E-skin has important applications in areas such as soft robotics, prosthetics, biomimicry, and biosensors because of its ability to mimic the soft properties and tactile functions of human skin. After decades of research, the flexible pressure sensor, one of the key components of the electronic skin, has made significant progress in the sensitivity and detection range of the sensor, but it still faces many challenges, especially the problem that the pressure sensing signal of the flexible pressure sensor is interfered with by the stretched state in the stretched state, which has not been well solved.
In past studies, scientists have explored the sensing mechanisms of different pressure sensors, such as piezoresistive, piezocapacitive, piezoelectric, optical, ionic, and magnetic. Among them, capacitive pressure sensors have attracted attention due to their high sensitivity, wide operating range and good linearity. However, in the face of simultaneous tension and compression, the accuracy of the pressure signal of these sensors is still a major challenge. When the capacitive sensor is stretched, the capacitance value increases due to the Poisson effect of the dielectric, the electrode area increases and the gap decreases, and if the capacitance value changes due to the out-of-plane pressure, it will be difficult to accurately measure the out-of-plane pressure through the capacitance value. This phenomenon is called tension interference of the pressure signal. Capacitive pressure sensors that are intrinsically stretchable and accurately measure pressure in the tensile state have not yet been realized, although capacitive pressure sensors that can be corrected to some extent by means of strain isolation or capacitance compensation can be corrected to a certain extent.
In order to solve the interference of stretching on the pressure readings of pressure sensors, Professor Lu Nanshu and his team at UT-Austin developed a flexible Stretchable Hybrid Response Pressure Sensors, or SHRPS, with high sensitivity and intrinsically stretchable. The sensor can be stretched to 70%, with a sensitivity of 1.25 kPa-1 over the operating range of 0-10 kPa, and the pressure readings of the sensor show good consistency in five cases: no in-plane stretch and uniaxial in-plane stretch (10%, 20%, 30%, 40%), thus effectively solving the problem of interference of tensile pressure readings. In addition, the research team has also attached the sensor to an inflatable flexible probe and realized different application scenarios. For example, the sensor can accurately detect the pulse waveform even when the flexible probe is inflated (the sensor is stretched in a biaxial plane with a 41% strain), and the sensor can be reliably grasped when the flexible probe is not inflated (the sensor is in the biaxial plane and not stretched).
This research not only provides a potential solution for e-skin, soft robotics and bio-integrated electronics, but also demonstrates the ability of SHRPS to perform a variety of tasks on the surface of smart inflatable probes with adjustable shape and controllable stiffness.
Figure 1. A) Schematic diagram of SHRPS structure, B) Scanning electron microscope diagram of conductive porous nanocomposites (PNCs), C) Electric field and potential distribution of three different types of capacitive sensors: (i) capacitive pressure sensor with dielectric layer material, (ii) SHRPS, (iii) capacitive pressure sensor with electrode material, and electric field (streamline) and electric potential (color profile) distribution in the undeformed (middle), compressed (left) and stretched (right) states. D) Schematic diagram of the deformation of the 3x3 SHRPS array: from left to right, it corresponds to the morphological changes of the SHRPS array under no load and different mechanical stresses (tensile, bending, torsion).
The research team first demonstrated the structural details of SHRPS, as shown in Figure 1A, in which a carbon nanotube (CNT) layer composite sprayed on the surface of a polydimethylsiloxane (PDMS) is used as an electrode, a thin layer of PDMS acts as an insulating layer for a capacitive sensor, and a conductive porous nanocomposite (PNC) is used as a dielectric layer material. Scanning electron microscopy (SEM) characterization of PNCs is shown in Figure 1B. Figure 1C illustrates the operation of three different types of capacitive sensors in both in-plane stretch and out-of-plane compression, where SHRPS behaves similarly to conventional capacitive sensors when stretched in-plane, and SHRPS exhibits a composite response when the PNC resistance changes during out-of-plane compression. This unique composite response greatly enhances the sensitivity of SHRPS to pressure changes while effectively reducing capacitive signal interference caused by stretching.
Figure 2. Compression and tensile response analysis of three different types of capacitive pressure transducers; A) Solid ecolfex, B) Porous ecolfex, C) SHRPS, (i) Schematic, (ii) Capacitive response of pressure alone (blue) vs. stretch alone (red), and (iii) capacitive response when pressure and stretch are applied at the same time.
To demonstrate the unique characteristics of SHRPS, the researchers compared and analyzed the response of three different types of capacitive pressure sensors in both pressure and tensile conditions.
For solid ecoflex capacitive pressure sensors:
The change in capacitance signal (ΔCs) at 40% uniaxial stretch is 500% of the change in capacitance signal (ΔCp) due to pressure at 50 kPa (see Figure 2A(ii)). In this case, when the sensor undergoes varying degrees of in-plane stretch and out-of-plane compression, the pressure-capacitance curve is discrete, resulting in inaccurate pressure readings at the same capacitance signal (see Figure 2A(iii)).
For porous ecoflex capacitive pressure sensors:
Similarly, the 40% uniaxial stretch-induced capacitance signal change (ΔCs) is only 12% of the capacitance signal change (ΔCp) due to the pressure of 50 kPa (see Figure 2B(ii)). In this case, the pressure-capacitance curves of the sensor begin to coexist under different in-plane tensile strains, making the pressure readings for the same capacitance signal more accurate (see Figure 2B(iii)).
For SHRPS:
At 40% uniaxial stretch, the change in capacitive signal (ΔCs) accounts for only 1.8% of the change in capacitive signal (ΔCp) due to pressure at 50 kPa (see Figure 2C(ii)). Moreover, the pressure-capacitance curves of the SHRPS almost exactly coincide when subjected to in-plane tensile strain (0-40%), ensuring a high degree of accuracy in the sensor's pressure readings under the same capacitance signal (see Figure 2C(iii)).
Figure 3. A) Pressure resistance curves of PNCs under different CNT doping ratios, B) Pressure response of SHRPS under different CNT doping ratios (△C/C0 curves), C) Resistance changes of PNCs with 0.4 wt% CNT under uniaxial and biaxial tensile tensile strain, D) Variation of △C/C0 under uniaxial and biaxial tensile strain of SHRPS with 0.4 wt% CNT, E) Pressure sensitivity (0 - 10 kPa) between SHRPS and existing capacitive sensors range) and Ashby diagram comparison of stretchability, F) lower limit of pressure detection for SHRPS, G) response and recovery time of SHRPS, H) Repeatability and durability test of SHRPS at (i) 0 - 10 kPa pressure, (ii) 0 - 40% uniaxial tensile strain, (iii) repetitive loading 0 - 10 kPa pressure at 40% constant uniaxial tensile strain, and (iv) bending from flat to 7.2 mm radius.
Subsequently, the research team adjusted the doping ratio of CNT in PNC to study the specific effects of different ratios on the performance of SHRPS. As shown in Figures 3A and 3B, the experimental results show that the optimal doping ratio of CNT is 0.4%. Figure 3C further illustrates the resistance curves of the PNC material in uniaxial and biaxial tensile states at this doping ratio. Figure 3D shows the △C/C0 variation of SHRPS in uniaxial and biaxial stretch for this doping ratio.
Figure 3E compares the performance of SHRPS with other existing capacitive sensors in terms of pressure sensitivity (0 - 10 kPa range) and stretchability, showing that SHRPS has high pressure sensitivity and superior stretchability. In addition, Fig. 3F&3G details the lower limit of pressure detection and response response time of SHRPS, respectively. Figure 3H presents the repeatability and durability test results of SHRPS for the following loads:
(i) 0 - 10 kPa pressure,
(ii) 0 - 40% uniaxial tensile strain,
(iii) repetitive loading of 0 - 10 kPa pressure at 40% constant uniaxial tensile strain, and (iv) bending from a flat state to a radius of 7.2 mm.
Figure 4. Theoretical modeling and analysis of SHRPS;A) Schematic diagram of the capacitance resistance of each component of PNC; B) Equivalent circuit diagram of SHRPS; C) Prediction of compressive strain and capacitance of SHRPS under different CNT doping ratios;D) Prediction of compressive strain and △C/C0 of SHRPS under different CNT doping ratios;E) Comparison of theoretical and experimental results of SHRPS with different CNT doping ratios.
To quantitatively evaluate this sensing mechanism, the research team performed detailed modeling and analysis. They first determined the capacitance and resistance of each component of the PNC (see Figure 4A), and then built a simplified equivalent circuit model of SHRPS and compared it to the actual circuit (see Figure 4B-C). As can be seen from the capacitance change results in Figure 4D-E, the simplified equivalent circuit model is in good agreement with the experimental results to a certain extent, thus confirming the validity of the model.
Fig. 5: Smart probe pressure sensor consisting of flexible inflatable probe and 3x3 SHRPS array;A) Schematic diagram of the 3x3 SHRPS array and inflatable probe assembly;B) Schematic diagram of the uninflated and inflated state of the probe;C) Finite element simulation of SHRPS under biaxial tensile strain;D) Schematic diagram of human artery measurement and wrist cross-section;E) Schematic diagram of pulse detection process;F) Capacitance signal changes of 9 SHRPS units in pulse diagnosis:(1) Capacitance change during inflation (6 seconds) ;(2) Capacitance change after touching the wrist (17 seconds), G) Pulse wave raw data, H) Pulse data filtered by 1-4 Hz bandpass, I) Diagram of the glass grabbing operation: showing the process of grasping the glass with the same external force using the smart probe pressure sensor in the uninflated and inflated states. J) Schematic diagram of the operation of grabbing tortilla crusts: The process of grasping tortilla crusts using the intelligent probe pressure sensor is shown under the same external force in both uninflated and inflated states.
Finally, the research team designed two experiments to demonstrate the advantages of SHRPS in terms of high sensitivity and high stretchability, as well as its potential application in areas such as soft robotics and biosensors. The researchers assembled a 3x3 SHRPS array with a flexible inflatable probe into a smart probe pressure sensor (see Figure 5A). The probe is available in both uninflated and inflated states (see Figure 5B), with the SHRPS of the central unit at 41% biaxial planar tensile strain in the inflated state. Thanks to the high insensitivity of the SHRPS to tensile interference signals, the probe accurately senses pressure signals in both inflated and deflated states.
In order to demonstrate the application effect of flexible probes in different states, the researchers conducted the following operation demonstrations. As shown in Figure 5D, there is a complex tissue structure near the radial artery in the human wrist, and to detect the pulse signal of the radial artery, the probe needs precise point contact to detect the pulse signal. This can be achieved using a flexible probe in inflation mode (see Figure 5E). Figure 5F shows that the capacitance signal changes very little during planar stretching of SHRPS, and significantly increases when detecting the pulse wave during out-of-plane compression. In addition, Figure 5G shows that SHRPS is able to detect the pulse signal clearly, without the need for any filters to detect regular pulse waves. The pulse signal filtered by bandpass filtering of the respiratory signal is shown in Figure 5H. Figures 5I and 5J mainly illustrate the advantages of the uninflated mode, which provides a larger contact area and a more uniform pressure distribution under the same force, which makes it more suitable for tasks such as gripping.
Supplementary Video 2 shows that SHRPS produces only a weak response during stretching caused by probe inflation, but a strong response to contact pressure after touching the wrist. Even in the presence of pre-pressure, SHRPS is sensitive enough to extract pulse fluctuations.
Supplementary Video 4 shows how the inflatable probe makes "point" contact with the cup instead of the inflatable probe forming "face" contact in the task of grasping a round cup, resulting in a more reliable grip.
The first authors of the paper are Kyoung-Ho Ha, a Ph.D. student in the Department of Mechanical Engineering at the University of Texas at Austin, who is currently a postdoctoral researcher at Northwestern University, followed by Zhengjie Li, a Ph.D. student in the Department of Engineering Mechanics at the University of Texas at Austin, Ph.D. students Sangjun Kim and Heeyong Huh, and Dr. Zheliang Wang, who graduated from Johns Hopkins University Dr. Hongyang Shi, who graduated from Michigan State University and is currently engaged in postdoctoral research in Prof. Nanshu Lu's group, is currently engaged in postdoctoral research in Prof. Nanshu Lu's group, and Chase Block, an undergraduate student. Ph.D. student Sarnab Bhattacharya and Professor Lu Nanshu from the Department of Aerospace Engineering and Engineering Mechanics at the University of Texas at Austin are the sole corresponding authors of the paper. This newsletter was co-authored by Li Zhengjie and Shi Hongyang.
Paper Links:
https://authors.elsevier.com/a/1j0dk9Cyxd6qri