Initial Inclusion Investigations

It has been suggested by the works of Ref 1, Ref 2, and Ref 3 that inclusions are often the controlling feature
for the fatigue response of materials such as Al-7075. Some previous work has been done in the McDowell
Group to model inclusions including the previously mentioned work by Salajegheh [28]. Much of this work
is not currently incorporated into the pipeline structure, but the results of this research as well as those of
Ref 4 will be useful in verifying the ongoing implementation of inclusions in the pipeline.
The work of Ref 4 investigated a wide range of inclusion parameters on stress fields including particle size,
shape, spacing, configuration, alignment, grouping, and microporosity. Out of these parameters, the
researchers found that inclusion shape and inclusion alignment with the loading direction had the biggest
effect on the stress intensity. The objective of the initial inclusion implementation in the current modeling
scheme is to determine if the current Al-7075 modeling capabilites can qualitatively reproduce these results
found in literature.
The inclusions often observed in the Al-7075 alloy are Al7Cu2Fe, Al23Fe4Cu, and Mg2Si which have
respective Young’s Modulus values of 160.2, 139.5, and 94.8 GPa [Ref 5]. The relatively large Young’s
Modulus of the inclusions compared to the Al-7075 matrix (71.7 GPa), combined with the possibility of
debonding and cracking make inclusions a life limiting factor in many aluminum parts. The ABAQUS
UMAT, which contains the constitutive modeling behavior for Al-7075, currently distinguishes inclusions
by defining a specific grain with a different phase identifier, ultimately leading to different material
properties being assigned to these inclusion “grains”. The UMAT defines inclusions using the material
properties of Al7Cu2Fe as the extreme case of Young’s Modulus disparity.
The majority of existing computational inclusion research uses analytically defined shapes such as
ellipsoids and spheres. In order to compare with these existing results, a new microstructure generation
capability to create an analytically defined grain arbitrarily oriented in a single crystal has been created. An
example of an ellipsoid and sphere created with this tool is provided in Figure 1 and Figure 2.

Figure 1. Ellipsoidal inclusion representation rotated 45˚ about Z-axis

 Figure 2. Spherical inclusion representation in pipeline.

The first simulation set used the tool described above to determine if the pipeline could capture the influence
of shape, or aspect ratio, in cyclic stress-strain tests. The test parameters are described in Table 1 below:
Table 1. Simulation parameters for aspect ratio investigation.

Simulation Box Side Length 0.15 µm

Elements per Side 30
Number of Cycles 3
Applied Strain 0.007
Material Al-7075
Boundary Conditions Periodic
Inclusion Volume 6.54E-5 µm3
Aspect Ratio Range .09 - 2.4

Using these test parameters, it was found that the current model could capture the qualitative results found
in literature as shown in Figure 3.The inclusion in these simulations was defined as an ellipsoid with a
primary axis aligned with the loading direction whose length differed from the other two equivalent length
secondary axis. The ellipsoid aspect ratio was defined as the ratio between the primary and secondary axis
lengths. Throughout the simulations, the inclusion volume was held constant as the aspect ratio was varied
by changing the primary and secondary axis lengths. It should be noted that the aspect ratio of zero in the
graph below corresponds to a simple single crystal cell without an inclusion. The stress plotted in Figure 3
corresponds to the maximum Von Mises stress encountered outside the inclusion in the matrix.
Furthermore, the stress was normalized to the maximum stress found in the single cell without an inclusion.
 Figure 3. Normalized Von Mises stress around inclusion as the aspect ratio is varied.

As shown in Figure 3, the pipeline is able to capture the same shape trends expected from the analytical
solution: ellipsoids with more extreme aspect ratios tend to demonstrate higher stress concentration factors
Ref 6. As the primary axis of the ellipsoid increases in length, the maximum Von Mises stress around the
inclusion increases.
The increase in stress around inclusions based on alignment and aspect ratio consequently influences the
strain localization in the area. These inclusion parameters are important for this project because the strain
localization will ultimately correspond to larger FIPs in these areas and an increased propensity for crack
formation. This can be seen from Figure 4 where the maximum equivalent plastic strain in the matrix
increases with the aspect ratio. Similar to the stress plots, the equivalent plastic strain was normalized to
the equivalent plastic strain found in the simulation without an inclusion.

Figure 4. Normalized maximum plastic strain as the inclusion aspect ratio is varied.
The second simulation used the same set of tools to investigate if the pipeline could capture the effect of
inclusion alignment on the material response. The test parameters are described in Table 2.
Table 2. Simulation parameters for the alignment investigation.

Simulation Box Side Length 0.15 µm

Elements per Side 30
Number of Cycles 3
Applied Strain 0.007
Material Al-7075
Boundary Conditions Periodic
Inclusion Volume 6.70E-5 µm3
Inclusion Aspect Ratio 2
Angle 0˚-90 ˚
These simulations consist of an ellipsoid of fixed dimensions with varying angles between the inclusion’s
primary axis and the loading direction. By definition, 0 ˚ corresponds to the primary axis being
perpendicular to the loading direction while 90 ˚ corresponds to the primary axis and the loading
directions coinciding. The analytical solution suggests that inclusions with their primary axis aligned with
the loading direction will experience the highest stress concentration while inclusions with their
secondary axis aligned with the loading direction will experience the least stress concentration Ref 6. The
model was able to capture the effect of inclusion alignment based on Figure 5 where the maximum Von
Mises stress in the matrix was normalized to the stress found in a single crystal without an inclusion.

Figure 5. Normalized Von Mises stress variation as the angle between the inclusion’s primary axis and
loading direction is varied.

Based on the results presented above, the pipeline is able to qualitatively capture the inclusion effect trends
suggested in literature, but a more rigorous verification of the protocols is required. The next step for
inclusion implementation in the pipeline will be an analytical verification that the inclusion representation
matches the elastic solution provided through [Ref 6] and [Ref 7]. Furthermore, the influence of inclusion
debonding and cracking must be investigated and implemented into the modeling framework.
References:
Ref 1 - https://www.sciencedirect.com/science/article/pii/0142112389900546
Ref 2 - https://www.sciencedirect.com/science/article/pii/S2211812814001515
Ref 3 - http://onlinelibrary.wiley.com/doi/10.1111/j.1460-2695.1982.tb01251.x/full
Ref 4 - https://link.springer.com/article/10.1007/s11661-999-0218-2
Ref 5 - https://asu.pure.elsevier.com/en/publications/3d-microstructural-characterization-and-mechanical-
properties-of-
Ref 6 -
https://www.researchgate.net/profile/G_Weng/publication/245358251_Stress_Distribution_in_and_Arou
nd_Spheroidal_Inclusion_and_Voids_at_Finite_Concentration/links/00b4953b31a2693b2b000000/Stress
-Distribution-in-and-Around-Spheroidal-Inclusion-and-Voids-at-Finite-Concentration.pdf
Ref 7 – https://www.sciencedirect.com/science/article/pii/S0020768304003464