A Fully Inkjet-Printed Strain Sensor Based on Carbon Nanotubes

Hsuan-Ling Kao,Cheng-Lin Cho,Chun-Bing Chen,Li-Chun Chang,Yun-Chen Tsai,Wen-Hung Chung

doi:10.3390/coatings10080792

Hsuan-Ling Kao, Cheng-Lin Cho + Show 4 more

Open Access

https://doi.org/10.3390/coatings10080792

Copy DOI

Abstract

A fully inkjet-printed strain sensor based on carbon nanotubes (CNTs) was fabricated in this study for microstrain and microcrack detection. Carbon nanotubes and silver films were used as the sensing layer and conductive layer, respectively. Inkjet-printed CNTs easily undergo agglomeration due to van der Waals forces between CNTs, resulting in uneven films. The uniformity of CNT film affects the electrical and mechanical properties. Multi-pass printing and pattern rotation provided precise quantities of sensing materials, enabling the realization of uniform CNT films and stable resistance. Three strain sensors printed eight-layer CNT film by unidirectional printing, rotated by 180° and 90° were compared. The low density on one side of eight-layer CNT film by unidirectional printing results in more disconnection and poor connectivity with the silver film, thereby, significantly increasing the resistance. For 180° rotation eight-layer strain sensors, lower sensitivity and smaller measured range were found because strain was applied to the uneven CNT film resulting in non-uniform strain distribution. Lower resistance and better strain sensitivity was obtained for eight-layer strain sensor with 90° rotation because of uniform film. Given the uniform surface morphology and saturated sheet resistance of the 20-layer CNT film, the strain performance of the 20-layer CNT strain sensor was also examined. Excluding the permanent destruction of the first strain, 0.76% and 1.05% responses were obtained for the 8- and 20-layer strain sensors under strain between 0% and 3128 µε, respectively, which demonstrates the high reproducibility and recoverability of the sensor. The gauge factor (GF) of 20-layer strain sensor was found to be 2.77 under strain from 71 to 3128 µε, which is higher than eight-layer strain sensor (GF = 1.93) due to the uniform surface morphology and stable resistance. The strain sensors exhibited a highly linear and reversible behavior under strain of 71 to 3128 µε, so that the microstrain level could be clearly distinguished. The technology of the fully inkjet-printed CNT-based microstrain sensor provides high reproducibility, stability, and rapid hardness detection.

Highlights

Strain sensors have great potential for application in wearable electronic devices and smart skin to monitor human motion, personal health monitoring, and human–machine interfaces
Chun et al reported highly oriented and free-standing hydrophobic Carbon nanotubes (CNTs) sheets synthesized by chemical vapor deposition and transferred onto polyethylene naphthalate substrate; the CNT sheets exhibited sensitive piezoresistive responses to applied pressures of 0.1–40 kPa [10]
The results demonstrate the mechanical deformation and electrical resistance of the fully inkjet-printed CNT-based strain sensor; the sensor features high reversibility, reproducibility, and stability to afford microstrain detection and classification

Summary

Introduction

Strain sensors have great potential for application in wearable electronic devices and smart skin to monitor human motion, personal health monitoring, and human–machine interfaces. Bu et al reported drop-casting-prepared multi-walled CNT films with a sensitivity of 2.5 for the strain below 0.1% [7]. Lee et al proposed a spray-coated single-walled CNT film with good linearity over a microstrain range of 0 to 400 με [8]. Sahatiya et al deposited multi-walled CNTs on an eraser by rod-coating and obtained a strain-and-pressure sensor with a strain gauge factor (GF) of 2.4 and capacitive pressure sensor sensitivity of 0.135 MPa−1 [9]. Previous studies have presented the fabrication technologies of CNT films such as drop-casting, spray coating, rod-coating, and chemical vapor deposition, which are complex and laborious processes results in high manufacturing costs and generate large material waste. Effectively controlling the precise amount of sensing materials using the above processes is difficult

Methods

Discussion

Conclusion