An out-of-autoclave monitoring system using carbon nanotube yarns that sense temperature, pressure, and curing behavior for quality control

The quality control of composite materials requires precise monitoring of temperature changes, pressure profiles, resin flow, and curing times. Autoclave tooling is critical because it is designed with complex sensing systems that provide the necessary parameter feedback that facilitates high-quality composite parts. Though the autoclave is standard, it is expensive to purchase and maintain, complex, energy intensive, and limited by its pre-defined volume. The intention of this research is to demonstrate new approaches to quality control in out-of-autoclave (OOA) manufacturing. By integrating minimally invasive sensors, that exhibit cross-sectional areas less than 0.1μm2, into the dry fiber or prepreg layup, various process parameters can be monitored without the need for complexity. In this research, carbon nanotube yarns (CNTys), which possess robust electromechanical and electrothermal couplings, were harnessed into a unique production monitoring system for OOA manufacturing. In addition to the temperature and pressure of the mold, CNTys can also monitor the curing behavior of the impregnated resin. As a result, this sensing system can potentially provide early indications of lost process control and defects. To accomplish the study, three types of CNTys were explored in terms of microstructure, electrical and mechanical properties, and overall sensing performance. The goal of these exploratory efforts was to systematically choose the most optimal yarn type before conducting lengthy curing experiments. The optimal CNTy type was deployed as a curing sensor, which generates a piezoresistive response to the internal stress that accompanies resin shrinkage. Data from CNTy sensors was collected and compared with differential scanning calorimetry (DSC) data to uncover correlations between the sensing responses and cross-linking degree of the cured resin. To validate these correlations, two distinct types of resins were used. The early detection of cross-linking degree is a transformative advantage in the quality control of composites. After the yarns were co-cured in a composite part, they were connected as strain gauges. Three-point bending experiments were conducted to showcase the strain sensing responses. This research combines manufacturing, material science, statistical analysis, and innovative data acquisition to establish a foundation for novel out-of-autoclave tooling with early artificial intelligence (AI).