Metallurgical approach for the development of heat treatment applied to 316L stainless steel cold spray coatings

Tuesday, May 25, 2021: 12:00 PM
Mrs. Laury-Hann Brassart , Mines ParisTech, PSL Research University, Evry, France, EDF-Lab Les Renardières, Moret-Sur-Loing, France
Anne-Françoise Gourgues-Lorenzon , Mines ParisTech, PSL Research University, Evry, France, Mines ParisTech, PSL Research University, Evry, France
Jacques Besson , Mines ParisTech, PSL Research University, Evry, France
Francesco Delloro , Mines ParisTech, PSL Research University, Evry, France
David Haboussa , EDF-Lab Saclay, Palaiseau, France
Gilles Rolland , EDF-Lab Les Renardières, Moret-Sur-Loing, France
Industries developing cold-spray processes aim at producing dense and resistant coatings. Controlling microstructure and interparticular fracture characteristics of as-sprayed coatings is essential to improve their properties. To do so, post-spraying heat treatment is a promising approach.

This work focuses on the development of such heat treatment based on the recovery and recrystallization phenomenon analysis. Different heat treatment parameters were explored, namely, holding temperature and time, heating rate and heating method. This approach revealed a competition between recrystallization and other microstructural evolution mechanisms, such as precipitation and development of porosities. An optimized heat treatment, allowing microstructural softening and adequate mechanical properties, was sought for.

DSC (Differential Scanning Calorimetry) measurements applied to as-sprayed coatings enabled to target recovery and recrystallization temperature ranges. Then, a variety of heat treatments was applied, involving long-time isothermal holding and short cycles. Microstructure analysis and hardness measurements were conducted to make a first selection of treatment conditions. Mechanical properties were evaluated after selected heat treatments using 3-point bending tests coupled with numerical simulation to provide better understanding of coating fracture behavior. Finally, to help selecting the best candidates, a deep learning algorithm was set up and used to quantify the ductile (dimpled) area fraction of fractured coatings.