Date of Award

2021-05-01

Degree Name

Master of Science

Department

Engineering

Advisor(s)

Yirong Lin

Abstract

The following research focuses on the additive manufacturing and characterization of epoxy syntactic foams that utilize hollow carbon microspheres (CMS) and graphene-coated hollow carbon microspheres (GCMS) as fillers. The matrix that is utilized in this study is a thermoset epoxy resin, chosen for its high strength, low shrinkage, excellent adhesion to various substrates, effective electrical properties, chemical and solvent resistance, low cost, and low toxicity.1 CMS and GCMS were used as fillers for the properties they contribute to the overall epoxy syntactic foam, and can be seen in Fig.1.1. CMS consist of outer stiff carbon and inner gas, giving the sphere its hollow nature. Hollow carbon microspheres were chosen for their high thermal conductivity, light weight, and low dielectric constant. Graphene was synthesized onto the spheres for the high surface area it contributes to the carbon spheres, as well as its high thermal conductivity and chemical stability. The following research focuses on the manufacturing of carbon fiber composites with fabricated defects for identification using ultrasonic non-destructive testing (NDT). These identified defect images are then used in the training of a neural network to identify defects independently. In industry settings, parts are always tested by an NDT expert to identify the presence of defects, and if they are detrimental to the part. The application of an artificial intelligence network that would do the initial identification of defects would reduce time an NDT expert would need to apply to a single part, allowing them to focus on more critical applications. This increase in productivity would not only be beneficial to the NDT expert, but to any industry that uses composite materials as a whole. While there are a variety of NDT methods available today, ultrasound will be the focus of this research for its wide range of application.

Language

en

Provenance

Received from ProQuest

File Size

36 p.

File Format

application/pdf

Rights Holder

Victoria Centeno

Share

COinS