Evaluating the effect of variable fiber content on mechanical properties of additively manufactured continuous carbon fiber composites

Fiber volume fraction is a driving factor in mechanical properties of composites. Micromechanical models are typically used to predict the effective properties of composites with different fiber volume fractions. Since the microstructure of 3D-printed composites is intrinsically different than conventional composites, such predictions need to be evaluated for 3D-printed composites. This investigation evaluates the ability of the Voigt, Reuss, and Halpin-Tsai models to capture the dependence of modulus and strength of 3D-printed composites on varying fiber volume fraction. Tensile coupons were printed with continuous carbon fiber-reinforced Onyx matrix using a Markforged Mark Two printer. Specimens were printed at five different volume fractions with unidirectional fibers oriented at either 0 degrees, 45 degrees, or 90 degrees to obtain longitudinal, shear, and transverse properties, respectively. It is shown that the Voigt model provides an excellent fit for the longitudinal tensile strength and a reasonable fit for the longitudinal modulus with varied fiber content. For the transverse direction, while the Reuss model fails to capture the transverse modulus trend, the Halpin-Tsai model provides a reasonable fit as it incorporates more experimental parameters. Like conventional composites, addition of fibers degrades the transverse strength, and the transverse strength decreases with increasing fiber volume fraction. The shear modulus variation with fiber content could not be fitted reasonably with either Halpin-Tsai model or Reuss model.

Automated summary

The study evaluated the ability of Voigt, Reuss, and Halpin-Tsai models to capture the relationship between strength and modulus of 3D-printed composites and varying fiber volume fractions. Results showed that the Voigt model provided a good fit for longitudinal tensile strength, the Halpin-Tsai model provided a reasonable fit for transverse modulus, while the Reuss model failed to capture the trend of transverse modulus.