Lu WangFollow

Date of Award

Fall 12-15-2017

Level of Access Assigned by Author

Open-Access Dissertation

Degree Name

Doctor of Philosophy (PhD)


Forest Resources


Douglas J. Gardner

Second Committee Member

Yousoo Han

Third Committee Member

Mehdi Tajvidi

Additional Committee Members

William M. Gramlich

Douglas W. Bousfield


Compared to conventional manufacturing process, additive manufacturing (AM) offers free-form design, lighter and more ergonomic products, short lead time and less waste. Extrusion-based AM can be used to print thermoplastics. However, extrusion-based AM has processing challenges in printing semi-crystalline thermoplastics, for instance, polypropylene (PP). Cellulose nanofibrils (CNF) are one type of cellulose nanofibers that are produced from pulp fibers. CNF has extraordinary properties which make it an ideal candidate to reinforce polymers. Spray-dried CNF (SDCNF) is able to be incorporated into thermoplastic matrices without modifying conventional processing procedures.

The mechanical properties of 3D printed plastic parts have been considered significantly weaker than injection molded parts because the former contains more pores. However, with proper printing parameter selection, the mechanical properties of 3D printed parts can be very close to that of injection molded parts. What’s more, 3D printed parts have lower density than injection molded parts, making it an ideal material for light-weight applications.

The shear rate involved in extrusion-based AM is reported to be much smaller than that during injection molding. Adding fillers would empirically increase the viscosity. However, the incorporation of SDCNF, up to 10 wt.%, did not significantly increase the viscosity of PP melts, even at a smaller shear rate.

Isotactic polypropylene (IPP) crystallizes much faster than other semi-crystalline thermoplastics used in extrusion-based AM. The overall crystallization rate depends on the nucleation rate and crystal growth rate. Adding fillers should increase the nucleation rate by providing more heterogeneous nucleation sites. Meanwhile the fillers decrease the crystal growth rate by reducing the available space for crystals to grow. At 10 wt.% loading level of SDCNF, the increase in nucleation rate was smaller than the decrease in crystal growth rate. So 10 wt.% SDCNF retarded the overall crystallization rate of iPP.

PP printed at a bed temperature of 120 ºC showed both α and β-crystal forms while injection molded PP only showed α-crystals. The heat deflection temperatures (HDT) of 3D printed parts were higher than injection molded parts because the voids in the 3D printed parts acted as thermal insulators that delayed heat transfer during the HDT test.