I am new in modelling of piezoelectric materials, though I have performed
indentation simulations on glasses using Abaqus. I am trying to reproduce
indentation load-depth curves reported by cheng et al. [*Cheng, G., &
Venkatesh, T. A. (2012). Nanoindentation response of anisotropic
piezoelectric materials. Philosophical magazine letters, 92(6), 278-287.*
*].* In this paper, authors have performed 3D spherical indentation
simulations using C3D8E (8-node linear piezoelectric brick) type elements
and C3D4E (4-node linear piezoelectric tetrahedron type elements) elements
for conducting as well as insulating indenter. They have stated followings
regarding boundary conditions (refer Fig. (a) which is taken from Cheng et
al., 2012) for these two cases:
1. Conducting Indenter:
“The contacting regions (i.e., BC) and the bottom surface (i.e., side FE)
are electrically grounded.”
2. Insulating indenter:
“No electrical boundary condition is imposed on the contacting regions or
I am conducting 3D FE simulations of spherical indentation on piezoelectric
material in Abaqus. Piezoelectric material is discretised by C3D8E elements
(refer Fig. (b)), while the indenter was modelled as an analytically rigid
indenter. I have imposed following mechanical boundary conditions for both
type of indenters.
Uz=0 for all nodes at bottom surface (plane z=0)
Indenter is constraint to move along x and y direction, while it is
subjected to a prescribed displacement rate along -z direction.
I have imposed following electrical boundary conditions:
1. Conducting Indenter:
The contact region on top surface and bottom surface are electrically
grounded by applying zero volt (φ=0 volt) in initial step.
2. Insulating indenter
No electrical boundary conditions are imposed anywhere in the model neither
voltage (φ) nor charge (q).
In the attached plots, I have compared indentation load-depth curves
obtained from my simulations with data of cheng et al. (2012) for
conducting and non-conducting indenter in Figs (c) and (d), respectively.
In both the figures, red solid line curve represents data from cheng et al.
(2012). As can be seen, my results are far away from the prediction of
cheng et al. (2012). I don’t know where I am going wrong.
Kindly help me in finding the mistake I am making in these simulations.
[Non-text portions of this message have been removed]