Question about the

[mailhexu]

Hello,
The list of spacegroup allowed in DWBuilder:
R3c (Polarization direction (1-11))
R3m (Polarization direction (001))
P4mm (Polarization direction (001))
p6_3cm (Polarization direction (-1-10))

Pmc2_1 (user-defined)
Amm2 (user-defined)

As I understand, Pnma is non-polar. Could you clarify how the domain wall is formed from it?
Thanks.
HeXu

[mzkhalid039]

Dear HeXu,

Thank you for your question.

Yes, you are correct that the Pnma structure is non-polar. In our example using DWBuilder, we explored the case of CaTiO3, which typically crystallizes in a non-polar Pnma structure under ambient conditions. This structure is characterized by antiferrodistortive (AFD) distortions, which involve tilting and rotating of the oxygen octahedra within the crystal. These distortions also lead to spontaneous ferroelastic strain.

In its parent cubic phase, CaTiO3 exhibits a secondary ferroelectric (FE) instability. This instability is due to the polar displacement of the titanium (Ti) ions within the oxygen octahedra, leading to ferroelectric polarization. DWBuilder generates twin ferroelastic domain walls between the [110] and [1-10] domains, these walls suppress the primary AFD order parameter. As a result, the secondary FE instability becomes active at the twin walls.

Therefore, these twin domain walls possess their own dipole moments within a ferroelastic paraelectric matrix, contributing to the overall properties of the material.

Here are some papers which explain and report the ferroelastic domain walls of Pnma:

[1] Barone, Paolo, Domenico Di Sante, and Silvia Picozzi. "Improper origin of polar displacements at CaTiO3 and CaMnO3 twin walls." Physical Review B 89.14 (2014): 144104.
[2] Goncalves-Ferreira L, Redfern SA, Artacho E, Salje EK. Ferrielectric twin walls in CaTiO3. Physical review letters. 2008 Aug 27;101(9):097602.