In-Situ TEM Nanoscratch of Hard Disc Drives

Table of Contents

Experimental Procedure

Perpendicular magnetic recording (PMR) hard disc drives (HDDs) are a ubiquitous technology that have extensively been used for economical and stable long-term data storage.

As such, these devices are highly engineered to improve their performance. For example, the flying height of the head is only a few nanometers above the platter to permit high-density data storage. Since this only allows limited protective material for the storage media in the platter, thin (2-4 nm) diamond-like carbon (DLC) films are utilized.

As the storage layer’s magnetic moment is perpendicularly oriented to the plane of the disc, even slight amounts of plastic deformation are sufficient to reorient the grains in the film and in turn effectively make the stored data unreadable [1]. It is difficult to evaluate the mechanical performance of a thin film stack where the layers are on the order of 1-10 nm, so sophisticated nanomechanical testing tools are required.

This article describes a recently developed 2D MEMS-based transducer exclusively developed to operate with the Hysitron® PI 95 TEM PicoIndenter®. The equipment was employed to test PMR HDD film stacks in-situ under scratch loading conditions. This technology facilitates high-precision µN-level tests to evaluate the material’s performance under loading conditions much closer to a head-disc collision than earlier possible, all while simultaneously visualizing the deformation mechanisms with high-resolution TEM imaging.

Experimental Procedure

To carry out in-situ scratch testing, hard disc drive device films were deposited onto a silicon wedge substrate. Fabricated by wet etching of silicon, these substrates are employed for in-situ TEM experiments, as they offer an electron transparent region at the apex of the wedge, in addition to a stable substrate for mechanical experiments. Also, the substrates are elevated a number of microns above the rest of the silicon wafer to prevent shadowing of the region of interest by mistilt. As measured by scanning probe microscopy with a Hysitron TI 950 TriboIndenter®, the deposited films had a radius of curvature around 300 nm.

The film stack is made of three primary layers: a 5 nm metallic orientation layer, a 2-3 nm protective DLC layer and a 12 nm recording layer, which is a stoichiometrically equivalent oxide of the orientation layer. Below this are a range of seed layers controlling the films grain size and other properties. This results in a film that consists of columnar grains 5-15 nm in diameter, which were then scratched with the Hysitron PI 95 fitted with a wedge-shaped diamond probe.


A constant normal force was applied for the scratch tests. A piezoelectric element actuates the lateral axis, and the resulting lateral forces and normal displacement are recorded. A range of normal forces from 1-20 µN were applied, resulting in different deformation behaviors and different scratch depths. The indenter went through stick-slip motion at low applied normal forces. During this process, the DLC was debonded from the substrate (as evidenced by the buckling of the film in advance of the tip) and the rounded tops of the columnar grains were plastically deformed and flattened.

In Figure 1, this effect is shown by an example 1 µN normal force scratch. In this case, as soon as a completely buckled mound of DLC is formed in front of the tip, a relatively constant average value is maintained by the lateral force. However, there are various load drops as the tip passes by each individual asperity in the recording layer grains beneath.

Figure 1. - An example 1 µN scratch test: a) Normal and lateral loads and displacements versus time and (b-e) corresponding frames from the in-situ TEM video showing the buckling of the DLC film in advance of the tip and flattening of the asperities in the tops of the grains.

The tip penetrated past the depth of the grain asperities and changes were seen in the deformation mechanism at higher normal forces. Here, the tip plastically ploughed deeper into the material and the grains were forced to rotate and/or bend as the tip passed. This secondary deformation behavior is illustrated by an example 10 µN normal force test. Although the lateral force is more reliable, it continues to increase as the lateral displacement increases, while the tip ploughs further in the wedge structure of the deposited film and the contact area grows.

Depending on the measured results, a transition between the two deformation regimes occurs at roughly 10 µN normal force. In the second regime, the bending of the grains in the recording layer may lead to loss of data in the device. A more extensive analysis is shown in [2].

Figure 2. - An example 10 µN scratch test: a) Normal and lateral loads and displacements versus time and (b-i) corresponding frames from the in-situ TEM video, where here the tip penetrates past the asperities and produces plastic deformation in the recording layer below.


The 2D MEMS transducer is a strong tool for investigating a wide range of mechanical phenomena in-situ TEM when coupled with the - Hysitron® PI 95 TEM PicoIndenter®. In this case, different applied normal forces allowed for different depths and, likewise, different deformation mechanisms to be investigated in an intricate industrial multilayer. Other promising applications are in-situ shearing of nanostructures or particles, basic studies of adhesion and friction and more.


  1. S.C. Lee, S.Y. Hong, N.Y. Kim, J. Ferber, X. Che, and B.D. Strom, J. Tribol. 131, 011904 (2009).
  2. E.D. Hintsala, D.D. Stauffer, Y. Oh, and S.S. Asif, JOM 69, 51-56 (2017).

This information has been sourced, reviewed and adapted from materials provided by Bruker Nano Surfaces.

For more information on this source, please visit Bruker Nano Surfaces.

Ask A Question

Do you have a question you'd like to ask regarding this article?

Leave your feedback