Prediction and characterization of surface roughness using sandblasting and acid etching process on new non-toxic titanium biomaterial: adaptive-network-based fuzzy inference System

Abstract

An adaptive neuro-fuzzy system (ANFIS) model was employed to predict the surface roughness. Surface roughening of titanium biomaterials has a crucial effect on increasing the biocompatibility. For this purpose, sandblasted, large-grit, acid-etched (SLA) has been introduced as an effective method to change the surface texturing and roughness. Subsequent processes--polishing, sandblasting and acid etching or SLA--were employed to modify the surface. Alumina particles for surface blasting and Kroll's etchant (3 ml HF + 6 ml HNO3 + 100 ml H2O) for acid etching were utilized in this experiment. This was performed for three different periods of time (10, 20 and 30 s) and temperatures (25, 45 and 60 centigrade). Correspondingly, the Ti-13Zr-13Nb surfaces were evaluated using a field emission scanning electron microscope for texturing, contact mode profile meter for the average surface roughness (Ra) (nm) and atomic force microscopy for surface texturing at the nano-scale. In addition, the surface roughness was reduced in each condition, particularly in extremely high conditions. Significantly, the ANFIS model predicted the Ra amount of textured surface with an error band of 10 %. This research presents an idea to use the ANFIS model to obtain proper biological signs on the roughened surface in terms of surface roughness.