W. M. Shewakh, Majed Moosa, Zainab Hussain, Osama M. Irfan
Fused deposition modeling (FDM) of polylactic acid (PLA) produces parts whose weak interlayer bonding and low as-printed crystallinity limit their tensile performance. This work used a Taguchi L9 orthogonal array with five replicates per cell (n = 5; N = 45 annealed specimens plus five non-annealed controls) to study how annealing temperature (70, 80, and 90 °C) and holding time (40, 60, and 80 min) change the tensile response of a commercial PLA grade (eSUN PLA+) printed on a desktop FDM machine. Differential scanning calorimetry (DSC) and X-ray diffraction (XRD) were used in parallel to measure total crystallinity, and XRD was deconvoluted to estimate the α'/α polymorph fractions; the DSC α'→α exothermic shoulder was used as an independent cross-check. Every annealed condition exceeded the non-annealed baseline ultimate tensile stress (UTS) of 39.75 ± 1.28 MPa. The optimum, 47.00 ± 0.97 MPa at 70 °C/60 min, gave an 18.2% gain. Total crystallinity rose from 8.6% (DSC baseline) to 41.8% (DSC, 90 °C/80 min), with DSC and XRD ranking the conditions consistently. ANOVA confirmed both temperature (30.0% contribution) and time (24.2%) as significant at α = 0.05. The new contribution is a combined strength-crystallinity-polymorph map for desktop FDM-printed PLA: the best-performing specimens are dominated by the disordered α' form, while the stiffer but weaker high-temperature specimens shift toward α. A partial least squares regression on all 50 specimens supports the polymorph-composition role beyond what total crystallinity alone explains. The practical conclusion is that moderate annealing just above the glass transition gives the best balance of crystal content, polymorph character, and geometric stability for FDM-printed PLA.