect the shear force, different IL concentrations were applied to the IL/polymer networks: the shear force-sensitive IL-based layer (denoted as the X-sensitive layer) and the normal force-sensitive IL-based layer (denoted as the Z-sensitive layer), as shown in Figure 1c. Both taxels were connected to a simple potential divider to collect voltage readings and detect the applied force, as shown in the simplified wiring diagram in Figure 1e. The voltage output V₂ primarily shows the X-taxel response (denoting the applied shear), while the voltage output V₁ mainly shows the Z-taxel response (denoting the applied normal force).

Materials

A commercially available photocurable resin, TangoPlus (Stratasys, Eden Prairie, MN), has previously been used to fabricate sensors for normal force detection,³² and it showed reliable performance when used for stretchable force sensors (Supplementary Table S1). In this study, sensors were fabricated from six different material formulations (Form1 to Form6, as listed in Table 1), in which different ratios of a photoinitiator and a crosslinking agent were incorporated into acrylate-based monomers that served as a base polymer matrix (Supplementary Fig. S1). Preliminary investigation and mechanical testing were conducted on all six formulations (the results are reported in the Results section and the Supplementary Figs. S2, S3, S4, and S5).

Table 1.

Formulations for Different Basic Elastomers (All Components Are Described with wt.%)

Ingredient	Type	Form1	Form2	Form3	Form4	Form5	Form6
2-[[(Butylamino)carbonyl]oxy]ethyl acrylate	Monomer	73.0	74.0	73.1	74.6	83.8	83.0
Di(ethylene glycol) ethyl ether acrylate	Monomer	5.0	5.0	4.0	5.0	15.0	15.0
Isobornyl acrylate	Monomer	20.0	20.0	20.0	20.0	0.0	0.0
Phenylbis(2,4,6-trimethyl-benzoyl)phosphine oxide	Photoinitiator	0.4	0.4	0.4	0.4	0.4	0.4
Glycerol propoxylate (1PO/OH) triacrylate	Crosslinker	1.6	0.6	2.5	0.0	0.8	1.6

Of the six formulations, Form1 was selected as the primary material for the proposed sensor based on its sensing response and mechanical properties. The top and bottom insulation layers in Figure 1c were directly fabricated using Form1. An IL, 1-ethyl-3-methyl-imidazolium tetrafluoroborate (EMIMBF₄), was mixed with Form1 to prepare the IL/polymer blend for the X-sensitive and Z-sensitive IL-based layers. For the CNT-based electrodes, a CNT/polymer composite was prepared based on the Form1 composition, and 2 wt.% of a thermal initiator (TRIGONOX 125C75; Akzo Nobel Functional Chemicals LLC, Chicago, IL) was added to make the sensor thermally curable.

Fabrication

The sensor in this study was fabricated using an aluminum mold and involved ultraviolet (UV) curing of prepolymer liquids, screen-printing of the CNT-based electrodes, followed by thermal curing of the electrodes. The step-by-step process to fabricate a combined X-taxel/Z-taxel sensor is shown in Figure 2. First, the insulation material (Form1 prepolymer) was poured into the mold to make a thin layer (<1 mm in thickness). The prepolymer liquid was exposed to UV light to polymerize it and create a soft layer. Next, a photopaper mask was used to screen print a CNT-based electrode. The electrode was thermally cured by placing the mold in a convection oven at 80°C for 5 min. The IL/prepolymer blend was poured into the mold and UV-cured to make the Z-sensitive layer.

FIG. 2.

A step-by-step fabrication method of the sensor: (a) UV curing of the first insulation layer; (b) screen printing the first electrode layer using a photopaper mask; (c) thermal curing of the electrode; (d) UV curing of the Z-sensitive layer; (e) screen printing the second electrode layer using a photopaper mask; (f) thermal curing of the electrodes; (g) UV curing of the X-sensitive layer; (h) UV curing the final insulation layer; (i) sensor removed from the mold. UV, ultraviolet.

Electrodes for X-taxel were screen printed using another paper mask and were thermally cured in the oven at 80°C for 5 min. Next, another IL/prepolymer blend was poured and UV-cured for the X-sensitive layer. Finally, an insulation layer was created and UV cured, and the sensor was removed from the mold after cooling it to room temperature. The UV curing for all the layers was conducted by a UV lamp with a 365 nm wavelength for 60 s.

Experimental setup

The test setup for a combined X-taxel/Z-taxel sensor is shown in Figure 3. A Mark-10 ESM 303 motorized force testing stand with an M5-2 force gauge (Mark-10 Corporation, Copiague, NY) was used to control the vertical (Z) motion and to collect force data. For testing, the sensor was placed on a motorized linear stage (A-LSQ075AE01; Zaber Technologies, Vancouver, BC, Canada), which was used as an X-motion stage and was used to apply shear. The X-taxel and Z-taxel of the sensor were connected to a potential divider with an external resistor of 20 MΩ. A data acquisition system (DAQ, USB-6212; National Instruments, Austin, TX) and an external power supply (E3630A; Keysight Technologies, Santa Rosa, CA) were used to read the sensor response and the supply voltage, respectively. The supply voltage was maintained at 24 V during the test. The data acquisition was conducted using LabVIEW software.

FIG. 3.

Setup for testing the combined X-taxel/Z-taxel sensor under shear and normal force.

Results

Material formulation selection

To obtain the optimum base formulation, mechanical and sensing tests were performed on the six formulations mentioned in Table 1. The base formulations include three monomers, a photoinitiator, and a crosslinker in different weight percentages. The mechanical properties of the six formulations were obtained as well as the properties of a commercial photocurable resin, TangoPlus, which was used as a reference stretchable polymer. Figure 4c and d shows the tensile and hardness test results for all seven specimens. For the tensile test, dog bone-shaped specimens were fabricated using a mold according to the dimensions given in ASTM D638 for a Type V tensile test specimen. An Instron 5582 universal testing machine was used to conduct tensile testing at a crosshead speed of 100 mm/min, and the hardness of all the polymerized samples was measured using a Shore A durometer.

FIG. 4.

Results of different material tests. (a) The Form6-based sensor cracked under deformation, while (b) the Form1-based sensor did not crack; (c) tensile tests; (d) hardness test; (e) Z-taxel sensor performance with different materials under normal force; (f) X-taxel sensor performance with different materials under shear; (g) electrical conductivity of CNT/polymer composites with different percentages of CNT. CNT, carbon nanotube.

The specimens made from formulations without isobornyl acrylate (Form5 and Form6) showed lower strength than those of the other formulations, and they were observed to crack under deformation (Fig. 4a). The specimen prepared from Form4 (the formulation with no crosslinking agent) was too soft, resulting in a very long recovery time after elastic deformation.

Figure 4e shows the performance of a Z-taxel sensor under normal force, and Figure 4f shows the performance of an X-taxel sensor under applied shear for sensors fabricated from Form1, Form2, and Form3. For normal force, a 2-mm probe displacement (around 66% compressive strain) was applied on the Z-taxel sensor at a speed of 0.5 mm/s in the vertical (Z) direction. For shear, a 2-mm probe displacement was applied on an X-taxel sensor at a speed of 0.5 mm/s in the horizontal (X) direction with an initial compressive load of 2 N. While the sensor based on Form2 showed good sensitivity, it showed a long recovery time while unloading (especially for the shear test with the X-taxel sensor). The sensor based on Form3 showed lower sensitivity than those based on Form1 and Form2. Therefore, Form1 was selected as the base material for sensor fabrication for the subsequent experiments.

Figure 4b shows a sensor fabricated using Form1. Different weight percentages of CNT (1 wt.%, 3wt.%, 5wt.%, and 10 wt.%) were dispersed in Form1 to prepare conductive inks, and the inks were investigated to examine the conductivity and percolation threshold of the CNT-based electrodes. The effect of the CNT loading ratio on the electrical conductivity of the electrodes with and without strain (Fig. 4g) was investigated using screen-printed CNT/polymer composite strips with dimensions of 30.0 × 1.0 × 0.2 mm. Resistance was measured using a multimeter without any strain as well as under 30% compressive strain. The 5 wt.% CNT showed the optimum electrical resistance; thus, the 5 wt.% CNT ratio was used to prepare CNT/polymer composites for fabricating the electrodes for the proposed stretchable sensor.

Z-taxel sensor and applied normal force

For Z-taxel characterization, a sensor primarily for normal force detection was designed with a Z-taxel but no X-taxel, as shown in the 3D model in Figure 5a. The Z-taxel sensor had only one IL-based layer that was sandwiched between two electrodes, as can be seen in the exploded view of the sensor layers in Figure 5b. The photograph in Figure 5c shows the fabricated Z-taxel sensor. To test the sensor, a force gauge was mounted on a motorized linear stage that moved in the Z axis. A probe attached to the force gauge was used to apply force or strain on the sensor, and the displacement of the probe in the Z direction is denoted by δ_z. To apply normal compressive deformation, a δ_z of 2 mm was applied using Z-stage movement. Figure 5d shows the test setup for a normal compression test on the Z-taxel sensor.

FIG. 5.

Z-taxel sensor evaluation for applied normal force: (a) 3D model of the sensor; (b) exploded view that shows the sensor layers; (c) the fabricated sensor; (d) experimental setup with a motorized Z stage and a force gauge; (e) Z-taxel sensor performance for different IL concentrations; (f) Z-taxel sensor performance for different electrode distances.

First, sensors with different IL concentrations were investigated. Z-taxel sensors with 1 wt.%, 2 wt.%, and 5 wt.% of IL were fabricated from Form1 material. The sensor response for 66% compressive loading and unloading cycles is shown in Figure 5e. The sensor response was measured in terms of, where V is the initial voltage output from the sensor system and ΔV is the voltage deviation under deformation. While all three sensors showed reliable responses, the ΔV/V did not significantly vary for IL concentrations between 1 wt.% and 5 wt.%. Therefore, the IL concentration for the Z-taxels was maintained at 2 wt.% for the next set of experiments.

The distance between electrodes, L_z, was also investigated for the Z-taxels. Sensors were fabricated with L_z values of 0.5, 0.9, 1.8, and 2.7 mm. A fixed compressive force of 15 N was applied on the four sensors at a probe speed of 0.5 mm/s. Figure 5f shows the test results for the sensors with different IL layer thicknesses. As can be noticed from this figure, a shorter distance between electrodes will provide higher sensitivity in the Z-taxel. However, it is difficult to obtain a layer height of 0.5 mm with any consistency. Thus, the distance between electrodes was maintained as 0.9 mm for the Z-taxels in the proposed sensor in this study.

X-taxel sensor and applied shear force

X-taxel sensors with different IL concentrations and electrode distances were characterized in terms of their response to applied shear force. In the X-taxel sensor, two electrodes are placed in the same horizontal layer as the IL/polymer network (as shown in Fig. 6c), but no Z-taxel is included. The X-taxel sensor, which was designed primarily to respond to applied shear force, is shown in Figure 6b. The sensor for applying shear was glued to the base platform on one side and to the force probe on the other side. A force gauge was mounted on a motorized Z-stage, which was mounted on an X-stage (as shown in Fig. 6d). X-taxel sensors with IL concentrations of 5 wt.%, 10 wt.%, and 15 wt.% were evaluated.

FIG. 6.

X-taxel sensor evaluation for applied shear: (a) 3D model of the sensor; (b) fabricated sensor glued with the probe; (c) exploded view and sensor layers; (d) experimental setup with X and Z stage; (e) X-taxel sensor performance for different IL concentrations; (f) X-taxel sensor performance for different electrode distances.

To test the shear response, a probe displacement in the X direction (denoted by δ_x) of 2 mm was applied with an initial compressive load of 0.5 N. The X-stage was moved back and forth at a speed of 0.5 mm/s. Of the three X-taxel sensors tested, the sensor with 15 wt.% of IL showed the highest relative change in voltage. The sensor with 10 wt.% IL concentration also showed a reliable response, while the sensor with 5 wt.% IL showed significant electrical noise. To limit the cost of the proposed sensor, 10 wt.% of IL was selected for the X-taxel sensors used in the combined flexible sensor.

The effect of the distance between electrodes, L_x, on the X-taxel sensor was also examined. Sensors with 10 wt.% IL were fabricated with L_x values of 0.5, 1.0, 2.0, and 3.0 mm and were also tested using a δ_x of 2 mm. Figure 6f shows that ΔV/V decreases with the increase in distance between electrodes. The sensor with 0.5 mm L_x showed the highest relative change in voltage, and the sensor with 1.0 mm L_x demonstrated a reliable response. The sensors with larger distances between electrodes (L_x = 2.0 and 3.0 mm) were not sensitive.

Combined X-taxel/Z-taxel sensor

A combined X-taxel/Z-taxel sensor was fabricated for simultaneous normal and shear measurements. The combined sensor had two IL layers: one IL layer was fabricated using 10 wt.% IL and an electrode distance of 1.0 mm (for an X-taxel and shear detection), while the second layer was fabricated with 2 wt.% IL and an electrode distance of 0.9 mm (for a Z-taxel and normal force measurement). The circuitry of the combined X-taxel/Z-taxel sensor is shown in Figure 1e.

Voltage outputs V₁ and V₂ are mainly affected by the strain applied in the Z and X direction, respectively. By inspecting these two voltage outputs, the nature of strain can be defined. Figure 7 shows the experimental results for a combined X-taxel/Z-taxel sensor. Initially, ∼1 mm of δ_z was applied (Fig. 7a). The normal compressive force increased, as shown in Figure 7c. Figure 7d and e shows the Z-taxel and X-taxel output, respectively. Figure 7d shows the initial δ_z and compressive normal force as well as the normal taxel output where the voltage increases in response to the applied normal strain.

FIG. 7.

Combined X-taxel/Z-taxel sensor test: (a) applied normal compressive deformation in the Z direction; (b) applied shear deformation in the X direction; (c) compressive force as measured using a force gauge; (d) Z-taxel output; (e) X-taxel output with time.

Next, a set of three shear cycles was applied using a δ_x of 2 mm, as shown in Figure 7b. The δ_x affected both the Z-taxel and the X-taxel. Because of the shear effect, the normal taxel voltage dropped due to the increase in resistance. However, the X-taxel voltage increased as one electrode was moved closer to the other in the same layer. Voltage reduction only in a normal taxel does not provide enough information to allow the shear to be detected, since the reduction can come from unloading in the Z direction. However, combining the outputs of the Z-taxel and the X-taxel (Fig. 7d, e) can provide enough information to define the nature of the strain.

Discussion

The IL/polymer network in the sensor works as the primary sensitive layer, where the electrical resistance changes in response to deformation. Figure 8 illustrates the molecular structure and proposed operating mechanism of the investigated IL/polymer-based sensor. Because the polymer network contains abundant urethane functional groups that can act as hydrogen bond donors,^43,44 we reason that a noncovalent interaction based on hydrogen bonds between urethane groups and tetrafluoroborate (BF₄⁻) would be formed. Before the application of external force, most of the cations and anions (i.e., EMIM⁺ and BF₄⁻) are confined inside the polymer network due to hydrogen bonds and Coulomb forces (Fig. 8a). The reduced mobility of the ion pairs accounts for the initial high electrical resistance of the investigated device. Under external pressure, hydrogen bonds between ILs and polymer side chains (i.e., urethane groups) can serve as energy dissipation sites by cleaving the hydrogen bonds (Fig. 8b).^45,46

FIG. 8.

Molecular structure and proposed operating mechanism of the IL/polymer network-based sensor: (a) ion confinement via hydrogen bonds. Before applying the pressure, ions are confined in the polymer network via hydrogen bonds between the BF₄⁻ anions and urethane functional groups. EMIM⁺ cations interacted with BF₄⁻ anions via the Coulomb force; (b) pumping of ions from the polymer network due to the pressure-induced breakage of hydrogen bonds. With the application of pressure, the ion pairs (EMIM⁺ and BF₄⁻) become detached from the polymer network through ion pumping into the electrode surfaces.

Moreover, the distance between the two electrodes decreases, which would increase the intensity of the electric field and thus promote an ion pumping process⁴⁷ during which the EMIM⁺ and BF₄⁻ ion pairs are detached from the polymer network. The overall hydrogen-bond breakage that is induced by pressure as well as the ion-pumping process will increase the electrical conductivity of the sensor. This deformation-based conductivity change serves as the core sensing mechanism for both normal and shear forces on the sensor.

The results presented in the Material Formulation Selection section show reliable performance for the sensor made using Form1. Supplementary Figure S2 shows the performance results for different material combinations. TangoPlus was previously used as a base polymer and showed a consistent response.³² While the performance of Form1 is close to that of TangoPlus, other formulations demonstrated issues such as cracks, long recovery time, and unreliable response. Thus, the selection of the base polymer is crucial for the CNT/polymer composite, as it can very easily affect the sensor performance. In this study, Form1 was selected as the base formulation, and 5 wt.% CNT was dispersed into it; this CNT/prepolymer was then used for the screen printing of electrodes. The resulting sensors showed good electrical conductivity and mechanical strength.

Other materials and compositions could be explored depending on the final application of the sensor. For example, the amount of crosslinker can be varied slightly to obtain sensors with different hardness values. The isobornyl acrylate ratio could also be varied to modulate the strength of the polymer and the sensor. The sensor response depends on the IL ratio in the IL/polymer membrane, especially for shear sensing. The results presented in the Z-Taxel Sensor and Applied Normal Force and X-Taxel Sensor and Applied Shear Force sections provide a good guideline for adjusting the IL ratio according to the need. However, a very low IL ratio generates electrical noise that could be detrimental to the sensor performance. In addition to the IL ratio, the geometric configuration will also significantly influence the sensor response.

As shown in Figures 5f and 6f, a lower distance between electrodes provides higher sensitivity, whereas a greater distance shows lower sensitivity and more electrical noise. Hysteresis loss during sensor unloading, which can be observed in Figure 5g, is common in viscoelastic elastomers because of their time-dependent elastic properties.^48,49 Overall, the Z-taxel sensors and the X-taxel sensors showed very predictable and consistent paths in loading and unloading over multiple cycles.

The results presented in the Z-Taxel Sensor and Applied Normal Force and X-Taxel Sensor and Applied Shear Force sections provide important insights for designing a combined X-taxel/Z-taxel sensor. The IL ratio and geometric configuration were carefully chosen for different IL layers in the combined sensor, and the combined sensor was evaluated under normal and shear deformations. Initially, a normal deformation (δ_z) of 1 mm was applied to the sensor; this resulted in a force increase (see the red dashed line in Fig. 7c). Due to the normal deformation on the Z-taxel, V₁ significantly increased, which indicates applied compression. This δ_z did not show a considerable effect on V₂ except for some electrical noise, which could be filtered out by incorporating some capacitors into the circuit.

Next, a three-cycle set of shear deformation (δ_x = 1 mm) was applied to the sensor. The V₂ signal from the X-taxel clearly showed three spikes due to the applied shear, while three troughs appeared in the V₁ signal of the Z-taxel resulting from the shear influence in the Z-direction. However, a combined analysis of V₁ and V₂ clearly indicates an applied shear deformation. The same deformation was applied one additional time to check the consistency of the sensor response. The X-taxel output V₂ and Z-taxel output V₁ showed a similar and consistent response. In future research, shear deformation in the Y-direction, strain-rate dependence, and angular motion will be studied. Future studies also include applications of the sensor such as sensor-embedded insole for clinical gait analysis and sensor-embedded tire to understand the tire mechanics.

Conclusions

This work presents a novel, flexible solid-state sensor capable of measuring normal and shear force. An IL was utilized to fabricate the sensor, and different material formulations were investigated for developing the base polymer of the sensor. A material composition was suggested for optimum mechanical and electrical properties, and different IL concentrations and geometric configurations were studied for normal and shear sensing. Based on the test results of separate normal and shear experiments, a combined sensor was designed and fabricated. The combined sensor performed reliably and was able to indicate the nature of the applied force. In addition to mechanical pliability, the proposed sensor is expected to provide unparalleled design freedom, as configurations such as geometry and IL ratio can be varied to achieve a need-specific sensor design. This study is believed to open avenues for newer applications of soft sensors in robotics, prosthetics, and wearables.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

Funding Information

This work was supported by a grant from the National Science Foundation (NSF Grant #1914165).

Supplementary Material

References

Wang

, Sun

, Zhao

, et al. A highly stretchable hydrogel sensor for soft robot multi-modal perception. Sens Actuator A Phys, 2021; 331:113006; doi: 10.1016/j.sna.2021.113006

Wang

, Tong

, Zhao

, et al. Flexible, high-sensitive, and wearable strain sensor based on organic crystal for human motion detection. Org Electron, 2018; 61:304–311; doi: 10.1016/j.orgel.2018.06.008

Sankar

, Nambi

, Bhat

, et al. Waterproof flexible polymer-functionalized graphene-based piezoresistive strain sensor for structural health monitoring and wearable devices. ACS Omega, 2020; 5(22):12682–12691; doi: 10.1021/acsomega.9b04205

Emon

MOF

, Choi

J-W

. Flexible piezoresistive sensors embedded in 3D printed tires. Sensors (Basel), 2017; 17(3):656; doi: 10.3390/s17030656

S-H

, Guo

P-W

, Pang

C-Y

, et al. Screen-printed flexible strain sensors with Ag nanowires for intelligent and tamper-evident packaging applications. Adv Mater Technol, 2020; 5(5):1901097; doi: 10.1002/admt.201901097

Arabagi

, Felfoul

, Gosline

, et al. Biocompatible pressure sensing skins for minimally invasive surgical instruments. IEEE Sens J, 2016; 16(5):1294–1303; doi: 10.1109/JSEN.2015.2498481

Luczak

, Burch

VRF

, Smith

, et al. Closing the wearable gap—Part V: Development of a pressure-sensitive sock utilizing soft sensors. Sensors (Basel), 2020; 20(1):208; doi: 10.3390/s20010208

, Dae-Hyeong

. Flexible and stretchable electronics paving the way for soft robotics. Soft Robot, 2013; 1(1):53–62; doi: 10.1089/soro.2013.0005

, Liu

, Chen

, et al. Ultrastretchable, transparent triboelectric nanogenerator as electronic skin for biomechanical energy harvesting and tactile sensing. Sci Adv, 2017; 3(5):e1700015; doi: 10.1126/sciadv.1700015

10.

, Wang

, Drack

, et al. A bimodal soft electronic skin for tactile and touchless interaction in real time. Nat Commun, 2019; 10(1):4405; doi: 10.1038/s41467-019-12303-5

11.

Herbert

, Lim

H-R

, Yeo

W-H

. Printed, soft, nanostructured strain sensors for monitoring of structural health and human physiology. ACS Appl Mater Interfaces, 2020; 12(22):25020–25030; doi: 10.1021/acsami.0c04857

12.

Emon

MOF

, Choi

J-W

. A Preliminary Study on 3D Printed Smart Insoles with Stretchable Piezoresistive Sensors for Plantar Pressure Monitoring. American. Society of Mechanical Engineers Digital Collection: Tampa, FL;. 2018.

13.

Banerjee

, Ponraj

, Kirthika

, et al. Hydrogel-shielded soft tactile sensor for biocompatible drug delivery monitoring. J Med Devices, 2019; 13(4):4114; doi: 10.1115/1.4044114

14.

Iguchi

, Kudo

, Saito

, et al. A flexible and wearable biosensor for tear glucose measurement. Biomed Microdevices, 2007; 9(4):603–609; doi: 10.1007/s10544-007-9073-3

15.

Zhang

, Ke

, Jiang

, et al. Fabrication and functionality integration technologies for small-scale soft robots. Adv Mater, 2022; 2022:2200671; doi: 10.1002/adma.202200671

16.

Dilibal

, Sahin

, Danquah

, et al. Additively manufactured custom soft gripper with embedded soft force sensors for an industrial robot. Int J Precis Eng Manuf, 2021; 22(4):709–718; doi: 10.1007/s12541-021-00479-0

17.

Hosseini

, Chakraborty

, Roe

, et al. Porous elastomer based wide range flexible pressure sensor for autonomous underwater vehicles. IEEE Sens J, 2022; 22(10):9914–9921; doi: 10.1109/JSEN.2022.3165560

18.

Kim

N-I

, Chang

Y-L

, Chen

, et al. Piezoelectric pressure sensor based on flexible gallium nitride thin film for harsh-environment and high-temperature applications. Sens Actuator A Phys, 2020; 305:111940; doi: 10.1016/j.sna.2020.111940

19.

Kenry, Yeo

, Lim

. Emerging flexible and wearable physical sensing platforms for healthcare and biomedical applications. Microsyst Nanoeng, 2016; 2(1):1–19; doi: 10.1038/micronano.2016.43

20.

Souri

, Banerjee

, Jusufi

, et al. Wearable and stretchable strain sensors: Materials, sensing mechanisms, and applications. Adv Intell Syst, 2020; 2(8):2000039; doi: 10.1002/aisy.202000039

21.

Amjadi

, Kyung

K-U

, Park

, et al. Stretchable, skin-mountable, and wearable strain sensors and their potential applications: A review. Adv Funct Mater, 2016; 26(11):1678–1698; doi: 10.1002/adfm.201504755

22.

Zhang

, Wang

, Peng

, et al. Flexible highly sensitive pressure sensor based on ionic liquid gel film. ACS Omega, 2018; 3(3):3014–3021; doi: 10.1021/acsomega.7b01575

23.

Scaffaro

, Maio

, Citarrella

. Ionic tactile sensors as promising biomaterials for artificial skin: Review of latest advances and future perspectives. Eur Polym J, 2021; 151:110421; doi: 10.1016/j.eurpolymj.2021.110421

24.

Cao

, Tan

, Li

, et al. Self-healing electronic skins for aquatic environments. Nat Electron, 2019; 2(2):75–82; doi: 10.1038/s41928-019-0206-5

25.

Choi

, Kim

, Oh

, et al. Highly stretchable, hysteresis-free ionic liquid-based strain sensor for precise human motion monitoring. ACS Appl Mater Interfaces, 2017; 9(2):1770–1780; doi: 10.1021/acsami.6b12415

26.

Zhou

, Fei

, Tian

, et al. A ionic liquid enhanced conductive hydrogel for strain sensing applications. J Colloid Interface Sci, 2022; 606:192–203; doi: 10.1016/j.jcis.2021.07.158

27.

Lee

, Emon

MOF

, Vatani

, et al. Effect of degree of crosslinking and polymerization of 3D printable polymer/ionic liquid composites on performance of stretchable piezoresistive sensors. Smart Mater Struct, 2017; 26(3):035043; doi: 10.1088/1361-665X/aa5c70

28.

Ohno

. Design of ion conductive polymers based on ionic liquids. Macromol Symp, 2007; 249–250(1):551–556; doi: 10.1002/masy.200750435

29.

Mishra

, Devi

, Siwal

, et al. Ionic liquid-based polymer nanocomposites for sensors, energy, biomedicine, and environmental applications: Roadmap to the future. Adv Sci (Weinh), 2022; 9(26):e2202187; doi: 10.1002/advs.202202187

30.

, Li

, Shi

, et al. Recent developments for flexible pressure sensors: A review. Micromachines, 2018; 9(11):580; doi: 10.3390/mi9110580

31.

Kanoun

, Bouhamed

, Ramalingame

, et al. Review on conductive polymer/CNTs nanocomposites based flexible and stretchable strain and pressure sensors. Sensors (Basel), 2021; 21(2):341; doi: 10.3390/s21020341

32.

Emon MdOF, Lee

, Choi

, et al. Characterization of a soft pressure sensor on the basis of ionic liquid concentration and thickness of the piezoresistive layer. IEEE Sens J, 2019; 19(15):6076–6084; doi: 10.1109/JSEN.2019.2911859

33.

Garkani Nejad

, Sheikhshoaie

, Beitollahi

. Simultaneous detection of carmoisine and tartrazine in food samples using GO-Fe3O4-PAMAM and ionic liquid based electrochemical sensor. Food Chem Toxicol, 2022; 162:112864; doi: 10.1016/j.fct.2022.112864

34.

Wolterink

, Sanders

, van Beijnum

B-J

, et al. A 3D-printed soft fingertip sensor for providing information about normal and shear components of interaction forces. Sensors (Basel), 2021; 21(13):4271; doi: 10.3390/s21134271

35.

Lee

H-K

, Chung

, Chang

S-I

, et al. Normal and shear force measurement using a flexible polymer tactile sensor with embedded multiple capacitors. J Microelectromechan Syst, 2008; 17(4):934–942; doi: 10.1109/JMEMS.2008.921727

36.

, Zhu

, Zhe

. An inductive sensor for real-time measurement of plantar normal and shear forces distribution. IEEE Trans Biomed Eng, 2015; 62(5):1316–1323; doi: 10.1109/TBME.2014.2386136

37.

Guggenheim

, Jentoft

, Tenzer

, et al. Robust and inexpensive six-axis force–torque sensors using MEMS barometers. IEEE ASME Trans Mechatron, 2017; 22(2):838–844; doi: 10.1109/TMECH.2017.2654446

38.

Charalambides

, Bergbreiter

. A novel all-elastomer MEMS tactile sensor for high dynamic range shear and normal force sensing. J Micromech Microeng, 2015; 25(9):095009; doi: 10.1088/0960-1317/25/9/095009

39.

Chen

, Bai

, Zhang

, et al. Flexible piezoresistive three-dimensional force sensor based on interlocked structures. Sens Actuator A Phys, 2021; 330:112857; doi: 10.1016/j.sna.2021.112857

40.

Agcayazi

, McKnight

, Sotory

, et al. A Scalable Shear and Normal Force Sensor for Prosthetic Sensing. In: 2017 IEEE Sensors. IEEE: Glasgow, UK; pp. 1–3; doi: 10.1109/ICSENS.2017.8233977

41.

Emon

MOF

, Alkadi

, Philip

, et al. Multi-material 3D printing of a soft pressure sensor. Addit Manuf, 2019; 28:629–638; doi: 10.1016/j.addma.2019.06.001

42.

Emon

, Alkadi

, Kiki

, et al. Conformal 3D printing of a polymeric tactile sensor. Addit Manuf Lett, 2022; 2:100027; doi: 10.1016/j.addlet.2022.100027

43.

Sun

, Li

, Gao

, et al. Self-healable, highly stretchable, ionic conducting polymers as efficient protecting layers for stable lithium-metal electrodes. ACS Appl Mater Interfaces, 2022; 14(22):26014–26023; doi: 10.1021/acsami.2c04166

44.

, Chassé

, Zick

, et al. The effect of hydrogen bonding on diffusion and permeability in UV-cured polyacrylate-based networks for controlled release. J Control Release, 2020; 327:150–160; doi: 10.1016/j.jconrel.2020.07.039

45.

Montero de Espinosa

, Meesorn

, Moatsou

, et al. Bioinspired polymer systems with stimuli-responsive mechanical properties. Chem Rev, 2017; 117(20):12851–12892; doi: 10.1021/acs.chemrev.7b00168

46.

Amoli

, Kim

, Jee

, et al. A bioinspired hydrogen bond-triggered ultrasensitive ionic mechanoreceptor skin. Nat Commun, 2019; 10(1):4019; doi: 10.1038/s41467-019-11973-5

47.

Amoli

, Kim

, et al. Ionic tactile sensors for emerging human-interactive technologies: A review of recent progress. Adv Funct Mater, 2020; 30(20):1904532; doi: 10.1002/adfm.201904532

48.

Tschoegl

. Time dependence in material properties: An overview. Mech Time Depend Mater, 1997; 1(1):3–31; doi: 10.1023/A:1009748023394

49.

Wang

, Yu

, Zhang

, et al. Highly stretchable, soft, low-hysteresis, and self-healable ionic conductive elastomers enabled by long, functional cross-linkers. Macromolecules, 2022; 55(17):7845–7855; doi: 10.1021/acs.macromol.2c00563

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.11 MB

0.23 MB

0.15 MB

0.16 MB

0.21 MB

0.03 MB

0.00 MB