Effect of Chemical Composition (Cr/Ni) on the Hysteresis of 17-4PH Stainless Steel

The resistance strain gage was adopted to investigate hysteresis property of 17-4PH steel treated with different proportions of Cr/Ni under the same heat treatment process. The relationship between mechanical properties, microstructure and hysteresis of the material under different proportions of elements was established. The results show that the residual austenite content has an important effect on hysteresis of the material, while the δ-ferrite content has little effect on the hysteresis of the material.


Introduction
The accuracy of force measurement is constantly improved, and the material of elastic components is the key to determine the high accuracy and stability of the sensor that requires elastomer materials with high strength limit, high yield limit, stable elastic modulus, low hysteresis, and excellent fatigue and impact performance [1] .It is known that hysteresis error is one of the most important characteristics of force transducers [2] .17-4PH precipitation-hardened martensitic stainless steel could be an ideal substitute material because of its high mechanical properties and good corrosion resistance.The difference between the values obtained with increasing force and with decreasing force determines the relative hysteresis error which can be calculated by using the Eq. ( 1): where v is the relative hysteresis (reversibility) error of the force transducer, ' is the reading on the indicator with decreasing test force,  is the reading on the indicator with increasing test force and XN is the average reading on the indicator with maximum test force [5] .A schematic representation of hysteresis error is also given in figure 1.
It was determined that the performance characteristics of force transducers were mostly dictated by the heat treatment that was applied to the spring element.The hardest material seemed to exhibit the best hysteresis performance [3,4]   .These researches focused on improving the hysteresis properties of materials by changing the heat treatment.The relationship between the hysteresis properties and microstructure caused by the change of chemical composition has not been discussed.This research intends to study the influence of the change of the micro-structure of the material with different proportions of Cr/Ni on the hysteresis under the same heat treatment process.

Materials and experimental procedures
The content of Cr and Ni in the material are changed to obtain different material structures on the basis of the chemical composition of 17-4PH.The specific material configuration is shown in Table 1.The ingot was forged a size of Φ25mm and then treated by solution (1040℃, 1.5h, air cooling) and aging (480℃, 4h, furnace cooling).The residual austenite content was measured by Bruker D8 ADVANCE X-ray diffractometer with Co target materials.The scanning angle was from 45° to 115° with 1°/min scanning rate.The microstructure of selected area was observed by the JEM-2100 transmission electron microscope (TEM).Nano-indentation experiments are performed on Agilent G200 equipment.The sample was etched to make the microstructure distinguished in the nano-indentation testing equipment.In this paper, resistance strain gauge is used to measure the strain change of the elastomer, so the signal difference of the sensor during the unloading and loading process is the hysteresis, expressed in mv/v.In sensor calibration, the hysteresis of sensor is usually represented by the ratio between the hysteresis and the output of the sensor's full scale.In this test, φ5mm tensile sample was used.The full scale of the sensor is 1500με while the biggest load is 6KN.The FE-J10 standard force machine is used.Force accuracy tolerance is smaller than 0.02%.Ambient temperature is 24℃.Step load is 1KN during loading and unloading.2. For elastomer materials, the yield strength determines the upper limit of the elastomer's pure elastic phase [6] .Therefore, this paper focuses on the change of yield strength of materials.With the increase of Cr/Ni ratio, the yield strength of the material decreases first and then increases, as shown in figure 2.  groups of materials contain a certain amount of residual austenite.The residual austenite content of C# specimen is significantly higher than that of the other 3 specimens, as shown in figure 4. The relationship of the yield strength and content of residual austenite in different Cr/Ni is showed in figure 2. The content of residual austenite is the key role to determine the yield strength.While the δ-ferrite content in D# materials was highest, but the tensile and yield strength of D# materials decreased little, indicating that the δ-ferritic content in the materials is not the key role to determine the yield strength of the materials.Figure 5 presents the TEM micrographs of microstructure of the 4 specimen.No significant ε-copper could be observed in A#, B# and C# specimen, which was due to the small size of ε-copper under the current heat treatment process.The δ-ferritic is observed in D# samples, as showed in figure 5 (d), and ε-copper phase could be observed in the δ-ferritic, as showed in figure 5 (e).Some free of dislocation loops around ε-copper particles can be observed that will be changed into Orowan loops with the dislocation keep on moving.The accumulated Orowan loops will shorten the interparticle distance and increase the driving force for the dislocation to across the ε-copper particles.And this is a typical strengthening mechanism in 17-4PH steel which is known as the source-shortening effect [10] .The nano-hardness of δ-ferrite, residual austenite and martensite were tested on C# sample and D# sample respectively, and the results are shown in Table 3.The indentation morphology on δ-ferrite and residual austenite is shown in the figure 6.The nano-hardness of δ-ferrite is higher than the one of residual austenite because of the strengthening effect of ε-copper particles.The relative hysteresis error test results are shown in figure 7. It can be seen that the maximum hysteresis errors of the four groups of materials all appear in 20%~70% of the full scale range.The maximum hysteresis error of C# specimen is significantly higher than that of the other three groups.
The maximum hysteresis error of B# specimen is the smallest.The maximum hysteresis error of A# materials is slightly lower than that of D#.The variation of relative hysteresis error is consistent with the content of residual austenite in the 4 samples.The content of the δ-Ferrite has little effect on the relative hysteresis error.
The study [7][8] shows that dislocation motion exists when materials are loaded in the elastic stage, and the interaction between dislocation and point defects produces nonlinear anelastic internal friction, and the internal friction of grain boundaries also occurs when the material is stressed.For elastomer materials, reducing the internal friction of the material and reducing the hysteresis error are always the purpose, so it is necessary to inhibit the movement of the dislocation at a lower strain level, while reducing the internal friction level at the grain boundary.The interaction between the dislocation and copper-rich phase is orowan mechanism and shear mechanism [9] , so increasing the dispersion of copper-rich phase can effectively increase the motion resistance of the dislocation and reduce the elastic hysteresis error of the material.The nano-hardness of residual austenite is lower than that of ferrite, indicating that the dislocation motion resistance of residual austenite is lower than that of ferrite phase.At the same stress level, the residual austenite has low resistance to produce internal friction.This is the reason that the residual austenite content almost determines the relative hysteresis level.There are dispersed ε-copper particles in the δ-ferrite phase, which strengthen the matrix obviously and hinder dislocation motion, so the content of δ-ferrite phase has little influence on the hysteresis.

Conclusions
The conclusions of the study can be outlined as follows: 1 With the increase of Cr/Ni ratio, the strength of the material decreases first and then increases.The correlation between yield strength and residual austenite content is strong, but not strong with ferrite content.
2 Microstructural changes play an important role on the hysteresis error of 17-4PH steel spring element.Reducing the residual austenite content can effectively reduce the hysteresis error of the material, and the ferrite content has little effect on the hysteresis.
3 The 0.49 of Cr/Ni ratio is a good chose to design the material for relatively low residual austenite content and small hysteresis error.

Figure 2 .
Figure 2. The yield strength and content of residual austenite in different Cr/Ni

Figure 7 .
Figure 7. Hysteresis test results of spring elements

Table 2
shows the mechanical properties of materials with different Cr/Ni content ratio.The yield strength of the four groups of materials can all reach 1200MPa.The tensile strength and yield strength of B# material are the highest and the tensile strength and yield strength of C# material are lowest.The plasticity is out of consideration.When Cr content is constant and Ni content is increased, the strength of the material decreases.When Ni content is constant, the strength of the material decreases with the increase of Cr content.In order to evaluate the effect of the relative contents of Cr and Ni on the microstructure and properties of the material, the ratio of Cr and Ni content (Cr/Ni) was used, as shown in figure

Table 2 .
Test results of mechanical properties of material