Martensitic twin boundary migration as a source of irreversible slip in shape memory alloys

Ahmed Sameer Khan Mohammed, Huseyin Sehitoglu

Research output: Contribution to journalArticle


The mechanistic origin of fatigue in Shape Memory Alloys (SMAs) is addressed using atomistic simulations. A causal explanation is proposed for the known agreement between the fatigue-activated slip system and the martensitic twinning system. As a model system, the Type II twin boundary (TB) in NiTi B19′ martensite phase is analyzed. TEM-based models have established the presence of disconnections on the TB. Topological models establish the TB migration to depend on the motion of twinning partials on these disconnections. A disconnection is setup within a Molecular Statics (MS) framework. A twinning partial is positioned on it by enforcing continuum displacement fields external to a prescribed core of atoms which is subsequently relaxed under governance of the interatomic potential. The displacement fields are calculated from the anisotropic Eshelby-Stroh formalism and enforced in a non-Cauchy-Born adherent manner to obtain the right core structulre. TB migration is simulated as a motion of this disconnection under applied load. In the presence of a barrier to this motion, a dislocation reaction occurs where a stacking fault emits at the TB while returning a residual negated partial. The emissary fault partial is proposed as a precursor to the resulting slip observed in reverse-transformed austenite.

Original languageEnglish (US)
Pages (from-to)50-67
Number of pages18
JournalActa Materialia
StatePublished - Mar 2020


  • Interface structure
  • NiTi
  • Shape Memory Alloys
  • Slip emission
  • Twin boundary migration

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Ceramics and Composites
  • Polymers and Plastics
  • Metals and Alloys

Fingerprint Dive into the research topics of 'Martensitic twin boundary migration as a source of irreversible slip in shape memory alloys'. Together they form a unique fingerprint.

Cite this