Effects of alloying, heat treatment and nanoreinforcement on mechanical properties and damping performances of Cu–Al-based alloys: A review

3.2.1 Mechanical alloying (MA)

MA technology is one of the most important ways to prepare new high-performance materials. The materials prepared by the MA process have a uniform and fine microstructure and dispersed strengthening phase, and their mechanical properties are often better than similar materials prepared by the traditional process [38]. Studies have shown that MA in the preparation of Cu-based alloys can form supersaturated solid solutions to promote particle refinement and improve mechanical properties [41,42]. Its working principle is to use a high-energy ball mill to make all kinds of powders mix evenly in the process of ball grinding, promote the mutual diffusion of elements, and finally, realize the function of mechanical alloying [43,44], as shown in Figure 2(a–b). The appearance of lamellar structure marks the beginning of alloying, as shown in Figure 2(c). In the process of ball milling, the elemental alloying accelerated, and finally, the fine-grain alloy powders were formed [43]. During this process, reasonable selection of milling parameters plays an important role in obtaining an ideal microstructure [44]. Shafeeq et al. [40] investigated the effects of ball-milling parameters, such as ball/powder ratio, ball-milling speed, and ball-milling time, on the microstructure and phases of powder metallurgy Cu–Al–Ni–Ti alloys. It was found that grinding in a ball/powder ratio of 40:1 and at a speed of 200 rpm for 40 h achieved a uniform distribution of alloying elements, which were the best conditions for MA. Mayahi et al. [45] prepared Cu–13.2Al–4Ni alloy by ball milling and found that high-energy collisions during the mechanical alloying process would lead to an increase in the density of structural defects, lattice strain, and a decrease in the crystallite size. In addition to the above parameters, the choice of ball-milling medium also plays an important role in the final powder formation. Alcohol and stearic acid are usually added as ball-milling media to avoid oxidation and cold welding, whereas alcohol is more readily available and used more widely [46]. In addition, many research works on Cu-based alloy powder metallurgy processing focused on mechanical alloy powder obtained by ball milling [47]. Kaouther et al. [48] promoted metallurgical bonding of nanosized Cu–Nb–Al alloy powders after ball milling for 200 h at a speed of 600 rpm, resulting in desirable mechanical strength and ductility of sintered alloys. In addition, the proper mechanical alloying process is beneficial to the uniform distribution of elements and shortens the processing time [40]. Therefore, MA as an effective way to promote powder alloying is widely used in the whole powder metallurgy field.

Figure 2

(a) Schematic diagram of ball milling, (b) enlarged view of ball milling tank, and (c) alloying process during ball milling.

3.2.2 Hot-press sintering (HP)

HP is a near-net forming technology for preparing highly compact alloys [47,49]. The method also has the advantages of high production efficiency and uniform product organization [50]. Xiao et al. [51] prepared a Cu–11.9Al–5Ni–2.2Mn alloy with high density and uniform element distribution by HP technology under the following conditions: temperature, 950°C; pressure, 40 MPa; holding time, 120 min; and vacuum degree, 10⁻¹ Pa. Liu et al. [52] obtained a TiNiNb alloy with a relative density of 98.8% under the hot-pressing conditions: temperature, 900°C; pressure, 35 MPa; vacuum degree, 1.6 × 10⁻³ Pa; and holding time, 1 h. Rodrı́guez et al. [53] prepared a Cu–14.2Al–4.2Ni alloy with good ductility after hot pressing at 900°C for 1 h. Vajpai et al. [54] prepared a Cu–Al–Ni alloy with reduced grain size and improved fracture strength by this method, which is salutary to the improvement of the comprehensive properties of the alloy. In addition, the selection of hot-pressing temperature has an important effect on the properties of the alloy [55]. Wakai et al. [56] also showed in their study that extremely high hot-pressing temperature will lead to excessive growth of grains, resulting in weakening of the effect of fine-grain strengthening. Wang et al. [57] prepared a Ti–22Al–25Nb alloy by vacuum hot-pressing sintering process and found that the alloy showed good comprehensive properties under the sintering process of 1,250°C, with the tensile strength and elongation reaching 620.53 MPa and 7.44%, respectively. Therefore, it is of great significance to select suitable technology in the process of HP to obtain materials with ideal properties.

3.2.3 Hot isostatic pressing (HIP)

HIP is another important method to obtain high-performance compact materials. It is a process of applying equal pressure and high temperature to the product in all directions and obtains the process products with a high internal density under the condition of high temperature and high pressure [58,59,60]. By adjusting the experimental parameters of the technology, temperature, and pressure, the microstructure and performance of the final prepared material can be controlled [61,62,63]. Cai et al. [61] prepared a compact and defect-free Ti6Al4V titanium alloy after HIP treatment at a temperature of 900–930°C and a pressure greater than 100 MPa. The strength and fatigue life of the alloy were significantly improved. Ran et al. [58] used hot isostatic pressing at 520°C and 100 MPa for 2 h to prepare an A356 alloy with a fine grain structure, which significantly reduced the porosity of the sample and improved the strength of the alloy. Jiang et al. [64] achieved complete densification and homogenization of the Cu/Ti₃SiC₂/C/graphene nanocomposites between the direction parallel and perpendicular to the pressure direction through hot isostatic pressing. Many studies have shown that the materials prepared by this method generally have a uniform fine grain structure, good process performance, and mechanical properties [65,66,67]. Therefore, due to its superiority, this technology has been widely used in many fields such as machine manufacturing, aerospace, and aviation.

3.2.4 SPS

With the increasing types and demands of new functional materials, SPS has been widely used as a new technology [68,69,70]. SPS has the characteristics of fast heating speed, short sintering time, controllable structure, energy saving, and environmental protection, making it suitable for the preparation of metal materials, ceramic materials, and composite materials [71,72]. The technology is affected by sintering temperature, holding time, pressure, and current, and the powder can be rapidly prepared into dense blocks with appropriate process parameters [73]. Richard et al. [74] obtained Cu–13.01Al–3.91Ni–0.37Ti–0.24C alloys with excellent properties by using SPS technology, which provided rapid heating for powder sintering through the graphite mold and the alternating current, and the sample placed therein at a maximum uniaxial pressure of 99.5 MPa. Eze et al. [75] prepared a Cu–Ti-based alloy with a strength of 749 MPa through SPS at 650°C and 50 MPa for 5 min. Qiang et al. [76] studied the size and distribution of reinforcers in the matrix of mechanical alloying and powder metallurgy processes, as shown in Figure 3. It was found that the hard chromium particles were further deformed after mechanical alloying, and the broken chromium was then embedded into the soft copper particles, regardless of the pretreated state of the powder. Finally, the material with high mechanical properties was obtained by SPS. In addition, Tan et al. [77] successfully prepared a high density and high strength 7056 Al alloy by SPS technology and confirmed that the diffusion of elements between the aluminum alloy matrix and the reinforcing body occurred during the SPS process. As a novel technology, SPS plays a guiding role in efficiently obtaining high-quality materials [78,79].

Figure 3

Size and distribution of reinforcement in the matrix after mechanical alloying and SPS processes [76].

4.1.1 The influence of alloying elements on the microstructure

As well known, the microstructure of the material will be affected by the composition and then affects the performance of the material [89]. This was mainly because the addition of alloying elements had low insolubility and would generate fine compounds due to its reactions with the parent phase elements, thus hindering the growth of grains and changing the microstructure of the alloy [90]. The properties of Cu–Al-based alloys mainly depend on the high-temperature β phase, and the behavior of martensite also strongly depends on the type and content of the alloy. Their properties are largely affected by the grain size, phase composition, and thermoelastic martensite [88].

Many studies have shown that modulating the microstructure of alloys by alloying is a very effective method [89,90,91]. Saud et al. [88] added Mn, Ti, and Co elements to the Cu–11.9Al–4Ni alloy, and the alloy was affected by the accumulation and precipitation phases at grain boundaries, which hindered the grain growth and reduced the grain sizes to 450, 650, and 320 μm, respectively. Zhang et al. [91] found that the grain size decreased significantly after adding different contents of Nb to the Cu–13Al–4Ni alloy, which was mainly related to the newly formed (Al,Ni)Cu₄Nd. In addition to the above elements, Ti is often added to alloys to improve the microstructure of substrates due to its low solid solubility in the β phase (less than 0.05 wt%), which tends to form fine and evenly distributed precipitates to refine the grain size [92,93,94]. Saud et al. [90] added different contents of the Ti element to the Cu–Al–Ni alloy; the decrease of the grain size and the increase of the precipitate phase reduced the mobility of the interface and changed the morphology of martensite. Similarly, Sampath [89] added Ti and Zr to Cu–Al–Ni, and the fine precipitates rich in Ti and Zr elements prevented the nucleation and growth of grains through the action of pinning. These results indicate that alloying plays an important role in microstructural adjustment.

In addition to the addition of alloying elements, numerous rare earth elements can also obtain good properties [95]. Rare earth elements such as Ce, Gd, Sn, and Te have been widely used in alloys and satisfactory results have been obtained [96,97,98]. The addition of rare earth elements in Cu–Al alloys has also been developed in recent years. Yang et al. [99] found that the addition of 0.43 wt% mixed rare earth to the Cu–Zn–Al alloy reduced the grain size of the alloy. Dalvand et al. [95] added 0.04 wt% rare-earth elements (Ce + La) to the Cu–12Al–3Ni–0.6Ti alloy, resulting in a reduction of about 30% in the average grain size due to the formation of rare-rich precipitated phases, as shown in Figure 5. Meanwhile, the addition of Gd and other alloying elements in the Cu–Al alloy also plays a crucial part in refining the grain size [100].

Figure 5

SEM micrographs of the alloys: (a and b) Cu–12Al–3Ni–0.6Ti alloy, (c–e) Cu–12Al–3Ni-0.6Ti–RE alloy, (f) table of element distribution in numbered areas in (b and e), and (g) simple schematic diagram of second phase precipitation [95].

Apart from the change in the grain size, the phase composition of the alloy is also affected by the alloy composition. The lattice parameters of the orthogonal element of the 18R structure martensite with thermoelastic effect can be expressed by ref. [88].

where d is the distance between the planes; (h, k, l) is the Miller index; a, b, and c are the axis length; and β is the interaxial angle.

Zhang et al. [100] found that the Cu–Al–Ni alloy with a Gd element was mainly composed of 18R martensite and the second phase, and the 18R martensite with high thermoelastic behavior and controlled growth in the adaptive group existed in a typical zigzag morphology [86]. Zhang et al. [91] added different Nb contents to the Cu–13Al–4Ni alloy, in addition to the 18R structure and 2H structure martensite, a newly formed (Al, Ni) Cu₄Nd phase appeared in the substrate, as shown in Figure 6. Sari [82] found that γ 1 ′ and β 1 ′ martensite appeared in the Cu–11.9Al–3.8Ni alloy with the addition of 2.5 wt% Mn. γ 1 ′ martensite came out in the coarse variant, while β 1 ′ martensite showed up in a typical zigzag shape. With the addition of Ag in the Cu–Al–Ni alloy [81], the Ag element is easily diffused into the matrix to react and form the precipitate phase, which will promote the formation of 18R martensite. Yildiz [101] studied the effect of Ta content on the microstructure of the Cu–12.4Al–4Ni–1Mn alloy. The results showed that with the addition of the Ta content, the microscopic composition of the alloy changed from 18R and 2H double martensite phase to 18R single martensite phase, and the volume fraction of the Ni₂Ta phase increased with the increase of the Ta content. These changes in the phase composition are of great significance for obtaining rational comprehensive properties of alloys.

Figure 6

TEM images of Cu–13.0Al–4.0Ni–1Nd alloy: (a) bright field image, (b) electron diffraction of region A in (a), (c) bright field image, (d) electron diffraction of region A in (b), (e) bright field image, and (f) electron diffraction of region C in (e) [91].

It must be noted that electron probe X-ray microanalysis (EPMA) is also one of the commonly used methods for modern microscopic analysis in the study of alloys. It has great advantages in the quantitative analysis of elements, distribution of trace elements, and analysis of inclusions and precipitated phases. It is even possible to measure compositional differences in microscopic phase distributions at the micron or even nanometer scale [102]. Lin et al. [103] found that in addition to the second phase discovered by SEM, the Al₆Fe phase was also detected when EPMA was used to measure the corrosion resistance of trace Sc and Y in the high-conductivity cold-drawn Al–0.2Ce base alloy. Shang et al. [104] used EPMA to successfully determine whether Mg content can be ignored in the composition calculation in the study of the second phase of the AZ31 magnesium alloy containing Ca, Si, and Ce. Meanwhile, it can fully identify each phase present in the inclusions. The composition of Al₉Cu_11.5 and Al_0.939Cu_0.987 phases is very close and they were usually identified as the AlCu phase in previous studies. However, Liu et al. [105] successfully differentiated them using EPMA. In addition, in the analysis of the second phase entrapped in the tungsten wire, it was found that the elements Y, La, and Hf, which cannot be detected in the second phase by EDS, could be detected by using EPMA [106]. EPMA can also detect the type of deviations of elements on martensite to determine the cause of martensite formation, which is not obvious if the energy spectrum detection is used [107]. In conclusion, EPMA helps to improve the precise positioning of the analysis area and the accuracy of the analysis. The analysis results are highly intuitive and will play an increasingly important role in future trace analysis. Therefore, researchers are encouraged to use EPMA for more accurate microzone analysis in future studies involving alloys.

In addition to the use of alloying elements, the addition of inoculants will significantly affect the microstructure and performance of the alloy while refining grains [108,109,110]. In order to achieve a better optimization effect, researchers cleverly combined alloying elements and grain refiners and studied the effect of their combined action on the alloy. Ding et al. [111] made use of the compound refining effect of the Cu₅₁Zr₁₄ incubator and Ti element to significantly refine the grain size of the Cu–Al–Ni alloy and improve the properties of the alloy. The inoculant refines the grain mainly by providing a heterogeneous nucleation core to the melt, while Ti refines the grain mainly by preventing grain growth. In addition, researchers also showed that the orientation relationship between the inoculant and the matrix alloy, the atomic matching, and the interface bonding state had significant effects on the refining effect of the alloy [112]. When the matrix alloy and the inoculant had similar atomic arrangement rules at the interface, the wettability and lattice matching between the two were better, which could achieve a good refining effect [113,114]. Figure 7 shows a simple schematic diagram of lattice matching between the matrix alloy and the inoculant. Wang et al. [115] reported that the morphology and the stress state of the second phase/matrix alloy interface depend on the difference in physical properties between the base alloy and the second phase. The different thermal expansion coefficients between the two phases would cause different shrinkage of the specimen during the solidification and cooling process, resulting in internal stress and leading to the appearance of the second phase. Moreover, only inoculant particles with appropriate average size had the best refining effect on the target alloy, while inoculant particles with inappropriate size might be pushed to the grain boundary [116]. Therefore, to obtain better performance, factors such as lattice matching degree and interface state between the inoculant and matrix alloy are very important in the selection of the inoculant [117,118,119].

Figure 7

Schematic diagram of lattice matching between the base alloy and the wetting agent. (a) Coherent interface (b) semi-coherent interface.

4.1.2 The influence of alloying elements on mechanical properties

The mechanical properties of alloys are usually controlled by many parameters, such as the grain size, vacancies, grain boundaries, phase sequence, precipitation, structural morphology, and dislocations [82]. To maintain the continuity of strain, Cu-based alloys with large elastic anisotropy are usually prone to stress concentration [120]. Meanwhile, the coarse grain size and the grain boundary segregation of impurity elements make the alloy prone to brittle fracture. In order to improve the above problems, researchers have achieved good results in improving the properties of the alloy by achieving fine-grain strengthening through alloying [80,81,82]. It is worth noting that grain refinement can significantly affect the strength of the alloy [121,122]. The Hall–Petch formula gives the relationship between the yield strength and the average grain diameter, as shown in equation (2) [117]:

Among them, σ s is the yield strength of the alloy, σ o is the resistance to deformation within the grain, K is the influence coefficient of the grain boundary structure on the deformation, and D is the average diameter of the grains. It can be seen from the above formula that the yield strength of the alloy is closely related to the grain size. As the grain size decreases, the yield strength increases, that is, the fine-grain strengthening is confirmed.

With the addition of the Nb content in the Cu–13Al–4Ni alloy, the grain size of the alloy was reduced, and the pinning effect of the second phase (Al, Ni) Cu₄Nd prevented the movement of dislocations, resulting in the increase of the compressive strength from 580 to 787 MPa, and finally reaching 940 MPa [91]. The compression fracture strength of the Cu–Al–Ni alloy is obviously improved by adding Gd and Fe to the alloy because the grain boundary is strengthened [112]. Moreover, some results showed that the addition of appropriate alloying elements could improve the ductility by strengthening grain boundaries without forming γ-phase boundaries [113]. Besides, other researchers also made some progress in this regard. After the addition of different Gd elements to Cu–13.0Al–4.0Ni, the newly formed AlCu₄Gd hindered the growth of grains, refined the grains, and improved the mechanical properties of the material [100]. The fracture strength of the alloy was significantly increased from 580 MPa to 1,200 MPa. The hardness of the Cu–14Al–4.5Ni alloy increases with the addition of the hard substance Ce, which is difficult to dissolve in the matrix, and the effect of grain refinement is achieved so that the strength of the alloy is also improved [114]. When 2.5 wt% Mn is added to the Cu–11.9Al–3.8Ni alloy, the grain refinement is realized, and the high fracture strength and fracture strain are obtained [85]. Similarly, the addition of Ag in Cu–Al–Ni also increases the fracture strength of the alloy and changes from the brittle fracture mode to the mixed fracture mode. It indicates that Ag precipitation at the grain boundary, on the one hand, impedes the dislocation movement; on the other hand, it reduces the stress concentration and improves the mechanical properties of the material [81]. In addition, Dalvand et al. [95] found that when Ti and rare earth elements were added to the Cu–12Al–3Ni–0.6Ti alloy, the fracture showed a combination pattern of brittle fracture and ductile fracture, as shown in Figure 8. It was found that the dimples appeared around the Ti element and rare earth rich elements, which inferred that the addition of alloying elements and rare earth elements were likely to delay the intergranular fracture and improve the ductility of the alloy.

Figure 8

(a–d) SEM images of tensile fracture of Cu–12Al–3Ni–0.6 Ti–Re alloy and (e) EDS microscopic analysis results of fractured numbering points [95].

4.1.3 The influence of alloying elements on the martensite transformation

The typical thermoelastic martensite structure of Cu-based alloy highlights a number of properties. The alloy is mainly composed of a high-temperature austenite phase and a low-temperature martensite phase. Martensite is the product of the austenite parent phase under extreme cooling or stress-induced under high-temperature conditions [123]. The phase transition temperatures of the two are A _s, A _f, M _s, and M _f, respectively, where A _s stands for the beginning temperature of the austenitic transformation, A _f for the ending temperature of the austenitic transformation, M _s for the beginning temperature of MT, and M _f for the ending temperature of MT. The value of A _s − M _s represents the hysteresis degree [123]. The phase transition temperature is affected by the element and element content, and the slight deviation of the composition will cause the increase and decrease of the alloy phase transition point [124]. The relationship between M _s and components is shown in equation (3) [24]:

where C is a constant; A i is the number of elements; and a i is a constant corresponding to the appropriate quantity of the element.

For the convenience of calculation, the typical phase transition temperature of the Cu–Al–Ni alloy is calculated by equation (4) [24]:

In addition, the dependence of the M _s temperature on the composition is related to the free energy difference between the parent phase and martensite. According to the Salzbrenner and Cohen equilibrium temperature calculation formula, the temperature T 0 at which the chemical energies of the austenite and martensite phases are equal can be calculated. As shown in equation (5) [125],

The phase transformation process of the alloy can be reflected by the different scanning calorimetry (DSC) curves, in which the endothermic peak and exothermic peak can be observed. The endothermic peak refers to the reverse MT during heating, and the exothermic peak refers to the MT during cooling. The main reason for the curve change is the influence of thermodynamic parameters, such as the change in enthalpy and entropy. The total area under the peak gives the change in enthalpy, ∆H, and the entropy of the intermediate phases can be calculated by dividing the change in enthalpy by the equilibrium temperature (T ₀). It has been found that the enthalpy of martensite to austenite transformation is greater than that of reverse transformation, indicating that the forward transformation requires more heat than the reverse transformation [126]. Taking the anti-martensitic process as an example, the change in entropy can be calculated according to equation (6) [127]:

where ∆S is the change in entropy; ∆H is the change in enthalpy; and T 0 is the equilibrium temperature between martensite and austenite phases. The higher the entropy, the more martensite will be transformed, which is beneficial to the damping performances and practical application of the alloy.

At the same time, the change in entropy will also affect the heat capacity of the alloy. The specific heat capacity can be calculated by equation (7) [128]:

where d Q / d t is the heat flow of the DSC curve, m is the mass of the sample, d T / d t is the heating rate of the sample, T is the temperature, and t is the time.

Therefore, the grain size and thermodynamic parameters of the alloy have an important effect on the transformation temperature of the alloy. Zhang et al. [91] found that the resulting (Al,Ni)Cu₄Nd changed the Al content in the matrix alloy and reduced the MT temperature after adding different contents of Nb element to the Cu-14Al–3Ni alloy. Adnan et al. [114] added different contents of Ce to the Cu–14Al–4.5Ni alloy, and the phase transition temperature of the alloy moved toward the endothermic direction. Waitz et al. [129] reported that grain boundaries in TiNi-based alloys can provide an additional transition energy barrier for the martensitic transition, thereby reducing the martensitic transition temperature. When 2.5 wt% Mn is added to the Cu–11.9Al–3.8Ni alloy, the phase transition temperature decreased due to grain refinement and the stable parent phase, and multiple peak values appeared due to the transition of multiple interfaces [82]. Mallik and Sampath [128] showed that after adding Zn and Ni to Cu–Al–Mn, since the two elements can dissolve in Cu to form a solid solution, the hardening of the solid solution increased the phase transition temperature. Due to the low solubility of Fe, Cr, Ti, Si, Mg, and Cu, fine precipitate particles were formed to reduce the phase transition temperature. At the same time, the addition of Pb would hinder the phase transition due to the formation of a large amount of precipitate phase. In addition, the addition of alloying elements also affected the thermodynamic parameters. The enthalpy and entropy of the Cu–11.9Al–4Ni alloy phase transition increased gradually with the addition of the Ti element. When the Ti element content reached 1 wt%, the enthalpy and entropy reached the maximum value. Because the thermal stability of the alloy was inversely proportional to the enthalpy value, the stability of the alloy decreased gradually with the addition of the Ti element. This might be related to the content and behavior of the precipitated phase and the morphology of martensite [90].

4.1.4 The influence of alloying elements on damping performances

As a new kind of functional material, Cu-based alloy has become a research hotspot due to its excellent damping performance. In addition to the dissipated energy caused by the hysteretic motion of defects and dislocations, its high damping capacity is also due to the frictional energy dissipated between martensite/martensite interfaces, parent/martensite interfaces, and twin interfaces [5,130]. Damping characteristics are generally composed of the following three parts, as shown in equation (8) [131]:

where IF Tr ( T ) is the transient damping, which only appears in the heating or cooling process; IF PT ( T ) is a phase change term, which is closely connected to the process of phase change, leading to the appearance of peaks in the damping spectrum; and IF Int ( T ) is an intrinsic term, which is related to the background internal friction of the parent phase and the new phase.

The maximum damping value of the alloy appears in the thermoelastic martensite transformation temperature range since there is usually plenty of transformed martensite in this temperature range [93]. The peak value within the phase transition region can be expressed by equation (9) [131]:

where V m is the volume fraction of the transformation martensite, ω is the angular frequency where stress is applied, and φ ( V m ) is a monotonic function of the volume fraction of the phase transition. Assuming that ( d φ / d V m ) is a constant for all thermoelastic martensite, the equation states that the damping capacity of the alloy is proportional to the number of transformed martensite.

Therefore, both the amount of martensite and the density of the different phase interfaces play a vital role in the damping properties [93]. Increasing the interfacial density can be achieved by refining the grain size and martensite, for which researchers have done tremendous research. According to the Cu–Al binary phase diagram, the content of the Al element plays an important role in the properties of the alloy. For example, an increase in the Al content decreases the damping properties. This is due to the decrease in the amount of martensite and the formation of γ₂ that inhibits the mobility of the interface [131]. Sutou et al. [132] reported that the addition of B, Ni, and Si elements can significantly reduce the grain size of the Cu–Al–Mn-based alloy, which is very important for enhancing the damping performance. After adding the Co element to the Cu–Al–Ni alloy, due to the increase in the number of grain boundaries, there was compressive stress at the adjacent interface, which was pernicious to the movement of the interface [133]. In addition, the addition of rare earth elements can also improve the damping characteristics of the alloy. Lu et al. [86] showed that the appropriate content of Ce elements can significantly improve the damping properties of the alloy, as shown in Figure 9. The results showed that the alloy obtained the best damping performance when the Ce content was 0.05 wt%.

Figure 9

Damping characterization of Cu–12Al–5Mn alloys with different Ce contents: (a) the variation trend of tan δ with strain amplitude and (b) the variation trend of storage modulus with strain amplitude [86].

4.1.5 Summary of the alloying mechanism

Alloying plays a crucial role in the adjustment of the microstructure, which in turn is closely related to the properties of the alloy. The common alloying elements or rare earth elements are added for the purpose of reducing the large grain structure of the Cu–Al-based alloy itself and thus optimize the comprehensive properties.

Fine-grain strengthening, a common strengthening mechanism, has the characteristic of simultaneously improving the strength and ductility of the alloy [86]. This is because the increase in the number of grain boundaries after grain refinement can effectively prevent dislocation movement and crack extension [86]. When the material resists stress, the same amount of deformation will be uniformly distributed in more grains. The length and number of dislocation clusters within the grains will be reduced, which effectively reduces the stress concentration and improves the ductility of the material [133]. In addition, the inhibition of dislocation motion achieves an effective improvement in mechanical properties.

Grain size also has an important effect on the phase transition temperature and damping properties. In general, as the average grain size decreases, the nucleation rate of austenite increases, which will promote the reverse MT. At the same time, the martensitic phase transition temperature moves to a lower temperature. The former is due to the stabilizing effect of grain boundaries on austenite, while the latter is due to the fact that both the martensite/martensite interface and grain boundaries are preferred locations for austenite nucleation [117].

The strong sensitivity of the phase change temperature to grain size provides us with a relatively simple way to control the phase change temperature [113]. At the same time, grain refinement has a double-sided effect on the damping capacity of the alloy [93,131]. On the one hand, grain refinement can reduce the interlayer spacing of martensite, which increases the number of interfaces per unit volume and facilitates the improvement of damping performance; on the other hand, with the excessive reduction of interlayer spacing, the compressive stress at adjacent interfaces and the bonding force at grain boundaries increase, leading to a decrease in the interfacial slip and a decrease in damping performance. Therefore, adjusting the composition of the alloy is an effective way to improve the mechanical properties and damping properties of the alloy.

4.2.1 The effect of heat treatment on the microstructure

The unique microstructure of Cu-based alloys, such as grain size, martensite type, and the presence of second-phase particles, provide excellent mechanical properties and high damping performances [140,141]. These microstructures are largely affected by the heat treatment process. When the alloy is heated to the high-temperature β-phase region for solution treatment, the increase of temperature and entropy will promote grain growth. When the grains grow, there will be a large slip on the grain boundary during the deformation process, and then the stress concentration will be generated at the grain boundary to maintain the coordination between the grain boundaries. When these stress concentrations accumulate gradually and reach the bearing limit of grain boundary, cracks will occur at the grain boundary and lead to fracture. For the sake of reducing negative effects, researchers have carried out a lot of studies on the heat treatment process.

Ren et al. [142] obtained Cu–Al alloys with different size grains by controlling the heat treatment temperature, as shown in Figure 10. It is found that the grain size of the alloy increases with the increase of temperature, especially from 600 to 800°C. Dalvand et al. [143] studied the effect of isothermal aging on the microstructure of Cu–12Al–3.5Ni–0.7Ti–0.05RE (RE = Ce, La) alloy, and found that the effect of aging at 250°C was small, but the long aging at 350°C led to the decomposition of the parent phase into equilibrium γ₂ and α phases. Qader et al. [123] studied the microstructure evolution of Cu–13Al–3Ni–4Hf during aging and found that β-phase (Cu₃Al) would be generated after aging above 973 K, which would change the microstructure of the alloy. Suresh et al. [140] conducted aging treatment on Cu–13.4Al–4Ni alloy at 473–573 K and found that γ₂ precipitates appeared at 473 K and the volume fraction of γ₂ precipitates increased with the increase of aging temperature. Yildiz [144] observed the microstructure of Cu–Al–Ti–Ta alloys at different cooling rates. The results showed that water cooling and air cooling mainly produce a large amount of 18R and a small amount of 2H martensite, and furnace cooling produced a sharp increase in the amount of 2H martensite. Wang et al. [145] treated the Cu–11.9Al–2.5Mn alloy with different cooling methods and showed that all samples except furnace cooling clearly showed lath martensite and twin structures. The excessively fast cooling rate would also increase the dislocation density and provide a heterogeneous nucleation location for martensite formation [22]. In short, through the control of the heat treatment process (temperature, holding time, cooling rate, etc.), the comprehensive performance can be improved by adjusting the microstructure.

Figure 10

The microstructure of Cu–Al alloy at different annealing temperatures increases from left to right: (a1–7) Cu–5Al alloys, (b1–7) Cu–8Al alloys, and (c1–7) Cu–11Al alloys [142].

4.2.2 The effect of heat treatment on mechanical properties

Mechanical properties are the basic properties of materials, which ultimately determine the bearing capacity, service life, and safety of structures. Due to the large anisotropy and grain size of Cu–Al alloy, brittle fracture, and low fatigue strength will reduce its mechanical properties and service life [123]. Therefore, how to improve the mechanical properties of Cu-based alloys has become the focus of many researchers.

Ren et al. [142] conducted heat treatment of Cu–Al alloy at different temperatures, and the results showed that with the increase of temperature, the grain was refined and the hardness and strength were improved, as shown in Figure 11. Qader et al. [123] studied the change in mechanical properties of Cu–13Al–3Ni-4Hf during aging and found that β-phase (Cu₃Al) would be generated after aging above 973 K, which would increase the brittleness of the alloy. In other studies, it was also confirmed that the increase of the α phase would lead to the increase of thermal stability and the decrease of the hardness of the alloy [146]. Saud [146] also showed that aging treatment improved the ductility of Cu–Al–Ni alloy. In addition, mechanical failure of Cu-based alloys was also related to brittle intergranular fracture caused by high shear stress concentration at grain boundaries [24]. High shear stress concentration occurs due to high elastic anisotropy and incompatible deformation of adjacent grains, as shown in Figure 12(a) [147]. At an appropriate cooling rate, the Cu-based alloy consists of two layers of columnar grains. A single layer of columnar grains will form at a cooling rate that is high enough, as shown in Figure 12(b). Monolayer structures reduce the number of potential crack initiation sites and shear stress concentration caused by elastic anisotropy. Therefore, in order to obtain excellent mechanical properties, the selection of appropriate heat treatment process parameters can achieve the purpose.

Figure 11

Microhardness and tensile strength curves of the Cu–Al alloy annealed at 400, 450, and 500°C: (a–c) microhardness and (d–f) tensile strength [142].

Figure 12

(a) Multilayer equiaxed structure and (b) single-layer columnar fibre textured structure [147].

4.2.3 The effect of heat treatment on martensite transformation

The MT mechanism of Cu-based alloys is usually related to the change in structure/properties during heat treatment [22,81]. Both the solution treatment time and the holding temperature can affect the phase transition temperature. The longer the solution time or higher the holding temperature will increase the phase transition temperature. It is mainly affected by the size of the crystal grains; the larger the crystal grains, the higher the phase transition temperature. However, defects such as vacancy and dislocation will be formed during the process of solid solution and quenching. The pinning effect of vacancies and dislocations on the interface could also increase the reverse MT temperature of the alloy [145]. In the long-range ordered parent phase, the diffusion of atoms through vacancy migration pointed to energy-favorable routes [148]. However, this was not the case in the martensitic structure. The migration of vacancies near the martensite interface did not occur sufficiently, resulting in the retention of disordered atomic pairs or locally disordered structures in the martensite. Thus, excessive vacancies were gathered around dislocation and martensite interface, resulting in martensite stability [144]. The increase of martensite stability is baleful to the phase transformation of the alloy. The kinetics of the phase transformation is reflected in the Kissinger and Flynn–Wall–Ozawa models [149].

The Kissinger model is given by equation (10) [150]:

where β is the heating/cooling rate; T p is the temperature of maximum intensity of the reaction; E a is the activation energy; and R is a constant.

The Ozawa model is given by equation (11) [151]:

Meanwhile, the relaxation energy during MT can be expressed as equations (12–14) [145]:

where ∆S is the change in entropy during MT; ∆H is the enthalpy change during reverse MT; and T 0 is the thermodynamic equilibrium temperature where the Gibbs free energy is equal to zero.

Undesired structural defects in the alloy (such as quenching stress, quenching vacancy, etc.) can be eliminated by aging, which is usually the process that determines the final physical properties [20]. Therefore, the relationship between heat treatment and MT has been extensively studied in recent years [130]. After Cu–13Al–4Ni–4Fe is heat-treated at 950°C, the increase of the thickness and grain size of martensite lath relaxes the microstrain, leading to an increase in the phase transition temperature of the alloy [152]. Wang et al. [137] studied the effect of aging on porous CuAlMn alloy and found that after aging at 350°C, due to the refinement of the martensitic structure and the elimination of quenching vacancies, the damping capacity of martensitic alloy and the height of the internal friction peak caused by reverse MT increased. In addition, the cooling medium also plays an important role in the process of solid solution and aging because different cooling media determine different cooling rates [22]. Generally, the cooling rate can be determined by the ratio of ∆T and ∆t, where ∆T is the change in the quenching temperature and ∆t is the time cost during the cooling process. Yildiz [144] studied the effect of different cooling rates on the phase transition temperature of the alloy. Water cooling was a one-step transformation in both MT and reverse MT processes, while air cooling and furnace cooling were characterized by a two-step transition. Studies have shown that the reduction of the cooling rate will move the phase transition temperature to high temperatures, and the effect of the heat treatment temperature on the change of the phase transition temperature and enthalpy is different, as shown in Figure 13 [123]. Therefore, the temperature and cooling rate during heat treatment have important effects on the phase transformation behavior of the alloy.

Figure 13

(a) Phase transition temperature of Cu–13Al–3Ni–4Hf alloy at different temperatures and (b) temperature hysteresis for each case [123].

4.2.4 The effect of heat treatment on damping performances

The damping capacity of Cu-based alloys comes from the inelastic strain of the phase interface and martensite twin boundary [140]. These interfaces can move continuously under alternating external stresses, thereby releasing stress and consuming mechanical energy. There were two peaks in the damping of Cu-based alloy, the high-temperature phase was due to anti-martensitic phase transformation, and the low-temperature phase was due to pseudo-first-order phase transformation during twinning thinning [145]. The density and distribution of inherent structural defects (vacancies, dislocations, grain boundaries, and precipitates) of the martensite phase-controlled the overall damping peaks of these alloys [140].

The damping peaks of low-temperature twins play an important role in the damping mechanism of Cu-based alloys. According to Zhang and Liu [112], when the twins were relatively rough or the space between the twin walls was relatively large, the twins were very likely to move due to the relatively weak interaction between the twin walls. However, if the space was small enough, the elastic strains caused by the twin walls would overlap, making the movement of adjacent twin walls difficult, resulting in reduced damping. The high-temperature damping peak is very sensitive to temperature [114]. Thermoelastic martensitic stability could be obtained by tempering the quenching alloy below the As for a long time. This led to a decrease in the thermoelastic behavior of martensite, which greatly reduced the damping performances of the material [151,153]. For quenched samples, the quenching vacancies were highly concentrated and randomly distributed [136]. When aging starts in the martensite phase, the vacancy will gradually accumulate toward the dislocation or interface, impeding the migration of dislocation and interface, and reducing the internal friction related to dislocation and interface movement [152]. In addition, the reverse MT will also be hindered, leading to an increase in the MT temperature and a decrease in the peak strength of phase change damping caused by the reverse MT [149].

In consequence, researchers have done a lot of work on the relationship between temperature and damping. Studies showed that as the ambient temperature increased, the damping increased at first. This was mainly because though the amount of martensite was reduced, more phase interfaces and twin boundaries could move at a higher temperature, resulting in higher energy consumption. When the ambient temperature was higher than M _s, with the further increase of the ambient temperature, the damping performances decreased sharply, mainly because there was little martensite remaining in the alloy [154]. Wang et al. [138] studied the effect of aging treatment on Cu–11.7Al–2.49Mn alloy and found that with the increase of aging temperature, the strength of the internal friction peak began to increase, and then the reverse martensite transformation decreased, as shown in Figure 14. In addition, the shift of the peak position from low temperature to high temperature was attributed to the thinning of martensite lath and the disappearance of quenching vacancy. Li et al. [136] studied the effect of aging treatment at 250–400°C on the damping properties of columnar Cu–Al–Mn shape memory alloys and their mechanism. The results showed that the maximum damping peak is 0.11. With the increase of aging temperature and time, the precipitates of bainite increased but the damping capacity of columnar Cu–Al–Mn shape memory alloy decreased at first, then increased, and finally decreased again. The more complete the MT is, the more energy is consumed, which increased the inherent damping of the alloy [20]. In addition, intermetallic precipitated particles formed in the aging process have a pinning effect on martensite variants, which will reduce the damping capacity [136,152]. Therefore, in the choice of heat treatment, only the appropriate heat treatment process can achieve the ideal effect.

Figure 14

Cu–Al–Mn alloy treated at different temperatures: (a) XRD patterns and (b) the damping properties [138].

4.2.5 Summary of heat treatment mechanisms

The unique martensitic structure of Cu–Al-based alloys determines that the heat treatment process plays an important role in its mechanical and damping properties. The solid solution treatment is mainly to obtain a supersaturated solid solution. The aging treatment is to eliminate the defects such as vacancies formed in the solid solution treatment. In the process of heat treatment, the grain size can be adjusted through reasonable control of the process parameters, and then the mechanical properties of the alloy can be improved through the fine grain strengthening effect. Structural defects in martensite have a significant impact on damping properties. The ultimate purpose of both solid solution and aging treatments is to improve the damping performance by modulating the energy-consuming movable interfaces and defects. The two-sided effect of grain refinement on damping properties was also mentioned earlier so a proper choice of the heat treatment process is also needed to achieve the ultimate improvement.