Search results

Additive manufacturing (AM) methods such as material extrusion (ME) are becoming widely used by engineers, designers and hobbyists alike for a wide variety of applications. Successfully manufacturing objects using ME three-dimensional printers can often require numerous iterations to attain predictable performance because the exact mechanical behavior of parts fabricated via additive processes are difficult to predict. One of that factors that contributes to this difficulty is the wide variety of ME feed stock materials currently available in the marketplace. These build materials are often sold based on their base polymer material such as acrylonitrile butadiene styrene or polylactic acid (PLA), but are produced by numerous different commercial suppliers in a wide variety of colors using typically undisclosed additive feed stocks and base polymer formulations. This paper aims to present the results from an experimental study concerned with quantifying how these sources of polymer variability can affect the mechanical behavior of three-dimensional printed objects. Specifically, the set of experiments conducted in this study focused on following: several different colors of PLA filament from a single commercial supplier to explore the effect of color additives and three filaments of the same color but produced by three different suppliers to account for potential variations in polymer formulation.

A set of five common mechanical and material characterization tests were performed on 11 commercially available PLA filaments in an effort to gain insight into the variations in mechanical response that stem from variances in filament manufacturer, feed stock polymer, additives and processing. Three black PLA filaments were purchased from three different commercial suppliers to consider the variations introduced by use of different feed stock polymers and filament processing by different manufacturers. An additional eight PLA filaments in varying colors were purchased from one of the three suppliers to focus on how color additives lead to property variations. Some tests were performed on unprocessed filament samples, while others were performed on objects three-dimensional printed from the various filaments. This study looked specifically at four mechanical properties (Young’s modulus, storage modulus, yield strength and toughness) as a function of numerous material properties (e.g. additive loading, molecular weight, molecular weight dispersity, enthalpy of melting and crystallinity).

For the 11 filaments tested the following mean values and standard deviations were observed for the material properties considered: p_a = 1.3 ± 0.9% (percent additives), M_w = 98.6 ± 16.4 kDa (molecular weight), Ð = 1.33 ± 0.1 (molecular weight dispersity), H_m = 37.4 ± 7.2 J/g (enthalpy of melting) and = 19.6 ± 2.1% (crystallinity). The corresponding mean values and standard deviations for the resulting mechanical behaviors were: E = 2,790 ± 145 MPa (Young’s modulus), E’ = 1,050 ± 125 MPa (storage modulus), S_y = 49.6 ± 4.93 MPa (yield strength) and U_t = 1.87 ± 0.354 MJ/m^³ (toughness). These variations were observed in filaments that were all manufactured from the same base polymer (e.g. PLA) and are only different in terms of the additives used by the manufacturers to produce different colors or different three-dimensional printing performance. Unfortunately, while the observed variations were significant, no definitive strong correlations were found between these observed variations in the mechanical behavior of the filaments studied and the considered material properties.

These variations in mechanical behavior and material properties could not be ascribed to any specific factor, but rather show that the mechanical of three-dimensional printed parts are potentially affected by variations in base polymer properties, additive usage and filament processing choices in complex ways that can be difficult to predict.

These results emphasize the need to take processing and thereby even filament color, into account when using ME printers, they emphasize the need for designers to use AM with caution when the mechanical behavior of a printed part is critical and they highlight the need for continued research in this important area. While all filaments used were marked as PLA, the feedstock materials, additives and processing conditions created significant differences in the mechanical behavior of the printed objects evaluated, but these differences could not be accurately and reliably predicted as function of the observed material properties that were the focus of this study.

The testing methods used in the study can be used by engineers and creators alike to better analyze the material properties of their filament printed objects, to increase success in print and mechanical design. Furthermore, the results clearly show that as AM continues to evolve and grow as a manufacturing method, standardization of feedstock processing conditions and additives would enable more reliable and repeatable printed objects and would better assist designers in effectively implementing AM methods.