Study of microplasticity in 304L
stainless steel
P. Vaucheret, A. Galtier
(USINOR R&D-IRSID, Maizières-lès-Metz)
Discontinuities of microstructure in steels induce stress
concentrations. Microplasticity is the plasticity which
exists locally in the locations of these stress concentra-
tions, whereas the remainder of the material is in the
elastic range. The initiation of fatigue cracks thus takes
place in the areas where there is some microplasticity.
The microplasticity threshold is the nominal stress in
uniaxial symmetric loading from which microplasticity
exists.
Still applying uniaxial symmetric stress varying stepwise,
some small hysteresis loops are observed under the elas-
tic limit. The area of these loops corresponds to dissipa-
ted energy per unit volume of steel, and would hence be
characteristic of microplasticity. When measurable, the
graph of these hysteresis loop areas versus the stress
amplitudes has the same aspect as the temperature varia-
tions graph. The slope break occurs around 60 MPa,
which would be another measurement of the microplasti-
city threshold.
Experimental methods
The microplasticity threshold can also be determined by
direct observations of the microstructure at the surface
during fatigue tests. Indeed, after at least 2,000,000
cycles, the polished surfaces in the middle of specimens
present discontinuities (at low magnification) or slip lines
(at higher magnification) only for an applied uniaxial
symmetric stress above or equal to 65 or 70 MPa. This
gives a third measurement of the microplasticity thre-
shold.
The material for which microplasticity has been studied
is the AISI 304L stainless steel (austenitic), with a grain
size of 20 µm, an elasticity modulus of 188,000 MPa, an
elastic limit at 0.2 % residual plastic strain of 318 MPa,
and an endurance limit at 200,000 cycles of 230 MPa.
Using fatigue tests, the microplasticity threshold of this
steel has been determined by temperature variation
measurements, hysteresis loop areas measurements and
observations of the surface microstructure of polished
specimens.
In all the above measurements, no progressive cyclic
hardening or softening of the steel is taken into account.
Finally, the elastic limits, the endurance limits and the
microplasticity thresholds of several steels are compa-
red. No apparent relationship has been found. This rela-
tionship should in addition depend of the microstructure
of the material.
Results
Applying uniaxial symmetric stress varying stepwise, tem-
perature steps are observed for high enough frequencies.
Reporting these temperature steps versus the stress ampli-
tude, the curve obtained presents a break of its slope
around 75 MPa. By considerations of dislocation displa-
cements, this stress is the microplasticity threshold. This
threshold has also been identified by studying the slopes
of the previous graph.
Les observations expérimentales de l’échauffement d’éprou-
vettes lors d’essais de fatigue sont cependant différentes de
ces résultats fondés sur l’élasticité linéaire.
■ DÉTERMINATION DU SEUIL
DE MICROPLASTICITÉ PAR
DES MESURES D’ÉCHAUFFEMENT
En effet, pour des fréquences suffisamment élevées (10 Hz
par exemple), deux cas sont possibles. L’échauffement
moyen par cycle peut présenter au cours du cyclage :
Mise en évidence de la microplasticité
– soit une augmentation rapide lors des premiers cycles, un
long palier, puis une seconde augmentation juste avant la
rupture finale (fig. 1),
L’équation de la chaleur, appliquée au cas d’une sollicita-
tion uniaxiale sur un matériau isotrope élastique (2) (3),
donne les résultats suivants :
– soit une courbe en cloche.
– en traction monotone : une variation linéaire et décrois-
sante de la température en fonction de la déformation ;
Ce deuxième cas se produit pour des aciers subissant un
durcissement cyclique.
– en fatigue alternée symétrique : une variation sinusoïdale
de la température, en opposition de phase avec la sollici-
tation, et d’amplitude proportionnelle à l’amplitude de
contrainte.
Pour les aciers présentant une saturation de la température,
l’échauffement seuil, noté θs (stabilisé), varie faiblement et
64
La Revue de Métallurgie-CIT
Janvier 2002