E-Archive
BACK
Science Update
in Vol. 23 - May Issue - Year 2022
Accurate Residual Stress Predictive Models Developed for Carburized Steel Samples
Figure 1. BCC versus FCC crystalline structures
Figure 2. An extensometer on a steel sample being loaded axially
Figure 3. Stress-strain curves for the through-carburized case samples (A) and the composite samples (B). Credit: W. Liang, J. Pineault, F. Albrecht Conle, and T. H. Topper, “Retained Austenite Transformation-Induced Residual Stress Change in Carburized 16MnCr5 Steel,” Journal of Testing and Evaluation. https://doi.org/10.1520/JTE20210457
Figure 4. RA content of through-carburized case samples (A) and composite samples (B) before and after loading. Credit: W. Liang, J. Pineault, F. Albrecht Conle, and T. H. Topper, “Retained Austenite Transformation-Induced Residual Stress Change in Carburized 16MnCr5 Steel,” Journal of Testing and Evaluation. https://doi.org/10.1520/JTE20210457
Figure 5. RS profiles generated from the compatibility and FE models versus actual XRD results. Plots (A) and (B) were generated from the one-cycle load cases, while plots (C) and (D) are from the one-cycle-plus-taper load cases. Credit: W. Liang, J. Pineault, F. Albrecht Conle, and T. H. Topper, “Retained Austenite Transformation-Induced Residual Stress Change in Carburized 16MnCr5 Steel,” Journal of Testing and Evaluation. https://doi.org/10.1520/JTE20210457
Carburization is a heat treatment method that involves placing steel materials in a carbon-dense environment inside a furnace or other enclosed space, with the final goal of increasing the hardness, fatigue resistance, and wear resistance of the surface material, as well as introducing compressive residual stress (RS) in the surface layer. For case hardening, the end material performs better under certain cyclic loading scenarios because the heating and quenching of the material and the absorption of carbon result in a hardened case, providing an increase in material strength, while the core layer stays tough. Similarly, through-hardening provides an increase in strength through the entire thickness of the material. In both processes, beneficial compressive RS is also imparted at and near the surface, further improving fatigue performance. This heat-treating method is often utilized in the automotive and metal stamping industries, as well as other industries that use steel components, in order to make parts stronger and more reliable so they can better withstand applied stresses in service.
Because many of the components that undergo carburization are safety-critical parts, it is extremely important to be able to understand the RS distribution inside the materials. Being able to measure and predict the strain and stress in a component with a particular retained austenite (RA) percentage or with certain surface treatments applied can help manufacturers to understand and improve their parts’ performance.
After carburization takes place, materials are left with altered RS states.1 In addition, although the austenite in the material usually remains untransformed after carburization, transformation of RA may occur during service loading;2 when it transforms into untempered martensite after an input of plastic strain energy, the case layer undergoes volume changes, which in turn changes the RS profile of the material.
In previous studies in which steel components underwent carburization, researchers found that the materials’ compressive yield stress (the stress at which plastic deformation began) was higher than their tensile yield stress due to retained austenite transformation.3 When RA transforms to martensite, the case layer of the carburized sample experiences a volume change. This happens because an austenite-to-martensite transformation involves an atom structural change from a face-centered cubic (FCC) to body-centered cubic (BCC) pattern, which results in a volume expansion. The change in volume also affects the RS of the component.4
In relation to cyclic loading in particular, it has been observed that RA transformation occurs during fatigue cycling, and the rate of transformation depends on the amount of plasticity imparted during fatigue cycling. However, based on different loading spectra, RA and RS can be affected in different ways. Manufacturers would benefit from knowing the resulting RA and RS after various loading conditions so that they can better understand the performance and fatigue life of their components.
In a new study by Liang, Pineault, Conle, and Topper (2021),5 case-hardened carburized 16MnCr5 steel samples underwent axial load testing to examine the effects of RA transformations on RS in the material. The samples’ RA and RS were determined via x-ray diffraction (XRD) measurements on a Proto LXRD instrument. Both a finite element model (FEM) and compatibility model were used to generate predictions of the case-hardened material’s stress-strain behaviour and its RS profile after loading.
The metal samples were heat-treated in three different ways, forming the following groups: case-hardened composite, through-carburized case, and simulated core. Upon testing for hardness, the case and core layers of the composite sample were confirmed to be similar in hardness to the through-carburized case and simulated core.
First, tension and compression strain tests were conducted on the through-carburized case and the case-hardened composite material. When RA transforms, the corresponding volume expansion can be seen in the stress-strain curves. When samples underwent RA transformation, they displayed a high strain at the same stress level during tensile loading, but under compressive loading, there was a lower strain at the same stress level. Moreover, as demonstrated by the deviation of the stress-strain curves in Figure 3, RA transformation begins when plastic deformation begins. This finding is consistent with those of previous studies.6-11
In the through-carburized case experiments, the positions of the tensile and compressive stress-strain curves, together with the curve of the deep-freeze sample (which contained no RA), indicate that much more RA transformed during tension straining than it did during compression straining. This is clear from the tensile curve having a lower yield stress than the deep-freeze sample, while the compressive curve had a higher relative yield stress. Similarly, the case-hardened composite material also showed a greater percentage of RA decomposing under tensile loading (Figure 4).
From Figure 4 below, it is clear that the composite sample (B) showed a greater percentage of RA transformation than the case sample (A) after being placed under tensile loading. In order to understand why this would be the case, we must first consider the RS state of the composite sample after carburization and strain-induced RA decomposition. With biaxial compressive stresses (in the longitudinal and hoop directions) in the case layer, the composite specimen would have higher shear stress than the through-carburized case specimen. With longitudinal strain applied, the composite sample would have a higher shear stress exerted upon it than the through-carburized case. This higher shear stress would cause an earlier yield for the case layer of the composite sample, which could explain the significant change in RA in the composite sample’s case layer, since RA transformation is related to the degree of plasticity a sample experiences.
Next, the researchers wanted to determine how the RS was generated during the carburization process. Using a 25-layer finite difference model that included the simulated carbon profile of the sample after carburization in the case-hardened sample, the RA volume fraction profile prior to any loading was determined. This RA profile, along with the carbon profile, was extrapolated and used to simulate the strain change in each layer of the case-hardened sample.
Once they had the simulated strain data from each layer, the researchers set out to calculate the RS present in the case-hardened specimen. To do so, they developed both a compatibility model and a FEM model that used their previously obtained strain values as inputs for the initial RS calculation. With the compatibility model, they determined the overall longitudinal strain and then the stress present in each layer of the case-hardened sample. With the FEM, they were able to estimate the longitudinal and hoop stresses in the sample.
Based on their findings, the compatibility model provided a relatively accurate longitudinal RS profile, while the FEM was able to accurately predict surface hoop RS.
To study the changes in RA and RS due to loading, Liang et al. subjected the composite samples to four different loading spectra: a ±1% one-cycle load case, a ±0.5% one-cycle load case, a ±1% one-cycle-plus-taper load case, and a ±0.5% one-cycle-plus-taper load case. Proto performed XRD measurements of the case layer after each loading pattern, which allowed them to determine how much RA transformed under each loading sequence. Assuming an FCC to BCC microstructure change (during austenite-to-martensite transformation), the transformation-induced strain was calculated, and the strain expansion in the case layer was used to generate a modified initial RS input.
From the measured RA results before and after four loading scenarios, it was observed that more RA transformed under the ±1% strain cycle than the ±0.5%. In the tapered loading scenario, in which subsequent load cycles decreased incrementally in strain magnitude, the majority of the RA transformed after the first cycle.
For both strain tests, the models predicted asymmetric stress-strain behaviour despite the symmetric tensile and compressive stress-strain inputs. This prediction was due to the RS present in the case-hardened sample. The measurements were in relatively good agreement with the models but displayed minor differences as a result of material biases. However, in the ±0.5% strain test, the FE model more accurately represented the measurements of the sample since it considered RS in the hoop direction.
After the four loading tests, measured RS results were plotted and compared with the predictions of the compatibility and FEMs (Figure 5). As the RA in the case layer decomposed under stress, it was found that the compressive surface RS of the sample increased, which is consistent with previous studies. Based on the RS profiles generated, the study concludes that using a modified initial RS to account for RA-transformation-induced strain change is a promising strategy.
In light of the important safety and performance implications associated with a component’s RS state, the development of models that can predict RS profiles post-carburization could be extremely useful. Carburization is a commonly used method in various metal industries that is known to impart residual stresses, and manufacturers must have a complete understanding of the RS states of their parts, particularly post-loading, in order to make critical safety decisions. By developing two different models that predict reasonably accurate RS profiles in carburized steel after RA transformation, Liang et al. demonstrated that RS can be closely approximated and provide helpful guidance about the behaviour of carburized steel components. Used in conjunction with RA and RS measurement techniques such as XRD, these predictions enable manufacturers and engineers to be more aware of the RS present in their safety-critical parts at any point in their service life.
References
1 P. Hiremath, S. Sharma, G. M.C., M. Shettar, and G. B.M., “Effect of post carburizing treatments on residual stress distribution in plain carbon and alloy steels – a numerical analysis,” Journal of Materials Research and Technology.
https://doi.org/10.1016/j.jmrt.2020.05.104.
2 V. Bedekar, “Effect of nickel on the kinematic stability of retained austenite in carburized bearing steels – In-situ neutron diffraction and crystal plasticity modeling of uniaxial tension tests in AISI 8620, 4320 and 3310 steels,” International Journal of Plasticity 131 (August 2020), https://doi.org/10.1016/j.ijplas.2020.102748.
3 R. W. Neu and H. Sehitoglu, “Stress-Induced Transformation in a Carburized Steel-Experiments and Analysis,” Acta Metallurgica Et Materialia 40, no. 9 (September 1992): 2257–2268, https://doi.org/10.1016/0956-7151(92)90144-4.
4 K. A. Venkata, C. E. Truman, D. J. Smith, and A. K. Bhaduri, “Characterising Electron Beam Welded Dissimilar Metal Joints to Study Residual Stress Relaxation from Specimen Extraction,”
International Journal of Pressure Vessels and Piping (2016), https://doi.org/10.1016/j.ijpvp.2016.02.025.
5 W. Liang, J. Pineault, F. Albrecht Conle, and T. H. Topper, “Retained Austenite Transformation-
Induced Residual Stress Change in Carburized 16MnCr5 Steel,” Journal of Testing and Evaluation. https://doi.org/10.1520/JTE20210457.
6 G. B. Olson and M. Cohen, “Kinetics of Strain-Induced Martensitic Nucleation,” Metallurgical Transac-tions A 6, no. 4 (April 1975): 791–795, https://doi.org/10.1007/bf02672301.
7 M. Y. Sherif, C. G. Mateo, T. Sourmail, and H. K. D. H. Bhadeshia, “Stability of Retained Austenite in Trip-Assisted Steels,” Materials Science and Technology 20, no. 3 (July 2004): 319–322, https://doi.org/10.1179/026708304225011180.
8 S. Thibaud, N. Boudeau, and J. C. Gelin, “TRIP Steel: Plastic Behaviour Modelling and Influence on Func-tional Behaviour,” Journal of Materials Processing Technology 177, nos. 1–3 (July 2006): 433–438, https://doi.org/10.1016/j.
9 C. Müller-Bollenhagen, M. Zimmermann, and H.-J. Christ, “Very High Cycle Fatigue Behaviour of Austenit-ic Stainless Steel and the Effect of Strain-Induced Martensite,” International Journal of Fatigue 32, no. 6 (June 2010): 936–942, https://doi.org/10.1016/j.ijfatigue.2009.05.007.
10 C. García-Mateo and F. G. Caballero, “The Role of Retained Austenite on Tensile Properties of Steels with Bainitic Microstructures,” Materials Transactions 46, no. 8 (November 2005): 1839–1846, https://doi.org/10.2320/matertrans.46.1839.
11 T. Reti, “Residual Stresses in Carburized, Carbonitrided, and Case-Hardened Components,” in Handbook of Residual Stress and Deformation of Steel, ed. G. Totten, M. Howes, and T. Inoue (Materials Park, OH: ASM International, 2002), 189–208, https://doi.org/10.1361/hrsd2002p189.
Proto Mfg. Inc.
12350 Universal Drive
Taylor, Michigan
USA, 48180
Tel. +1.734.946-0974
E-mail: info@protoxrd.com
www.protoxrd.com
