TL;DR
This paper investigates the indentation of hard tungsten films on soft aluminum substrates, revealing that the observed decrease in hardness with increasing indentation depth is primarily due to the strain gradient effect within the tungsten film at shallow depths and the influence of the soft substrate at deeper depths. The authors extend a plasticity theory to accurately model these phenomena, providing a more nuanced understanding of micro-indentation behavior in layered materials.
What Problem Does This Paper Address?
Micro-indentation is a crucial technique for characterizing the mechanical properties of materials at small scales, particularly for thin films. However, the interpretation of indentation hardness can be complex, as it often deviates from macroscopic behavior. For bulk metals, hardness typically decreases with increasing indentation depth and then plateaus. Thin films, especially when layered on different substrates, exhibit even more intricate responses. Specifically, researchers have observed that a hard tungsten film on a soft aluminum substrate shows a monotonic decrease in hardness with indentation depth, never reaching a macroscopic plateau. This behavior is attributed to "strain gradient hardening," a phenomenon where plastic deformation is influenced by the gradient of strain, becoming significant at small scales. However, disentangling the contributions of the hard film's strain gradient effects from the soft substrate's influence has been challenging. This paper aims to address this by developing and applying a refined plasticity model to quantitatively analyze these competing effects.
Methods
The authors employ the conventional theory of mechanism-based strain gradient plasticity (CMSG). To account for the specific material properties of tungsten (a body-centered-cubic metal), they extend CMSG to include the effect of friction stress, which represents intrinsic lattice resistance and is significant in BCC metals. Furthermore, recognizing that experimental indentation depths can be comparable to film thickness, they generalize CMSG to a finite deformation theory. This is crucial because infinitesimal deformation theories can struggle with large deformations and require remeshing.
The core of their approach involves developing an indentation model and implementing it using a finite element method (FEM). They utilize the ABAQUS software with a USER-MATERIAL subroutine (UMAT) to model the contact between a rigid indenter and the tungsten film on the aluminum substrate. This allows for a rigorous treatment of contact, including finite sliding and frictionless conditions. The FEM analysis is performed with a fine mesh in critical areas to capture the localized deformation and strain gradients accurately.
Datasets
The study relies on experimental data from Saha and Nix (2002) for micro-indentation hardness measurements of:
- Bulk aluminum (no thin film).
- A 640 nm-thick tungsten film on a sapphire substrate.
- A 640 nm-thick tungsten film on an aluminum substrate.
These experimental datasets are used to validate the developed CMSG model and to determine material parameters. No new datasets are introduced in this paper.
Key Insights and Ideas
- Decoupling Film vs. Substrate Effects: The study successfully differentiates the contributions of strain gradient hardening in the tungsten film from the influence of the soft aluminum substrate. This is a significant step towards understanding layered material behavior.
- Dominance of Film Strain Gradient at Shallow Depths: For shallow indentations (h << film thickness), the strain gradient effect within the tungsten film is the primary driver for the observed hardness behavior.
- Substrate Dominance at Deeper Depths: As indentation depth increases and approaches the film thickness, the soft aluminum substrate's yielding and relaxation of constraint become the dominant factor influencing hardness.
- Tungsten's Small Intrinsic Length: The strain gradient effect in tungsten diminishes rapidly with increasing indentation depth because tungsten has a relatively small intrinsic material length. This contrasts with materials having larger intrinsic lengths where strain gradient effects might persist longer.
- Finite Deformation is Crucial: The use of a finite deformation theory is essential for accurately modeling indentations where the depth is comparable to the film thickness, avoiding issues with remeshing and capturing large strain effects.
- Friction Stress Importance: Incorporating friction stress is vital for accurately modeling the behavior of BCC metals like tungsten, which have significant intrinsic lattice resistance.
Key Results
- The CMSG model, extended for finite deformation and friction stress, accurately captures the experimentally observed monotonic decrease in indentation hardness with increasing depth for a hard tungsten film on a soft aluminum substrate.
- The model shows that the strain gradient effect in the aluminum substrate is insignificant for the studied indentation depths.
- The strain gradient effect in the tungsten film is important at shallow indentation depths but disappears as the depth increases due to tungsten's small intrinsic material length (lw = 0.23 µm).
- The numerical results for the tungsten film on aluminum substrate are generally higher than the experimental data, which the authors attribute to potential inaccuracies in material properties derived from limited data or assumptions about perfect film/substrate bonding.
Limitations and Future Work
- Material Property Determination: The material properties for tungsten were determined from only three experimental data points for shallow indentations on a sapphire substrate. This limited data might affect the accuracy of the derived parameters.
- Film/Substrate Bonding: The numerical analysis assumes perfect bonding between the tungsten film and the aluminum substrate. Real interfaces might exhibit debonding or imperfect adhesion, which could influence the results.
- Future Work: The authors suggest that future work could explore the effect of interface properties and investigate the influence of indenter tip radius on the indentation hardness, building upon their current framework. Further experimental validation with more precise material property determination would also be beneficial.
Summary updated