Citation: Rickhey, F.; Hong, S. Stress
Triaxiality in Anisotropic Metal
Sheets—Definition and Experimental
Acquisition for Numerical Damage
Prediction. Materials 2022, 15, 3738.
https://doi.org/10.3390/
ma15113738
Academic Editor: Francesco Freddi
Received: 6 May 2022
Accepted: 20 May 2022
Published: 24 May 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
materials
Article
Stress Triaxiality in Anisotropic Metal Sheets—Definition and
Experimental Acquisition for Numerical Damage Prediction
Felix Rickhey and Seokmoo Hong *
Department of Automotive and Mechanical Engineering, Kongju National University, Cheonan 31080, Korea;
frrickhey@kongju.ac.kr
* Correspondence: smhong@kongju.ac.kr
Abstract: Governing void growth, stress triaxiality (η) is a crucial parameter in ductile damage
prediction. η is defined as the ratio of mean stress to equivalent stress and represents loading
conditions. Attempts at introducing material anisotropy in ductile damage models have started only
recently, rendering necessary in-depth investigation into the role of η here. η is commonly derived
via finite elemnt (FE) simulation. An alternative is presented here: based on analytical expressions,
η is obtained directly from the strains in the critical zone. For anisotropic materials, η associated
with a specimen varies with yield criterion and material (anisotropy). To investigate the meaning of
triaxiality for anisotropic materials, metal sheets made of dual phase steel DP780, and zirconium alloy
Zirlo are chosen. Analytical expressions for η are derived for three popular yield criteria: von Mises,
Hill48 and Barlat89. Tensile tests are performed with uniaxial tension, notch, and shear specimens,
and the local principal strains, measured via digital image correlation (DIC), are converted to h.
The uniaxial tension case reveals that only the anisotropic yield criteria can predict the expected
η = 1/3. The ramifications associated with anisotropy become apparent for notched specimens,
where η differences are highest; for shear specimens, the yield criterion and material-dependence is
relatively moderate. This necessitates η and, consequently, the triaxiality failure diagram (TFD) being
accompanied by the underlying yield criterion and anisotropy parameters. As the TFD becomes
difficult to interpret, it seems more advantageous to provide pairs of principal strain ratio β and
failure strain. Suggestions for deriving representative β and η are made.
Keywords: stress triaxiality; anisotropy; sheet metal; ductile fracture; triaxiality failure diagram;
digital image correlation
1. Introduction
For the numerical prediction of damage in a sheet metal structure during forming
or in crash events, a plethora of macroscopic ductile damage models has been developed.
To reduce weight and consumption of resources, and thus increase efficiency, material
of often-low ductility is exploited up to close to their strain limit, which renders these
models increasingly important. Despite the large variety of damage models available today,
what they share is the incorporation of the strong influence of triaxiality η on damage
and fracture. This insight can be traced back to the early works by McClintock [1,2], Rice
and Tracey [3], Hancock and Mackenzie [4], and Beremin [5], who already pointed out
the importance of η in ductile damage. The inclusion of the anisotropic yield in damage
models has gained attention recently.
Ductile damage models can be divided into uncoupled and coupled models. Un-
coupled, in this context, means that failure is judged based on the current state (e.g.,
equivalent strain, maximum principal stress, plastic dissipation, or some other measure)
of a “virgin” material, i.e., without accounting for the damage-induced material softening
with increasing deformation. Early prominent models are those by Rice and Tracey [3]
and Johnson and Cook [6]; more recent models were presented, for example, by Bai and
Materials 2022, 15, 3738. https://doi.org/10.3390/ma15113738 https://www.mdpi.com/journal/materials