3. Results and Discussion
Figure 2 shows the XRD patterns of the Mn–Cu alloy subjected to four different heat treatment regimes. As shown by analysis, the alloy is predominantly made up of the γ-Mn phase and has a face-centered cubic (fcc) structure when it is at room temperature. Its diffraction peaks are located on the (111), (200), (220) and (311) planes. Upon aging treatment, a unique phase transformation takes place. The diffraction peaks of the aged samples are all associated with the (411) plane, which implies that the α-Mn phase has been precipitated. With the rise in the age temperature, the α-Mn diffraction peaks become more intense and the intensity of these peaks reflects the growing volume of precipitating Mn-rich phases.
At the first stage of aging, the supersaturated γ-Mn phase cannot directly precipitate the α-Mn. Rather, it is decomposed by a particular type of diffusion-controlled phase transformation called spinodal decomposition [
20,
21]. In that process, the alloy naturally splits into two phases with distinct compositions, and each phase has the same fcc structure as the parent phase:
(1) Mn-rich zones (γ-Mn-enriched): they are defined by a high Mn concentration, which is used as the precursor of later stages of α-Mn.
(2) A Cu-rich region (Cu enriched zone): with significant levels of Cu this would eventually be a member of the final matrix.
With the changing aging temperature, the decomposition continues to further form the α-Mn phase. The spinodal decomposition leads to the formation of regions (γ-Mn) with a high concentration of Mn, which is much higher than the nominal composition of this alloy. Under specific conditions of composition and structure, those regions become unstable themselves. After a long period of aging, these unstable fcc areas with excess Mn (γ-Mn) achieve an equilibrium, attaining a more structurally stable body-centered cubic (bcc) α-Mn phase through a martensitic transformation or precipitation mechanism. It is a shear type of transformation, which means that no diffusion occurs, but it leads to alteration of the lattice structure, namely, the transition of fcc to bcc. Thus, the α-Mn phase can be regarded as the end product, which was obtained due to transformation of the initial Mn-rich regions, instead of being precipitated directly out of the starting γ phase.
Here, we present the EPMA results of the samples. The surface scanning images of the solution-treated specimens (
Figure 3a) reveal a uniform distribution of Mn and Cu elements, demonstrating a homogeneous composition of the alloys prior to spinodal decomposition. During the aging treatment, spinodal decomposition occurs in the alloy specimens, and the degree of compositional fluctuation varies markedly with aging temperature (
Figure 3b–d).
Figure 3 intuitively characterizes the spinodal decomposition behavior of the samples during aging. In this process, the initial single-phase homogeneous solid solution transforms into a compositionally modulated solid solution, and the microstructural morphology is strongly governed by the scale of compositional modulation.
The specimens aged at 400 °C and 425 °C (
Figure 3b,c) exhibit coarse Mn-rich domains with intense compositional fluctuation. In comparison, the Mn-rich regions of the sample aged at 450 °C (
Figure 3d) are relatively refined. This phenomenon is primarily attributed to the weakened spinodal decomposition of Mn–Cu alloys at such temperatures. Combined with the Mn–Cu binary phase diagram in
Figure 1, this temperature is inferred to be close to the single-phase stable solid solution region.
Figure 4 shows the variations in mechanical properties of the samples as a result of the solution treatment and aging at different temperatures, and the resulting statistics are captured in
Table 2. The solution-treated sample has a high degree of elongation to fracture and relatively low values of tensile strength (UTS, Rm) and yield strength (YS, Rp0.2). Following the aging process, the tensile and yield strengths are much higher, whereas the elongation (A) and reduction in cross-sectional area (Z) are markedly lower, which means that the aging process improves strength at the cost of a certain amount of ductility.
After further examination, it is found that as the aging temperature reaches 400–425 °C, the tensile strength rises by about 57 MPa, and the elongation is reduced by 5%. At the temperature of 425–450 °C, tensile strength is increased by approximately 46 MPa and elongation by 3%. Such a tendency indicates that the strength of the rise in magnitude and elongation reduction both decline in line with an increase in the temperature used in the aging process. In general, the specimen aged at 450 °C demonstrated the highest overall mechanical behavior, having a tensile strength of 561 MPa, a yield strength of 297 MPa and an elongation of 34%, which is a reasonable compromise between strength and ductility.
The results presented in
Figure 2,
Figure 3 and
Figure 4 reveal that during the aging process, strain energy within the γ-Mn matrix is relieved via decomposition, leading to the formation of Mn-rich and Cu-rich regions and a characteristic spinodal (vermicular) microstructure [
22]. This texture, resembling a woven fabric, reflects the internal compositional modulation and microstructural evolution. Investigations into the spinodal structures formed under varying aging temperatures indicate that the degree of decomposition exhibits a trend of initial increase followed by a decrease as the temperature rises. This decomposition process not only facilitates the martensitic transformation but also exacerbates lattice distortion, thereby substantially enhancing the mechanical properties of the alloy. Consequently, the spinodal structure plays a pivotal role in strengthening the material, providing a critical microstructural theoretical basis for the performance optimization of damping alloys.
Figure 5 presents the damping capacity (tanδ) versus frequency at room temperature (25 °C) of alloys with various aging treatments. The damping capacity of each alloy is typically lowered with an increase in the frequency of testing. There is a notable decrease in tanδ in the low frequency region between 0.1 Hz and 0.2 Hz. After this, the downward trend slows until the minimum is reached at around 2 Hz. Such behavior can be expected of the most popular model of the interface or dislocation damping mechanism, in which the damping capability is greater at lower frequencies.
It is interesting to note that the damping ability of all the alloys in the frequency range of 2–3 Hz can be observed to recover sharply to a peak value and then begin to decrease gradually. The sudden shift in the damping behavior around 2 Hz can be explained by the fact that the motion of internal micro-interfaces can also vary with the frequency. High damping ability in Mn–Cu alloys depends greatly on the reciprocal motion and energy loss between the phase boundaries, twin boundaries, and antiferromagnetic domain walls in the thermoelastic martensite. Of all the samples, the ones with the highest damping capacity are the two samples 3# and 4# followed by sample 2#.
Figure 6 illustrates the temperature dependence of the physical expansion coefficient and loss tangent tanδ for Mn–Cu alloys subjected to different heat treatment regimes.
On average, α_phys is plotted in an increasing direction, which is the expected behavior of thermal expansion (
Figure 6 dashed line). All samples are observed to have an accelerated rise in the region of −150 °C to −130 °C. The growth then decelerates and reaches its maximum at about 200 °C with a value of about 30 × 10
−6 °C
−1. Sample 3# has a peculiarly low α_phys at about 85 °C, which decreases sharply to about 100 °C, then gradually rises with increasing temperature again. Sharp changes in slope or an inflection point on the α_phys curve generally suggest the existence of a solid-state phase transformation. In this work, this phase transformation is the martensitic transformation, and the temperature of the transformation is named T
M. The T
M of the samples 1#, 2#, 3#, and 4# are about 57 °C, 61 °C, 85 °C, and 100 °C respectively.
In
Figure 6, the dashed lines represent the temperature dependence of the loss tangent tanδ for the alloys under different heat treatment conditions over a temperature range from −100 °C to 200 °C, measured at a frequency of 1 Hz. The specimen which was treated only with a solution (1#) has a fairly flat damping response at all temperatures and relatively low tanδ values. Conversely, the aged samples have a gradual reduction in tanδ values in the temperature interval of −100 °C to −75 °C, which reaches its lowest point at around −75 °C. Then, the damping capacity sharply rises and peaks at about 0 °C. The peak temperatures for samples 2#, 3#, and 4# are −5 °C, 0 °C, and 10 °C, respectively. Of these, sample 3# has the maximum damping capacity with a peak tanδ of about 0.05. Further heating causes the damping capacity of the aged samples to reduce even more. It is worth noting that sample 3# displays a small improvement in tanδ in the temperature range of 50–100 °C instead of further decreasing. Above T
M, as the paramagnetic transition begins, the tanδ value drops sharply to a very low level.
Based on the properties of Mn–Cu alloys, commonly displaying volume anomalies because of antiferromagnetic transitions or spinodal decomposition, and using the information obtained in XRD analysis and EPMA results, aging treatment facilitates the development of Mn-rich phase precipitation and thus raises the paramagnetic-to-antiferromagnetic transition temperature. A large antiferromagnetic transition temperature means that the alloy may be maintained in an antiferromagnetic state at a wider temperature interval, which is very important to ensure high damping behavior in room temperature regions and over-room temperature regions.
Figure 7 depicts the morphology of the tensile fracture surface. The solution-treated specimen (1#) shows typical properties of a ductile fracture that are characterized by the existence of dimple areas (dimples) within most of the fracture surface. The distribution of dimples is not even and it is closely connected with the mixed grain of the alloy structure, where the bigger grains represent deep dimples and the smaller ones shallow and more closely packed dimples. Conversely, the number of dimples present at the fracture face with increasing aging temperature decreases, whereas their size distribution grows less heterogeneous. In other regions, tear ridges are evident. Such a morphological transformation is probably related to a greater amount of brittle α-Mn second-phase precipitates in the aged alloys. These brittle particles, when subjected to tensional deformation, may serve as regions where cracks initiate and propagate preferentially, thus decreasing the ductility of the alloy. It is in agreement with the experimentally determined reduction in the elongation and area reduction after aging.
The graphs in
Figure 8 show inverse pole figures (IPF) along with orientation distribution function (ODF) images of the alloy subjected to various heat treatment conditions. The average grain size of the solution-treated sample is 27.06 μm. After aging at 400 °C, 425 °C, and 450 °C, the average grain sizes are 19.25 μm, 26.14 μm, and 23.29 μm, respectively, indicating relatively uniform grain sizes after aging. Combined with the above results, the formation of α–Mn phase caused by aging heat treatment leads to the phenomenon that the grain size of the alloy does not change significantly with an increase in effective temperature, but the strength increases and the elongation decreases. Wang et al. [
23] also discovered the same phenomenon and pattern.
The ODF of the solution-treated sample is randomly oriented with a wide distribution of grain orientations and does not have any significant preferred orientation. It indicates an elimination or reduction in dendrite segregation and compositional heterogeneity in the initial as-worked condition by diffusion in the course of high-temperature solution treatment, giving rise to a homogeneous solid supersaturated solution of all alloying components in the fcc γ phase.
Following aging at 400 °C the ODF has a somewhat spread-out pattern of orientation, low in texture intensity, and hence it is concluded that the alloy is in the recovery or early recrystallization phase, and there is no prominent orientation formation. With rising aging to 425 °C the ODF shows more concentrated clusters of orientation, especially at certain Euler angles (e.g., at sections where φ2 = 0° and 45°), which implies the development of regular high-cube texture components. It demonstrates the fact that the recrystallization process is underway, and grains start to transform into lower-energy orientations, and deformation texture turns into recrystallization texture. At 450 °C, as the temperature continues to increase, texture intensity in ODF achieves the maximum and the distribution is more condensed with a definite preferred orientation present. The ODF analysis in conjunction with the appropriate IPF maps reveals the preferred [001] orientation post-aging at 425 °C changing into a preferred [101] orientation when the aging temperature is increased to 450 °C.
The computed Schmid factor in
Figure 9 represents the relaxation motion of {011} twin boundaries on the {011}<0
1> system with various aging temperatures. The average Schmid factors are 0.33 and 0.29, respectively, in the solution-treated sample and in the sample aged at 400 °C, 0.38 in the sample aged at 425 °C and 0.34 in the sample aged at 450 °C. The sample aged at 425 °C has the highest average Schmid factor.
In terms of damping mechanisms, the dissipated energy capability of the Mn–Cu alloy depends largely on the effectiveness of reversible {011} twin boundary motion during the alternating external stress [
24]. According to Schmid factor analysis, the response efficiency of different crystallographic orientations to applied stress varies significantly. The [001] orientation possesses the highest Schmid factor, meaning that under the same applied stress, the resolved shear stress acting on the twin boundary is maximized, making the twin boundaries most susceptible to motion and consequently dissipating the most vibrational energy. In contrast, the [011] orientation has an intermediate Schmid factor, while the [111] orientation has the smallest, making twin boundary motion the most difficult. Therefore, when EBSD results show an increase in the proportion of [001]-oriented grains with extended aging time, it implies that the fraction of grains in the optimal damping orientation within the bulk material is increasing, enhancing the overall mobility of twin boundaries under alternating stress. Hence, increasing the proportion of [001]-oriented grains through texture evolution during aging is an effective strategy for improving the macroscopic damping capacity. This aligns with the study by Shungui Zuo et al. [
25] on the orientation dependence of damping behavior in Mn–Cu alloys, where multi-directional damping tests on single crystals showed the highest damping capacity in the [001] direction and the lowest in the [111] direction.
Figure 10 shows TEM images of the Mn–Cu alloy samples. In the as-solidified sample (1#), no twins are observed within the grains, while some dislocation tangles are present, and no precipitates are found at the grain boundaries. After aging treatment, twins appear inside the grains, and the twin density shows a trend of first increasing and then decreasing with rising aging temperature (see
Figure 10b–d). Meanwhile, blocky α-Mn precipitates are present at the grain boundaries in all aged samples (see
Figure 10f–h), and their size gradually increases as the aging temperature rises. The possible reasons for these observations are as follows. At lower aging temperatures, atomic diffusion is limited, and the driving force for twin nucleation is insufficient, resulting in a low twin density. As the temperature increases, thermal activation becomes stronger, and atomic migration rates rise, which facilitates twin nucleation induced by stacking faults or localized stress concentrations. In addition, an appropriate amount of precipitates may provide heterogeneous nucleation sites or local stress fields that promote twin growth, leading to an increase in twin density. However, when the aging temperature becomes excessively high, atomic mobility is too strong, making dislocation cross-slip and climb easier. As a result, recovery and recrystallization processes dominate within the grains, and the pre-existing twins may be annihilated or transformed. Furthermore, precipitate coarsening destroys the coherency with the matrix and weakens the local stress fields, causing the twin density to decrease. These combined factors give rise to the observed trend of twin density first increasing and then decreasing with aging temperature.
Aging treatment first induces spinodal decomposition in the Mn–Cu alloy, causing the supersaturated solid solution to separate into a nanoscale alternating structure of Mn-rich and Cu-rich regions. This compositional separation process significantly elevates the martensite start temperature (M
s) of the Mn-rich regions, enabling the formation of a greater number of fine {011} twins at room temperature—the microstructural origin of the high damping capacity in Mn–Cu alloys. Meanwhile, as the aging temperature rises, the growth of grains promotes the formation of movable twinning [
26], resulting in a higher density of movable twinning in larger grains. However, excessively high temperatures during the processing period will also lead to an increase in the amount of α-Mn precipitates, reducing the Mn content within the grains and resulting in a decrease in damping performance.
At the same time, there is a preferential evolution in the grain orientation under the aging of the material. With processes like twin reorientation and grain boundary migration, the fraction of [001]-oriented grains slowly increases.
To sum up, the aging treatment optimizes the twin structure as a result of spinodal decomposition and at the same time enhances the fraction of [001]-oriented grains by developing texture. These two factors, acting together, have a significant positive impact on the damping capability of the alloy.