3D-Printed Conductive Nanocomposites for Liquid Sensing Applications

Abstract

Additive manufacturing (AM) offers a new and unique method for the fabrication of functional and smart material and structures. In this method, parts are fabricated directly from a 3D computer model layer by layer. Fused deposition modeling (FDM) is the most widely adapted AM method. In this method, the feedstock is usually a thermoplastic-based material. In recent years, flexible smart materials have gained unflagging interests due to their promising applications in health monitoring, sensing, actuation, etc. However, the 3D printing of flexible materials is recent with its own challenges and limited sources of feedstock. Conductive polymer nanocomposites (CPNs) have many promising uses within sensing filed including liquid sensing. Sensing chemical leakage is one the important capabilities of liquid sensors. There is a good number of studies on the fabrication and sensitivity characterization of CPN-based liquid sensors. However, the sensitivity and response time of CPN-based liquid sensors do not yet meet the industrial demands and should be further enhanced for their practical and widespread applications. This study presents an attempt to integrate the tunability of CPN’s conductivity behavior and the design flexibility of 3D printing to explore the benefits that their coupling may offer toward more sensitive and/or faster liquid sensing. Thermoplastic polyurethane/multiwalled carbon nanotube (TPU/MWCNT) nanocomposites were selected as a model material system and their filaments were first fabricated using melt-mixing by twin-screw extruder at 1, 2 and 3 wt.% of MWCNT. Flexible U-shaped TPU/MWCNT specimens were designed and successfully 3D-printed as a liquid sensor. Specimens fabricated at three different raster patterns of linear, 0–90, and 45/−45 and three infill percent levels of 100, 75, and 50%. Ethanol was used as the model chemical and the resistivity change of the sensors was measured as a function of time when immersed in ethanol. The results revealed that the printed sensors greatly outperformed the pressed bulk counterparts. This enhancement in the 3D printed sensors was primarily due to the increased surface area, and thus higher surface/volume ratio, enabling faster liquid uptake. In addition, MWCNT content, raster pattern, and infill percent all affected the overall response time as well as the sensor sensitivity. This work suggests that highly sensitive liquid sensors can be developed by material and structure optimizations via FDM 3D printing.