Introduction

Wire arc directed energy deposition (WADED) has found extensive application in advanced contexts, characterized by large-scale (meter-level) fabrication and intricate metallic structures. This utilization is attributed to its notable characteristics, which encompass a high deposition rate, efficient material utilization, cost-effectiveness in both production and equipment and a high degree of equipment versatility and scalability1,2,3,4,5. WADED components can be directly manufactured using computer-aided design models and metal wire. This process involves the localized injection of metal wire into a molten pool formed by a focused arc through a feed nozzle, adding successive layers to incrementally construct the three-dimensional component6.

Within the WADED process, metallic elements are incrementally deposited to a substantial metallic substrate. This substrate not only serves to efficiently dissipate heat generated during the deposition but also provides essential support for the workpiece throughout the fabrication process. As a consequence of the layer additive characteristic in the WADED fabrication process7, diverse regions of the deposited materials encounter a complex thermal profile. The thermal histories within the WADED procedure encompass an initial rapid quenching from the liquid state, succeeded by an inherent heat treatment (IHT) involving cyclic reheating characterized by numerous brief temperature spikes approaching values proximate to the melting point8. The intricate thermal dynamics in the WADED process give rise to a multifaceted evolution of the microstructure. In particular, the employment of a tightly focused arc for the formation of a swiftly transiting molten pool can yield relatively elevated solidification velocities and cooling rates.

In the past decade, the utilization of aluminum alloys has experienced a significant upsurge in various automotive and portable device applications, primarily due to their favorable combination of low density and high strength9. Among heat-treatable and high-strength Al alloys, Al-Cu alloys are notably attractive because of their superior creep strength at elevated temperatures10. Aluminum-copper alloys are a type of precipitation-hardened Al-based alloy, forming multiple metastable phases during the phase decomposition process11. The phase decomposition process of aluminum-copper alloys is commonly concluded as follows: supersaturated solid solution - G.P.(I) - θ’’ phase - metastable θ’ phase - stable θ phase12. Besides, the θ’ (Al2Cu) phase, as an important intermetallic compound, significantly influences subsequent mechanical properties based on its growth, size, morphology, and particle size distribution during thermodynamic processes13.

The aging temperature of Al-Cu alloy is relatively low (165 °C), leading to a certain degree of IHT frequently occurring during the repetitive melt-cooling deposition process. IHT is notably observed in WADED. Philipp14 et al. have documented that IHT has the capacity to induce the precipitation of nickel aluminide (NiAl) in Fe-Ni-Al alloy. Zheng15 et al. conducted a simulation of the thermal dynamics involved in laser deposition using laser-engineered net shaping. Their findings revealed that the deposited material undergoes a pronounced rapid quenching phenomenon in the early stages of deposition, with an exceptionally high cooling rate. Krakhmalev16 et al. proposed that the temperatures reached through the process of Selective Laser Melting (SLM) were sufficiently able to initiate diffusional transformation in AISI 420 steel.

Until now, there are still few studies on the effect of intrinsic heat treatment on the WADED process of aluminum alloys. Therefore, this paper fabricated thin-walled components of Al-Cu alloy using FS205 (Al-Cu) as the filling wire and systematically investigated the microstructural evolution, phase transformations, and mechanical properties during the WADED process. And heat treatment processes were applied to eliminate the heterogeneous effects caused by IHT.

This paper is organized as follows: Sect. 2 introduces the experimental materials, equipment and methods used. Section 3 shows the experimental results including temperature evolution, phase composition, microstructure and mechanical properties. The effect of IHT on microstructure and mechanical properties is discussed in Sect. 4. The conclusion of this paper is shown in Sect. 5.

Materials and methods

The WADED system in the present research mainly consisted of a six-axis FANUC robot, a Fronius cold metal transfer (CMT) power source, a torch, an external wire feeder and a working platform. The nominal compositions of the commercial ZL205A wire (1.2 mm in diameter) and the deposit are listed in Table 1. The 6061 substrate plate had dimensions of 200 mm×200 mm×20 mm and was washed in alkaline water and dried in air. This was then followed by mechanical cleaning and degreasing using acetone immediately before deposition. As shown in Fig. 1b, the wall with dimensions of 100 mm×155 mm×20 mm was deposited layer by layer along the centerline of the substrates. The use of wall components primarily serves to simplify the deposition process, facilitating a more focused investigation into the effects of intrinsic heat treatment on the material’s microstructure. The CMT-ADV mode was applied for building the wall. In the deposition process, mean current (80 A), voltage (11.5 V), torch travel speed (0.5 m/min), wire feed speed (4.3 m/min), inter-layer cooling time (20 s), shielding gas flow rate (25 L/min) of pure argon (99.99%) and contact tip to workpiece distance (15 mm) were all kept constant for the wall sample.

A 45-layer Al-Cu linear wall structure was deposited to study the microstructural evolution and mechanical properties of the as-deposited WADED component and its response to a standard aging heat treatment. The heat treatment applied to the as-deposited wall structure consisted of two steps, which were solution treatment and artificial aging. The samples were ramped in the furnace at 535℃ for 2 h and subsequently quenched into water. Artificial aging treatment was then carried out at 165 ℃ for 16 h. Wall samples were then cooled down to room temperature in the furnace.

The wall sample’s ends (10 mm) were cut off and discarded. The sampling positions of the wall were sectioned as shown in Fig. 1a. Samples for the microstructure analysis and hardness test were taken from the middle part of the wall. The tensile test samples as shown in Fig. 1c were machined to a final surface roughness less than 1.6 μm. Tensile tests were carried out at ambient temperature by an electro-mechanical universal testing machine following the BS EN ISO6892-1:2009 standard. The specimen was loaded with an initial strain rate of 0.001 s− 1. Tensile tests were conducted three times at each ___location (SH1, SH3, etc.), and the average value was taken. Vickers microhardness testing machine (Zwick Roell Z100) was employed with a 10 kg load for 10s for measuring the hardness. The hardness test started 7 mm from the bottom surface of each wall. 40 tests were conducted along the vertical direction, with a 2.5 mm interval between each position. Three tests were performed at each position, and the average value was taken.

Samples were transverse cross-sectioned from the as-deposited wall for microstructure investigation. Scanning electron microscopy (SEM) (NOVA NanoSEM 230) was employed to take images of the microstructure. The SEM specimens were ground with 180, 400, 800, 1200 and 2500 grit SiC papers under flowing water, then were polished with 5 μm,1.5 μm and 0.5 μm diamond paste to a mirror finish. Energy dispersive spectrometry (EDS) was used for micro-area composition analysis. Electron back scattered diffraction (EBSD) with a step length of 1.718 μm was carried out for grain size analysis by using the commercial software HKL CHANNEL 5. Fracture surface morphology of the tensile specimens was performed using SEM. The phase constituents near the substrate, at the middle buildup section and top section, were identified using a Rigaku D/max 2500 X-ray diffractometer (XRD) with CuKα radiation (λ¼1.5418Å) and a scanning speed of 1°/min. The scanning angle (2θ) was set from 20° to 90° and the step size was set at 0.02°. Jade 6 software was used to analyze the XRD patterns. The temperature value of the substrate was measured by a thermometer (LINI-T UT321) after each layer was deposited. Transmission electron microscopy (TEM) (JEM-2100) was used for further characterization of the microstructure. The TEM samples were initially ground on both surfaces using 400, 800, 1200 and 2500 grit SiC papers. Once the sample thickness reached 70–90 μm, discs with a diameter of approximately 3 mm were cut. The samples were then pre-thinned to perforation using a twin-jet electropolishing device, with an electrolyte solution composed of a 3:7 volume ratio of nitric acid to methanol. During the electropolishing process, liquid nitrogen maintained the electrolyte temperature at -30 °C. Subsequently, ion thinning was performed on the perforated samples using an ion thinning apparatus for 20 min, at an angle of 3° and a voltage of 3 kV.

Fig. 1
figure 1

(a) The sampling positions of the wall; (b) the view of the thin-wall sample; (c) the size of the tensile samples (the figure (a) and (c) were created using PowerPoint software. Version number: LTSC 2021; URL link: https://software.sjtu.edu.cn/En/List/Office/2021).

Full size image
Table 1 Chemical composition of the ZL205A wire and the WADED Al-Cu wall (wt %).
Full size table

Results

Temperature evolution

The WADED process is essentially an additive multi-pass welding process17. The layer-by-layer additive manufacture is different from the traditional manufacturing process in a unique manner characterized by local melting and directional thermal conduction. In the WADED process, the heat accumulated during deposition is primarily dissipated through thermal radiation and heat transfer to the substrate. The substrate temperature can intuitively reflect the intrinsic heat treatment (IHT) effect on the solidified state at the bottom, resulting from the remelting-solidification process at the top.

Figure 2 shows the temperature evolution of the substrate in the WADED process. The initial temperature of the substrate is 31.2℃. It can be seen the temperature of the substrate ranged from 126℃ to 193℃ in the layer-by-layer manufacture of the linear wall. The average temperature value is 157.7℃. With the increase of deposited layers, the temperature of the substrate firstly reaches maximum temperature (193℃) at the fifth layer. Then the temperature value of the substrate starts to fluctuate between the sixth layer and the thirtieth layer. From the thirty-first layer, the temperature value gradually decreases from 162.2℃ to 136.2℃. It can be seen that the thermal cycling history experienced by SH1-SH7 is different, which will lead to variations in their microstructures. This is because the temperature level (averaging 157.7 °C) can induce solid-state phase transformation in the Al-Cu alloy, affecting the microstructure of the solidified layer at the bottom.

Fig. 2
figure 2

The temperature of substrate.

Full size image

Phase characterization

Figure 3a presents the XRD results of the as-deposited WADED Al-Cu alloy. It can be seen that the main phase is α-Al. Besides, Al2Cu also can be confirmed to exist in the four as-deposited components. The top part in the as-deposited condition does not undergo successive thermal cycles, thus only tiny Al2Cu peaks are observed in the diffractogram. In contrast, more Al2Cu peaks are found in the middle and bottom parts, indicating a larger amount of Al2Cu forming. Table 2 summarizes the corresponding Al2Cu content. From SH1 to SH5, the volume fraction of Al2Cu gradually increases from 7.21 to 8.17% with the increase in the sample height. However, the top part decreases to 6.66%. After heat treatment, α-Al still existed in the components, while Al2Cu could not be detected (Fig. 3b).

Table 2 Volumetric fraction of the phases in the as-deposited WADED Al-Cu alloy (%).
Full size table
Fig. 3
figure 3

XRD results of samples extracted from the bottom near the substrate section, middle buildup section and the top near-surface section of the buildup wall: (a) as-deposited condition; (b) heat treatment condition.

Full size image

Microstructure

The microstructure of the WADED components in the as-deposited condition is shown in Fig. 4. In the bottom part, network-like second-phase particles with light color scatter along as-deposited grain boundaries and dendritic arms, as shown in Fig. 4a. A small number of spherical particles are embedded in intra-grain regions. As shown in Fig. 4b and c, the white network-like eutectic particles scattered along the grain boundaries and similar circular particles distributed in the intra-grain regions in the middle part; whereas the top part (Fig. 4d) contains less network-like eutectic particles and more similar circular particles.

In the WADED process, the as-deposited material undergoes repeated thermal cycles from subsequent depositions and IHT could occur. Table 3 lists the EDS analysis of the secondary phases in Fig. 4. There are four kinds of secondary intermetallic phases in the experimental alloy in Fig. 4, including an irregular grey phase (marked A), a needle-shaped or short rod-like phase (marked B), a circular grey particle (marked C), and a block-like white phase (marked D). The detected EDS data indicated the irregular grey phase (marked A) dominated in grains is θ phase (Al2Cu). The θ-Al2Cu particles, in accompaniment with α-Al, connect with each other to establish a network in three-dimensional space (called as eutectic network). It can be seen that a very high area fraction of coarsening θ-Al2Cu/α-Al eutectic network is found with different positions (Fig. 4a–d). Meanwhile, some large cubic particles, needle-shaped particles and small needle-shaped precipitates (marked B) are observed inside the α-Al matrix grains (Fig. 4a–c). Zhang18,19 et al. have reported that these needle-shaped particles are θ″ or θʹ phase while the large cubic particles are θ-Al2Cu phase20. Spectrum C and D on the circular phase contain Al, Cu, Fe, Mn, Si and Cd elements. According to the ICP results of the WADED Al-Cu components (4.64–5.74% Cu, 0.36–0.44% Mn), these circular phases contain Al2Cu phase and T(Al20Cu2Mn3). Moreover, according to ref21. , the iron-containing phase is Al6(Mn, Fe). These Cu-rich particles easily coarsen and are non-uniformly distributed due to the decreased cooling rate during the repeated thermal cycles of the WADED process. Besides, this segregation generally causes Cu solute atom depletion zone around the phases, which reduces the mechanical properties22. Further investigation will be undertaken to investigate the influence of those particles on the mechanical properties of this alloy.

Fig. 4
figure 4

Microstructure of the as-deposited WADED Al-Cu alloys: (a) SH1; (b) SH3; (c) SH5; (d) SH7.

Full size image
Table 3 Chemical compositions of secondary phases in Fig. 4 (wt%).
Full size table

Figure 5 illustrates the microstructure of the WADED Al-Cu alloy sample after heat treatment. It can be observed that, following the heat treatment, the majority of the coarse eutectic Al2Cu phases at the grain boundaries have dissolved and solid-solution into the matrix. The remaining eutectic Al2Cu phase has transformed into a particulate form. In Al-Cu alloys, apart from Ostwald ripening, the dissolution of eutectic phases formed in the dendritic arms and the oversaturation of hydrogen in the matrix are the primary factors contributing to the significant increase in pore defects in the samples post-heat treatment23. The dissolution of the eutectic phase during heat treatment generates excess vacancies and voids. At elevated temperatures, when hydrogen atoms are released from dislocations or interfaces, microvoids will heterogeneously nucleate on the surface of the particulate Al2Cu phase and grow within the voids. Additionally, the number of precipitates at the bottom section is significantly higher than that observed in the middle and top sections.

The TEM results of the WADED Al-5Cu alloy in both the as-deposited and heat-treated states along the [100] axis are presented in Fig. 6. As shown in Fig. 6, two primary types of precipitate phases are identified: the θ′ phase and the T phase (Al20Cu2Mn3). The elongated, perpendicular precipitates correspond to the θ′ phase. This phase serves as the primary strengthening phase in the heat-treated samples, as it impedes the motion of dislocations and subgrain boundaries. As a result, the volume fraction of the θ′ phase in the heat-treated alloy is considerably higher than that in the as-deposited state. Additionally, the T phase (Al20Cu2Mn3) precipitates at the tips of the θ′ phase and along the grain boundaries. The T phase typically forms during solidification and is primarily located at the grain boundaries24. It acts as a pinning agent, restricting both grain boundary migration and dislocation motion during deformation, thereby contributing to the overall strengthening of the alloy. Moreover, as illustrated in Fig. 6d, the matrix contains substructures that have gradually developed through the accumulation of dislocations. These subgrain boundaries, which are generated by dislocation pile-up, can further hinder the movement of dislocations.

Fig. 5
figure 5

Microstructure of the heat-treated WADED Al-Cu alloys: (a) SH1; (b) SH3; (c) SH5; (d) SH7.

Full size image
Fig. 6
figure 6

TEM results of as-deposited (a), (b) and heat-treated (c), (d) WADED Al-Cu alloys in a [100]Al zone axis.

Full size image

Coloring inverse pole figures (IPF) from cross sections (Y-Z plane) of the as-deposited samples are depicted in Fig. 7. For the WADED process, heat accumulation and thermal cycling induce changes in the microstructure as the different layers are deposited. Fig. 7a–d illustrates that all the grain morphologies are equiaxed and have no evident orientation distribution. The blue region in Fig. 7e–h is the distribution of the grains with a size smaller than 20 μm, and the area ratios are 24.4%, 8.6%, 8.7% and 3.5%, respectively. The grains smaller than 20 μm in the samples SH1 are mostly concentrated at the interlayers, which produces finer grain zones. The grain size is evenly distributed in the middle and top regions, and the concentration phenomenon disappears.

The histograms of grain size distribution for different WADED specimens are shown in Fig. 8. If grain or sub-grain sizes are to be determined from an EBSD map then consideration must be given to the pixel step size about the grain size. According to Humphreys criterion25, 5 pixels (1 in 4) per grain are required to obtain an accuracy of 10%. This means there should be at least 3 pixels in a row within a grain or sub-grain region. Therefore, only grains larger than 5 μm are counted for this research due to the limitation of EBSD scanning resolution. It can be seen that the average grain size increases from 26.1 μm to 40.1 μm as the distance from the wall bottom surface increases from 10 mm to 130 mm, with an increase of 53.6%. This may be caused by residual heat from the previous layer. As the previous layer cools down, the dissipation of the heat will effectively provide a preheat for the second layer, decreasing the temperature gradient and creating a cyclic thermal field. The solidification environment promotes the coarsening of some grains in the layer corresponding to samples in the middle and top regions. Furthermore, the average grain size of the samples SH3 and SH5 are 37.9 μm and 38.2 μm, respectively. As shown in Fig. 9, after the heat treatment, the average grain size increases to between 37.2 μm and 44.3 μm and the variation from the bottom to the top layers becomes less pronounced.

Fig. 7
figure 7

EBSD results: (a)–(d) inverse pole figure (IPF) map; (e)–(h) grain map with grain size less than 20 μm; (a) (e) SH1; (b) (f) SH3; (c) (g) SH5; (d) (h) SH7 (the figures were created using HKL CHANNEL 5 software. Version number: 2019v5.12; URL link: https://nano.oxinst.cn/products/ebsd/post-processing-software).

Full size image
Fig. 8
figure 8

Grain size distribution for the as-deposited WADED Al-Cu alloy: (a) SH1; (b) SH3; (c) SH5; (d) SH7.

Full size image
Fig. 9
figure 9

EBSD results of the heat-treated WADED Al-Cu alloy: (a)–(d) inverse pole figure (IPF) map; (e)–(h) grain size distribution; (a) (e) SH1; (b) (f) SH3; (c) (g) SH5; (d) (h) SH7 (the figures (a)–(d) were created using HKL CHANNEL 5 software. Version number: 2019v5.12; URL link: https://nano.oxinst.cn/products/ebsd/post-processing-software. The figures (e)–(h) were created using OriginPro software. Version number: 2019v5.12; URL link: https://www.originlab.com/OriginProLearning.aspx).

Full size image

Mechanical properties

Figure 10a presents the microhardness variation from the bottom to the top part of the as-deposited Al-Cu alloy. The WADED structure is not uniform in microhardness from the bottom to the top: the material is harder in the bottom part (ranging from 58.5 to 64 HV) but gets softened slightly near and above the bottom part to around 56.9 HV. The difference could be attributed to the different cooling rates between the upper section and the lower section. Combined with the XRD results, it can thus be inferred that Al-Cu alloy produced by WADED is non-uniformly age hardened to different extents from the bottom to the top along the building direction with the bottom aged more. As shown in Fig. 10b, after heat treatment, the hardness increases to between 134.2 HV and 155.5 HV and the average value is 145.3 HV. The fluctuation of the microhardness value is little, indicating stable microhardness distributing characteristics after heat treatment. The microhardness of WADED Al-Cu alloy can be remarkably improved by solution treatment and the artificial aging process.

The ultimate tensile strength (UTS), yield strength (YS) and elongation (EL) results of WADED-CMT Al-Cu alloys along the building direction with increased distance from the wall bottom surface are shown in Fig. 10c and d. In the as-deposited condition, the UTS, YS and elongation generally reduce from the bottom to the top: from 213 MPa to 191.3 MPa, from 151.7 MPa to 147.7 MPa and from 11.1 to 9.1%, respectively, which is in line with the microhardness measurements and the grain size distribution. After heat treatment, the UTS and YS values at each corresponding ___location increase noticeably. The UTS increased by 128.2% and the YS increased by 175.4%. The variations in UTS and YS values from the bottom to the top are virtually eliminated. However, there is a significant fluctuation in elongation from the bottom to the top. Compared to the as-deposited state, the elongation of the heat-treated samples has decreased.

Fig. 10
figure 10

Mechanical properties: (a,b) the microhardness distribution of the WADED Al-Cu alloy; (c,d) tensile properties for the as-deposited WADED Al-Cu alloys; (a,c) the as-deposited samples, (b,d) the heat-treated samples.

Full size image

Fracture surface

Figure 11 displays the SEM images of the tensile test fractography morphology in the horizontal direction for the WADED Al-Cu alloy. All the fractured surfaces consist of dimples, which are regarded as an indication of ductile fracture. The dimple sizes are different from the specimens with various distances from the bottom surface.

As shown in Fig. 11a, small and deep dimples and cracked second-phase particles can be observed on the fractured surface of the as-deposited Al-Cu alloy. The second phase particles are found embedded at the center of each dimple, which can initiate primary cracks of fracture21. Generally, the Cu solute atom depletion zone (SDZ) in the as-deposited alloy is formed around grain boundaries due to the concentrated scattered θ phase particles along the grain boundaries26. These SDZ areas are weaker in strength compared with the other areas. Under static tensile stress, the tearing edges surrounding the particles are generated at these SDZ areas, showing a typical trans-granular fracture mode. Compared with the bottom part of the as-deposited specimen, the dimples of the lower middle part (Fig. 11b) become coarse and the number of the dimples decreases with the increased distance from the bottom surface. The solidification cracks on this surface are responsible for the rupture along the horizontal direction, exhibiting a brittle fracture mode. These cracks may be developed by the inter-grain defects or from the fractured brittle phase particles.

Most of the fracture surface of the middle-upper part (Fig. 11c) and top part (Fig. 11d) is taken by coarse and shallow dimples and micro-voids. These features are typical characteristics of ductile fracture mode. In addition, smooth cleavage planes, cleavage steps and tearing ridges can also be observed, suggesting that cleavage fracture also exists in the specimens of the middle-upper part and top part. Therefore, the fracture transforms from a ductile fracture to a mixture of ductile and cleavage fractures in the middle and top parts.

Figure 12 shows the SEM images of the tensile fracture surfaces of heat-treated WAAM-printed Al-Cu alloy samples. From Fig. 12a and c, it can be observed that the fracture surface of the bottom sample contains numerous pores and a few dimples, with second-phase particles present at the bottom of the dimples. Some of these second-phase particles are fractured, and cracks appear in the matrix. After heat treatment of the top sample, as shown in Fig. 12b and d, the tensile fracture surface exhibits a large number of pores, with very few dimples. A significant amount of second-phase particles can be seen along the pore walls and boundaries, and there is evidence of pore wall fragmentation as well as second-phase particle fracture.

Fig. 11
figure 11

SEM fractography morphology for tensile specimens of the as-deposited WADED Al-Cu alloys: (a) SH1; (b) SH3; (c) SH5; (d) SH7.

Full size image
Fig. 12
figure 12

SEM fractography morphology for tensile specimens of the heat-treated WADED Al-Cu alloys: (a,c) SH1; (b,d) SH7.

Full size image

Discussion

WADED process and IHT

The layer-by-layer building manner in the WADED process results in heat and mass transfer from the arc, molten wire re-melted a certain amount of the previous layer and the liquid metal solidifies upon it to form a new layer. As shown in Fig. 2, the average temperature value of the substrate is 157.7℃ during the WADED process. In the layer-by-layer manufacture of the linear wall, the heat mostly flows from the top layer (where the melt pool is located) to the bottom. Hence, the temperature of the linear wall during the WADED process is higher than the substrate. Since the aging temperature of Al-Cu alloy (165℃) is low, materials are nearly reheated up to such an aging temperature during the whole WADED process. Due to the rapid aging response of Al-Cu alloy, the accumulation of such durations will result in an IHT, which results in all the differences in microstructure evolution and mechanical properties with increased distance from the wall bottom surface.

Effect of IHT on microstructure in the WADED process

According to the solidification principle, the actual liquid temperature is below the liquidus temperature at the solid/liquid interface, which will generate supercooling and a mushy zone. At a high degree of supercooling, the mushy zone becomes so wide that it is easier for the equiaxed dendrites to nucleate than for the columnar dendrites to stretch. A fine grain zone forms finally. The WADED interlayer is located in the molten pool boundary, and the solidification process is similar to the above process. During each deposition, the upper part of the former layer is re-melted and leaves the equiaxed grains in the former layer partially melted. When solidification occurs, the grain growth direction would be perpendicular to the solid/liquid interface in order to follow the maximum temperature gradient. During the deposition process, the bottom part has the highest degree of supercooling so there are evident interlayer fine-grain zones. In the case of the successive layer depositions where the IHT are enhanced and more heat is transferred to the solid of the former layer, which decreases the temperature gradients on the solid/liquid interface and the degree of supercooling, so the fine grain zone disappears and the average grain size gradually increases with the increased distance from the wall bottom surface. Furthermore, the IHT facilitates the diffusion, solute atoms diffuse out from the unit cell and form precipitates. Different parts undergo a different number of thermal cycles in the WADED process and will have different amounts of intermetallic compounds precipitated: the more thermal cycles, the more precipitates.

Combined with the XRD results, the SEM and EDS of the deposits with different distances from the bottom surface show that the precipitated phases are mainly the massive Al2Cu phases. In Al-Cu alloy, the number of eutectic structures combined with Scheil’s law can be calculated from Eq. (1)27:

$${f_s}=({C_e}{C_1}^{{ - 1}})/(1 - {k_0})$$
(1)

Among them, Ce represents the content of Cu in the eutectic structure, and its value is 0.54 for Al2Cu; C1 is the mass fraction of Cu in the alloy, its value is 0.05, and k0 is the equilibrium distribution coefficient of the Cu element, taking the value of 0.1728.

From Eq. (1), it can be seen that the content of the eutectic phase increases with the mass fraction of Cu. According to the results in Table 1, during the WAAM process, the mass fraction of Cu increases with the number of deposited layers, leading to a gradual increase in the volume fraction of the eutectic Al2Cu phase, which is consistent with the XRD results. For SH7, due to its position at the top of the sample, the cooling rate is rapid, and there is no IHT from subsequent deposition layers, which does not meet the premise for the validity of this formula. Therefore, the mass fraction of Cu and the volume fraction of the eutectic Al2Cu phase in SH7 do not follow this rule. In addition, it can be seen from Fig. 4 that IHT has a greater effect on the morphology and distribution of the Al2Cu phases. As the number of thermal cycles increases, the precipitated Al2Cu phases in the grain gradually increase, and their number on the grain boundaries increases. After heat treatment, the coarse eutectic phases dissolve, and solute atoms enter into the solid solution. The influence of thermal cycling is almost eliminated.

Effect of IHT on mechanical properties in the WADED process

The microstructures such as grain size, precipitation strengthening phase, and high-density dislocations can hinder dislocation motion and crystal face slip, which is a decisive factor in improving the mechanical properties of metals. The comprehensive analysis of the microstructures shows that the most obvious change of the microstructures with increased distance from the wall bottom surface is the reduction of intermetallic compounds precipitated and grain growth. The decreased mechanical strength is mainly attributed to these two factors. The IHT from the successive deposits heat treats the previously deposited materials during the WADED process, thus the deposits closer to the bottom underwent numerous cycles of a different magnitude to age more and get harder than the top part.

It is evident that grains greatly grow with continuous subsequent heat input due to the imbalance between the thermodynamic driving forces of the grains and the pinning forces impeding grain boundary migration in the subsequent deposition29. Grain size is one of the most important microstructural factors in controlling the mechanical properties of aluminum alloys30. In this research, grains are defined as fine grains (FG) and coarse grains (CG) according to two diameter scales of < 20 μm and > 20 μm, respectively. As shown in Fig. 7, the proportion of FG reaches 25.3% in the top part from 46.1% in the bottom part. There is an obvious increase in CG from 53.9 to 74.7% after IHT. Along with increasing transient thermal cycles, FG scales in the lower middle part and middle-upper part continuously decrease to 22.7% and 22.3%, respectively. CG grains take up 77.3% and 77.7% respectively for the lower middle part and middle-upper part. It can be concluded from the above analysis that IHT in the WADED process is bad for grain refinement in Al-Cu alloy. IHT can significantly decrease the FG proportion and increase the CG proportion, which could decrease the strength properties.

This indicates that grain refinement may be the main strengthening mechanism for the WADED Al-Cu alloy because it is well-known that the mechanical properties of a metal are strongly influenced by the grain and sub-grain sizes. The Hall-Petch equation depicts an increase in yield stress (σs) with a decrease in grain size (d)31,32:

$${\sigma _s}={\sigma _0}+k{d^{ - 1/2}}$$
(2)

where σ0 and k are constants determined experimentally.

There are some other presumed strengthening mechanisms for the Al-Cu alloy due to alloying elements addition. Hu et al.33 pointed out that the addition of Cd in Al-Cu alloys can trap vacancies and extend their effect on the diffusion of Cu in the Al matrix. Moreover, Cd tends to segregate at the interfaces between θ′ (Al2Cu) and α (Al) phases, as well as reduce the interfacial energy and promote the nucleation of the θ′ phase34. Consequently, the resistance of dislocation movement is increased and leads to solution strengthening. Xu et al.35 reported that the Mn and Si elements are added to break up brittle and needle-like Fe-rich intermetallics or to transform them into blocks and this significantly improves the mechanical properties36. The dispersed second phases such as θ′ (Al2Cu) normally provide grain boundaries pinning forces in stabilizing the grain structures.

Conclusions

A wall of Al-Cu alloy has been successfully fabricated using wire arc directed energy deposition process and the specific characterization of the buildup material was divided into four regions along the vertical build direction to investigate the material properties as a function of ___location. In the layer-by-layer manufacture of the linear wall, the heat mostly flows from the top layer (where the melt pool is located) to the bottom, which results in an IHT that results in some extent of diffusion and precipitation to strengthen the material. During the WADED process, the transient thermal cycling from the successive deposits heat treats the previous materials, thus the deposits closer to the bottom underwent numerous thermal cycles of a different magnitude to be aged more.

The bottom part contains mainly fine and equiaxed α-Al grains with Al2Cu precipitates distributed both inside grains and along grain boundaries. The grain refinement and the existence of Al2Cu precipitates induced significant fine grain strengthening and solution strengthening respectively in the bottom region. However, the average grain size increases from the bottom part (26.1 μm) to the top part (40.1 μm) to the top part, which is caused by a variation in the specific cooling rate of the deposited layers. From the bottom part to the top part, the microhardness and strength properties are gradually decreased and the fracture transforms from a ductile fracture to a mixture of ductile and cleavage fracture because the average grain size increases and less intermetallic compounds are precipitated, as indicated by EBSD grain size measurements and XRD phase statistics results. Furthermore, the major strengthening mechanism of the as-deposited WADED Al-Cu alloy is fine grain strengthening and solution strengthening.

Solution and aging heat treatment have been applied to WADED Al-Cu alloy, which eliminates the microstructural heterogeneity from the bottom to the top and gives an increase of 128.2% in UTS and 175.4% in YS.